I am a professional machinist and I had never heard of this method, (until about 2 years ago) thought it was pretty hacky to be honest. After messing around with it I completely endorse it! It makes all the difference in the world with small tools like those under 0.5mm
@@Trident_Euclid He very most likely was talking about the end mills diameter. 1mm runout would be catastrophic! At my work place we have tool holders consistently capable of less than 3 microns of runout, and that's probably not even as good as it gets.
Greasing the ER thread and nut taper improves a lot the runout, with increased holding force. Also is interesting to use a simetrical tool to screw the nut, so you only aply torque instead of torque plus side load on the nut.
John have you tried just tapping the nut instead of the tool? I find tapping the nut with a small brass hammer is enough to move everything without having to hit the tool itself. That indeed makes sense, as the nut both has a thread, and is engaged in a thread with a required amount of clearance. Tapping the nut will move the nut, the collet, and the tool across on the spindle thread. Not much, but enough.
It would be interesting to see the same tool runout in different types of holders: ER, hydraulic, steep taper like an emuge apc (uberchuck), shrink fit.
I like this stuff. I have used this method more times than I can remember because it is common sense, but was not aware how much it is done in other shops. But it dang sure works! Brass against steel instead of tape. It reminds me of "Staking" steel with a punch in order to get it to expand and fit tightly against another piece of steel or cast iron. Yup, many times...
Really good video. Interesting to hear about the small tools possibly being ground in multiple operations, I never thought of that but it makes sense. I've also found that many times simply loosening the collet nut and rotating the collet or the tool within the collet will make a big difference and only then tapping it in after rotating works pretty well (after rotating sometimes I no longer even need to tap).
Great video, just now found it and I follow NYCCNC and have been for years. Been doing this for years. I use a piece of plastic since we machine LOTS of plastic (delrin). One thing I recommend is to thoroughly clean your tool holder, collet and collet chuck. Apply a light film of oil to allow everything to slip together easy, torque your collet to the correct torque. More often than not I get 0 to 2 tenths with the above right off the bat. If it is out, a gentle tap is all that's needed to bring it in.
Do this all the time with .015 and .020 endmills. They will break in a couple minutes if you don't and leave nasty burrs before the go. If the runnout is bad clean and reseat the collet before you start hammering.
Thanks again for another great video. Been doing a bit of surfacing with ball end mills in aluminium and I've been finding the biggest variable in surface quality was the size of the endmill. reckon it'd be pretty night and day if you did the same part with the same stepover / feeds and speeds only with the split being between a 1/8" and a 1/2" ball end mill. Pretty sure I remember you touching on the advantage of using a larger ball end mill in another video.
Guys! Your videos are the best one on CNC. A suggestion for a possible new video. Please explain conventional and climbing milling and moreover when and if you advice to use one instead of the other. Maybe making distinction in case for big CNC (really rigid) or mini mill, like Grizzly G0704, BF20 or the smaller Tormach, when the rigidity and HP can be an issue. Thank you vey much if you an do it! I think it's really important and interesting.
Generally I do not pick this much but occasionally I do have tasks i will knock my cutter around to get it as perfect as I can. More so on lathe tasks i get annoyed by anything off.
very informative! I'm assuming that if you did the same procedure in a part that could be measured accurately, such as a diametrical part, then the adjusted mill (or a reamer) would perform better.
Thank you for this video. I ask you this very question months ago, so I'm glad to see you tested it and found it "sticks." Regarding that tolerance stackup: I'd be curious to know: if you optimize runout int he Speroni and then transfer it to the Tormach (ideally to a PCNC1100, because that's what I have :P), do you see a degradation in runout? Are you able to put a TTS holder in the Speroni?
a person also has to factor in deflection that takes place during a cut. Example-picture an extreme for visualization only-- 3/8 long end mill with .002 runnout. However picture it taking a .005 depth of cut side on the tangent and you are feeding along pretty fast. That runnout now is canceled out and the entire cutter bends true. However note the test. there is little deflection of the cutter but its also cutting way down about 50% inward on the ball not on the periphery. So its good points and a good microscope picture but in most cases you would not notice much.
the points are well made regardless but looking again at the shape of the part--if they had measured the runout about 50% of the way on the sphere--I believe the runnout would not be detectable. Keep in mind that you checked on the periphery and that the sphere of the ball is ground totally independently as you noted on the other parts of the cutter therefore its hard to say.
