Видео

Aaah, moest er even over nadenken wat je bedoelde, die heb ik bij SMT montagetechnik vandaan.

@@gremlin60 machine is going through some changes now, need to make a new movie soon

I had a round one on my previous machine, (water cooled) they are more complicated to install, because they need more parts to mount them. the round spindles also use smaller sized bearings than the square spindles potentially making them less stable. I personally prefer the square ones over the round ones.

@@berendlucasvanderweide Thank you for the reply. Where can I buy the same JMC square spindle you are using? I am in USA. Thanks.

@@sto2779 the driving motors on xyz are the jmc 180w versions. The square spindle is from Aliexpress. I'm unsure if there is a local vendor in the us selling these motors. Mine comes from a company called Rattm.

@@berendlucasvanderweide Thanks for the reply. That makes sense. You like the JMC for motion control? What kind of controller board are you using?

@@sto2779 I've had the first version of these jmc motors (V0.5) they had no auto tune, lacked anti resonance filters and had no inertia settings other than plain PID tuning. I've changed them to OMC stepperonline motors, (also nema23 180w. they probably come from the same factory but have better control and software with the former mentioned features (filter, inertia). I like them a lot for their money. I'm using an edingcnc 760 controller board now. (Dutch company).

It's a rattm 2.2 kw square spindle with ceramic bearings and a Huan Yang vfd. It served me good. I bought one as spare but never needed it. I'm currently using an atc spindle. Need to make an update video..

@@berendlucasvanderweide okay thanks👌🏻

Hi, i bought the connectors for the aluminium profiles at smt-montagetechnik.

Feel free to ask!

Thank you for your positive comment! I need to make a new movie soon on the upgrades it got during the time of use.. thank you for subscribing!

Nice to hear! Thank you! I am currently (suffering a broken foot) upgrading the machine. ATC spindle, upgraded the machine with a camera for positioning, re-doing the cabinet/pneumatics. I'm using a different controller now. If it's finished or nearing completion i'll make sure to get an updated video again.

Thank you for your kind words Nicolas! I need to make another movie if the entire machine is done again. Also replaced the JMC servomotors for stepperonline (OMC) 180W servomotors.

​@@berendlucasvanderweide These are indeed some interesting news. Belt errors should be pretty small though, interesting. I do have some JMC non-integrated servos I'm planning to use for my design however aiming at lower accuracy, bigger space (1250x1200mm²).. Will reconsider belts then.. Though I'm not sure this is managable. I'd love to see any updat from your machine. Concerning the mechanics with the profile: you're still happy with the al profiles you've used? Steel linke the printnc is something I'm constantly having in mind as well.

@@dernicolas6281 The problem is that the beams aren't straight due to their production process. They are stretched to specification, which leads to tension in the beam and a deformation of the squareness. I've machined the reference flats on my machine on my larger machine. Within 0,02 straightness. But combining metal (lineair rails) and the aluminium profiles create a bi-metal with an interesting difference in length when the temperature changes. The belts themselves don't introduce the biggest error, but the runout of the holes in the pulleys do. Jmc makes great products in my opinion, but these omc I now have also have two filters for resonance dampening, which enable a stiffer setup without resonances. (I was using V5 jmc's, that simple didn't had this feature). They worked fine though, but I wanted to test the OMC's.

​@@berendlucasvanderweide yea - I'm aware of that. No higher end industrial machine is mixing materials and even non mixing they're keeping the temperatures constant. My plan is to mainly process wood anyway. ;) Resonances in long Al parts need to be treated then? Ok. Makes sense I guess... Thanks for the detailed answer! Will follow up whatever you're releasing on this chanel.

