If you'd like to help make future projects like this possible please consider supporting them / us by becoming a channel member: ruclips.net/channel/UCHrFvnP1EEEZHNam_Nk_5rQjoin/join A big thanks to those who already support the projects!
I didn't see this below, but you may want to change the power connection from your idler bearing to one of the tension spools. Bearings do not provide a stable / reliable mechanism to commute the power/current.
I have been running a few wire edm's at work for a few years including a big Fanuc Robocut 600. Here is just a few surgestions . 1.To prevent wire snaps your are looking at a movement speed (speed of cut) of about 0.1-0.5mm per minute depending on what your are cutting and the thickness. 2. Speed up the speed of your wire (speed at wich the wire pass thru the cut , is being used) to 4-6meters per minutes. The faster the wire the cleaner and smoother the cut. 3. If your are using a large cap on your output power lines, back off on the size, this is scalpel not a broadsword like a sinker edm the initial large spark will snap the wire. 4. Current, from what I have seen you do not need more that 5amp on that wire, again scalpel not broadsword. 4.1 If I was cutting that round bar on the fanuc, typical setting would be tavel speed I would get 0.3mm per minute, wire speed of 8m/min, current 3amp. With a voltage of about 80v. keep in mind the wire cut moves forward and back depending on the gap voltage reading, you cannot just set it at a speed and plouge into the material. 5. Try swappingthe polarity of the output line, it sounds stupid but is does matter. I designed my own Small hole edm under the name Radnor Technologies a few years back and if I remember correctly I used - on the electrode and + on the work piece this results in a quick material removal rate but eats your electrode for breakfast. In your case with the wire that would be bad. 6. Remember when in doubt flush flush flush loose carbon can mess with your cut. Here is just a few other surgestions not related to the wire snaps. 1. Filter your water else things goe pear shaped very quickly. the 20 inch big blue spun polyprop filters used for home water systems works well. Fit a 10 micron filter the 5 micron does not last very long. The fanuc has a hugh 5micron filter as standard. Aluminium will clog a filter like you will not believe. 2.Deionize the water if your want to cut for longer than a few minutes. The waters optimal resistance should be 70 kilo ohm per cm2. Doin ght deionizing is easy, you buy the dionizing resin (small balls) from the same place that sells the filters, normally in a 25Kg bag. It is in a bucket with a inlet and outlet. Let the outlet drain into your tank and hook up a pump to the inlet to switch on when the resistanc is below 60K. This is a easy job for a Nano with a ADC and 2 probes on the water, we used 2 stainless rods with a surface area of 1cm2 3.The rewinder unless you want to try and reuse the wire... bad idea... ditch the roll winder just make a puller that pulls the wire and dumps it in a box, this is what the fanuc does. Some other brands chops the wire wich will be better since it takes less space. Hope this helps, all and all great project I wish you the best of luck and look forward to the next video... subscribed 🙂
do I understand this right, to make a 10mm long cut would take around 100 minutes and use around 500 meters of wire? is the wire re-usable? seems it's an extremely high cost. unsure of the cost of the wire, but.. seems a lot for a cut that is less than half an inch long. did I make a silly math mistake somewhere?
@@CrudelyMade your maths is a little off. But only a little. Based on those numbers I get 33 minutes and 266m of wire. Still seems high wire usage and very slow feed.
@@frankcarter8399 To measure the gap voltage you have to measure the line voltage, as soon as spark is made the voltage will drop to something like 3-5V depending on the system, since the rest of the power is being used (routed thru the cut). If the voltage is too high you can move a bit forward into the cut and if it is too low ..aka short circuit move back from the material. You need to monitor this with a ADC circuit and move the wire accordingly. You do us alot of wire but just for referance we paid about 70USD for a 8KG roll of wire and that contains about 19 kilometers of wire, so not to bad.
@@msbealo Jip my boss use to get annoyed when I run the wire at 10-14meters per min and set the speed down until I just coved the screen so he can't see the speed. It is better to waste a bit of wire than to have to restart a 8 hour job if the wire snapped in the middle of the work piece. Things normall goes fine in the beginning but a soon as you get to section that is hard to flush the wire like to snap when it is being run slowly. The wire cost works out at about 0.0036 USD per meter. Wire EDM is a very slow process, we had 3 machines a 30 year old Fanuc that works similiar to this home machine, slow and manual wire loading, that one used to cut at 0.02-0.05mm per min for 50mm thick stainless. We had the new Fanuc that ran at about 0.1-0.6mm per min on 50mm stainless. Lastly we had some Fong Kong machine that reused it's wire. It used a 0.19mm Moly wire that last very long but is very expensive and a complete "B" to load. Can take up to a hour to get up and running if the wire snaps.
It’s your spark generator. BAXEDM makes a modern one, but EDM has been around wwii and you can find some patents that basically use a big resistor for good current control. Most switching power supplies have poor “instant” current control, but A simple resistor should be smooth. You also need feedback between controller and current generator, water conductivity increases as you cut.
Thanks Ethan, that makes sense. Ill try your suggestion and stick a big resistor at the tip of my power supply. Ill get back to you on that. On the note of BAXEDM, I would love to collaborate with him. Hopefully he reaches out to me now that I have asked him. Cheers 😃
@@pappaflammyboi5799 I have reached out to him asking for a collaboration but he has not responded yet. I hope he does eventually 😊. Thanks for your input and hope you enjoyed the video. Cheers Dave
@@ActionBOX be a man, make power supply. Use 110v for supply, 2-20 ohm resistor, IGBT with 2-10 us on and 10-20 us off timing. Now you use very long pulses
It's pretty amazing how much more people can accomplish now that 3d printers and arduino boards are so popular/affordable. Kids these days dont realize how good they have it with all the technology at their fingertips... I wish all this stuff existed when I was a kid because I would be so many years ahead of where I am now.
That’s very true! 3D printing technology has definitely helped us make our ideas a reality. It’s never too late to start playing around with this technology 😃
@@ActionBOX That's what everyone tells me, but I find myself staying up till 3am trying to learn new things... There isn't enough time in the day for work and play. BTW, I subscribed. I appreciate when a channel takes the time to respond to comments like you did.
@@user-lx9jm1wo3hYes, good point. Juggling work and play is never easy. We appreciate you subscribing! We always respond to our viewers' comments. Looking forward to your comment on our next video (coming out Jan 21) 😉. Cheers, Alan
Great point.. If I would have had access to all this as a kid I would be working all day building new stuff. Kids these days are more interested in videogames and instagram than utilizing the technology they have at their finger tips.
In the late 1970s I learned to operate one of the first Wire EDM machines in the US. It was a Elox 100 with 2 axis Fanuc NC that was commanded by paper tape with punched holes. This was so high tech at the time that our tool & die shop became a stop for VIP visits to our city. Wire EDM machines have advanced a very long way from the Elox 100. Although many decades have permitted the concepts and components to become more normal I am blown away by the idea of an individual being able to DIY their own wire EDM machine. Very kewl and I'm impressed and envious of any of you that have the ability, resources and space to do this at home. Awesome.
The wire is snapping because of the lack of a real spark generator, bax edm makes one, but thats about it, spark generators are a serious technology and not really something you can buy off a shelf.
This brings me back a few years back when I was interning at a metalwork company where I was operating a wire EDM. It was boring as heck because the whole morning we would set up 1 or 2 plates and then we wait for them to cut finish in the next 3 days. I was setting up the parts too fast that the full time guys told me to slow down so that it will look I'm doing work. I find it really counter intuvtive. I wanted to learn more but they didn't let me because I'm only interning for 3 months. What great memory that was.
Im glad my machine could bring you back to those days 😃. Hope you enjoyed the video. Ps. feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
@@ActionBOX hi Dave. First video I saw of yours wire machines have a list of parameters for the machine know as E-PACKS for Example Ip,Vo,Sa,WS,Wt and many more. Epacks control voltage, amp, off times, wire speed, tension and Liquid quantity , stability controls and other finer setting. Second feed rates are important not only linear travel but wire speed will drastically reduce your consumption of your electrode (wire). Resin filters will help with your conductivity. last thing I saw that looked odd was the distance of the wire to your work piece. It could be angle of the video but the electrode should Not contact work piece. Wire machine will not start with a contact. Sorry for confusing comment hope some of the info helps
I can see were you got your inspiration from, looks very familiar :-) I can give you a ton of advice if you are interested. There is a long list of improvements to be made on your current design before you can make any wire EDM cuts, and a very long list before you can make high speed efficient and accurate cuts. Feel free to contact me.
I was just gonna comment asking him to watch baxedm to understand how the XY movement has to be controlled as a function of the spark current. But now he has a direct line to you :)
@@scooteracers the power supply would be the only limitation really. I’ll give you a hint, our V2 is meant to solve this problem. I like the idea though, I might start doing that myself. Cheers
@@scooteracers you want to cut metal stock at less than 1mm per minute for 50mm thick stainless? why would you want to do that? any saw will do at 10 times the speed i guess
EDM Engineer here. First of all, this is absolutely outstanding work. Secondly, we call it dielectric fluid, not coolant Thirdly, Wire EDM is much easier when feedback from the generator is used to control the feedrate. Fixed feedrates aren't very useful. I can't go into the feedback mechanism. You also would benefit from top and bottom flushing to ensure the spark gap is kept free of debris.
Thank you, this is an awesome reply to get from a professional. We will incorporate all of these suggestions into our V2. Feel free to help out and share the video with others in your field who may be interested, we are trying to grow the channel. Cheers, Dave
@@ActionBOX I've already shared it with my colleagues 😊 Can you drop the amperage at all? In general with wire erosion you're looking for a 300v ignition voltage, so you can drop your current. That spark looks way too big for 0.25mm brass wire. If you increase your wire feedrate it might resist the damage better. You can also try reversing the polarity - that can help with electrode wear. Using coated wire might help too, but it's quite a bit more expensive. You might also get some benefit by reducing the duty cycle. EDIT: Don't forget that when eroding a round workpiece, all of the energy at the start of the process goes into a single point. Try eroding something like a square blank - it'll be more consistent. Good luck!
Dude, this is literally the most amazing project I’ve seen in the past year. And your CNC mill?!?! That thing looks seriously impressive. Incredible engineering!
@@AndrewBoraas Do you mean it needs to change it operating parameters dynamically as the engaged wire length (area?) changes? Using a constant feed rate and the object's geometry to dictate the power is a fairly simple programming exercise BUT your power now needs to be software controlled and integrated with the other components. A dial and alligator clips won't cut it! (😅) Starting to see why a 'real' one will cost more than $100! 😊 Cheers, Michael 🇦🇺
I worked on a wire edm machine years ago. That machine monitored the EDM current and reversed the feed whenever the current got to high. This backed the material off of contact with the material being cut, As soon as the current dropped to an in tolerance value it would switch back to normal feed direction again. Also Without current limiting wire failure seems certain. Series resistors seem more important than parallel capacitors. Good luck with this thing Wonderful video quality and great narrative.