Does runout still has an impact on finishing if you use single flute end mills? The effective diameter would slightly change, but the feed per tooth would be constant.
Feed per tooth may be constant, but chip thickness would vary along the cut. If you took the run out to an extreme and say had .010 out of concentricity, and took an .010 finish pass, your have vertical lines along the surface protruding out .020.
It seems to work really well, but it leads me to ask 2 questions. Is the adjustment only valid at the tip of the tool or would there be wobble from the tip to the shank? In such case where the tool is in fact perfectly made and the runout is from your machine,then doing this to fix either an angle alignment or an eccentric alignment would misalign the tool such that it would be even more out of alignment when next used (I suppose if you have a witness mark on the spindle then you could mark the tool to match). I'd made an adapter for my jig saw so that I could hold and operate needle files in it. The file holding side had 8 adjustment screws (2 sets of 4 upper and lower). Because my files were so varied in the handle no simple method would suffice. Needless to say my thoughts drifted to this topic when I saw this video. How difficult would it be to apply the same sort of total control over axis and angle for a milling tool?
Low runout for mold type work would be 3 to 6 micron, not sure that 7 tenths 0.0007" runout improved upon with a screw driver in the Tormach spindle will show that much difference. The micrograph does show some differences. What's the runout on the spindle then ?
I am wondering what it is that bends or moves when you are hammering on the milling cutter. Presumably it is the seating of the tool in the spring collet, and not the milling cutter bending, or rotating in the collet? Is this an argument in favour of using shrink-fit instead of ER collets? Or maybe just buying better quality grades of ER collet? Then again, if the flutes of the milling cutter aren't perfectly concentric with its shaft, then even the best quality ER collet in the world won't magically make them concentric with the spindle.
I believe it is more the collet being moved around in the holder. Lower quality tooling will have a squishy feeling as it gets tighter, really good stuff is more like hitting a wall when it gets tight, the wrench just stops moving. Not surprisingly the squishy stuff responds to the tapping process more then "better" tooling. I have some HPI MX holders and collets where if everything is clean you get what you get there is no tapping it in, admitedly not an ER style collet.
What is the setup towards the end where you rotate the finished item on the black background? Looks interesting/unusual since the background is so block but the item is pretty well lit. Super curious... Thanks!
Are these corrections being made to the tool itself (i.e. bending/deflecting the tool), or is it moving the collet inside the collet holder when you tap on the cutter? Or a combo of both?
I know you said after the tool is of such a diameter tool run out of these numbers are negligible. I'd have to think after 1/8 th inch it's not necessary. I know I rarely use 1/8th inch. Ocasionly 1/4. But mostly 1/2 3/8 or 5/8 end mills& ball end mills. And never check these
If the clamping nut consists of 2 parts (with pressure ring), it is always a more accurate clamping of the cutter. Interesting to hear about the inaccuracy in production of the tools, but you can distinguish the quality according to their origin western or asian.
Solid holders with set screws are pretty bad for causing runout. Shrink fit holders may be better but in certain applications are 'too rigid' that can contribute to vibration.
@@EUnit111986 solid holders are pretty bad for causing run out? Are you mad :) never had a problem using solid holders for end mills. The only time I've had a problem with run out, chatter, deflection, etc. with an end mill is when I used a collet holder . Especially during a side cutting operation . Sometimes you have to decrease the feed slightly and increase the rpms slightly. Using a collet holder does enable you to indicate the tool, but to what extent? Will this ensure stabilization of the tool during the process throughout?
@@gusjohnson6646 that also depends how bad is the runout. If it's less than .0005 which in most cases it is then the tool will be fine. Speaking about aluminum that is
Great video! But you're still using a small endmill for finishing. A bigger ball will give better finish at a given stepover. You only need the small ones if you have very steep creases. (And then, only for those narrow channels/pencils)
@@thegribbs you're not wrong, but maybe this test isn't the right test then? I see them using small ball mills on big low curvature surfaces in many videos. How about parting lines as a demonstration? Cuts faster, too :-)
0.00762... I don’t know if imperial people state decimal places past the micron as a bad habit of having more than 3 numbers as normal or just completely misunderstand a superior standard of units!