@@dernicolas6281 Yes, exactly. Not saying it can't be accurate, it just takes time to get there. (adjusting can take a lot of time, consider it a rainy sunday afternoon task.. ) :). When designing a machine take the posibility for adjustment into account, that you don't have to tear the machine down to reach a screw. A few tactical holes and flanged guides on the lineair rails can help a lot. When machining stiffer material, the machine is most stiff in the corners, where it sits as close to the siderails as possible. Aluminium is very convenient if you don't have the equipment to weld etc. That was my main reason i went that route. Just buy, and screw. Welded steel also requires tempering. I'm using edingcnc, and it can do ballscrew mapping (currently using preloaded C5 ballscrews) and squaring it to perfection is also possible with this software/controller. Te be honest, i choose the controller also based on that, but haven't been using it. It's good enough as it is, but you could make temperature depended profiles if you have the urge to go the extra mile.. I am using an 80160 'heavy' profile for the x axis and the Y, i haven't bothered filling it with concrete or polymer concrete, i don't know if it will add anything. I can machine birch at 25mm depth with 2,5mm overstep, with a 12mm cutter with ease, without chattering. I thought i have a little clip of that as well, have too look it up..

Hi, i am currently using a different spindle (total weight of the moving Z axis is now around 30kg, the mass itself is not a large factor when calculating the inertia. (since it moves at a relative slow speed) Rotational weight at the full RPM of the motor is the biggest factor for your inertia. I have just upgraded this machine to an ISO30 spindle and bought 3pcs of the OMC stepperonline 180W servo's for X and Y.The Z is still an issue dropping down when unpowered. I have to make a brake for it. The Yaskawa's sigma 7's are premium. Their auto tune works miracles, also possible to tune 2 servo's in tandem. their extremely high resolution and high frequentie feedback/control loop makes their real-time tuning very usable. For our application (CNC) the real time tuning of the OMC is also appropriate. Delta makes nice affordable servos too. The OMC also include a notch filter that works very well. (but you have to find the resonance yourself). Just do some inertia calculations and determine what servo you need. The 180W versions are potent for sure. direct drive with light aluminum couplers are the best for the innertia. 25mm ballscrews with 40T HTD5 15mm pulleys in steel are bad for the inertia. 63M/min sounds like a rack/pinion? These cheap servo's are great at what they do for the step/dir situations, but in theory there are better solutions where the controller and servo are connected in a way they are aware of eachother (the system knows when to anticipate for a fast inaccurate G0 move, or when to increase the P to a maximum stiffness when the tolerances in cutting are getting tight). But that being said; this ancient and incorrect way of doing it with cheap servo's is working great for me. I have the OMC's inline auto-tune in position mode instead of CNC mode, which make them run smoother (though a larger tracking error of commanded position) but this can be compensated for in the CAM strategy with ease.. (if you are not on the hunt for the last 0,001mm, this method is fine). It makes the machine run smoother at higher acceleration. (so your setup shouldn't be set too tight for the best possible positioning accuracy, this makes the entire machine shake and vibrate). It is not the best way in theory, but it works the best in my case if this makes sense. Here is a video of the software with the OMC motor; ruclips.net/video/Uxx4bMXpRKw/видео.html 3000rpm, 15msec/krpm acceleration / deacceleration time and a very small short distance test; ruclips.net/video/ZuuerD23QOk/видео.html

@@berendlucasvanderweide I dont know why I miss the notification for this comment! Im sorry for that! Thank you greatly for such an elaborated and comprehensive reply which I dont usually get online!. Life finally get less busy so I am building my CNC as well, I am getting to the stage of really buy the motor. My partner says 200w wont be enough to lift a 30kg moving Z the speed we need (0.7g), but seeing your vids (particularly the "very small short distance test"), I feel more confident in the motor. Lets see who will buy other beer haha Thanks for other advice, will definitely look into those, especially the motor inertia

@@thanhnhanhuynhnguyen3652 no worries, it happens to me from time to time too. At these relative slow speeds The moving mass is almost non existing against inertia of the parts rotating at a higher rpm. Especially large pulleys affect the moment of inertia. I don't know what pitch leadscrew is used for z and what diameter of ballscrew. I'm using 1605 direct drive on the z and that can move fast. I don't tune the servos too tight for minimum tracking error. But suitable for my cnc application.