One of the best videos so far, lot of work you put in. Unfortunately this is real example how most DIY projects end up, you put in a lot of work and in 95% of cases you end up to have no use of them. Great lesson to all DIY guys, I am one of them.
@@ActionBOX Woow thank you Dave, you really are something different and really working on it. Wish you lot of success on your project, and to people following you. Wire EDM is really useful and finding a good recipe for it would make a big deal. Good luck.
I think the problem is with the motion control. The wire shouldn't touch the workpiece, the motion should stop right before touching it, or sometimes even reverse the path when the wire gets stuck to the metal. You need a feedback loop, that measures voltage between the positive and negative leads ( the wire and the workpiece). Or perhaps a powersupply that limits the current when the wire touches the metal. No need to control the speed of the wire spooling, I've seen machines with constant wire feed rate, the only downside of that is that you are wasting wire
This is exactly it, your motion control will even need to reverse during normal operation if the gap gets too small. It doesn't happen often (every 10 mins or so) but it does happen (see the BaxEDM videos and the Applied Science EDM video). If your spark generator is an old school capacitor + resistor combo that could also be an issue, you'll need fairly precise current control for small wire diameters which usually means a linear or switch current source.
ffoska and Matthew Anderson, Thanks so much for your input, this is exactly the type of answer I was looking for. I'll definitely do my research and come back with a subsequent video on the power supply. I wonder if. could develop my own for cheap, such that its perhaps not industrial grade but would work at a low cost for us hobbyists? Do you know of any open source build perhaps that I can work off of? How hard would it be to just create a pulled square DC wave anyways 😜. I really appreciate your detailed inputs. Thanks again 😊
@@ActionBOX As far as i know, the spark generator power supply is very much closed source, manufacturers who make EDM cutters do not provide too much info. You can find some stuff on youtube, as mentioned BAX EDM is a great source. I've worked on a machine from the '80 so somehow they managed to make it work using old tech, so it is probably much easier to do nowdays. The current is pulsed, around 120 - 150Hz ( the old machine, new machines probably have higher freq - probably the higher freq makes the cuts faster and gives better surface finish, but don't quote me on this...), the voltage is low, like around 3 to 5V. I would personally start with a welder that can do AC and DC pulsed and hack it ( or build some sort of high amp, low voltage powersupply with IGBT or power MOS FET), measure the voltage with some kind of high resoulution ADC and use the motion controll that is shown in BAXEDM's videos.
@@ffoska a think sodick have a lot info on it a worked on a320 a500 and a750 but y didn't remember the parameter , just selec materail thinknees number of pass, dia, and use c777 to align (teach ) :(
@@ActionBOX I'd be happy to collaborate on this. I have a design in my head for a unipolar spark generator that would be easy to make. How do I get in touch?
Thanks Mike. The two of us are working hard to make an easy to replicate DIY Wire EDM Machine. We aren't trying to compete with industrial machines, but just give you all something to work with at a low cost. Feel free to help us out on our mission to boost our channel, and share with any of your friends who may be interested. Cheers, Dave.
Wow! That’s an awesome wire machine build you’ve created there! Well done! I’ve built solid sink EDM machines in the past which I find are pretty easy to build in comparison to this especially as my machine design is old school and doesn’t require any code to work etc. I don’t know your power generator design but it does need to be different from a solid sink pulse EDM generator, it needs to be more like an RC design with a sine wave type output. This type of generator gives a superior surface finish but the electrode wear rate is very high which is not a problem on wire EDM as the electrode is constantly being replaced. You could try placing a low value capacitor across the output leads to emulate this more but you might need a generator redesign. Also your wire feed rate is too slow it needs to be around 8m/min with a cutting speed of 3 or 4mm/min and your software needs to be able to pause and reverse the g code if there’s a short circuit etc which Mach3 can’t do, but I believe Linux CNC can and it’s free software.
EDMGuru, WOW. This is the exact comment I was hoping for. I think you are totally right, and I will definitely try your suggestion of adding a resistor to the power supply. I appreciate you taking the time to write such a detailed response to my video, and I will get back to you all with a future video on perhaps developing my own workable power supply for Wire EDM. Cheers, Dave
This is very interesting and I love the creativity, that said I bought a 1994 wire edm that has millions of yen in development costs. It works great for a $2500.00 machine and is still commercially viable for limited use . And it is 4 axis.
I don't know what your power supply is like, but regular units have adjustable voltage, power, off time , plus compensating circuits that fine tune the sensitivity, also as some have said, the de ionizing system, 5 micron filters, etc.
Great work. I'm an engineer myself and most people don't realize it takes many failures before you reach a point of success. Despite the failure, I still learned a lot in this video. Hope you'll keep working on this machine though, it's very interesting.
I really like the content! It has enough information to explain what the thought behind the process is, good visuals, and some light humor. Can't wait to see this project evolve.
Thanks Jonah, Im glad you enjoyed our video and picked up some of the humor 😜. Dont forget to Subscribe and share with at least one friend. It helps us out 😃
Beautiful work. I developed a wired Edm in my garage. After a lot of study I learned the fundamental concepts so that the wire does not break. The open and closed load time must be critically controlled for high cutting efficiency. I achieved 90% erosion on the part versus 10% erosion on the electrode or wire. The junction between the advance pulse and the cutting and loading time is the key point for perfect operation. On my channel there is a small video cutting with penetration electrode. When possible I will make some EDM Wire videos to maybe help friends who follow the same path. A hug and good projects broo.
I use these at work, may I suggest the contacts for the upper and lower be solid carbide, and 0.3mm hard brass 900mpa is used for best results, your wire is braking as it is making contact with the work piece as said by other viewers, you should have a spark gap of aprox 0.15mm , our machines automatically sense the cut via an electrical ark which will never touching the workpiece, (short circuit) All the metal rollers have contact points which I assume are to prevent the wire arking on them
got my sub for the awesome video and got my like and bell for offering the parts for sale but also the prints for free thats dope good on you guys! its refreshing to see people actually wanting to share and teach others instead of just trying to grab as much cash as quickly as they possibly can
The money would definitely help our channel grow if anyone prefers to save the time or just cant make it at home, but on top of making a living we also want to help others, so hopefully all of our future videos will be the same. Feel free to help us out on our mission to boost our channel, and share with any of your friends who may be interested. Cheers, Dave.
Thank you so much Azure. I really appreciate it 😃. Feel free to help me out by sharing the video with anyone interested. We are trying to grow our channel. Thanks again, Dave
Just to be clear. There were tons of test made that said making parts with full infill does not make the part as strong as you think. Its all about how many walls you use and what kind of infill pattern you use. Gyroid 30% does the trick everytime. And well wall count is up to you
Thank you, this comment made my day 😊. I look forward to your comment in our future video. Something in plastic injection is coming up in two weeks 😉. Cheers, Dave
I am a toolmaker, I run sodick wire Edm machines from 1995. Currently trying to update them to modern drivers and motors, as the components it originally came with are deprecated and only available on eBay and likely don’t work anyway. This is fun stuff,
Awesome build and great video presentation. This video is one of the reasons why I liked and subscribed ages ago. Keep up the great work. Looking forward to more.
I have worked with EDM at a few places and have never seen wire reused. At one place where they did EDM with electrodes (graphite copper) they would be machined in a specially ventilated cnc . Then each hole had a rougher and a finisher electrode. When the rougher wore out the finisher became the rougher . At my current place were run 2 wire edm machines 20 hrs a day and ever 2-3 weeks it takes a whole pallet of brass wire
One way to solve this sort of wire break issue is with closed-loop speed control based on the cutting voltage/current, so the machine can automatically adjust the feed rate as cutting conditions change -- slowing down when the current increases due to more material thickness and worse flushing. You should also plan on treating those two idler ball bearings as consumables requiring regular replacement, since you're passing all of your cutting current through them and will be eroding their rolling contact area over time. They're certainly cheap enough to be treated as consumable, after all. (If you do want a contact roller that won't need replacement though, I'd recommend a mercury slip ring bearing like a Mercotac 205.)
Awesome job, I have at least a decade running wire and cnc and manual sinker edm. You getting this far is great. Some tips I would tell you is wire is its own beast and where your flush cups are located in relation to work is crucial .005 above and below plus good flush and filter. I have no idea how your controlling the power but lower power until fully into cut switching polarity from table to electrode and adjusting on and off times in your spark. In my opinion finding an older machines power supply with the preloaded power settings would be the only way I could ever even attempt something like this. Great job again, very very impressive. PS you got me to sub.
@@ActionBOX Have you ever thought of making a 5-axis? Your 30 taper machine inspired me and i've been tinkering in CAD about making a horizontal 5-axis. But achiveing minimal perpendicularity error on the table I think will be difficult. Also I think it will be difficult to achieve minimal backlash in the rotary joints without using a harmonic drive type gearbox.
@@Chris-yk5rm Hi Chris, great question. We've actually already built a trunnion table for our DIY CNC machine, which sits right on the flat table, enabling the 4th and 5th axis. The backlash is indeed an issue still which I have not come up with a solution for yet. Great suggestion, and glad you enjoy our content 😊. Cheers, Dave
I am really impressed with your toy. I believe the current and feed rate are too high. I do not know your power supply frequency but if it is like straight DC voltage you probably want to back it way off to not exceed 1/4 amp (maybe 30-40 volts, this has some connection to the frequency of the power supply) and then drive forward really slowly. Our first machine was less than 30 mm per hour in 25 mm material. I believe the frequency was about 2000 HZ and the machine had a vibrator on the wire to help keep the wire from shorting. This was before adaptive control was available so you have to keep the feedrate below the shorting speed. The wire should never touch the workpiece but you should try to maintain a spark gap of about .025-.050 mm. Additionally if the wire is not held tightly it will bow, which will be evident when all your sharpe corners look washed out.... You can get other weirdness. S-shaped walls. Almost all the current machines are capable of backing up if they hit the workpiece and short out, turn power off, and will not reapply power until the short clears. (most macines have a short circuit sensor for that. They first thing I would suggest is to go to a v-style groove for your guides. Something hard and permanent would be preferred but you could cut a v-groove in a carbide pin. Or if you really clever maybe a carbide cutting tool. Ultimatley, something non-conductive and wear-resistant would be you best bet. Using your roller and small hole you are also; going to EDM your bearing, eventually it will not run round, and the small hole will eventually enlarge especially after soemtime in the water as the metal parts become part of the curcuit. Aslo as an asiide the small rollers are really hard on the wire. Most of the production machines don't have small diameter tight rollers because it puts a bend into teh wire. If you unroll five feet of wire it should go almost straight to the floor without any cbvious stress. If the wire tends to lean in a direction like it has been on the roll too long it can cause the s-shape along the wall of your cutting path. If you can finally circulate water and can pump it coaxially along the wire it will help flush the cut slot (but only after you have entered the material) and you can pick up a small amout of speed. Good luck.
