I'm a retired tool & die maker, that has worked in the auto, pulp & paper, mining, and oil and gas industries, and the only center finder that I would use is the wiggler. It's the most versatile and burrs and imperfections or dirt won't give you a false readings. You can place the ball on one spot, and not get false readings from nearby defects. I purchased mine in 1971, from Starrett 828A, and it is still as accurate as the day I bought it. All precision goes out the window, if the spindle, or workpiece is not square. I've used the other type, the one that has a piece of roundbar, held in with a spring, and found it great for finding the edge of burrs, and edge dents, and I can't use it to pick up layout lines.
Hi Rick; I agree with you in all of the comments except on the dial indicator. I believe that even if the indicator run out 1mm still if you clock both sides of a gap .always will give you the center.
I was taught that you pick up the edge with a wiggler at the moment the end of the wiggler goes skating off the edge. That's also what the instructions that came with my set of wigglers say. That seems to happen about 0.01mm after it appears to run true. Gives a very clear and binary indication that you have the edge. My favourite of the options discussed.
I have used every one of those from time to time over the past 60 or so years, I have owned one of those scopes for most of 40 years and now I find thanks to you that the cross hairs are adjustable, I should have read the instructions that came with it. Now that I am retired I find out how to do properly how to do things that I should have learned as a kid. Thank you I do enjoy your videos and I have learned something from every one. By the way I saw a target painted on your head when you bent over looking in that scope. Don't know where the shadow comes from but I believe it was pointing to where much knowledge comes from. Again thanks for the video and please keep them coming.
I use a 10mm dowel with smoking paper, to get roughly within 0.004" then I use the verdict clock method with a 10mm inspection slip to get bang on centre of my Horizontal Devlieg and sip machine. Excellent video
Ive been a car-mechanic for 30 yrs.The electronic one reminds me of how I use to set the ignition-point by using a testlight on the breaker-points. You only would need a pin with known diameter and a battery hooked up to pin and wise. Enjoying your videos. //Kjell Carlsson Gothenbourg Sweden
Hi Don, Great video. Several years ago I designed and built myself a CNC router primarily for wood working (I'm a violin maker) and I've been looking for a good explanation of edge finders in order to bring more accuracy to my setups. Now at least I know where to start.
Very informative video,thank you. I also use an optical centre finder but mine fits directly to the quill shaft on my mill,means always going to be true to axis regardless of chucks or collets. A quick tip I find useful is to store a datum point in my DRO memory that can be worked from right away. I have stored the X-y coordinates of the fixed jaw and my stop block on it. Once the job is clamped up in vice and against the stop then coordinates for holes can be worked from that point. I have checked it for accuracy and it is within less than half a thou every time.
Don, thank you so much for making these videos. I have learned a great deal by watching them. I know they take time and effort but they are very necessary.
I most frequently use a double sided Starrett edge finder, but I have a magnetic edge finder like you showcased that I use when I need to be more accurate. My issue with the magnetic is that when I purchased it, it was not ground correctly. The centerline was off by almost .002" but I was able to grind it in, and can now very accurately hold .0002" with it using a .0001" indicator.
there are two schools of thought: where it sits still and where it kicks out. after exhausting repetitive tests. its within 5 tenths both ways. either way ur gonna be alright: if u gotta be closer, letter kick. if you gotta be backed off let her sit still as she touches the edge. though i must note that the repeatability of letting it sit still had a negligible amount of error, possibly my own eyes not seeing the complete stillness so i tossed those two events from the data. tested on a mori DMG duravertical with two fairly new edgefinders in a collet and a solid holder, edgefinding against a 123 block and 150 tests for repeatability. took me quite a while but i had to get to the bottom of it. its negligible other than the minute amount of time taken to double check if its still versus observing the kick. if anything the kick is a bit quicker for me because i didnt have to scrutinize its stillness with such intensity. btw we have an angle plate at work that was made by suburban tool co!!! excellent and perfectly precise tool. im proud to use it and thrilled to see this channel.
Great video. I have and use an electronic one and have noticed the error you were talking about on occasion. I just accepted it as my error. Thanks for the info.
Good job on the video demonstration. I was looking into a electronic edge finder because the company I work for will not allow door to be open while spindle is running (CNC mill). It is difficult to see the kick-out to find edge on edge finder. After viewing the video, it appears that the electronic version are very accurate but not after rotating spindle. Chucks do vary "off center" to spindle as you have illustrated. So with that in mind, if I manually rotate the spindle 180 degrees and touch off opposite side of part so it touches on same side on probe, than it should be pretty darn accurate. Same concept as using a test indicator to find a center of part or hole. When finding edge, rotate 180 degrees and split the difference. I found using the kick-out method repeats better than waiting for it to stop wiggling, but only when it barely kick's out. I don't use wigglers so I can't comment on them. The point of your video in "picking out the right edge finder" is very good. For your readers, I use edge finder with .200 on one end and V-point on other end. V-point picks up small holes and grooves. I also test .0001 increment indicator with fine adjustment to easily gauge it in for precision if needed. After watching your video, The 180 degree rotation entered my mind and I may reconsider trying the electronic versions.
I use the wiggler, and I alway take the point at which it gets thrown off the edge - that way I don't have to use any judgment as to whether it is still.
Great vid, as always. It is very nice that you share your knowledge and experience with the rest of us, many of which have not had the opportunity to work with/for/near someone with your skill set. RUclips is a great venue for furthering education. As a side note, we picked up a nice angle plate of yours from MSC, not much difference in price from the other offerings, but really well finished. The maroon crinkle paint is easy to hold onto and nice to look at too. Nice to have lifetime tools to work with, 👍👍
hello Don, thank you for the information. one point i would like clarified is that all other people i have seen using edge finders wait for it to kick out for the centre, i see that you find the point before kick out as true centre ! I assume that your way is more accurate best regards Andrew Norman Guernsey,Channel Islands, UK
Great video! Just the right amount of theory and shop practical demo. I have a concentric edge finder and an electronic edge finder. I just bought a Haimer Zero Master. It should be here in a day or two.