Hi Seth; Yes, still using them! but i'm using the V5's, if you bought them recently you'll probably have the V6's which tune very different, they have some rudimentary version of autotune, since i have the V5's (which required quite some tuning) the V6 is more user friendly. I've recently upgraded the spindle so needed to move the servo from belt to direct drive (DD is way more silent than belt) on the Y axis i'm using 10mm pitch ballscrews with a 2:1 reduction and on the X/Y it's a 1:1 to a 5mm pitch ballscrew. There are users that use 10mm pitch in a 1:1 ratio. Just keep in mind that you always need to tune a servo to match the inertia. For our application (dynamic load) a set inertia number with a known stiffness and soms sort of autotune should be the best option (since the load tends to differ most of the time).

@@berendlucasvanderweide Awesome! I'm glad the new ones have some level of auto tune as tuning was my biggest concern. It is also great that they have enough torque to drive a machine with a direct drive. Thank you for the helpful information and fast reply!

@@SethAltobelli i don't know what you want to push around with the motors, but they are pretty strong in comparison to my previously used 4,5Nm steppers. (I dare to claim that they are way more powerful) when setup correctly they move the machine around quicker and more silent.

@@berendlucasvanderweide the machine is largely aluminum extrusion similar to your machine. I was originally planning on 3Nm NEMA 23 stepper motors before I discovered these. It seems these will be great.

Heya! missed this one; That's a 10mm cutter, 26mm flute length, i had a bunch of custom cut cutters to do some testing in plywood, i used to machine a lot of it and need a good edge finish. The machine is performing well, Perhaps i could upgrade the current liquid ring vacuum pump to a claw pump. But this one is still running, so i'll probably wait till this one fails. (makes the most sense). I did made some upgrades to the pump last year to give it a better performance throughout the day.

This one is used for the plywood. Had some fun today at a friends workshop, adjusted a few things on his machine (he bought a simular one) and the machine was able to rapid at 90m/min. It's scary to see such a relative large machine moving this fast.

@@berendlucasvanderweide We once had a client where we trained them up on a large machine - and when we pressed home all - the client ran out the door scared it was going to crash into him :)

I'm currently replacing the spindle with an iso30 atc.

I have stupidly attached the same 2.2kw spindle to a chinese 3040 cnc lol I'd like atc on my AMA25LV

@@dougyt261 that's a Frankenstein creation! My machine looks out of proportion with the ATC, it's about a factor 5 bigger in volume than the 2.2, so it needs some 'getting used to' time.

Still going strong! I bought one extra as spare, that's still on the shelve. I like that spindle a lot for my work. Upgraded it with a high flow fan, running at 1/2 the speed till 35 degrees and after that it kicks in at 100 percent. If you are planning on drilling, get some carbide drills, the spindle needs more RPM to get into a more ideal torque range. HSS drills will burn on these higher RPM's.

Yes. Many thanks bro! Can you give me link to order it? I search and get many item look same it but price different 😅

@@pham7878 This is it, but they changed the design of the connectorbox, it's no longer a plug, but has a junction box. item 2038803546 on Alie

@@berendlucasvanderweide many thanks bro! 👍👍

I would still recommend them, and it would be my choice if i would need to buy new ones for a machine like this. They are by no means high end servo's. They don't have a very advanced encoder, they don't support an absolute position, and their integrated controller might an issue if you are using flood coolant. The JMC tuning software and auto tune is not as advanced as the more expensive ones. Their documentation is relative limited. The current ones (V6 over V5 that i have) have a more advanced tuning and even have some notch filters if you need those. They are better/smoother than any (closed loop) stepper combo i've came accross. That said; I would never use steppers on a machine like this anymore with these available. even though they have their shortcomings; looking at how smooth i was able to tune them and how happy i am with them; i don't care if there are better servo's available, for this application they are perfect. On my larger machine are Yaskawa servo's, they are more advanced, but that machine is just heavy and for fast 3d work, this machine is my preferred machine. How heavy is your gantry and z-axis? Don't underestimate the power of the servo's. Perhaps the 200W is more than enough.