There is some awesome information here. Thanks so much for taking the time to write such a detailed response, I really appreciate it. I will consider all of this while designing our second version of the machine. Ps. feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
Hey just for fun the very first wire EDMS were attachment's to regular sinker EDM's. They ran in oil, they were more forgiving to poor electrical conditions. The gap was smaller and voltages were higher). You might turn your assembly 90 degrees with the wire facing downwards and see if you move it in .0025mm (that's .0001" right?) straight down with across your workpiece and see if it cuts, which might be easier than shelling out for pumps, filters and other stuff power supplies, etc.
I wonder, if the power supplies for wire EDM and die-based EDM should be completely different.. I'm pulling this out of my ... but it seems that for die-based EDM high current is ok, but for wire EDM current and feed speed should be really low, while the voltage is much higher. Like barely touching surface to create a bunch of tiny sparks, instead of completely rubbing against the surface. P.S. Props for sweet cad skills. Good CAD skills feels like super power (I'm late to the party, but I'm learning this now).
Thank you, I really appreciate your comment 😃. I'll look into this as you suggested and hopefully make a mention of it in my subsequent video. Thanks again 😊
@@ActionBOX correct me if I’m wrong but I thought the wire never makes contact with the part, just the arc. Maybe run the wire at a faster rate and your xy movements slower until you find the Goldilocks zone.
Excellent Build. To avoid wire breaking try using arc voltage/current as feedback to determine axis feed rate instead of a fixed feedrate .Similar to THC on plasma.
I once trainee at the Japanese renown wirecut edm factory( Its the compulsory in faculty final semester). I noticed in wiring diagram ,the submersible solution also had ion concentration control. By the way , I see some China adhoc CNC wirecut the workpiece in the air. No need submersible. A lttile bit more , I think to electrical the wire, remind me of the sparkling circuit might have close loop control in watching the spark behavior. I mean they can control wire current change in millisecond by feedback sensor on how to movement not to make wire touch to the workpiece.
So I’m watching this video thinking you need to have a sinker Edm machine and bam! You made one of those already. Both are on my list but I do need hellish accuracy so don’t know if I can do it diy. That said, for the cost of these systems, if you can work within their tolerances then this is golden!
HAHAHAH, I love this comment. Its doubtful that we can compete with professional industrial grade machines, but It sure does the trick at the cost of some accuracy. Ps. If you enjoyed this video, feel free to help us out by sharing with friends who may be interested. We are trying to drive up our views to get more exposure 😊 Cheers, Dave
Absolutely great project! I really like it. I think your problem with the wire snapping is mostly related to your spark generator. Just from the sound of the sparks I assume that you have very high peak currents (>50A) and probably no short circuit detection to switch off the current in case of a short (touching workpiece or simply having a conducive particle in spark gap). This might be acceptable for a sinker EDM since it 'only' increases your electrode wear, but for a wire EDM it leads to an immediate breakage of the wire. I built an electronic EDM spark generator some time ago (which I so far only used in an prototype sinker application). The sparks sound much less aggressive and more uniform if you have a proper short circuit detection and current control. If you have the chance take a measurement with an oscilloscope either with a high frequency current clamp or carefully insert a shunt resistor in the circuit to visualize the current (but dont blow up your system/scope). Another thing is that you need a feedback loop for your feed which is based on the spark gap (breakdown voltage or statistics between 'open circit'/'good sparks'/'short circuits'). Not having that, your feed rate of 1mm/min seems to be very aggressive in my opinion. Also I feel that your wire speed is very low compared to professional machines. Seeing how fast industrial machines feed the wire in combination with the cost of those spools was the main reason why I never continued to build a wire EDM even though that would be a very useful tool. Keep up your content!
Wow, this comment summarized all our issues. Thanks for taking the time to type this all out. I definitely believe you re 100% correct after having read other comments. I will take this into account when building our second version, although its seems like there is lots of work to do before this is comparable to a real industrial grade machine. I have never seen these in person, and Ive never actually looked at videos of Wire EDM on RUclips so I probably should have started there 😆. Thanks again for an awesome response. Ps. If you enjoyed this video, feel free to help us out by sharing with friends who may be interested. We are trying to drive up our views to get more exposure 😊 Cheers, Dave
I agree, one of the biggest issus should be power source and therefor that the sparks are way to violent, coupled with no "spark gap" control loop. THB i have only been exposed to EDM as "sinker EDM" in University for like 1-2h in a practical workshop(?), but the sparks back there were way way softer and smaller. Also they really pushed that spark gap control onto us as a important part, including that "sinker EDM" seems to have "pull backs" to flush the "hole".
Thank you Jonas. Much appreciated. Ps. Feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
I wanted to build one a few years ago. I learned some place that industrial machines use pulsed current using PWM. I experimented with low voltage ac and a brass strip mounted in a hack saw and a bucket of water. By putting an induction coil with air core in series with ac transformer supply the spark happens on contact release rather than contact touch. The induction coil I used was the high voltage secondary coil removed from a MOT (had to cut transformer open to get it). Was able to saw into aluminum fairly easy. The ac voltage I was using was low like about 24 volts. Some guys use diesel in the tank because its cheep 😅.
Yeah Filter the water down to about 10 micron (You can use a pool sand filter, but use 10 micron cotton instead of sand), also maybe pump the water through deionization beads to keep the conductivity of the water more stable (A little table salt helps, but just a pinch. Also there is a common saying in Wire EDM "Flushing, flushing and flushing", its important to flush away that residue quickly, flushing is key.
Yay! We’re glad the video was at a good pace to follow along. Feel free to share photos with us if you do end up building it - we’d love to see it! Thanks for supporting our channel! 😃 Cheers, Alan
Well your on the right track. Coming from an industrial standpoint, your wire feed is way too slow, your cut feed is too fast and your flushing pressure is way too low, your not controlling the resistance of the DI fluid. Also oddly enough round stock is actually pretty hard to cut in an edm, relatively speaking. Since it is not a constant thickness they resistance from the part is constantly changing and thus your settings technically should be as well. Add to that the DI flow around the round stock which will actually divert the DI Fluid away from the cut and thus make it less effective. Also depending on the grade of stainless it may have a rather high resistance rating which will make it more difficult to cut, something like brass or aluminum with a lower resistance rating may be a better start point. Your excessive feed rate is obvious from the deflection of the wire. As weird as it sounds the wire should never actually touch the part during the cut. So seeing it bow before it breaks is an obvious sign. There should always be a "spark gap" between the wire and the detail, the DI fluid will transfer the current to the part. Which brings up another point, your water may be too pure and not passing enough current. Industrial machines have sensors and use filtration and DI resins to keep the resistance level of the water very narrow window. Ignoring all the potential resistance changes from passing the current thru a bearing, the contacts are sacrificial. Unless you want to be replaceing the bearing all the time you really should make the contacts a separate piece. A cheap 6mm carbide endmill trapped in an aluminum block a few mils in the wire path would be an easy cheap option. Also it is good to have them as close to the wire guides as possible as the further they are apart the more resistance you have in the wire, and industrial machines will generally have them in the low pressure flushing to increase voltage transfer and keep the wire and contact cool. May want to consider getting some coated wire. The wire cost more and will cause more wear on the system but it works better in poor cut conditions, which is pretty much the best you will see on a DIY machine, so it will give you a bit more wiggle room on dialing in your settings.
Thanks Mark! We will for sure be releasing another video with our next iteration in the future. Meanwhile, stay tuned for some really cool upcoming machine videos. Cheers, Alan
I guess those stepper will be reused in other projects, but they are way overkill for this project. The smallest Nema17 would work as good. Just in case someone wants to build this project. They can get away with 30$ instead of 300$. I like the design, can't wait to see it working!
Flushing is key for edm. Get your wire guides as close to the material as possible. 0.2mm or closer. And maybe higher pressure on the nozzles. Every wire edm I ever ran varied the feedrate into the part to match how fast it was actually cutting, especially as it's entering the part starting the cut.
@@Jkirk3279 You could just have a sealed chamber and purge the lot with Argon- arduinos don't need to breathe. That or use a multi pass compatible flux core
The tension sensor are actually "Load Cell" They detect pressure being exerted on the cell when you press on the end. the load cell is located in the open area in the middle where all that white stuff is. They usually look like a bare IC but are not made like an IC. They are usually resistive elements in a zig zag pattern and change value when stressed.
Awesome! As your power supply gets more sophisticated you’re going to need to shield those strain gauges and their wiring. Also using set screws to transfer torque to cylindrical shafts is bad practice. Keep up the great work!
I can't speak for Chris-bg8mk but I would think that for torque transfer a woodruff key or a taper to shaft and hub/wheel is preferred over a set screw. I'm curious whether and how quickly this comment, 2 years after, will draw (your) attention 😀. Apart from all that I'm impressed by both your design and the video production. The positivity and the useful contributions in the comments is great too. ETA: subscribed!
Interesting project. Another thing to consider is the spool collecting the wire may build up in the center only. Make a feeder that travels back and forth like a baitcaster fishing reel 👌
Idk much about this technology but I do think that the wire isn't supposed to contact the workpiece which is why the need for the dialect water, it acts as an insulator while at the same time keeps the wire cool, so I don't know what kind of power supply you would need but it would need to be high enough volts to be able to create a spark gap much like how plasma cutters work. if you ever used one the first thing they tell you is don't let the tip touch the workpiece. Might be a good place to start. Seems like a cool project I would love to see it work for us diy guys.
Ok so I find this technology really fascinating and found some interesting things out. So there is this guy who made one of these over at instucables, and for his power supply he uses incoming 110 ac, runs it through a 200w bridge rectifier, uses a 200w incadcent light bulb to limit the current, and looks like it has a couple capacitors to. The circuitry looks relatively simple, and he has posted videos of it working. The AC is converted into a pulsating DC voltage 60hz or 60 times a second, idk if this would work or not because I thought it had to be between 300 to 1000 a second. But maybe still does work, and is just slower. I was also right about the wire not supposed to touch the work piece you may have to touch it to start the arc but then pull it away much like Tig welding. But check his setup out because it seems relatively simple, and Id like to try this myself because this could open up a world of machining not normally accessable by us hobbyists. www.instructables.com/EDM-Electrical-Discharge-Machining/
Great project, & well presented! Applied Science did a video 2 years ago on DIY EDM drilling ( ruclips.net/video/rpHYBz7ToII/видео.html ). It covers many of the issues documented in comments including the requirements for current control (BAXEDM), motion control with backup (DynaMotion), water filtration / conductivity, etc. Some of the issues he covers don't apply (such as rotating the 'cutting tool' (/drill tube) to ensure uniform erosion; with wire EDM you are constantly replacing the 'cutting tool' with new wire), but the core concepts are still the same.