I enjoy these videos and it's interesting to watch people who really know what they are doing. I've always wanted one of those scopes. With a good readout on a good mill, it would make a handy measuring microscope.
theslimeylimey For flat stuff, scan it at the highest resolution on an inexpensive flat bed scanner. I pull the image into GIMP software and zoom in. GIMP has measuring capabilities. Draw a line segment and it'll tell you how many pixels it is. Zoom in and you have sub pixel resolution. 2400 DPI scan, subpixel measuring and some averaging and I'm measuring into the microns! I use this to measure the size of the divot pressed into metal for Brinell hardness tester. It's great to determine how hard a lead alloy is when I'm reloading ammo with my cast lead bullets.
Liberty4Ever I have actually done that very thing before circa 2001 :) It was for a small difficult extrusion I designed and I wanted to see how accurate the parts were so i scanned a slice, imported the scaled image into CAD and overlaid my designed profile so I could see where the shrinkage and warpage was. To make sure my output image was correct, I scanned an object of known exact size to determine the correct scaling in the X and Y axis. It works surprisingly well doesn't it!
Just use the electronic one with rotating spindle. It will blink very fast, but you would not be able to tell because of persistence of vision. That way you will not have any issues with collet not running true.
I love your videos Don. There is something to learn in every one. I was taught to pick up both edges and work from the center whenever possible. It will split the error in half from a given edge. My takeaway from this episode is how you touch off, aiming for the floating head to run true vs just breaking off axis how most do it. Going to give it a try and see if it is more consistent.
It seems in your technique for using the edge finders and wiggler that you are trying to stop when you visually see the tip running true. And your repeatability is not what I expect from a edge finder or wiggler. I have always used the exact point at which the edge finder jumps off center as the correct location and the same with the wiggler. If you redo your tests with that method I think all the methods will repeat within a few tenths except the electronic edge finder. The recommended speed for an edge finder according to Starrett is aprox. 1000 rpm. I have watched most of your videos and this is the first time I felt a need to constructively criticize one of your techniques.
correct. his testing leaves a lot unsaid. I never liked electronic edgefinders and I will always use a collect on mine, if for no reason but to preserve the tool. drill chucks put divots into the tool they clamp.
@@kelvinblair4012 Dittos. There are too many youtube shop "teachers" out there giving out advice on shop practice when they have yet to complete their own schooling.
@@garth849 Hahaha! "yet to complete their schooling" Brother you have no idea what nor who you're talking about. You might want to click on ROBERENZ and check out his channel before you go running your mouth. That aside I've never seen or even heard of someone using an edge finder without making it jump. Probably because that's how they are designed to used, if not just use a damn pointer.
Don, thank you so very much for another great video. I think it would be helpful? to see the edge finder that had the wood shim in it, (extreme runout) tested without the shim as well. None of the other styles had that kind of discrepancy. Just thinking that kind of vibration might effect the result?
I Don, An other important thing that makes an good edge finder is the diameter of the tip. I will prefer a .5 over a .2 because 'cause when you forget to subtract the half of it you have a lot more change to see your mistake before cutting off in your piece.
Don, again great piece. i think you and I are some of the last to know the "Secret Hand Shake" I like to travel light when it come to my tool box and I've found that the Starrett wiggler is most versatile. I'm so glad you did this Video. For me it just conferrers what I all ready knew. Keep making worthwhile videos ...and I'll keep watching and liking them. Good Job ! "Unemployed" Journeyman Toolmaker (A1)/ (Why I don't know) It's not because I'm not hansom ....or modest !?
Ok, I use the type of edge finder that you shimmed out of true...only no shim...I wished you would have removed that shim and tried it the proper way...I use mine and I get nowhere near the random results you got with yours...when I use mine properly chucked up, I can do it a dozen times and every time I look at the DRO I get the same result. i also advance the feed until I get the slight "kick out" as I was taught years ago.
If you made the edge finder into a square instead of 2 sides then you could check the centering from side to side as well as front to back. If there is runout that should show up I would think.
The electronic finder senses the edge much better than the others but run-out ruins the accuracy. It seems you could get rid of much run-out error with your cylinder square technique by finding the high point then turning the spindle 90 degrees and using it there. Maybe even a mark on the tool and spindle or collet so you can repeat it well.
Your point about a chuck or collet running out and giving a an error equivalent to the run out is wrong. If say the chuck was running out by say 0.020" at say 12 o'clock and the stylus of the DTI set at that point then as the spindle is rotated to say 3 o'clock that 0.020" would have turned with it so the actual radial point never changes regardless of chuck/collet run out. I hope this makes sense. Paul.
Can we do a test on 5 or 6 different 1/2inch endmills in hardened tool steels? For high speed machining purposes. See if any brand has any clear advantage.
Great video, I didn't fully understand why some instruments are better and others are not... it's my limit for sure. Maybe he means that if the spindle it's old then spins a bit around it's vertical axis. So all rotating instruments are doubling the error, while no-rotating tool are not. It's my interpretation correct???
Thank you very much for the excellent series of videos! I'm curious about how the mechanical edge finder that you shimmed would repeat with, vs without the shim while leaving the DRO at the same zero. The video clearly showed that the mechanical repeated itself well when shimmed, but didn't compare that zero to the same tool when un-shimmed.
The ultimate is a 3 D taster but mega expensive. When using a wiggler i wind in until the wiggler throws off not until it runs true and i use the electrical type when rotating also but use low speed and look for a constant light.
To set a datum, for example on a drawing of say a square part with some holes the datum will usually be from a corner so you set a 0 (datum on 2 edges in x & y which you can then use to accurately follow the drawing and put the holes where they need to be. Ofcourse this isn't always the case but essentially that's what edge finders are used for, they align the centre of the spindle with the edge of the work so that you can work accurately.
This was the first time I have seen the electronic edge finder, NOT SPINNING when in use. Sorta accommodates the run out the same way of the snap back spinamathingy...
I think the concept here is the same as with a lathe. You can have you workpiece off center by an inch, but when you turn the part, the cylinder that you create will be in line with the lathe spindle's axis, although it won't be concentric with the rest of your workpiece. I'm assuming edge finders follow the same principle.
Just discovered your videos a few days ago, and they are excellent! -Couldn't the electric finders areas of "On" or "Off" be marked, and then go 90 degrees to the mean result? -How accurate is the digital readout really? How does it work? Does it have to be checked with Jo blocks from time to time?