@@berendlucasvanderweide thanks for the reply! I was first looking for 200W versions but 400W is almost the same price. Gantry might weight about 60-70kg plus z-axis assembly about 30kg. I was thinking to put motor with brake for Z

@@MikkoKui That brake on the Z axis might be a good choice, i am using preloaded ballscrews and preloaded lineair rails, it doesn't move, unless force i applied after turning it off. I don't think my machine is that lighter compared to yours, my gantry is about 50kg, It all adds up fast. I wouldn't use 400W. On my large machine i'm using 850W servo's, and that gantry is about 650kg, and it's using a 9kw spindle, it can cut 25mm MDF in 1 pas with 16mm cutter at 25m/min. Acceleration on that machine is 0.4G. On my little machine i cut 20mm birch plywood with 2 layers of HPL at 8m/min at full depth running the 2.2kw spindle. (on a single phase 230vac 16A). Servo's are pretty potent, they can push 3x the rated torque for 10 seconds from 0-3000rpm, that's a lot of power when you reduce that motion to 5mm per revolution. (depending on the pitch/reduction of your ballsrew/drivetrain). If it feels more comfortable to buy 400W, go ahead, but it is probably overkill. even though it might just be a bit more expensive but the 200W is much shorter. (lower inertia too). So it requires less power to accelerate quickly. And tuning it might be more predictable with a 'fit for purpose' setup.

@@berendlucasvanderweide I have 1605 screws and going to use them 1:1. Maybe I need still to consider 200W motors, or at least Y axis could have 2pcs of 200w

​@@MikkoKui What are you planning on machining? What is the power of your spindle and is the structural base of the machine steel or aluminium? Willing to share some details of the design?

It's a ratm 2.2kw square spindle, air cooled with ceramic bearings. Don't search too long for ceramic bearings, good steel bearings are great too. There currently are two models available. One with a plug and one with a junction box. I prefer the model with the plug.

1605's. I don't know about the rest of the machine, but 100-180w is normally more than enough to turn them with authority.

1605's on x/z and 1610's on Y

@@berendlucasvanderweide which type do you think is best for torque and accuracy? I did the calculations and with the 5mm I would reach 15000mm / m of acceleration.

@@andreabonucci3834 speed you mean? 15000 is plenty fast, but I'm not pushing it that far. 3000rpm is a lot for a ballscrew and they need to be perfectly straight otherwise you'll introduce unwanted resonances/vibrations. (Especially above 700mm length). The thicker the ballscrew the higher the inertia and the more power is required for acceleration. I'm using p1 (preload 1, standard is 0) and c5 accuracy. The are perfectly straight. I'm using 2:1 reduction on the 1610's. The jmc motors are specified at 3000rpm, but can do 4000rpm

@@berendlucasvanderweide yes and a lot, but I'm not going to reach it. however I intend to use 20mm screws but I don't know if it is better to have a 10 or 5 pitch for precision and motor effort. I think that with a 10 pitch the engines try harder, but having never tried them I have no idea

@@andreabonucci3834 I don't know what motors you are planning on using. I'm using a 2:1 reduction, but the 180W motors also worked fine with a 1:1 on the 1610's.

I made some additional changes to it, it works perfect for my work, but i don't think i'll make another machine from aluminium profiles again. It's tedious to get it dialed in, you need flattened/referenced surfaces, as the profiles itself are not perfectly flat due to the production-process. All materials have their flaws and strengths. In terms of ease in use, these profiles score high.

@@berendlucasvanderweide I know your pain. Built mine from 60x60 steel profile and it's not perfect either. 😅🤭 Keep those videos coming please 😁👌🏻

@@pmsilvei Yeh, it's hard to get it fully flat/square. When you use the proper tools to measure, youre eyes will start watering sooner or later.. :) You need to invest serious time to get it dailed in. I've started with a precision straight of 1000mm within 0,02, from there determined the 90 degree angle with a 0,002 precision square. And used the straight again to follow that line. By the time you start measuring the lineair rails on 2 planes, you'll realize when tightening them down, steel is also somewhat like rubber and it starts to twist and bend in all directions you don't want... But in the end, patience, time and brass foil are all that's needed

@@berendlucasvanderweide ahahah I like how you explained it. It's a big challenge, indeed.

Did you get a change to finish your plasma?