Your feed rollers seem like they'll work fine, but I think a capstan roller might work even better with less required machining, fewer moving parts, and no cantilevered loading.
Its great to see a Canadian person doing this kind of DIY and education! Keep up the great work! Especially sharing your parts and designs! I hope to one day make all of these cool machines you have! Thank you!
Thanks a bunch. Feel free to help us out and share our video with a friend who may be interested. We are really trying to grow our views 😊. Cheers, Dave
I maybe missing something... but is there a feedback system for the arc? If the wire is making direct contact to the cut material ( resistance between them near zero) the wire must brake, mechanically and due to excessive current. Shouldn’t the power be disconnected, the wire and cut material be separated from the wire (high resistance), then gapped by an additional predetermined amount before the arc (power) and cutting is reestablished? Perhaps this is already being done, but I didn’t see it mentioned in the video.
Hi, great comment, you are absolutely correct that this needs to be done, and indeed it is not in this V1. We are working on the V2 which should tackle these issues. Cheers. Dave.
@@ActionBOXSide note: I'm always surprise when a great content creator, like you, are saying "thank you for watching"... I meant.. YOU put hours and hours of work to show us a FREEEEEE content. And you thanks us in the process!
@Alexandre Valiquette hahah, well we appreciate you helping us grow. We really want to hit 100k subscribers by 2023 so we appreciate every viewer like yourself. Feel free to help us out further and share the video with your friends who may be interested. Thanks so much again for your kind comment. I look forward to your comments in our subsequent videos. Cheers, Dave
Cutting with WEDM is a delicate balance between a dead short and a open circuit. The internal diagnostics of the cutting condition is everything. Ask yourself why Sodick puts linear motors on all their wire machines. Quicker response times during cuts, equates to better performance cuts. Be sure of ONE thing. If you can't get your path movement to "back up" slightly, then move forward again (per your cutting conditions requirements), you won't end up with much. I'll say it another way. What CNC machine do you know "backs up" slightly in its own path, then returns forward in its path. Answer: Only WEDM's. This is done to protect the wire from breakage. Good Luck. Lee
very nice. There's a different type of wire EDM called a Moly EDM that uses molybdenum wire instead of brass. The wire is more durable and cuts considerably faster on harder materials. It doesn't submerge either but uses an oil instead directly on the wire. It also uses a single wire spool that also takes up the used wire which is then reversed. The same length of wire can cut for hours. There's a huge market for that type of EDM but most people don't know about it. It saves days of cut time over brass.
This is awesome, I have not heard of it. Thank you for the tip, and I hope you liked the video 😊. Feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
You should detect when the arc happens and then rectract, I'd say it's a very visible short that is happening there. From what I've read, you need to supply the same current, if the wire touches the workpiece it will then short and of course.... blow up! :)
I ran a Sodick A280L for years. Mostly 0.010"wire with diamond wire feeds. I cut stock up to 7 inches thick with that machine. The filter and deionizing systems were substantial. Also, there needs to be feedback from the spark to the wire power system as the condition of the kerf is a continuous state of change what with variation in the spark line and excellence of flushing, which was powerful - lots of water at quite high pressure - I would guess 40 to 60 psi or more. Nice project. Visit a working wire and take some notes.
Thanks Ron. Admittedly I have never seen a wire machine while attempting this so I do appreciate all of the feedback. I will take all of these notes into account when building the V2. Cheers, Dave
@@ActionBOX Maybe two more things. Try cutting aluminum foil. It will confirm most of your operating ideas and little flushing will be need. You will need an auto fee mechanism tied to the wire condition circuit. This will slow the cut if conditions are tenuous at the wire and speed it up when conditions are good. These machines are quite complex I think with lots of feedback from the wire circuit going back to the motion control module. It will test your engineering skills quite seriously but you seem willing to throw yourself into the fray. Best of luck.
An incandescent light bulb (current limiter) in series works great for over current protection. The bulb will prevent burnout and is also an indicator for a short. Couple that with a sensor and custom code to reverse the axis direction if the light turns on.
Hahahah, your comment made my day. I hope you got to share with someone else who may find this interesting ey 😜. Glad you enjoyed the video, and thanks for helping us grow this channel.
What do your tension readings look like leading up to the break? Without gap voltage feedback adjusting your feed, you might be making contact with the sample. With constant feed I image your tension would start to increase as you make contact and start to push against the sample with the wire.
Is that nominal guide hole to make sure the wire diameter remains constant also like an extruder? And surely reel/wire speed increases cutting area, no idea typical rpms.
Hahaha, thanks A J, Much appreciated. Feel free to help us out on our mission to boost our channel, and share with any of your friends who may be interested. Cheers, Dave.
Considering the value of information and experience i get from this channel you gave me no choice but to ... subscribe immediately. Looking forward to learn a lot. Thank you so much !
Spark control ... needed. The nozzles should have the conductor tips.. you only charge the wire near the work area. The take up wheel could be rubberized .. running wire between 2.. when running wire you don't need to re spool. Most machines just collect in a bin. I've run and programmed Hitachi 203r and also a Makino wire edm. These have independent top and bottom wire guides to help align and square to work area..
What about crossing the wires between the brake rollers? instead of = looking wires, use x. zero is rollers, x is wire, 0x0? or would that be too much tension?
Few thoughts from a wire edm noob: 1, that 4 groove back and forth spooling approach seems like it'd be tedious to load. 2, rather than a pid tension control scheme you can simply use 2 servos to spool and unspool at the same rate and use a tensioner pulley to achieve a constant tension
If you'd like to help make future projects like this possible please consider supporting them / us by becoming a channel member: ruclips.net/channel/UCHrFvnP1EEEZHNam_Nk_5rQjoin/join A big thanks to those who already support the projects!
It's really not a good idea to grab the output lever of a servo and turn it like that. It's an output not an input. There are gears in there.
question did you reverse polarity
I didn't see this below, but you may want to change the power connection from your idler bearing to one of the tension spools. Bearings do not provide a stable / reliable mechanism to commute the power/current.
I have been running a few wire edm's at work for a few years including a big Fanuc Robocut 600.
Here is just a few surgestions .
1.To prevent wire snaps your are looking at a movement speed (speed of cut) of about 0.1-0.5mm per minute depending on what your are cutting and the thickness.
2. Speed up the speed of your wire (speed at wich the wire pass thru the cut , is being used) to 4-6meters per minutes. The faster the wire the cleaner and smoother the cut.
3. If your are using a large cap on your output power lines, back off on the size, this is scalpel not a broadsword like a sinker edm the initial large spark will snap the wire.
4. Current, from what I have seen you do not need more that 5amp on that wire, again scalpel not broadsword.
4.1 If I was cutting that round bar on the fanuc, typical setting would be tavel speed I would get 0.3mm per minute, wire speed of 8m/min, current 3amp. With a voltage of about 80v.
keep in mind the wire cut moves forward and back depending on the gap voltage reading, you cannot just set it at a speed and plouge into the material.
5. Try swappingthe polarity of the output line, it sounds stupid but is does matter.
I designed my own Small hole edm under the name Radnor Technologies a few years back and if I remember correctly I used - on the electrode and + on the work piece this results in a quick material removal rate but eats your electrode for breakfast. In your case with the wire that would be bad.
6. Remember when in doubt flush flush flush loose carbon can mess with your cut.
Here is just a few other surgestions not related to the wire snaps.
1. Filter your water else things goe pear shaped very quickly.
the 20 inch big blue spun polyprop filters used for home water systems works well.
Fit a 10 micron filter the 5 micron does not last very long. The fanuc has a hugh 5micron filter as standard. Aluminium will clog a filter like you will not believe.
2.Deionize the water if your want to cut for longer than a few minutes.
The waters optimal resistance should be 70 kilo ohm per cm2. Doin ght deionizing is easy, you buy the dionizing resin (small balls) from the same place that sells the filters, normally in a 25Kg bag.
It is in a bucket with a inlet and outlet.
Let the outlet drain into your tank and hook up a pump to the inlet to switch on when the resistanc is below 60K. This is a easy job for a Nano with a ADC and 2 probes on the water, we used 2 stainless rods with a surface area of 1cm2
3.The rewinder unless you want to try and reuse the wire... bad idea... ditch the roll winder just make a puller that pulls the wire and dumps it in a box, this is what the fanuc does. Some other brands chops the wire wich will be better since it takes less space.
Hope this helps, all and all great project I wish you the best of luck and look forward to the next video... subscribed 🙂
How does one gather the gap voltage reading? Awesome advice btw, sounds like they are going to run through wire much faster than they expected lol
do I understand this right, to make a 10mm long cut would take around 100 minutes and use around 500 meters of wire? is the wire re-usable? seems it's an extremely high cost. unsure of the cost of the wire, but.. seems a lot for a cut that is less than half an inch long. did I make a silly math mistake somewhere?
@@CrudelyMade your maths is a little off. But only a little. Based on those numbers I get 33 minutes and 266m of wire. Still seems high wire usage and very slow feed.
@@frankcarter8399 To measure the gap voltage you have to measure the line voltage, as soon as spark is made the voltage will drop to something like 3-5V depending on the system, since the rest of the power is being used (routed thru the cut).
If the voltage is too high you can move a bit forward into the cut and if it is too low ..aka short circuit move back from the material.
You need to monitor this with a ADC circuit and move the wire accordingly.
You do us alot of wire but just for referance we paid about 70USD for a 8KG roll of wire and that contains about 19 kilometers of wire, so not to bad.
@@msbealo Jip my boss use to get annoyed when I run the wire at 10-14meters per min and set the speed down until I just coved the screen so he can't see the speed.
It is better to waste a bit of wire than to have to restart a 8 hour job if the wire snapped in the middle of the work piece. Things normall goes fine in the beginning but a soon as you get to section that is hard to flush the wire like to snap when it is being run slowly. The wire cost works out at about 0.0036 USD per meter.
Wire EDM is a very slow process, we had 3 machines a 30 year old Fanuc that works similiar to this home machine, slow and manual wire loading, that one used to cut at 0.02-0.05mm per min for 50mm thick stainless. We had the new Fanuc that ran at about 0.1-0.6mm per min on 50mm stainless. Lastly we had some Fong Kong machine that reused it's wire. It used a 0.19mm Moly wire that last very long but is very expensive and a complete "B" to load. Can take up to a hour to get up and running if the wire snaps.
It’s your spark generator. BAXEDM makes a modern one, but EDM has been around wwii and you can find some patents that basically use a big resistor for good current control. Most switching power supplies have poor “instant” current control, but A simple resistor should be smooth. You also need feedback between controller and current generator, water conductivity increases as you cut.