+Steffen Dahlberg Most manufacturures guaranty .003 - .005 over 24 inches on their DRO's. It all depends on the manufacturer.Generally the DRO reads a glass scale and no , you should not have to check it with Jo Blocks.
+Steffen Dahlberg A glass scale DRO is extremely accurate, the one in this video looks to have a 1um resolution since it is reading in 0.0002" increments. When working slowly like he is, even an inexpensive Chinese 1um scale will have very good repeatability. One can argue that a better method would have been to put a 0.0001" DTI on it instead. Remember that he is just making relative measurements, so the accuracy of the system is not important, only it's precision. One would also argue that he is not using the ground edge finder properly. Most people advance the work until the puck kicks out, rather than advance the work until the puck runs true. Doing it his way depends on the eyes of the operator, which in some cases will be just as accurate as using the scope, but in others a crap shoot.
+Steffen Dahlberg the DRO is one of the best inventions I have seen. Even a low skill person like me can have confidence in finding the edge and can produce much more accurate work overall.
Hi Don: I would have liked more explanation of the stuff you were trying to prove. For instance, with the edge finders being unaffected by the collet runout, i think you mean that the moving part begins to run concentric(no apparent motion) when it is exactly on the edge, and that if the collet has runout, then the 2 pieces of the edge finder will NOT line up, but the moving part will have no apparent runout. I know that when I use this type of edge finder, I look for the 2 parts of it to run concentric-maybe that is not the correct way, but it is what I have always done. Also, when using the electronic edge finder, I think you zeroed it and then tested for repeatability-but you sort of downplayed the part that the collet runout played-I would have liked to see a measurement of what that runout was-pretty easy to do once you were rotating it and getting the light to flash on and off.
Andy Wander With the cylindrical type, axial run-out doesn't matter like Don said and I use mine in a (good) drill chuck all the time. However, if there is angular run-out, meaning the axis of the edge finder is not parallel to the spindle axis like you might see in a badly worn chuck, there will be an error but unless its really wonky, it will only be a few tenths. I have always used a Mitutoyo cylindrical type that I mirror lapped the mating faces of so it's smooth as silk and by bringing it in to the edge and going just _beyond_ tangency until it "kicks" I get accurate results. It will kick within tenths of going past tangency and is repeatable withing a couple tenths. It's very visible so you can see it kick from a distance without squinting to see if you are touching exactly tangent or still have a thou or two to go. It appears there are two schools of thought on how to use these things so I'm learning here. I have the same Starret ball type wiggler as Don is using but I have found it's not as accurate as the cylindrical type for picking up an edge by making it kick but it's fairly repeatable for finding the center because any error is applied to both touch faces equally. The thin sharp point attachment is very handy for visually aligning to a scribed line or center punch mark in a drill press or when accuracy is not critical because you can make it run dead true to the spindle axis even in a worn chuck.
Hi Don, Good explanations. I have been taught to keep cranking until the mechanical edge finders 'kick out' but this never made sense to me. Commonsense tells me if it is moving in an orbit, it is not on centre, as you approach centre, you are pushing the tip nearer the true centreline, when you are on centre the percieved movement stops, to get it to kick out you have to go past centreline and overcome the friction the spring creates between the body and tip, the tip is always 'advertised size' if you are supposed to kick it out, i would expect the tip to be a thou or two oversize, so the point of kick out occurs at the 'advertised dia' , this being the case, as the spring tension and friction varies so will the 'calibration', obviously centre finding would work if you use opposite edges, but working from one edge only does concern me, your guidance on my perception would be appreciated
One thing that should likely be mentioned here is to make darn sure your collet is clean in and out and that the bore of the spindle is clean. It would not take much trash to move the center
I wish Don had shown us the machine coordinate so edge finder accuracy could be compared. Furthermore, it is not hard to measure and cancel the electronic edge finder run-out which makes it the most accurate instrument presented.
I have to disagree about an indicator in an inaccurate collet. If the collet is tightened, and it must be, then the indicator will not rotate in the collet. So any error in the collet or in the edge finder will put the tip of the DTI at the same position relative to the center line at all angles of rotation.
I don’t understand the orbital thing. The edgefinders will find its own Center, however if you put a mill end in the collet itwill run ecentric as the collet is out, which means that your cut will be the ecentric + the mill size ? Please clarify. Maybe I’m missing something ?
9 times out of 10 the dowel pin and piece of paper works good enough for me lol. Used 3d tester which are great, quick and easy to use but seen them get broken alot aswell lol
Yeah, sometime I like this method also. Long dowel pin is good to get on the side of a large diameter. Works pretty good however your dowel pin needs to be the right diameter and in good shape.
You can run a dti on the dowel in the chuck by hand & find the two points that are between the highest & lowest clock readings. Mark the dowel at these points as they are excactly the dowel radius from the centreline of the spindle. Use either of these marks with your paper with the chuck stationary & any runout in the chuck won't affect the accuracy.
Your using an old bridgeport Miller... modern cnc machines are more stable, more accurate etc. ..i.e you could increment over 0.01mm instead of nudging the handwheel.
What kind of electronic edge-finder is that? It seems it has narrow body, not like most of them which have 20mm wide body. Could you please provide model number of that edge-finder?
there is one more method that is very accurate and only requires a indicator and flat ground block. that is to indicate zero on the edge of the piece, then rotate the indicator 180 degrees and indicate zero against the flat block the results will be as accurate as the indicator used.
Why did you use a drill chuck to hold some of the finders and use a collet with others. I think you should have used the same holder for all of the finders. The drill chuck will be out of round more than a collet. How does turning the spindle get the air out of the collet?
Hi!Thanks for all those educational videos. Changing from 40 years with electronics to a future with mechanics there are lots to learn.I bought some sets of wigglers from eBay. On the wiggler having balls in both ends, the first one You show, the sencing ball is off center. Spinning it between the fingers it wobbles. I don't believ it can be used the way You show. Can it be machined into a useful tool?The second one You show, double ended, the 200 thou end is like bent on my one. It is probably useless unless I can turn down the tip to be centered.A third guy has a conical tip, and the tip is like bent, is wobbling. Probably not useful.Thanks to Your education I claimed a refund and got half the money back.