Depends. 99 percent of the time it's set to 2500-3000mms2.

@@berendlucasvanderweide wew thats fast ><

Lol

Heya! Turns out to be very handy, upgraded some things on it, using a camera for x/y reference (lowered by a small pneumatic cilinder). Bought parts for a 4th axis. It's fast in x/y/z. It's convenient to have 2 machines for my work. The controller is capable of handling parabolic arcs smoothly, so all and all, i'm very happy with it. To be honest, i don't think i'll ever build another machine from aluminium for several reasons. But i'm still very happy with its smaller footprint and overall performance.

@@berendlucasvanderweide Sounds like maybe its time for a quick video update :)

I posted another video with more details about the machine, have you seen it?

@@berendlucasvanderweide Not yet

@@rafaesposito7796 that is the machine. It's an aluminum portal.

@@berendlucasvanderweide OK thank you very much

@@rafaesposito7796 Feel free to reach out if you have any questions.

PED is unused, ALM is fed by a power-source, i thought i'm using 12V and the output is going to this relay board (highly recommended board, it's very convenient in multiple situations) it can be setup active high, active low, output can be set to NO and NC. Item 32742580606 on Alie. So that ALM output is going to that relay board, i'm using 4 boards, but 1 with the signal input in series works also. I'm using the output of that relays board to go to my PAUSE input on my controller (don't have a dedicated motor ALM input on my controller) I had it connected to the ESTOP on my controller first, but that was horrible. When the ESTOP was active, the ENA signal of the servo's was terminated and the servo's alarm status was reset. (so no trouble-shooting was possible) Hope this helps.

@@berendlucasvanderweide Thanks a lot for the quick reply. So I already have a similar relay board in the enclosure 32970231202. I can not change the H/L activation but that can be changed in the JMC software if needed (at least I have read so). So in my case I would supply 5V to the ALM+ of the JMC with in between a 4K resistor like the manual says. Then the ALM- goes to the relay board in one of the inputs and the output of the relay board in either NC or NO configuration (depends what I then set in mach3) goes to one of the input of the BOB (CNC controller, I use smooth stepper). Then I will set in Mach3 the input of the BOB to stop the machine. Is this right?

They can all go below 8krpm, but they don't have much torque left. Use a sensorless vector VFD to keep some torque.

@@berendlucasvanderweide How much torque is it possible to conserve with this method? I know you can always use a belt drive but it sure would be nice if that wasn't necessary.

@@davynolan182 What do you want do do? drilling over 5mm with optimum feed/speed in aluminium is impossible for instance. Taking it slow might help (aluminium is relative forgiving on HSS cutters) Use carbide drills at higher RPM's to get more torque. These spindles need to make some RPM to get optimum torque. I don't know about your budget, but there are permanent magnet spindles out there that are able to do rigid tapping. These permanent magnet spindles have loads of torque. It's a debate you need figure out yourself. (not meant in a harsh way) My current machine is still using a 24k RPM spindle, it suits my work. But if you want to work with steel, then a high RPM spindle might not be ideal. For aluminium, wood and plastics this is very suitable. I machined some hardened steel on my other machine, with a 4mm 4 flute carbide cutter. It's doable too. there are calculators online that determine what torque you need for a defined diameter cutter in the material of choice. Most manufacturers of spindles can hand you the torque vs rpm plots. Watercooled is more suited for lower RPM's on a sensorless vector VFD than air cooled (unless you make it forced air-cooled)

@@berendlucasvanderweide yeah it’s primarily for steel parts I need mine, I’m currently designing a belt pulley system for my cnc, I couldn’t find spindles that can be reliable at low rpms (in my budget) so I had to bite the bullet and reduce the rpms the old fashioned way. I want to be able to use face cutters so I need pretty low rpm

Just make sure to ground the housing of the spindle properly and connect al the ground points at one star/central point . I'm sure mine has got ceramic bearings but for aluminum steel bearings are perfect.