Thanks Ethan, that makes sense. Ill try your suggestion and stick a big resistor at the tip of my power supply. Ill get back to you on that. On the note of BAXEDM, I would love to collaborate with him. Hopefully he reaches out to me now that I have asked him. Cheers 😃
@@ActionBOX I definitely recommend BAXEDM. He seems to be the pro on amateur wire EDM...
@@pappaflammyboi5799 I have reached out to him asking for a collaboration but he has not responded yet. I hope he does eventually 😊. Thanks for your input and hope you enjoyed the video. Cheers Dave
@@ActionBOX be a man, make power supply. Use 110v for supply, 2-20 ohm resistor, IGBT with 2-10 us on and 10-20 us off timing. Now you use very long pulses
Yes BAXEDM is the man to get it done, I remember he had a problem with wire breakage as well, something to do with polarity if I remember correctly.
It's pretty amazing how much more people can accomplish now that 3d printers and arduino boards are so popular/affordable. Kids these days dont realize how good they have it with all the technology at their fingertips... I wish all this stuff existed when I was a kid because I would be so many years ahead of where I am now.
That’s very true! 3D printing technology has definitely helped us make our ideas a reality. It’s never too late to start playing around with this technology 😃
@@ActionBOX That's what everyone tells me, but I find myself staying up till 3am trying to learn new things... There isn't enough time in the day for work and play. BTW, I subscribed. I appreciate when a channel takes the time to respond to comments like you did.
@@user-lx9jm1wo3hYes, good point. Juggling work and play is never easy. We appreciate you subscribing! We always respond to our viewers' comments. Looking forward to your comment on our next video (coming out Jan 21) 😉. Cheers, Alan
Great point.. If I would have had access to all this as a kid I would be working all day building new stuff. Kids these days are more interested in videogames and instagram than utilizing the technology they have at their finger tips.
In the late 1970s I learned to operate one of the first Wire EDM machines in the US. It was a Elox 100 with 2 axis Fanuc NC that was commanded by paper tape with punched holes. This was so high tech at the time that our tool & die shop became a stop for VIP visits to our city. Wire EDM machines have advanced a very long way from the Elox 100.
Although many decades have permitted the concepts and components to become more normal I am blown away by the idea of an individual being able to DIY their own wire EDM machine. Very kewl and I'm impressed and envious of any of you that have the ability, resources and space to do this at home. Awesome.
The wire is snapping because of the lack of a real spark generator, bax edm makes one, but thats about it, spark generators are a serious technology and not really something you can buy off a shelf.
Thanks for your insight. Ill look into making the spark generator another project of mine, and come back to post it here. I appreciate your input 😃,
@@alextreseder6274 please. 😊if you have a link for info, please share :)
@@alextreseder6274 that sounds really good. i am hoping for some opensource variant so we hobbyists can take part in this precisionworld
Could you use a tig welder and soft start?
@@engina1572 no
This brings me back a few years back when I was interning at a metalwork company where I was operating a wire EDM. It was boring as heck because the whole morning we would set up 1 or 2 plates and then we wait for them to cut finish in the next 3 days. I was setting up the parts too fast that the full time guys told me to slow down so that it will look I'm doing work. I find it really counter intuvtive. I wanted to learn more but they didn't let me because I'm only interning for 3 months.
What great memory that was.
Im glad my machine could bring you back to those days 😃. Hope you enjoyed the video.
Ps. feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
I just search in google for if someone has made a hobby wire EDM machine and I see you just released this yesterday! take my sub
Hahaha. Thanks. I appreciate that. Glad you enjoyed the video 😃
I have been running wire edm for 10years now and everything about this video scares me. Great job
Hahaha, it was our first attempt 😉. We are working on an improved version 😃. Thanks for your support. Cheers, Dave
@@ActionBOX hi Dave. First video I saw of yours wire machines have a list of parameters for the machine know as E-PACKS for Example Ip,Vo,Sa,WS,Wt and many more. Epacks control voltage, amp, off times, wire speed, tension and Liquid quantity , stability controls and other finer setting. Second feed rates are important not only linear travel but wire speed will drastically reduce your consumption of your electrode (wire). Resin filters will help with your conductivity. last thing I saw that looked odd was the distance of the wire to your work piece. It could be angle of the video but the electrode should Not contact work piece. Wire machine will not start with a contact. Sorry for confusing comment hope some of the info helps
I can see were you got your inspiration from, looks very familiar :-) I can give you a ton of advice if you are interested. There is a long list of improvements to be made on your current design before you can make any wire EDM cuts, and a very long list before you can make high speed efficient and accurate cuts. Feel free to contact me.
Hey, I love that you reached out. I would love to get in touch and work on a collaboration if you are interested 😃. Cheers, Dave
I was just gonna comment asking him to watch baxedm to understand how the XY movement has to be controlled as a function of the spark current. But now he has a direct line to you :)
@@scooteracers the power supply would be the only limitation really. I’ll give you a hint, our V2 is meant to solve this problem. I like the idea though, I might start doing that myself. Cheers
@@scooteracers you want to cut metal stock at less than 1mm per minute for 50mm thick stainless? why would you want to do that? any saw will do at 10 times the speed i guess
I would really like to hear your take on things as well.
a little bullet point list would be really cool to learn from.
EDM Engineer here.
First of all, this is absolutely outstanding work.
Secondly, we call it dielectric fluid, not coolant
Thirdly, Wire EDM is much easier when feedback from the generator is used to control the feedrate. Fixed feedrates aren't very useful. I can't go into the feedback mechanism.
You also would benefit from top and bottom flushing to ensure the spark gap is kept free of debris.
Thank you, this is an awesome reply to get from a professional. We will incorporate all of these suggestions into our V2. Feel free to help out and share the video with others in your field who may be interested, we are trying to grow the channel. Cheers, Dave
@@ActionBOX I've already shared it with my colleagues 😊
Can you drop the amperage at all? In general with wire erosion you're looking for a 300v ignition voltage, so you can drop your current. That spark looks way too big for 0.25mm brass wire. If you increase your wire feedrate it might resist the damage better. You can also try reversing the polarity - that can help with electrode wear.
Using coated wire might help too, but it's quite a bit more expensive.
You might also get some benefit by reducing the duty cycle.
EDIT: Don't forget that when eroding a round workpiece, all of the energy at the start of the process goes into a single point. Try eroding something like a square blank - it'll be more consistent.
Good luck!
Dude, this is literally the most amazing project I’ve seen in the past year. And your CNC mill?!?! That thing looks seriously impressive.
Incredible engineering!
Thank you so much 😊. I really appreciate this comment. Cheers, Dave
applied sciences edm videos have your answers, you need a control that manages the postion in the cut
@@AndrewBoraas Do you mean it needs to change it operating parameters dynamically as the engaged wire length (area?) changes?
Using a constant feed rate and the object's geometry to dictate the power is a fairly simple programming exercise BUT your power now needs to be software controlled and integrated with the other components. A dial and alligator clips won't cut it! (😅)
Starting to see why a 'real' one will cost more than $100! 😊 Cheers, Michael 🇦🇺
I worked on a wire edm machine years ago. That machine monitored the EDM current and reversed the feed whenever the current got to high. This backed the material off of contact with the material being cut, As soon as the current dropped to an in tolerance value it would switch back to normal feed direction again. Also Without current limiting wire failure seems certain. Series resistors seem more important than parallel capacitors.
Good luck with this thing Wonderful video quality and great narrative.
Thanks so much Bob. I am currently working on all those updates and will have a video of the V2 in a few months. Cheers, Dave
This project greatly reduces the barriers to entry for the hobbyist.. Thank you and I will be anxiously awaiting your next video on this project.
Thanks Karl 😃, we appreciate your compliment and support.
One of the best videos so far, lot of work you put in. Unfortunately this is real example how most DIY projects end up, you put in a lot of work and in 95% of cases you end up to have no use of them. Great lesson to all DIY guys, I am one of them.
Glad you enjoyed the video 😃. We have the V2 coming out soon and hopefully it will be the %100 😊. Cheers, Dave
@@ActionBOX Woow thank you Dave, you really are something different and really working on it. Wish you lot of success on your project, and to people following you. Wire EDM is really useful and finding a good recipe for it would make a big deal. Good luck.
@@icuric4077 Thank you 😃 I really do appreciate your support. Looking forward to your comment in our future videos 😊
I think the problem is with the motion control. The wire shouldn't touch the workpiece, the motion should stop right before touching it, or sometimes even reverse the path when the wire gets stuck to the metal. You need a feedback loop, that measures voltage between the positive and negative leads ( the wire and the workpiece). Or perhaps a powersupply that limits the current when the wire touches the metal. No need to control the speed of the wire spooling, I've seen machines with constant wire feed rate, the only downside of that is that you are wasting wire
This is exactly it, your motion control will even need to reverse during normal operation if the gap gets too small. It doesn't happen often (every 10 mins or so) but it does happen (see the BaxEDM videos and the Applied Science EDM video). If your spark generator is an old school capacitor + resistor combo that could also be an issue, you'll need fairly precise current control for small wire diameters which usually means a linear or switch current source.
ffoska and Matthew Anderson, Thanks so much for your input, this is exactly the type of answer I was looking for. I'll definitely do my research and come back with a subsequent video on the power supply. I wonder if. could develop my own for cheap, such that its perhaps not industrial grade but would work at a low cost for us hobbyists? Do you know of any open source build perhaps that I can work off of? How hard would it be to just create a pulled square DC wave anyways 😜. I really appreciate your detailed inputs. Thanks again 😊
@@ActionBOX As far as i know, the spark generator power supply is very much closed source, manufacturers who make EDM cutters do not provide too much info. You can find some stuff on youtube, as mentioned BAX EDM is a great source. I've worked on a machine from the '80 so somehow they managed to make it work using old tech, so it is probably much easier to do nowdays. The current is pulsed, around 120 - 150Hz ( the old machine, new machines probably have higher freq - probably the higher freq makes the cuts faster and gives better surface finish, but don't quote me on this...), the voltage is low, like around 3 to 5V. I would personally start with a welder that can do AC and DC pulsed and hack it ( or build some sort of high amp, low voltage powersupply with IGBT or power MOS FET), measure the voltage with some kind of high resoulution ADC and use the motion controll that is shown in BAXEDM's videos.
@@ffoska a think sodick have a lot info on it a worked on a320 a500 and a750 but y didn't remember the parameter , just selec materail thinknees number of pass, dia, and use c777 to align (teach ) :(
@@ActionBOX I'd be happy to collaborate on this. I have a design in my head for a unipolar spark generator that would be easy to make. How do I get in touch?
What kind of genius sorcery is this?.. Not the wire edm, that's easy. But the design.. My god man, Big Brain!