You didn't explain why you were so concerned with collet and chuck run-out. What is the point of finding the 'true' centre if you are then going to drill or bore with the very device you have rejected as too inaccurate?
@@SuburbanToolInc I am addicted to Metrology ! but my " Thang " is Voltage, if I had a mill, lathe and a surface plate I would be 'doing' not just watching you tutorials...Thanks again !
If your indicator is mounted in an inaccurate collet, it does not matter. If it is off .010 in one direction and you turn the spindle 180 degrees, it will be out by .010 in the opposite direction. Radial distance remains the same no matter the spindle position. You proved this when you showed the Indicol. If .010 runout in the collet would throw things out, what about the 1.5 inches off center of the Indicol?
ardvarkkkkk1 I agree with this.. It doesn't make any difference as to where the indicator is mounted as long as it is attached to the spindle someplace that doesn't move.
I agree too. The probe of the indicator is not on centre. It's not just that it doesn't need to be: it CANNOT be, if is to do its job. However I do agree that the electronic probe needs to be true running.
Ok: You are in my world now. Let's start with the electric " Flashlight " indicator. I found those to be quite good. And although they tell you to go with a dead spindle... I run maybe a couple hundred RPM and look for approximately 50% illumination. Great results. Spindle and holder error corrected. Watch out for non-conductive Dust-Mites. Don, just how big is a dust-mite? And the scope, reticle is adjustable in one respect only. The other is just fluff. You will wind up dizzy with a crick in yer neck using this thing. I do like it sometimes though. I use them all to best advantage. Not always are they all appropriate. It is easier on the neck if you place a PC Microscope eye on it and use the laptop to view, rather than crawl into the machine with it. Third: I like to run my Starrett Edge finder at 1200-1400 RPM, it seems kind of speed sensitive. I am comfy to a tenth of a thousandth with this. I never saw the mag groove before, not many thought except for hygiene. After 20+ Years in the shop, my buddy gave me a wiggler. All I can say is I do approve. I'd never met one, always Starrett edge finders. Then I have the Blake co-ax... run live... I can see well within ....0002" And now, the Haimer Taster. On my CNC, I can see .0001 as clear as day. Painfully obvious. You have to dial it in to your holder, always orient it the same way... Put money on it. It saves me soooooo much time. Bank it. I sometimes do 15 setups a day in the Job shop. But it is much larger than the trusty edge finder, so when on the little series 1 Bridgeport, the Starrett usually get's the job. And I have the laser pointer. Quick, cheap fast and loose. .010 er so. And then last, but not the least... anything you can chuck with a point on it. Heck, you don't even need a point. Hit x+ with one side of the 2" round stock, hit x- on the other side... split it. Additionally, I am quite fond of my Newall Inductive DRO with the inductive pickups and wonderful math killers. Love your videos. Mark
when I use my edge finder I run it at about 1000 rpm and when it finds the edge it shoots right or left that is then the edge not when it just runs true my way I will always get a repeat of .0001" in your money or .00254mm in mine that just my penny's worth
I have one of those dual-headed edge finders. They are, for the most part, useless to me because you can't use them on a chuck. If you do, the exposed end will come off because the other end is chucked firmly in the chuck hence it cannot rotate. Both ends are tied to each other with a spring and both need to rotate at the same time when using one end or the other. You would think they should scribe on the body of theses center finders: DO NOT USE ON A DRILL CHUCK OR DAMAGE WILL ENSUE. Needless to say, I will never buy a dual-headed edge finder.
way wrong IMO. If you used an indicator to sweep between a part that wide( or using an indicol or long extension) don't move the machine, just sweep high point side to side. same concept as jig boring. pickup hole without moving machine. His theory is flawed, if you use any size edge finger i can almost guarantee you'll be within .001 of the same result. if you used a fine point wiggler i bet you'd be just a close. If you used a collect and indicator( what he is saying is flawed) then sweep a gage ring and make sure it's round. then dedicate that collet. There's no one way to do it, it's how to get a good result in the end. To stick with one method and one method only for finding an edge is asinine.
Hi Don: Do you have any comment on www.lasercenteredgefinder.com/ I have one and I like it, but is there a problem hear as well ? Thanks Brian F. AKA GUSMIX
You demonstrate and imply there could be an error but what you need is an exaggerated example to make your point. I am still not convinced . The only REAL consideration is if your error is of a cosine nature. That is to say a non-linear variance.
I'm a retired tool & die maker, that has worked in the auto, pulp & paper, mining, and oil and gas industries, and the only center finder that I would use is the wiggler. It's the most versatile and burrs and imperfections or dirt won't give you a false readings. You can place the ball on one spot, and not get false readings from nearby defects. I purchased mine in 1971, from Starrett 828A, and it is still as accurate as the day I bought it. All precision goes out the window, if the spindle, or workpiece is not square. I've used the other type, the one that has a piece of roundbar, held in with a spring, and found it great for finding the edge of burrs, and edge dents, and I can't use it to pick up layout lines.
Hi Rick; I agree with you in all of the comments except on the dial indicator. I believe that even if the indicator run out 1mm still if you clock both sides of a gap .always will give you the center.
Don I am impressed. Thanks for being good teacher. You are adding real value to humanity. Thank you for giving your time.
what kind of drivel is that?
I was taught that you pick up the edge with a wiggler at the moment the end of the wiggler goes skating off the edge. That's also what the instructions that came with my set of wigglers say. That seems to happen about 0.01mm after it appears to run true. Gives a very clear and binary indication that you have the edge. My favourite of the options discussed.
Certainly is another way of doing it appreciate your comments.
@@SuburbanToolInc No, it's not "another way of doing it", it the RIGHT way of doing it!
I have used every one of those from time to time over the past 60 or so years, I have owned one of those scopes for most of 40 years and now I find thanks to you that the cross hairs are adjustable, I should have read the instructions that came with it. Now that I am retired I find out how to do properly how to do things that I should have learned as a kid. Thank you I do enjoy your videos and I have learned something from every one. By the way I saw a target painted on your head when you bent over looking in that scope. Don't know where the shadow comes from but I believe it was pointing to where much knowledge comes from. Again thanks for the video and please keep them coming.