Americans are generally not intimately familiar with the metric system, id bet youd make minor insignificant mistakes if you were using us customary as well
What country are you from? Just curious to see what you've done for the world. Then we can compare what the U.S. has done using the Imperial unit against the accomplishments of your superior system.
Those numbers are converted to be polite to metric-speaking people and that's what came out of the conversion device (he's built a custom arduino thingy that does this, there's a video somewhere).
![ร้อยพ่อพันแม่ - WanMai [Official MV]](http://i.ytimg.com/vi/XY2pNYASEbg/mqdefault.jpg)
![[LIVE] : ONE ลุมพินี 87 วันนี้!! คู่เอก "ก้องชัย vs โชคปรีชา"](http://i.ytimg.com/vi/YN6e7HVnekw/mqdefault.jpg)
I am a professional machinist and I had never heard of this method, (until about 2 years ago) thought it was pretty hacky to be honest. After messing around with it I completely endorse it! It makes all the difference in the world with small tools like those under 0.5mm
Everydays a learning day at the shop
@Daniel Denucci Why is that?
What were you doing to correct runout previously?
Interesting!
I had exceptionel good look doing this with small carbide drills
You mean luck?
@@dominikm1457 Thats correct
Isn't slightly less than a millimeter runout would lead to big tolerances deviations?
@@Trident_Euclid He very most likely was talking about the end mills diameter. 1mm runout would be catastrophic! At my work place we have tool holders consistently capable of less than 3 microns of runout, and that's probably not even as good as it gets.
@@matiaspetersen2644 how do u remove this error I mean If there is outrun of 1 mm than how it can be reduced to 3 microns in endmill
Greasing the ER thread and nut taper improves a lot the runout, with increased holding force. Also is interesting to use a simetrical tool to screw the nut, so you only aply torque instead of torque plus side load on the nut.
Very interesting, thanks.
@@JohnGrimsmo given how much I learnt from you, glad to help a little bit!
if you dont want things to run out then close the bloody doors.
...or put it in a container that doesn't have a hole in the bottom.
I had no idea that high quality tools would still have run out. Very cool demo on this one John.
John have you tried just tapping the nut instead of the tool?
I find tapping the nut with a small brass hammer is enough to move everything without having to hit the tool itself. That indeed makes sense, as the nut both has a thread, and is engaged in a thread with a required amount of clearance. Tapping the nut will move the nut, the collet, and the tool across on the spindle thread. Not much, but enough.
I did this in a set screw tool holder once on an old manual horizontal boring mill. I had to hold +- .0003 . It actually worked .
It would be interesting to see the same tool runout in different types of holders: ER, hydraulic, steep taper like an emuge apc (uberchuck), shrink fit.
I like this stuff. I have used this method more times than I can remember because it is common sense, but was not aware how much it is done in other shops. But it dang sure works! Brass against steel instead of tape. It reminds me of "Staking" steel with a punch in order to get it to expand and fit tightly against another piece of steel or cast iron. Yup, many times...
This was super helpful to me. Thanks John!
Really good video.
Interesting to hear about the small tools possibly being ground in multiple operations, I never thought of that but it makes sense.
I've also found that many times simply loosening the collet nut and rotating the collet or the tool within the collet will make a big difference and only then tapping it in after rotating works pretty well (after rotating sometimes I no longer even need to tap).
Great video, just now found it and I follow NYCCNC and have been for years.
Been doing this for years. I use a piece of plastic since we machine LOTS of plastic (delrin).
One thing I recommend is to thoroughly clean your tool holder, collet and collet chuck. Apply a light film of oil to allow everything to slip together easy, torque your collet to the correct torque.
More often than not I get 0 to 2 tenths with the above right off the bat.
If it is out, a gentle tap is all that's needed to bring it in.
Good to know that it stays!
Do this all the time with .015 and .020 endmills. They will break in a couple minutes if you don't and leave nasty burrs before the go. If the runnout is bad clean and reseat the collet before you start hammering.
Thanks again for another great video. Been doing a bit of surfacing with ball end mills in aluminium and I've been finding the biggest variable in surface quality was the size of the endmill. reckon it'd be pretty night and day if you did the same part with the same stepover / feeds and speeds only with the split being between a 1/8" and a 1/2" ball end mill. Pretty sure I remember you touching on the advantage of using a larger ball end mill in another video.
love the Metric info
I tap end mills in everyday....its a must down at .040" and smaller
Ah the old tappy tappy, well accustomed to this haha.