Thanks Mike. The two of us are working hard to make an easy to replicate DIY Wire EDM Machine. We aren't trying to compete with industrial machines, but just give you all something to work with at a low cost. Feel free to help us out on our mission to boost our channel, and share with any of your friends who may be interested. Cheers, Dave.
Wow! That’s an awesome wire machine build you’ve created there! Well done!
I’ve built solid sink EDM machines in the past which I find are pretty easy to build in comparison to this especially as my machine design is old school and doesn’t require any code to work etc.
I don’t know your power generator design but it does need to be different from a solid sink pulse EDM generator, it needs to be more like an RC design with a sine wave type output. This type of generator gives a superior surface finish but the electrode wear rate is very high which is not a problem on wire EDM as the electrode is constantly being replaced. You could try placing a low value capacitor across the output leads to emulate this more but you might need a generator redesign.
Also your wire feed rate is too slow it needs to be around 8m/min with a cutting speed of 3 or 4mm/min and your software needs to be able to pause and reverse the g code if there’s a short circuit etc which Mach3 can’t do, but I believe Linux CNC can and it’s free software.
EDMGuru, WOW. This is the exact comment I was hoping for. I think you are totally right, and I will definitely try your suggestion of adding a resistor to the power supply. I appreciate you taking the time to write such a detailed response to my video, and I will get back to you all with a future video on perhaps developing my own workable power supply for Wire EDM. Cheers, Dave
Very good reply! Only thing I'd like to add to not use distilled water. Dielectric fluid is needed.
@@MadHatter123456 as stated in the video, we are using deionized water which is a dielectric. We are not using distilled water.
@@ActionBOX ooops. Sorry, my bad!
@@MadHatter123456 no worries hahah. Thanks again for watching and for your insight
This is very interesting and I love the creativity, that said I bought a 1994 wire edm that has millions of yen in development costs. It works great for a $2500.00 machine and is still commercially viable for limited use . And it is 4 axis.
Plus I can cut 3/4 steel at almost .3" a min.
I don't know what your power supply is like, but regular units have adjustable voltage, power, off time , plus compensating circuits that fine tune the sensitivity, also as some have said, the de ionizing system, 5 micron filters, etc.
These are awesome suggestions. Thanks for sharing 😊
That looks impressive!
I'm also currently working on a wire EDM machine and think your project is very inspiring.
Thanks for sharing.
Thank you! We’re so glad this video is inspiring. It’s great that you’re working on your own machine, wish you all the best with that project!
Great work. I'm an engineer myself and most people don't realize it takes many failures before you reach a point of success. Despite the failure, I still learned a lot in this video. Hope you'll keep working on this machine though, it's very interesting.
Thanks for your awesome comment, I appreciate your support and will continue to work on this 😊
I really like the content! It has enough information to explain what the thought behind the process is, good visuals, and some light humor. Can't wait to see this project evolve.
Thanks Jonah, Im glad you enjoyed our video and picked up some of the humor 😜. Dont forget to Subscribe and share with at least one friend. It helps us out 😃
@@ActionBOX pp
@@ActionBOX I’ll
@@ActionBOX ⁶⁶6
Beautiful work. I developed a wired Edm in my garage. After a lot of study I learned the fundamental concepts so that the wire does not break. The open and closed load time must be critically controlled for high cutting efficiency. I achieved 90% erosion on the part versus 10% erosion on the electrode or wire. The junction between the advance pulse and the cutting and loading time is the key point for perfect operation. On my channel there is a small video cutting with penetration electrode. When possible I will make some EDM Wire videos to maybe help friends who follow the same path. A hug and good projects broo.
That was so fun! Looking forward to seeing the solution!
Thank you. Working on it already 😃. Don’t forget to help me out and share the video in the meanwhile 😊. Cheers, Dave
I use these at work, may I suggest the contacts for the upper and lower be solid carbide, and 0.3mm hard brass 900mpa is used for best results, your wire is braking as it is making contact with the work piece as said by other viewers, you should have a spark gap of aprox 0.15mm , our machines automatically sense the cut via an electrical ark which will never touching the workpiece, (short circuit)
All the metal rollers have contact points which I assume are to prevent the wire arking on them
Makes sense. Thank you 😊
got my sub for the awesome video and got my like and bell for offering the parts for sale but also the prints for free thats dope good on you guys! its refreshing to see people actually wanting to share and teach others instead of just trying to grab as much cash as quickly as they possibly can
The money would definitely help our channel grow if anyone prefers to save the time or just cant make it at home, but on top of making a living we also want to help others, so hopefully all of our future videos will be the same. Feel free to help us out on our mission to boost our channel, and share with any of your friends who may be interested. Cheers, Dave.
This is amazing! Its very inspriring what one can do with a garage full of tools and the internet.
Thanks! We’re glad you found this inspiring and helpful - mission accomplished! 😎 Stay tuned for some really cool upcoming projects. Cheers, Alan
Absolutely amazing build! Keep it up! That's top notch engineering.
Thank you so much Azure. I really appreciate it 😃. Feel free to help me out by sharing the video with anyone interested. We are trying to grow our channel. Thanks again, Dave
I almost have no idea what he's making, but I kept watching because the design and building process was so interesting.
Thank you, I love this comment 😃
Just to be clear. There were tons of test made that said making parts with full infill does not make the part as strong as you think. Its all about how many walls you use and what kind of infill pattern you use. Gyroid 30% does the trick everytime. And well wall count is up to you
You guys are awesome providing this knowledge for free I couldn't have subbd any quicker!!!
You guys have a great channel. Love watching each new upload
Thank you, this comment made my day 😊. I look forward to your comment in our future video. Something in plastic injection is coming up in two weeks 😉. Cheers, Dave
I am a toolmaker, I run sodick wire Edm machines from 1995. Currently trying to update them to modern drivers and motors, as the components it originally came with are deprecated and only available on eBay and likely don’t work anyway. This is fun stuff,
That’s awesome, maybe I can learn more from you as I’m clearly not an edm machinist. This is indeed fun stuff. Glad you enjoyed the video.
Awesome build and great video presentation. This video is one of the reasons why I liked and subscribed ages ago. Keep up the great work. Looking forward to more.
Awesome, thank you!
I have worked with EDM at a few places and have never seen wire reused. At one place where they did EDM with electrodes (graphite copper) they would be machined in a specially ventilated cnc . Then each hole had a rougher and a finisher electrode. When the rougher wore out the finisher became the rougher . At my current place were run 2 wire edm machines 20 hrs a day and ever 2-3 weeks it takes a whole pallet of brass wire
good to know, thank you for sharing.
One way to solve this sort of wire break issue is with closed-loop speed control based on the cutting voltage/current, so the machine can automatically adjust the feed rate as cutting conditions change -- slowing down when the current increases due to more material thickness and worse flushing.
You should also plan on treating those two idler ball bearings as consumables requiring regular replacement, since you're passing all of your cutting current through them and will be eroding their rolling contact area over time. They're certainly cheap enough to be treated as consumable, after all. (If you do want a contact roller that won't need replacement though, I'd recommend a mercury slip ring bearing like a Mercotac 205.)
Wow, that’s an amazing recommendation with the mercury bearing. Thanks so much for sharing. Your comment is definitely useful. Cheers.
Awesome job, I have at least a decade running wire and cnc and manual sinker edm. You getting this far is great. Some tips I would tell you is wire is its own beast and where your flush cups are located in relation to work is crucial .005 above and below plus good flush and filter. I have no idea how your controlling the power but lower power until fully into cut switching polarity from table to electrode and adjusting on and off times in your spark. In my opinion finding an older machines power supply with the preloaded power settings would be the only way I could ever even attempt something like this. Great job again, very very impressive. PS you got me to sub.
Thanks for all the awesome videos!
No worries! These awesome comments motivate us to make more great content for subscribers! Glad you enjoyed the videos 😄
@@ActionBOX Have you ever thought of making a 5-axis? Your 30 taper machine inspired me and i've been tinkering in CAD about making a horizontal 5-axis. But achiveing minimal perpendicularity error on the table I think will be difficult. Also I think it will be difficult to achieve minimal backlash in the rotary joints without using a harmonic drive type gearbox.
@@Chris-yk5rm Hi Chris, great question. We've actually already built a trunnion table for our DIY CNC machine, which sits right on the flat table, enabling the 4th and 5th axis. The backlash is indeed an issue still which I have not come up with a solution for yet. Great suggestion, and glad you enjoy our content 😊. Cheers, Dave
Love how open this guy is with the shortcomings of the design. very humble, I respect that.
Thanks Astro, I appreciate it. Cheers, Dave
I am really impressed with your toy. I believe the current and feed rate are too high. I do not know your power supply frequency but if it is like straight DC voltage you probably want to back it way off to not exceed 1/4 amp (maybe 30-40 volts, this has some connection to the frequency of the power supply) and then drive forward really slowly. Our first machine was less than 30 mm per hour in 25 mm material. I believe the frequency was about 2000 HZ and the machine had a vibrator on the wire to help keep the wire from shorting. This was before adaptive control was available so you have to keep the feedrate below the shorting speed. The wire should never touch the workpiece but you should try to maintain a spark gap of about .025-.050 mm. Additionally if the wire is not held tightly it will bow, which will be evident when all your sharpe corners look washed out.... You can get other weirdness. S-shaped walls. Almost all the current machines are capable of backing up if they hit the workpiece and short out, turn power off, and will not reapply power until the short clears. (most macines have a short circuit sensor for that. They first thing I would suggest is to go to a v-style groove for your guides. Something hard and permanent would be preferred but you could cut a v-groove in a carbide pin. Or if you really clever maybe a carbide cutting tool. Ultimatley, something non-conductive and wear-resistant would be you best bet. Using your roller and small hole you are also; going to EDM your bearing, eventually it will not run round, and the small hole will eventually enlarge especially after soemtime in the water as the metal parts become part of the curcuit. Aslo as an asiide the small rollers are really hard on the wire. Most of the production machines don't have small diameter tight rollers because it puts a bend into teh wire. If you unroll five feet of wire it should go almost straight to the floor without any cbvious stress. If the wire tends to lean in a direction like it has been on the roll too long it can cause the s-shape along the wall of your cutting path. If you can finally circulate water and can pump it coaxially along the wire it will help flush the cut slot (but only after you have entered the material) and you can pick up a small amout of speed. Good luck.
There is some awesome information here. Thanks so much for taking the time to write such a detailed response, I really appreciate it. I will consider all of this while designing our second version of the machine.
Ps. feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
Hey just for fun the very first wire EDMS were attachment's to regular sinker EDM's. They ran in oil, they were more forgiving to poor electrical conditions. The gap was smaller and voltages were higher). You might turn your assembly 90 degrees with the wire facing downwards and see if you move it in .0025mm (that's .0001" right?) straight down with across your workpiece and see if it cuts, which might be easier than shelling out for pumps, filters and other stuff power supplies, etc.