I use a 10mm dowel with smoking paper, to get roughly within 0.004" then I use the verdict clock method with a 10mm inspection slip to get bang on centre of my Horizontal Devlieg and sip machine. Excellent video
Ive been a car-mechanic for 30 yrs.The electronic one reminds me of how I use to set the ignition-point by using a testlight on the breaker-points.
You only would need a pin with known diameter and a battery hooked up to pin and wise.
Enjoying your videos. //Kjell Carlsson Gothenbourg Sweden
Hi Don, Great video. Several years ago I designed and built myself a CNC router primarily for wood working (I'm a violin maker) and I've been looking for a good explanation of edge finders in order to bring more accuracy to my setups. Now at least I know where to start.
Wow, violin maker! That's an interesting profession. Glad I could help, thanks again for watching!
For some reason I really like test indicators and edge finders. Loved your video
Excellent overview for a beginner like me. Thank you very much!
Very informative video,thank you. I also use an optical centre finder but mine fits directly to the quill shaft on my mill,means always going to be true to axis regardless of chucks or collets. A quick tip I find useful is to store a datum point in my DRO memory that can be worked from right away. I have stored the X-y coordinates of the fixed jaw and my stop block on it. Once the job is clamped up in vice and against the stop then coordinates for holes can be worked from that point. I have checked it for accuracy and it is within less than half a thou every time.
I wish you would have re- touched off the electronic one to the side that the light was off to show the difference
Don, thank you so much for making these videos. I have learned a great deal by watching them. I know they take time and effort but they are very necessary.
I most frequently use a double sided Starrett edge finder, but I have a magnetic edge finder like you showcased that I use when I need to be more accurate. My issue with the magnetic is that when I purchased it, it was not ground correctly. The centerline was off by almost .002" but I was able to grind it in, and can now very accurately hold .0002" with it using a .0001" indicator.
there are two schools of thought: where it sits still and where it kicks out. after exhausting repetitive tests. its within 5 tenths both ways. either way ur gonna be alright: if u gotta be closer, letter kick. if you gotta be backed off let her sit still as she touches the edge. though i must note that the repeatability of letting it sit still had a negligible amount of error, possibly my own eyes not seeing the complete stillness so i tossed those two events from the data. tested on a mori DMG duravertical with two fairly new edgefinders in a collet and a solid holder, edgefinding against a 123 block and 150 tests for repeatability. took me quite a while but i had to get to the bottom of it. its negligible other than the minute amount of time taken to double check if its still versus observing the kick. if anything the kick is a bit quicker for me because i didnt have to scrutinize its stillness with such intensity. btw we have an angle plate at work that was made by suburban tool co!!! excellent and perfectly precise tool. im proud to use it and thrilled to see this channel.
Great video. I have and use an electronic one and have noticed the error you were talking about on occasion. I just accepted it as my error. Thanks for the info.
Good job on the video demonstration. I was looking into a electronic edge finder because the company I work for will not allow door to be open while spindle is running (CNC mill). It is difficult to see the kick-out to find edge on edge finder. After viewing the video, it appears that the electronic version are very accurate but not after rotating spindle. Chucks do vary "off center" to spindle as you have illustrated. So with that in mind, if I manually rotate the spindle 180 degrees and touch off opposite side of part so it touches on same side on probe, than it should be pretty darn accurate. Same concept as using a test indicator to find a center of part or hole. When finding edge, rotate 180 degrees and split the difference.
I found using the kick-out method repeats better than waiting for it to stop wiggling, but only when it barely kick's out. I don't use wigglers so I can't comment on them. The point of your video in "picking out the right edge finder" is very good.
For your readers, I use edge finder with .200 on one end and V-point on other end. V-point picks up small holes and grooves. I also test .0001 increment indicator with fine adjustment to easily gauge it in for precision if needed.
After watching your video, The 180 degree rotation entered my mind and I may reconsider trying the electronic versions.
Just reading all the comments makes it very clear why some machinists are worth much more than other "machinists". :)
I use the wiggler, and I alway take the point at which it gets thrown off the edge - that way I don't have to use any judgment as to whether it is still.
Great vid, as always. It is very nice that you share your knowledge and experience with the rest of us, many of which have not had the opportunity to work with/for/near someone with your skill set. RUclips is a great venue for furthering education. As a side note, we picked up a nice angle plate of yours from MSC, not much difference in price from the other offerings, but really well finished. The maroon crinkle paint is easy to hold onto and nice to look at too. Nice to have lifetime tools to work with, 👍👍
hello Don, thank you for the information.
one point i would like clarified is that all other people i have seen using edge finders wait for it to kick out for the centre,
i see that you find the point before kick out as true centre ! I assume that your way is more accurate
best regards
Andrew Norman
Guernsey,Channel Islands, UK
Great video! Just the right amount of theory and shop practical demo. I have a concentric edge finder and an electronic edge finder. I just bought a Haimer Zero Master. It should be here in a day or two.
I enjoy these videos and it's interesting to watch people who really know what they are doing. I've always wanted one of those scopes. With a good readout on a good mill, it would make a handy measuring microscope.
theslimeylimey For flat stuff, scan it at the highest resolution on an inexpensive flat bed scanner. I pull the image into GIMP software and zoom in. GIMP has measuring capabilities. Draw a line segment and it'll tell you how many pixels it is. Zoom in and you have sub pixel resolution. 2400 DPI scan, subpixel measuring and some averaging and I'm measuring into the microns! I use this to measure the size of the divot pressed into metal for Brinell hardness tester. It's great to determine how hard a lead alloy is when I'm reloading ammo with my cast lead bullets.
Liberty4Ever I have actually done that very thing before circa 2001 :) It was for a small difficult extrusion I designed and I wanted to see how accurate the parts were so i scanned a slice, imported the scaled image into CAD and overlaid my designed profile so I could see where the shrinkage and warpage was. To make sure my output image was correct, I scanned an object of known exact size to determine the correct scaling in the X and Y axis. It works surprisingly well doesn't it!
Liberty4Ever Cool.
Amazing good lesson good explanation simple English enjoying your clips very informative and great refresher of how to do it right.
Just use the electronic one with rotating spindle. It will blink very fast, but you would not be able to tell because of persistence of vision. That way you will not have any issues with collet not running true.
Great info, thanks for sharing! I haven't seen that scope one before, pretty neat.