Great investigation!
It would be interesting to see the runout again after the run on the corrected tool.
CONFIRMED 2:55 Size Matters!
Guys! Your videos are the best one on CNC. A suggestion for a possible new video. Please explain conventional and climbing milling and moreover when and if you advice to use one instead of the other. Maybe making distinction in case for big CNC (really rigid) or mini mill, like Grizzly G0704, BF20 or the smaller Tormach, when the rigidity and HP can be an issue. Thank you vey much if you an do it! I think it's really important and interesting.
Good questions, keep up the good work.
Glad to see this video.
Generally I do not pick this much but occasionally I do have tasks i will knock my cutter around to get it as perfect as I can. More so on lathe tasks i get annoyed by anything off.
very informative! I'm assuming that if you did the same procedure in a part that could be measured accurately, such as a diametrical part, then the adjusted mill (or a reamer) would perform better.
Thank you for this video. I ask you this very question months ago, so I'm glad to see you tested it and found it "sticks."
Regarding that tolerance stackup: I'd be curious to know: if you optimize runout int he Speroni and then transfer it to the Tormach (ideally to a PCNC1100, because that's what I have :P), do you see a degradation in runout? Are you able to put a TTS holder in the Speroni?
a person also has to factor in deflection that takes place during a cut. Example-picture an extreme for visualization only-- 3/8 long end mill with .002 runnout. However picture it taking a .005 depth of cut side on the tangent and you are feeding along pretty fast. That runnout now is canceled out and the entire cutter bends true. However note the test. there is little deflection of the cutter but its also cutting way down about 50% inward on the ball not on the periphery. So its good points and a good microscope picture but in most cases you would not notice much.
the points are well made regardless but looking again at the shape of the part--if they had measured the runout about 50% of the way on the sphere--I believe the runnout would not be detectable. Keep in mind that you checked on the periphery and that the sphere of the ball is ground totally independently as you noted on the other parts of the cutter therefore its hard to say.
So quality stuff on this chanel! Thank you!!
Is there any equipment to check cutter runout offline ?
Does runout still has an impact on finishing if you use single flute end mills? The effective diameter would slightly change, but the feed per tooth would be constant.
Feed per tooth may be constant, but chip thickness would vary along the cut.
If you took the run out to an extreme and say had .010 out of concentricity, and took an .010 finish pass, your have vertical lines along the surface protruding out .020.
Anyone who chooses a hammer to work on an endmill while still in the Spindle is my hero 😍🤩 That’s my kinda fix!!
It seems to work really well, but it leads me to ask 2 questions.
Is the adjustment only valid at the tip of the tool or would there be wobble from the tip to the shank?
In such case where the tool is in fact perfectly made and the runout is from your machine,then doing this to fix either an angle alignment or an eccentric alignment would misalign the tool such that it would be even more out of alignment when next used (I suppose if you have a witness mark on the spindle then you could mark the tool to match).
I'd made an adapter for my jig saw so that I could hold and operate needle files in it. The file holding side had 8 adjustment screws (2 sets of 4 upper and lower). Because my files were so varied in the handle no simple method would suffice. Needless to say my thoughts drifted to this topic when I saw this video. How difficult would it be to apply the same sort of total control over axis and angle for a milling tool?
Low runout for mold type work would be 3 to 6 micron, not sure that 7 tenths 0.0007" runout improved upon with a screw driver in the Tormach spindle will show that much difference. The micrograph does show some differences. What's the runout on the spindle then ?
I am wondering what it is that bends or moves when you are hammering on the milling cutter.
Presumably it is the seating of the tool in the spring collet, and not the milling cutter bending, or rotating in the collet?
Is this an argument in favour of using shrink-fit instead of ER collets? Or maybe just buying better quality grades of ER collet?
Then again, if the flutes of the milling cutter aren't perfectly concentric with its shaft, then even the best quality ER collet in the world won't magically make them concentric with the spindle.
Yep need an answer to this ... please 🙂
I believe it is more the collet being moved around in the holder. Lower quality tooling will have a squishy feeling as it gets tighter, really good stuff is more like hitting a wall when it gets tight, the wrench just stops moving. Not surprisingly the squishy stuff responds to the tapping process more then "better" tooling. I have some HPI MX holders and collets where if everything is clean you get what you get there is no tapping it in, admitedly not an ER style collet.