@@berendlucasvanderweide ✲ understand What about spindle, I doubt motor is in centre of aluminium case. Do you think that spindle mounting alu. surface should be machined for perfect angle before mounting on steel Z axis. Im afraid of heating steel bearings if go high cutting speed for a longer time like you do. Can you tell why do you think steel bearing is better for aluminium milling? Thanks man, you already helped alot, I didnt even knew my spindle is good enough for aluminium milling.

@@eugenega7494 You can not trust the angle of the rotor in the motor. In my case it was OK, but just measure it when it's in place. there are several ways to align it to the machine, but if you are in the hunt for the last 0,005mm, you shouldn't build a machine from aluminium profiles to begin with :) I've used a 12x150 rod in my spindle to align it. Rotate it to determine the runout, move the Z in height while rotating it and take the average number of the dial indicator as a guide for the linearity of the Z axis. On my spindle it only needed side adjustment, so in my case the rotor was fine on the most critical direction. But it's not a ground/reference surface, so you can't trust it. I would prefer good steel bearings over lower grade ceramics. (with ceramic bearings the races are still made from steel. Only the balls are made from ceramic. And ceramic has a larger stability in size over temperature, but nothing to worry about. I used to machine in aluminium with a water cooled 2.2kw with steel bearings (the old style, 3pcs bearings) the round water cooled use smaller bearings to begin with, (which use smaller balls, and can withstand lower loads) but it machined aluminium just fine. You don't need ceramic bearings perse. I didn't say steel was better, I say it would work perfect too. If you look for Rattm 2.2kw air cooled on alie (2038270310) it mentions ceramic bearings.

@@berendlucasvanderweide Yes I found, Its 50$ more exspenisive then mine. I will weld strong construction with tick steel pipes and then give it to a company to machine places for rails and Z axis. I dont know how will they do it but will probably give them spindle also to machine it to fit on Z axis perfectly.

Hi, This machine is using a 9kw air cooled spindle, the other one is using a 2.2kw air cooled. The latter had a fan mounted on the shaft itself, but that was rather loud at higher RPM's. The 9KW uses a dedicated high-flow fan. On the smaller spindle I have now removed the shaft fan and installed a high flow fan (12V) it's by default running at 6V so it's more silent. I have bridged the 6V power supply with 2 parallel (redundant) 38 degree/C switches. So if the machine is being used for a longer period of time and the temperature reaches the 38 degree/C mark, the fan kicks in at full power. You won't hear any difference between the ceramic bearings and proper full steel bearings. When doing cuts in aluminium with 6mm cutter or larger, the noise from that is larger than the rest of the machine.

@@berendlucasvanderweide Oh yeh! I see thanks !

The Ooznest is somewhat of a shapeoko machine? Belt driven? No worries. 1.5Kw is more than capable enough.

@@berendlucasvanderweide No, It's with leadscrews. not sure but it seems a bit stiffer than shapeoko. ok thanks

Yes, that spindle is recommended. So far it's still going strong. Bought one as spare, but that is still unused.

1605 1:1

Software is set for 200 steps per mm. (1605 ballscrews using a 1:1 drive from the motor).The z uses the same accelerations and speed as x and y. Speed is set at around 10m/min and around 2700mm/s2 in acceleration

@@berendlucasvanderweide thanks. and what model of servo? ac or dc?

@@olivderpo jmc 180w, DC

@@berendlucasvanderweide with integrated driver on motor?

@@olivderpo correct. Have a look at the other vid on my channel about the machine itself

It depends, you can more or less drill on a higher RPM, but that requires a carbide drill. I wouldn't drill with anything bigger than 6mm. You could buy a permanent magnet spindel (only commercial available model known to me is with an ATC). What you could do is use a beefier spindle (something like 4KW) and use a sensorless vector based VFD. But you can also settle for a smaller spindle and make the holes using a smaller carbide cutter with a spiral down motion. It all depends on what you want and how fast you need it done.

@@berendlucasvanderweide Thanks for the fast reply. Looks like I'll be doing somethings similar with my g0704 when I get it. What feed and speed were you using for aluminum with this machine?

Hi Ulises; rattm 2.2kw, square air cooled. ER20 collet.

@@berendlucasvanderweide Great thank you.