I love your videos Don. There is something to learn in every one. I was taught to pick up both edges and work from the center whenever possible. It will split the error in half from a given edge. My takeaway from this episode is how you touch off, aiming for the floating head to run true vs just breaking off axis how most do it. Going to give it a try and see if it is more consistent.
It seems in your technique for using the edge finders and wiggler that you are trying to stop when you visually see the tip running true. And your repeatability is not what I expect from a edge finder or wiggler. I have always used the exact point at which the edge finder jumps off center as the correct location and the same with the wiggler. If you redo your tests with that method I think all the methods will repeat within a few tenths except the electronic edge finder. The recommended speed for an edge finder according to Starrett is aprox. 1000 rpm. I have watched most of your videos and this is the first time I felt a need to constructively criticize one of your techniques.
correct. his testing leaves a lot unsaid. I never liked electronic edgefinders and I will always use a collect on mine, if for no reason but to preserve the tool. drill chucks put divots into the tool they clamp.
Amazing to see someone talk with such authority who does not know how to use a wiggler edge finder properly
i always did it your way also for 14 years and worked great for me
@@kelvinblair4012 Dittos. There are too many youtube shop "teachers" out there giving out advice on shop practice when they have yet to complete their own schooling.
@@garth849 Hahaha! "yet to complete their schooling" Brother you have no idea what nor who you're talking about. You might want to click on ROBERENZ and check out his channel before you go running your mouth. That aside I've never seen or even heard of someone using an edge finder without making it jump. Probably because that's how they are designed to used, if not just use a damn pointer.
Thanks for the video, I am impressed with all the methods, but especially the optical one.
You're welcome, thanks again for watching.
Don, thank you so very much for another great video. I think it would be helpful? to see the edge finder that had the wood shim in it, (extreme runout) tested without the shim as well. None of the other styles had that kind of discrepancy. Just thinking that kind of vibration might effect the result?
I Don,
An other important thing that makes an good edge finder is the diameter of the tip. I will prefer a .5 over a .2 because 'cause when you forget to subtract the half of it you have a lot more change to see your mistake before cutting off in your piece.
Don, again great piece. i think you and I are some of the last to know the "Secret Hand Shake"
I like to travel light when it come to my tool box and I've found that the Starrett wiggler is most versatile.
I'm so glad you did this Video. For me it just conferrers what I all ready knew.
Keep making worthwhile videos ...and I'll keep watching and liking them.
Good Job !
"Unemployed" Journeyman Toolmaker (A1)/ (Why I don't know) It's not because I'm not hansom ....or modest !?
Ok, I use the type of edge finder that you shimmed out of true...only no shim...I wished you would have removed that shim and tried it the proper way...I use mine and I get nowhere near the random results you got with yours...when I use mine properly chucked up, I can do it a dozen times and every time I look at the DRO I get the same result. i also advance the feed until I get the slight "kick out" as I was taught years ago.
If you made the edge finder into a square instead of 2 sides then you could check the centering from side to side as well as front to back. If there is runout that should show up I would think.
The electronic finder senses the edge much better than the others but run-out ruins the accuracy. It seems you could get rid of much run-out error with your cylinder square technique by finding the high point then turning the spindle 90 degrees and using it there. Maybe even a mark on the tool and spindle or collet so you can repeat it well.
Your point about a chuck or collet running out and giving a an error equivalent to the run out is wrong. If say the chuck was running out by say 0.020" at say 12 o'clock and the stylus of the DTI set at that point then as the spindle is rotated to say 3 o'clock that 0.020" would have turned with it so the actual radial point never changes regardless of chuck/collet run out. I hope this makes sense. Paul.
Such a great lesson, thank you sir!
Thank you!
So then perhaps a dedicated collet setup for the electric edge finder so it ran true would be the fastest and most repeatable.
bcbloc02 With the perfect collet.... Only
Can we do a test on 5 or 6 different 1/2inch endmills in hardened tool steels? For high speed machining purposes. See if any brand has any clear advantage.
Very nice explaination... great demo
forgot to mention a broken drill and a piece of paper is my "go to" method.
respectfully submitted,
mike
Great video, I didn't fully understand why some instruments are better and others are not... it's my limit for sure.
Maybe he means that if the spindle it's old then spins a bit around it's vertical axis. So all rotating instruments are doubling the error, while no-rotating tool are not. It's my interpretation correct???
Good video Don! Been watching for some time. I linked this video in one of our videos. Keep them coming.
Best
Matt
+Centurial Inc Thank you for the support, and thanks for sharing!
Thanks for the video. I am new in the cnc world so this info wil help me on my way.
Thank you very much for the excellent series of videos!
I'm curious about how the mechanical edge finder that you shimmed would repeat with, vs without the shim while leaving the DRO at the same zero.
The video clearly showed that the mechanical repeated itself well when shimmed, but didn't compare that zero to the same tool when un-shimmed.
The ultimate is a 3 D taster but mega expensive.
When using a wiggler i wind in until the wiggler throws off not until it runs true and i use the electrical type when rotating also but use low speed and look for a constant light.
What is the point of an edge finder? I'm not a machinist obviously. By the way that looks so awesome at 16:27
To set a datum, for example on a drawing of say a square part with some holes the datum will usually be from a corner so you set a 0 (datum on 2 edges in x & y which you can then use to accurately follow the drawing and put the holes where they need to be.
Ofcourse this isn't always the case but essentially that's what edge finders are used for, they align the centre of the spindle with the edge of the work so that you can work accurately.
This was the first time I have seen the electronic edge finder, NOT SPINNING when in use. Sorta accommodates the run out the same way of the snap back spinamathingy...
I think the concept here is the same as with a lathe. You can have you workpiece off center by an inch, but when you turn the part, the cylinder that you create will be in line with the lathe spindle's axis, although it won't be concentric with the rest of your workpiece. I'm assuming edge finders follow the same principle.
Just discovered your videos a few days ago, and they are excellent!
-Couldn't the electric finders areas of "On" or "Off" be marked, and then go 90 degrees to the mean result?
-How accurate is the digital readout really? How does it work? Does it have to be checked with Jo blocks from time to time?
+Steffen Dahlberg Most manufacturures guaranty .003 - .005 over 24 inches on their DRO's. It all depends on the manufacturer.Generally the DRO reads a glass scale and no , you should not have to check it with Jo Blocks.