Thanks for the detailed comment. Ill give it a try 😃
I wonder, if the power supplies for wire EDM and die-based EDM should be completely different.. I'm pulling this out of my ... but it seems that for die-based EDM high current is ok, but for wire EDM current and feed speed should be really low, while the voltage is much higher. Like barely touching surface to create a bunch of tiny sparks, instead of completely rubbing against the surface.
P.S. Props for sweet cad skills. Good CAD skills feels like super power (I'm late to the party, but I'm learning this now).
Thank you, I really appreciate your comment 😃. I'll look into this as you suggested and hopefully make a mention of it in my subsequent video. Thanks again 😊
@@ActionBOX correct me if I’m wrong but I thought the wire never makes contact with the part, just the arc. Maybe run the wire at a faster rate and your xy movements slower until you find the Goldilocks zone.
@@gordon6029 that will very likely help, gauging by how popular this response is. Thanks for your input 😃
Great job, subscribed!
Welcome aboard 😃. Great channel by the way
Excellent Build. To avoid wire breaking try using arc voltage/current as feedback to determine axis feed rate instead of a fixed feedrate .Similar to THC on plasma.
Thanks for the comment, and good point. I’ll definitely keep that in mind. Hope you enjoyed the video! 😄
I once trainee at the Japanese renown wirecut edm factory( Its the compulsory in faculty final semester). I noticed in wiring diagram ,the submersible solution also had ion concentration control.
By the way , I see some China adhoc CNC wirecut the workpiece in the air. No need submersible.
A lttile bit more , I think to electrical the wire, remind me of the sparkling circuit might have close loop control in watching the spark behavior. I mean they can control wire current change in millisecond by feedback sensor on how to movement not to make wire touch to the workpiece.
Thanks for the details 😊. Much appreciated
So I’m watching this video thinking you need to have a sinker Edm machine and bam! You made one of those already. Both are on my list but I do need hellish accuracy so don’t know if I can do it diy. That said, for the cost of these systems, if you can work within their tolerances then this is golden!
HAHAHAH, I love this comment. Its doubtful that we can compete with professional industrial grade machines, but It sure does the trick at the cost of some accuracy.
Ps. If you enjoyed this video, feel free to help us out by sharing with friends who may be interested. We are trying to drive up our views to get more exposure 😊
Cheers,
Dave
This level of work deserves way more Subs!
Thanks Anthony 😅. Your support is much appreciated 😊
Absolutely great project! I really like it.
I think your problem with the wire snapping is mostly related to your spark generator. Just from the sound of the sparks I assume that you have very high peak currents (>50A) and probably no short circuit detection to switch off the current in case of a short (touching workpiece or simply having a conducive particle in spark gap). This might be acceptable for a sinker EDM since it 'only' increases your electrode wear, but for a wire EDM it leads to an immediate breakage of the wire. I built an electronic EDM spark generator some time ago (which I so far only used in an prototype sinker application). The sparks sound much less aggressive and more uniform if you have a proper short circuit detection and current control. If you have the chance take a measurement with an oscilloscope either with a high frequency current clamp or carefully insert a shunt resistor in the circuit to visualize the current (but dont blow up your system/scope).
Another thing is that you need a feedback loop for your feed which is based on the spark gap (breakdown voltage or statistics between 'open circit'/'good sparks'/'short circuits'). Not having that, your feed rate of 1mm/min seems to be very aggressive in my opinion.
Also I feel that your wire speed is very low compared to professional machines. Seeing how fast industrial machines feed the wire in combination with the cost of those spools was the main reason why I never continued to build a wire EDM even though that would be a very useful tool.
Keep up your content!
Wow, this comment summarized all our issues. Thanks for taking the time to type this all out. I definitely believe you re 100% correct after having read other comments. I will take this into account when building our second version, although its seems like there is lots of work to do before this is comparable to a real industrial grade machine. I have never seen these in person, and Ive never actually looked at videos of Wire EDM on RUclips so I probably should have started there 😆.
Thanks again for an awesome response.
Ps. If you enjoyed this video, feel free to help us out by sharing with friends who may be interested. We are trying to drive up our views to get more exposure 😊
Cheers,
Dave
I agree, one of the biggest issus should be power source and therefor that the sparks are way to violent, coupled with no "spark gap" control loop. THB i have only been exposed to EDM as "sinker EDM" in University for like 1-2h in a practical workshop(?), but the sparks back there were way way softer and smaller. Also they really pushed that spark gap control onto us as a important part, including that "sinker EDM" seems to have "pull backs" to flush the "hole".
clicked for cool edm music machine and invested in 3d printing great video!
😂 glad you enjoyed the video!
I would love more in depth videos like this! Keep up the good work guys😃
Thank you Jonas. Much appreciated.
Ps. Feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
I wanted to build one a few years ago. I learned some place that industrial machines use pulsed current using PWM. I experimented with low voltage ac and a brass strip mounted in a hack saw and a bucket of water. By putting an induction coil with air core in series with ac transformer supply the spark happens on contact release rather than contact touch. The induction coil I used was the high voltage secondary coil removed from a MOT (had to cut transformer open to get it). Was able to saw into aluminum fairly easy. The ac voltage I was using was low like about 24 volts. Some guys use diesel in the tank because its cheep 😅.
We all start somewhere 😃. Thanks for sharing. Dave
Yeah Filter the water down to about 10 micron (You can use a pool sand filter, but use 10 micron cotton instead of sand), also maybe pump the water through deionization beads to keep the conductivity of the water more stable (A little table salt helps, but just a pinch. Also there is a common saying in Wire EDM "Flushing, flushing and flushing", its important to flush away that residue quickly, flushing is key.
Thanks a ton, I will use your recommendations while building the V2. Cheers, Dave
AAAAAHHHH!!!!! MAKERS THAT LET YOU ACTUALLY MAKE ALONG WITH THEM!? Not only is this a project i wanted to do...i need it... sooo liked and subbed!
Yay! We’re glad the video was at a good pace to follow along. Feel free to share photos with us if you do end up building it - we’d love to see it! Thanks for supporting our channel! 😃 Cheers, Alan
Well your on the right track. Coming from an industrial standpoint, your wire feed is way too slow, your cut feed is too fast and your flushing pressure is way too low, your not controlling the resistance of the DI fluid.
Also oddly enough round stock is actually pretty hard to cut in an edm, relatively speaking. Since it is not a constant thickness they resistance from the part is constantly changing and thus your settings technically should be as well. Add to that the DI flow around the round stock which will actually divert the DI Fluid away from the cut and thus make it less effective. Also depending on the grade of stainless it may have a rather high resistance rating which will make it more difficult to cut, something like brass or aluminum with a lower resistance rating may be a better start point.
Your excessive feed rate is obvious from the deflection of the wire. As weird as it sounds the wire should never actually touch the part during the cut. So seeing it bow before it breaks is an obvious sign. There should always be a "spark gap" between the wire and the detail, the DI fluid will transfer the current to the part. Which brings up another point, your water may be too pure and not passing enough current. Industrial machines have sensors and use filtration and DI resins to keep the resistance level of the water very narrow window.
Ignoring all the potential resistance changes from passing the current thru a bearing, the contacts are sacrificial. Unless you want to be replaceing the bearing all the time you really should make the contacts a separate piece. A cheap 6mm carbide endmill trapped in an aluminum block a few mils in the wire path would be an easy cheap option. Also it is good to have them as close to the wire guides as possible as the further they are apart the more resistance you have in the wire, and industrial machines will generally have them in the low pressure flushing to increase voltage transfer and keep the wire and contact cool. May want to consider getting some coated wire. The wire cost more and will cause more wear on the system but it works better in poor cut conditions, which is pretty much the best you will see on a DIY machine, so it will give you a bit more wiggle room on dialing in your settings.
Excellent video. I own 2 Agie wire machines and you have it nailed. Awesome work can’t wait to see your next iteration.
Thanks Mark! We will for sure be releasing another video with our next iteration in the future. Meanwhile, stay tuned for some really cool upcoming machine videos. Cheers, Alan
I guess those stepper will be reused in other projects, but they are way overkill for this project. The smallest Nema17 would work as good. Just in case someone wants to build this project. They can get away with 30$ instead of 300$. I like the design, can't wait to see it working!
You are 100% correct. Thanks for your support and your comment. Cheers, Dave
Flushing is key for edm. Get your wire guides as close to the material as possible. 0.2mm or closer. And maybe higher pressure on the nozzles. Every wire edm I ever ran varied the feedrate into the part to match how fast it was actually cutting, especially as it's entering the part starting the cut.
Than you, I will add that to my to do list in the V2. Cheers
Can't wait to see you guys get this licked, looking forward to the part 2! Any plans for a DIY laser sintering metal powder 3d printer?
That would be a cool project 😎
Or a cnc controlled mig welder "3D" printer
@@WeighedWilson What shielding gas would you use? MIG welding isn’t going to work with oxygen around.
@@Jkirk3279 You could just have a sealed chamber and purge the lot with Argon- arduinos don't need to breathe. That or use a multi pass compatible flux core
The tension sensor are actually "Load Cell" They detect pressure being exerted on the cell when you press on the end. the load cell is located in the open area in the middle where all that white stuff is. They usually look like a bare IC but are not made like an IC. They are usually resistive elements in a zig zag pattern and change value when stressed.
Ooo, pretty cool, I like the concept. Thanks for sharing 😃
Awesome! As your power supply gets more sophisticated you’re going to need to shield those strain gauges and their wiring. Also using set screws to transfer torque to cylindrical shafts is bad practice. Keep up the great work!
Hi Chris, the set screws are a standard in all CNC machine parts I have ever purchased. Can you please elaborate on that point? Cheers.
I can't speak for Chris-bg8mk but I would think that for torque transfer a woodruff key or a taper to shaft and hub/wheel is preferred over a set screw.
I'm curious whether and how quickly this comment, 2 years after, will draw (your) attention 😀.
Apart from all that I'm impressed by both your design and the video production. The positivity and the useful contributions in the comments is great too.
ETA: subscribed!
Great work 👍
Thank you! Cool channel 😄
Just found this channel, great content. Interesting builds, keep them coming.
Thanks Darrel, much appreciated and welcome to the actionBOX. Cheers, dave
Interesting project. Another thing to consider is the spool collecting the wire may build up in the center only. Make a feeder that travels back and forth like a baitcaster fishing reel 👌
Cool idea! Ill add that to my list of upgrades for V2. Thanks for watching 😃
What ever happened with this project is there a part 2
Indeed.
I have been wanting to build one! Can't wait to start!
Awesome, feel free to reach out if you need help.