SMT in Germany. Keeo in mind that aluminium profiles don't come perfectly straight and are not cut at a perfect 90 degree angle.

Heya! thanks for stopping by, interesting question; you should check it on more than 1 angle and in several ways. This method only tells if the spindle is moving in a straight line on 1 point. This means there could still be an angle in the Z that you compensate to 0 by misaligning the spindle. Don't try this with a square spindle, this spindle was known to be true, since i used the parallel part of the outside of the spindle as a reference to grind and elliminate the runout.You should also check with the dial indicator attached to the spindle and measure the angle of the bed. In other words; eventually you do need something with a perfect 90 degree angle. I don't like the tramming tools with two clocks, since they only give you an insight if the spindle is square to the bed. So if the bed is uneven or has a sag, you compensate on that point. With a VMC this adjustment is easier than on a portal, since the latter also depend on the straightness of the X-axis. Alligning machines perfectly within 0,01 accuracy is an artform. It's hard to describe with only words. You also want to know if the Z is moving straight; final test can be done with a rod clamped in (on my large machine, i have a 250mm long rod that is 0,002 accurate with the proper taper that fits the ATC. This is easy if you have a hair angle and you can also measure the up/down error. But you can also do this with a simple rod that doesn't have to be perfectly straight. (it's easier if it is, but not mandatory) Just measure the runout on that rod by rotating the spindle, and make sure to take an average of the 2 peaks in the runout to determine the true position of the rod. Alligning is an adventure itself... ;):)

Berend Lucas van der Weide ah yes I understand. I am using a cnc router, and was wondering if I could check along the sleeve for both x and y? Doesn’t need to be perfect but I would hope I wouldn’t get a bad finish? Also, with this method did you get the criss-cross pattern left from the end mill in facing operations (usually indicates a good tram)?

@@aaronfonseca6031 yeh, this method gives quick insight, but you have to have a known reference to be true in order to have good results. For facing, you can slightly tilt the head for the best possible finish. (but it can also indicate that your machine has a difference in stiffness going from one side to the other). What is your portal made of? aluminium profiles or steel?

Berend Lucas van der Weide ok, I have a steel table on the machine and I would use it as a surface to measure from. I’m not too sure what you mean by portal? The frame of the machine is steel and the gantry is aluminium, it’s a Chinese 1212 machine, if you just give a google for “TEG1212 CNC” you will see my machine (yes my own machine before it was shipped from China). I cut aluminium parts for Rc cars and want to get a decent surface finish. I just want to know if you are happy with your results from this tramming method or weather I should invest into a tramming tool?

@@aaronfonseca6031 Tramming tools give you an insight, but perform some accurate measurements on your machine first. I use all sorts of tools, but no specific tramming tools. Just test the straight rod in the collet and measure to see if there is an error in moving up/down. The issues with the surface finish, are these on the facing opps i assume? Just face in 1 direction and determine what motion gives the best finish. What diameter tool are you using for surfacing the material? A slightly angled head gives the best surface finish possible with facing.

Yes, that is possible for sure. I'm using 2 ballscrews for the y axis. For a wider portal that's a must i think.

@@berendlucasvanderweide thanks i wanted a moving gantry too ,and planning to use a 180w servo but because of the shear force when using only 1 of them im kinda leaning on moving bed

@@RYU47376 you mean they produce too much force? Their peak torque is 3x their nominal rated torque. They are incredibly strong. They'll outperform a Nema34 4nm stepper with ease. Don't be fooled by the 0,56Nm torque rating.

@@berendlucasvanderweide im more scared of the frame twisting because of only using 1 servo to turn the entire gantry

@@RYU47376 That's the reason why you should use two of them. I couldn't agree more. One ballscrew is not a good solution.

Thanks for your comment! It's running at around 12m/min. Don't want to stress the ballscrews too much. I was using 1610's but switched back to 1605's for the y axis (in my case the dual ballscrew axis).