+Steffen Dahlberg A glass scale DRO is extremely accurate, the one in this video looks to have a 1um resolution since it is reading in 0.0002" increments. When working slowly like he is, even an inexpensive Chinese 1um scale will have very good repeatability. One can argue that a better method would have been to put a 0.0001" DTI on it instead. Remember that he is just making relative measurements, so the accuracy of the system is not important, only it's precision.
One would also argue that he is not using the ground edge finder properly. Most people advance the work until the puck kicks out, rather than advance the work until the puck runs true. Doing it his way depends on the eyes of the operator, which in some cases will be just as accurate as using the scope, but in others a crap shoot.
+Steffen Dahlberg the DRO is one of the best inventions I have seen. Even a low skill person like me can have confidence in finding the edge and can produce much more accurate work overall.
I use an electronic edge finder but I put a clock gauge on it and run the spindle to see if it's running true or not.
Hi Don:
I would have liked more explanation of the stuff you were trying to prove. For instance, with the edge finders being unaffected by the collet runout, i think you mean that the moving part begins to run concentric(no apparent motion) when it is exactly on the edge, and that if the collet has runout, then the 2 pieces of the edge finder will NOT line up, but the moving part will have no apparent runout. I know that when I use this type of edge finder, I look for the 2 parts of it to run concentric-maybe that is not the correct way, but it is what I have always done.
Also, when using the electronic edge finder, I think you zeroed it and then tested for repeatability-but you sort of downplayed the part that the collet runout played-I would have liked to see a measurement of what that runout was-pretty easy to do once you were rotating it and getting the light to flash on and off.
Andy Wander With the cylindrical type, axial run-out doesn't matter like Don said and I use mine in a (good) drill chuck all the time. However, if there is angular run-out, meaning the axis of the edge finder is not parallel to the spindle axis like you might see in a badly worn chuck, there will be an error but unless its really wonky, it will only be a few tenths.
I have always used a Mitutoyo cylindrical type that I mirror lapped the mating faces of so it's smooth as silk and by bringing it in to the edge and going just _beyond_ tangency until it "kicks" I get accurate results. It will kick within tenths of going past tangency and is repeatable withing a couple tenths. It's very visible so you can see it kick from a distance without squinting to see if you are touching exactly tangent or still have a thou or two to go. It appears there are two schools of thought on how to use these things so I'm learning here.
I have the same Starret ball type wiggler as Don is using but I have found it's not as accurate as the cylindrical type for picking up an edge by making it kick but it's fairly repeatable for finding the center because any error is applied to both touch faces equally. The thin sharp point attachment is very handy for visually aligning to a scribed line or center punch mark in a drill press or when accuracy is not critical because you can make it run dead true to the spindle axis even in a worn chuck.
Hi Don, Good explanations. I have been taught to keep cranking until the mechanical edge finders 'kick out' but this never made sense to me. Commonsense tells me if it is moving in an orbit, it is not on centre, as you approach centre, you are pushing the tip nearer the true centreline, when you are on centre the percieved movement stops, to get it to kick out you have to go past centreline and overcome the friction the spring creates between the body and tip, the tip is always 'advertised size' if you are supposed to kick it out, i would expect the tip to be a thou or two oversize, so the point of kick out occurs at the 'advertised dia' , this being the case, as the spring tension and friction varies so will the 'calibration', obviously centre finding would work if you use opposite edges, but working from one edge only does concern me, your guidance on my perception would be appreciated
Thank you for watching.
Sheds light on some common primary usage tools...
When I ran the electronic, I had it turning at maybe 200 RPM and looked for 50% illumination.
One thing that should likely be mentioned here is to make darn sure your collet is clean in and out and that the bore of the spindle is clean. It would not take much trash to move the center
Good point, clean, clean, clean. Thank you for watching!
What are your thoughts on laser edge finders?
I Like centering the part by sweeping booth sides "opposite "
Alec
We are talking about finding an edge. Referencing from an edge. Has nothing to do with finding the center of a part for this tutorial.
Thanks for the video, my question was answered at 1m04ec.
Glad to have been able to help. Thanks for watching!
Glen cracks me up!
Thanks for the helpful information!
I wish Don had shown us the machine coordinate so edge finder accuracy could be compared. Furthermore, it is not hard to measure and cancel the electronic edge finder run-out which makes it the most accurate instrument presented.
I have to disagree about an indicator in an inaccurate collet. If the collet is tightened, and it must be, then the indicator will not rotate in the collet. So any error in the collet or in the edge finder will put the tip of the DTI at the same position relative to the center line at all angles of rotation.
Very interesting. Thanks.
Glad you enjoyed it
I don’t understand the orbital thing. The edgefinders will find its own Center, however if you put a mill end in the collet itwill run ecentric as the collet is out, which means that your cut will be the ecentric + the mill size ? Please clarify. Maybe I’m missing something ?
9 times out of 10 the dowel pin and piece of paper works good enough for me lol. Used 3d tester which are great, quick and easy to use but seen them get broken alot aswell lol
Yeah, sometime I like this method also. Long dowel pin is good to get on the side of a large diameter. Works pretty good however your dowel pin needs to be the right diameter and in good shape.
You can run a dti on the dowel in the chuck by hand & find the two points that are between the highest & lowest clock readings. Mark the dowel at these points as they are excactly the dowel radius from the centreline of the spindle. Use either of these marks with your paper with the chuck stationary & any runout in the chuck won't affect the accuracy.
Really enjoyed don
Your using an old bridgeport Miller... modern cnc machines are more stable, more accurate etc. ..i.e you could increment over 0.01mm instead of nudging the handwheel.
I prefer a test/dial indicator personally.
I've been able to be within a a few millionths
What kind of electronic edge-finder is that? It seems it has narrow body, not like most of them which have 20mm wide body. Could you please provide model number of that edge-finder?
I keep looking for information on the angled edge finder and no one has it the sucks
there is one more method that is very accurate and only requires a indicator and flat ground block. that is to indicate zero on the edge of the piece, then rotate the indicator 180 degrees and indicate zero against the flat block the results will be as accurate as the indicator used.