Oh ok thanks ..
The collet fingers are moving inside the nut and taper. A carbide tool is either straight or broken. You can't really bend it.
What is your experience with round-out on hole diameters, have you been able to correlate the run out on the flutes to change in diameter?
What is the setup towards the end where you rotate the finished item on the black background? Looks interesting/unusual since the background is so block but the item is pretty well lit. Super curious...
Thanks!
Are these corrections being made to the tool itself (i.e. bending/deflecting the tool), or is it moving the collet inside the collet holder when you tap on the cutter? Or a combo of both?
the tool is not bending at all. You're just moving the collet a little bit.
does tir seem to be more a factor of axial angle deviation or axial offset?
Wouldn't it depend on the cause of the runout?
Great question.
I would imagine an even combo of the two🤷♂️
I know you said after the tool is of such a diameter tool run out of these numbers are negligible. I'd have to think after 1/8 th inch it's not necessary. I know I rarely use 1/8th inch. Ocasionly 1/4. But mostly 1/2 3/8 or 5/8 end mills& ball end mills. And never check these
Now do a comparison of 3D profiling at 0deg (as in this video) and 45 deg (load split over axes)
Do you consider 0.125" as a "small diameter" endmill?
Not as small as .040, which is that they used for the test.
Yes
If you guys need a good machinist lmk. I feel there is a lot of ideas and suggestions I could offer to help you in every day production.
If the clamping nut consists of 2 parts (with pressure ring), it is always a more accurate clamping of the cutter. Interesting to hear about the inaccuracy in production of the tools, but you can distinguish the quality according to their origin western or asian.
Grinding spindle taper
Можно еще синей изолентой примотать фрезу или сварочкой прихватить.)
As far as end mills are concerned, why are you not using a solid holder for this application?
Solid holders with set screws are pretty bad for causing runout. Shrink fit holders may be better but in certain applications are 'too rigid' that can contribute to vibration.
@@EUnit111986 solid holders are pretty bad for causing run out? Are you mad :) never had a problem using solid holders for end mills. The only time I've had a problem with run out, chatter, deflection, etc. with an end mill is when I used a collet holder . Especially during a side cutting operation . Sometimes you have to decrease the feed slightly and increase the rpms slightly. Using a collet holder does enable you to indicate the tool, but to what extent? Will this ensure stabilization of the tool during the process throughout?
@@EUnit111986 Check out the couple of videos this channel did on solid holders and TIR.
I almost never check runout on my tools unless i'm sticking out alot. Don't need to with everything being +-0.004 or 0.005
That depends. Runout = worse tool life.
@@gusjohnson6646 that also depends how bad is the runout. If it's less than .0005 which in most cases it is then the tool will be fine. Speaking about aluminum that is
Great video!
But you're still using a small endmill for finishing.
A bigger ball will give better finish at a given stepover. You only need the small ones if you have very steep creases. (And then, only for those narrow channels/pencils)
A larger endmill really wouldn't be the best to demonstrate what he's teaching.
@@thegribbs you're not wrong, but maybe this test isn't the right test then? I see them using small ball mills on big low curvature surfaces in many videos. How about parting lines as a demonstration? Cuts faster, too :-)
0.00762... I don’t know if imperial people state decimal places past the micron as a bad habit of having more than 3 numbers as normal or just completely misunderstand a superior standard of units!
Americans are generally not intimately familiar with the metric system, id bet youd make minor insignificant mistakes if you were using us customary as well
What country are you from? Just curious to see what you've done for the world. Then we can compare what the U.S. has done using the Imperial unit against the accomplishments of your superior system.
Very common to machine close tolerance metric parts here in USA.... Don't forget USA sets the world standards for everything ( except for Russia)
Those numbers are converted to be polite to metric-speaking people and that's what came out of the conversion device (he's built a custom arduino thingy that does this, there's a video somewhere).
@@keithhansen3963 That's a bit of a straw man argument. Unless you actually think the US would have accomplished less using the metric system.
I try to never use ER holders with endmills.
ไม่ควรเคาะ
Of course it works lol
First
What about tool push off when machining 🤔🤔🤔🤔 Not very scientific really 👎🏻
still forgot to add! But after such a tap and after processing, the runout returns to its place. sorry there was no check after processing