Berend Lucas van der Weide I tried 14 m/min once (I‘m talking G0 in X and Y of course :-) ) with my machine. I‘m using 1205 - smaller machine.. Such movements are a bit to quick for my taste. It exceeds the reaction time and that’s a bit dangerous and like you said: not good for the linear components - I also cranked it back. Machining („none professional“) is not this much of a race, is it? It’s fun to watch though :)

@@EngineeringSpareTime yes, agreed! The large machine i have is quicker (50m/min) and was accelerating at 0,4g, but that's tamed down, since it's just stressing the components. Although it should be able to handle it day and night, it just doesn't feel right. But on that machine it makes more sense considering the distance it has to travel.

@@berendlucasvanderweide 0.4g = 3920mm/s2 ?

@@olivderpo 1g=9806.65 mm/s2

Yes, stil working fine! Highly recommended. The current version is easier to tune. Just make sure you have some shielding around the jumper boards that are on them and to cover the wire connectors properly.

Thanks Nico!

I totally missed this post Aldito. Sorry. Don't know if it's still relevant. What spindle do you use? Air or water cooled? Is the stator or the rotor 75 degrees C? And are you sure it comes from the windings and not from the bearings? On higher RPM's the bearings tend to produce a lot of heat. And on lower RPM's the motor can't deliver much torque, although full current is flowing through the windings and with an air cooled model the airflow is dictated by the RPM of the spindle (since the fan is directly bolted onto the rotor). I am currently using a high flow fan, that turns on low RPM below 40 degreesC and goes into full RPM above. With a hysteresis of about 7 degrees C.

Hi Lars; thanks for your positive feedback! i've replaced it with a high flow fan (from a DELL pc, i can look up the part no if you need it). It's regulated through a tiny voltage regulator and it now starts at around 6.5-7V. I've placed this thermal connector over it that bypasses that converter and feeds it 12V directly at around 40 degrees. The original fan has been removed. It uses a screw. I've placed a small plastic lid (old plastic ER20 housing) over the bearings to avoid dust entering the bearings. I just made a hole in the rear duct and fitted the fan on top of it.

@@berendlucasvanderweide Thanks for your response. So the fan is sitting outside of the OEM rear duct? What about the run out - is the spindle still as good in quality as you expected?

@@larsx312 yes, the spindle is still in good shape. When i had a large assignment I bought a 2nd one as spare, just in case. But it's still sitting in the box. Runout was 0,002 and after the two years of use it hasn't degraded. I'll measure it again, last time was about a half year ago I've used it for hours and hours and hours. Also for 10 hour runs in beech, 3mm overstep, 30mm deep with a 12mm 3 flute with chip breakers. I have this warm up routine I'm using, can't exactly remember it, it's about 14k 1min, 16k 2 min 20k 2min 24k 2min. But i also have a regular customer for plastic, and I'm running it at 6000rpm without warmup.. (light task, 3mm single flute in plastic) so it'll warm up during the production run. I almost never run it above 18000rpm though, except for the warmup and light engraving. Indeed, the fan is mounted on top of the oem cover, protected with a fan cover

@@berendlucasvanderweide Perfect - unfortunately Rattm has the exaxt same model not anymore - found these de.aliexpress.com/item/32435696679.html but they hafe not that smart plug like yours

@@larsx312 Yes, that's the same motor. At the time I had the choice between this and the one with the plug. No particular reason why I choose to buy the one with the plug. (other than looks, i like the plug :) ) They are the same motors.

Hi Thomas, No, i didn't use dampening in the profiles. The overall construction is relative stiff. I'm using the thicker variant of the 80x160 profiles. Sure, as with everything, it can and will resonate. The 760mm X beam sounds acoustically dead if you hit it. No audible resonances. This more or less goes for all the beams used to carry the rails. My idea was to dampen the bed with sand, but never bothered. Btw, if you are planning on dampening with sand, make sure not to fill the horizontal beams fully. Tight packed sand will resonate better (at a lower frequency) than loose sand. The latter is more self dampening. (I use that 'trick' in my loudspeakers that have horizontal aluminum braces). A 2 flute at 18000 rpm generates 600hz.

@@berendlucasvanderweide Hi Berend, Thanks a lot for your answer. Im also planning to use a 80x160 profile for the gantry and i will test filling it loosen with sand. Best regards Thomas