Why did you use a drill chuck to hold some of the finders and use a collet with others. I think you should have used the same holder for all of the finders. The drill chuck will be out of round more than a collet. How does turning the spindle get the air out of the collet?
Not "air" I think he said "error". LOL
Hi!Thanks for all those educational videos. Changing from 40 years with electronics to a future with mechanics there are lots to learn.I bought some sets of wigglers from eBay. On the wiggler having balls in both ends, the first one You show, the sencing ball is off center. Spinning it between the fingers it wobbles. I don't believ it can be used the way You show. Can it be machined into a useful tool?The second one You show, double ended, the 200 thou end is like bent on my one. It is probably useless unless I can turn down the tip to be centered.A third guy has a conical tip, and the tip is like bent, is wobbling. Probably not useful.Thanks to Your education I claimed a refund and got half the money back.
Thanks!
You didn't explain why you were so concerned with collet and chuck run-out. What is the point of finding the 'true' centre if you are then going to drill or bore with the very device you have rejected as too inaccurate?
If your really intending to check the resolution of an edge finder your dro should resolve better than 0.0002...
Always wondered why the USA have never use metric measurements always imperial ,we use both where necessary but mostly metric
Very good question. Thank you for watching.
Great thanks.
You're welcome!
@@SuburbanToolInc I am addicted to Metrology ! but my " Thang " is Voltage, if I had a mill, lathe and a surface plate I would be 'doing' not just watching you tutorials...Thanks again !
Oh... forgot.
Next rig: Renishaw Wireless FM. Tool and probe.
Thank you sir
haha, just read into previous comments and I see I'm not the first:)
If your indicator is mounted in an inaccurate collet, it does not matter. If it is off .010 in one direction and you turn the spindle 180 degrees, it will be out by .010 in the opposite direction. Radial distance remains the same no matter the spindle position. You proved this when you showed the Indicol. If .010 runout in the collet would throw things out, what about the 1.5 inches off center of the Indicol?
ardvarkkkkk1
I agree with this.. It doesn't make any difference as to where the indicator is mounted as long as it is attached to the spindle someplace that doesn't move.
I agree too. The probe of the indicator is not on centre. It's not just that it doesn't need to be: it CANNOT be, if is to do its job.
However I do agree that the electronic probe needs to be true running.
Ok:
You are in my world now.
Let's start with the electric " Flashlight " indicator.
I found those to be quite good.
And although they tell you to go with a dead spindle... I run maybe a couple hundred RPM and look for approximately 50% illumination. Great results. Spindle and holder error corrected.
Watch out for non-conductive Dust-Mites.
Don, just how big is a dust-mite?
And the scope, reticle is adjustable in one respect only. The other is just fluff.
You will wind up dizzy with a crick in yer neck using this thing.
I do like it sometimes though.
I use them all to best advantage. Not always are they all appropriate.
It is easier on the neck if you place a PC Microscope eye on it and use the laptop to view, rather than crawl into the machine with it.
Third: I like to run my Starrett Edge finder at 1200-1400 RPM, it seems kind of speed sensitive. I am comfy to a tenth of a thousandth with this.
I never saw the mag groove before, not many thought except for hygiene.
After 20+ Years in the shop, my buddy gave me a wiggler.
All I can say is I do approve. I'd never met one, always Starrett edge finders.
Then I have the Blake co-ax... run live... I can see well within ....0002"
And now, the Haimer Taster.
On my CNC, I can see .0001 as clear as day. Painfully obvious.
You have to dial it in to your holder, always orient it the same way...
Put money on it. It saves me soooooo much time.
Bank it. I sometimes do 15 setups a day in the Job shop.
But it is much larger than the trusty edge finder, so when on the little series 1 Bridgeport, the Starrett usually get's the job.
And I have the laser pointer. Quick, cheap fast and loose. .010 er so. And then last, but not the least... anything you can chuck with a point on it.
Heck, you don't even need a point.
Hit x+ with one side of the 2" round stock, hit x- on the other side... split it.
Additionally, I am quite fond of my Newall Inductive DRO with the inductive pickups and wonderful math killers.
Love your videos.
Mark
thanks......I like the electronic, however I've had issues with it. now I know why. the wiggler also. presently I'm using sterret.
mike
amazing
I'm new and always hit the vice 2 or 3k.. annoyed
when I use my edge finder I run it at about 1000 rpm and when it finds the edge it shoots right or left that is then the edge not when it just runs true my way I will always get a repeat of .0001" in your money or .00254mm in mine
that just my penny's worth
Judging by a part I got from china, they haven't heard of any of these methods.
good one :)
That's why you should buy from Suburban Tool, Inc american made products
Oh I'm sure you're fed the trolls with the wiggler didn't you
oi m8 did you grab me wiggla again? lolol
When I wanna know where center is i just ask my wife cause she knows everything.
Ride the shai hulud!
I have one of those dual-headed edge finders. They are, for the most part, useless to me because you can't use them on a chuck. If you do, the exposed end will come off because the other end is chucked firmly in the chuck hence it cannot rotate. Both ends are tied to each other with a spring and both need to rotate at the same time when using one end or the other. You would think they should scribe on the body of theses center finders: DO NOT USE ON A DRILL CHUCK OR DAMAGE WILL ENSUE. Needless to say, I will never buy a dual-headed edge finder.
way wrong IMO. If you used an indicator to sweep between a part that wide( or using an indicol or long extension) don't move the machine, just sweep high point side to side. same concept as jig boring. pickup hole without moving machine. His theory is flawed, if you use any size edge finger i can almost guarantee you'll be within .001 of the same result. if you used a fine point wiggler i bet you'd be just a close. If you used a collect and indicator( what he is saying is flawed) then sweep a gage ring and make sure it's round. then dedicate that collet. There's no one way to do it, it's how to get a good result in the end. To stick with one method and one method only for finding an edge is asinine.
Hi Don: Do you have any comment on
www.lasercenteredgefinder.com/
I have one and I like it, but is there a problem hear as well ?
Thanks
Brian F. AKA GUSMIX
GUSMIX22 I have not tried this one, but it looks like it will have the same problem regarding run out.
You demonstrate and imply there could be an error but what you need is an exaggerated example to make your point. I am still not convinced . The only REAL consideration is if your error is of a cosine nature. That is to say a non-linear variance.