Idk much about this technology but I do think that the wire isn't supposed to contact the workpiece which is why the need for the dialect water, it acts as an insulator while at the same time keeps the wire cool, so I don't know what kind of power supply you would need but it would need to be high enough volts to be able to create a spark gap much like how plasma cutters work. if you ever used one the first thing they tell you is don't let the tip touch the workpiece. Might be a good place to start. Seems like a cool project I would love to see it work for us diy guys.
Ok so I find this technology really fascinating and found some interesting things out. So there is this guy who made one of these over at instucables, and for his power supply he uses incoming 110 ac, runs it through a 200w bridge rectifier, uses a 200w incadcent light bulb to limit the current, and looks like it has a couple capacitors to. The circuitry looks relatively simple, and he has posted videos of it working. The AC is converted into a pulsating DC voltage 60hz or 60 times a second, idk if this would work or not because I thought it had to be between 300 to 1000 a second. But maybe still does work, and is just slower. I was also right about the wire not supposed to touch the work piece you may have to touch it to start the arc but then pull it away much like Tig welding. But check his setup out because it seems relatively simple, and Id like to try this myself because this could open up a world of machining not normally accessable by us hobbyists.
www.instructables.com/EDM-Electrical-Discharge-Machining/
Great project, & well presented!
Applied Science did a video 2 years ago on DIY EDM drilling ( ruclips.net/video/rpHYBz7ToII/видео.html ). It covers many of the issues documented in comments including the requirements for current control (BAXEDM), motion control with backup (DynaMotion), water filtration / conductivity, etc. Some of the issues he covers don't apply (such as rotating the 'cutting tool' (/drill tube) to ensure uniform erosion; with wire EDM you are constantly replacing the 'cutting tool' with new wire), but the core concepts are still the same.
Very interesting! Thanks for sharing 😃 I’ll have a look at that video. Glad you enjoyed our project!
Your feed rollers seem like they'll work fine, but I think a capstan roller might work even better with less required machining, fewer moving parts, and no cantilevered loading.
After 2 years I am still waiting for part two in which everything will be fine tuned and working.
Its great to see a Canadian person doing this kind of DIY and education! Keep up the great work! Especially sharing your parts and designs! I hope to one day make all of these cool machines you have! Thank you!
Thank you very much! 😃. I appreciate your support. Cheers, Dave
A scenario of total engineering process. Urgently subscribed
Thanks a bunch. Feel free to help us out and share our video with a friend who may be interested. We are really trying to grow our views 😊. Cheers, Dave
I maybe missing something... but is there a feedback system for the arc?
If the wire is making direct contact to the cut material ( resistance between them near zero) the wire must brake, mechanically and due to excessive current.
Shouldn’t the power be disconnected, the wire and cut material be separated from the wire (high resistance), then gapped by an additional predetermined amount before the arc (power) and cutting is reestablished?
Perhaps this is already being done, but I didn’t see it mentioned in the video.
Hi, great comment, you are absolutely correct that this needs to be done, and indeed it is not in this V1. We are working on the V2 which should tackle these issues. Cheers. Dave.
Can't wait to see this working!!!
Me too!! Glad you like it 😃
I've hit the thumbs up 3 times for good mesures!
We appreciate viewers like you 🙏 thank you so much! Stay tuned for some really cool content coming soon.
@@ActionBOXSide note: I'm always surprise when a great content creator, like you, are saying "thank you for watching"...
I meant.. YOU put hours and hours of work to show us a FREEEEEE content. And you thanks us in the process!
@Alexandre Valiquette hahah, well we appreciate you helping us grow. We really want to hit 100k subscribers by 2023 so we appreciate every viewer like yourself. Feel free to help us out further and share the video with your friends who may be interested. Thanks so much again for your kind comment. I look forward to your comments in our subsequent videos. Cheers, Dave
Excellent work guys!
Thanks! 😃
Cutting with WEDM is a delicate balance between a dead short and a open circuit. The internal diagnostics of the cutting condition is everything. Ask yourself why Sodick puts linear motors on all their wire machines. Quicker response times during cuts, equates to better performance cuts. Be sure of ONE thing. If you can't get your path movement to "back up" slightly, then move forward again (per your cutting conditions requirements), you won't end up with much. I'll say it another way. What CNC machine do you know "backs up" slightly in its own path, then returns forward in its path. Answer: Only WEDM's. This is done to protect the wire from breakage. Good Luck. Lee
Thanks Lee.
First time watching your videos. Amazing content. Subscribed
very nice. There's a different type of wire EDM called a Moly EDM that uses molybdenum wire instead of brass. The wire is more durable and cuts considerably faster on harder materials. It doesn't submerge either but uses an oil instead directly on the wire. It also uses a single wire spool that also takes up the used wire which is then reversed. The same length of wire can cut for hours. There's a huge market for that type of EDM but most people don't know about it. It saves days of cut time over brass.
This is awesome, I have not heard of it. Thank you for the tip, and I hope you liked the video 😊. Feel free to help us out by sharing with a friend who may be interested. We are trying to increase our views and channel recognition. Thanks in advance 😊. Dave
You should detect when the arc happens and then rectract, I'd say it's a very visible short that is happening there. From what I've read, you need to supply the same current, if the wire touches the workpiece it will then short and of course.... blow up! :)
Thanks for the input, we are indeed working on this for our V2. I hope you are right, and I hope it works 😃. Thanks again.
I ran a Sodick A280L for years. Mostly 0.010"wire with diamond wire feeds. I cut stock up to 7 inches thick with that machine. The filter and deionizing systems were substantial. Also, there needs to be feedback from the spark to the wire power system as the condition of the kerf is a continuous state of change what with variation in the spark line and excellence of flushing, which was powerful - lots of water at quite high pressure - I would guess 40 to 60 psi or more. Nice project. Visit a working wire and take some notes.
Thanks Ron. Admittedly I have never seen a wire machine while attempting this so I do appreciate all of the feedback. I will take all of these notes into account when building the V2. Cheers, Dave
@@ActionBOX Maybe two more things. Try cutting aluminum foil. It will confirm most of your operating ideas and little flushing will be need. You will need an auto fee mechanism tied to the wire condition circuit. This will slow the cut if conditions are tenuous at the wire and speed it up when conditions are good. These machines are quite complex I think with lots of feedback from the wire circuit going back to the motion control module. It will test your engineering skills quite seriously but you seem willing to throw yourself into the fray. Best of luck.
@@ronrabie9638thanks a bunch, this will help 😃
Great video, love the soldering iron for inserting the inserts
Thanks, I appreciate your support 😃
An incandescent light bulb (current limiter) in series works great for over current protection. The bulb will prevent burnout and is also an indicator for a short. Couple that with a sensor and custom code to reverse the axis direction if the light turns on.
That’s a great idea hahaha. I’ll definitely be doing that. Cheers
Interesting finding you here. Your parts were instrumental in my airsoft grenade launcher build.
@@DaveSmith-cp5kj I’m not sure if that’s a joke or not hahaha. Are you referring to my nerf video? Or my Mandalorian video?
@@ActionBOX Oh sorry it was a reply to Palmers Pursuit. He owns a shop for obscure pneumatic parts for paintballers.
@@DaveSmith-cp5kj gotcha haha. My mistake. I made a rocket fuel powered nerf bullet so I thought you were referring to that video. Cheers.
Love the channel boys. Sending love from Asheville, NC!
Thanks Dax-Terry. The love has been received 😊. We appreciate your support. Cheers, Dave
this is really great. thanks for publishing it i've been wanting a wire edm but no time to design. you rock.
Thank you. I appreciate that a lot. I will continue to work on V2 and look forward to your comment in the future 😉. Thanks again for your support.
Something new and DIY.
I'm in!
The road to success is littered with failure.
Good tutorial, I'm inspired eh!
Hahahah, your comment made my day. I hope you got to share with someone else who may find this interesting ey 😜. Glad you enjoyed the video, and thanks for helping us grow this channel.
What do your tension readings look like leading up to the break? Without gap voltage feedback adjusting your feed, you might be making contact with the sample. With constant feed I image your tension would start to increase as you make contact and start to push against the sample with the wire.
That as a good point. Unfortunately we did not keep this data from our tests so we do not know. We will look out for this in the V2
Cheers,
Dave
One word: Awesome!
Hahah. Thanks a lot 😊
Is that nominal guide hole to make sure the wire diameter remains constant also like an extruder? And surely reel/wire speed increases cutting area, no idea typical rpms.
It’s actually to guide the wire with less variation. Perhaps straighten it out as well. Great question.
Uh.............. MAZING!!!!!! Subscribed.
Hahaha, thanks A J, Much appreciated. Feel free to help us out on our mission to boost our channel, and share with any of your friends who may be interested. Cheers, Dave.
Man brilliant work your attention to detail and quality of work is art for sure. Thank you for sharing this my friend !!!
Thanks for your support 😃. I really appreciate it. Cheers, Dave
Interesting topic. It went on and on and on.😀
I have no interest in making an EDM but I love your enthusiasm and determination.......so I subbed and liked.👍
We appreciate you subscribing, it really helps our channel grow! 🙏 Glad you found the video content entertaining 😃
very nice design long time i wish to have similar this machine.
thanks a lot
No problem. Thanks for watching 😊
Considering the value of information and experience i get from this channel you gave me no choice but to ... subscribe immediately. Looking forward to learn a lot. Thank you so much !
We’re so glad, and that is our goal! Stay tuned for some really cool content coming soon!
Spark control ... needed. The nozzles should have the conductor tips.. you only charge the wire near the work area. The take up wheel could be rubberized .. running wire between 2.. when running wire you don't need to re spool. Most machines just collect in a bin. I've run and programmed Hitachi 203r and also a Makino wire edm. These have independent top and bottom wire guides to help align and square to work area..
What power supply are you using? I suspect that probing voltage and current using an oscilloscope will reveal the problem
You are correct. Our next iteration will be using a proper power supply. Cheers
Great job!
Perhaps the wire break is due to the low melting point of brass compared to steel. Try cutting aluminum.
Great point. We did try that though with no luck 😞. Thank you for your comment 😃
What about crossing the wires between the brake rollers? instead of = looking wires, use x. zero is rollers, x is wire, 0x0? or would that be too much tension?
That would be far too much tension and the wire will snap, but great question. And thinking 😊
Few thoughts from a wire edm noob: 1, that 4 groove back and forth spooling approach seems like it'd be tedious to load. 2, rather than a pid tension control scheme you can simply use 2 servos to spool and unspool at the same rate and use a tensioner pulley to achieve a constant tension
This also eliminates the brake and the servo. Just a spring and a pulley, tho a servo there gives you computer control of the tension
Thanks for this, this is indeed an update that we are making in the V2. Thank you for your insight and great suggestion. Cheers, Dave