Thanks, Ade, your video was most inspiring. I did things very slightly different to you, and my lathe is a WM290V. Knowing that the taper on a collet is 8 degrees, I very accurately set my compound to that, and set about cutting the taper on a billet of aluminium, as you had done. I also secured the collet chuck with a tapped hole in the end of the taper. And I also found that the rear face of the chuck was running out, which I trued. When I assembled everything, I placed the aluminium taper into the chuck, and clocked it on the exposed taper. No runout at all. Then I tried a collet on a 20mm mill cutter. 2 hundreds runout. But then, I oiled the bore, and tightened the collet a bit harder. No runout at all...!!! (And my collets are from the same stable as the lathe). I must admit, I was surprised, and happy! So is the secret using 8 degrees, rather than "following the bore taper", and for the collets, oil and tight? Who knows, I only completed this today, but have a lot more checks to make. I should have mentioned... The backing plate was running out as well, (which I trued). Altogether, not great quality parts, although I was pleased to note that after 6 years, my lathe (taper roller head bearings) did not need adjusting, and it was all running true. I know you did yours some time back, is it still running good? Cheers!
I am an amateur self taught model engineer. A few years ago I bought an old ML7 of 1949 vintage. The taper inside the headstock although far from perfect, was running within 0.015mm even after all thes years. I spent a fortune trying to find an ER 25 collet to run accurately on that lathe. I tried chuck fitting, 2MT, Back plate etc. The best I got was just over 0.06mm . I think that most of these Chinese and Indian stuff are just rubbish. They once used to be cheap so one wouldn't mind so much if they turned out to be crap but have you seen the prices of them now a days? Thanks for a very entertainig video . You have more than enough skill to produce your own collet chuck from scratch knowing that it will run true. The quality of the collets also play a major role. I would say give Veterx or Bison a try but they are far from cheap. ATB.
Interesting series Ade, the top and bottom is you cannot get everything perfect even on the clock. As you demontrate pushing on the lathe headstock gives runout, when cutting this will do exactly the same so even if the collet bore ran perfect as soon as you start cutting the preasure created would create runout due to the bearings and on many of these smaller lathes they are not tapered so very little you can do to preload them like in a bigger lathe. Maybe changing the bearings would help but your chasing a fannies hair(to quote john mills, doubleboost)
An interesting series of videos. Unfortunately I doubt you'll find any ER collets that come with some type of guarantee of decent accuracy for less than 30 quid. Rego-Fix, Bison etc are expensive for a reason. My full set of Bison ER 40'S and an R8 collet chuck cost me close to $900. Even the best unless it's pure luck will show at least some measurable run out. It should be very small and around or under .0005" for the good one's but it will be there. High accuracy ER collets guaranteeing .0002" maximum runout or less are available, but there even more expensive. You did real well by getting the low run out numbers for the lathe spindle mounted chuck that you did. You mentioned you thought the chuck was case hardened, if it was you would have seen soft steel below the case hardening somewhere between .005" - .030" deep so I strongly suspect it was induction hardened since that's the usual method today. Given the cross section with most lathe faceplates it's also common for them to warp a bit after the finish machining at the factory with cast iron face plates. As far as I know it's pretty much expected to re-machine them on the lathe there to be used on just as you did. Unless you did so and forgot to mention it I think I would have tightened the spindle bearing preload a touch before doing anything else. It may or may not have made any difference in the finished results, but one of the first tests done when putting a new lathe into service and before it machines it's first part is checking the bearing clearances and adjusting if needed. That very sharp aluminum specific carbide tip worked to end your chatter problem. But a lot of times wrapping the outer rim of the part with a large elastic band, rubber strap, or even multiple wraps of wire solder can fix the chatter when nothing else does. It's a very old machinist trick that changes the harmonics out of the range it's chattering at. Reducing the rpm and / or upping the feed rate can also help sometimes.
the whole thing turned out to be a bit of an adventure . i did sort the spindle bearings out and i like the ideas for vibration suppression , cheers Ade.
I appreciate why he used a back plate, as he had to drill it to fit the chuck, why not drill the collet chuck, with the appropriate drills to fit the lathe spindle, thus eliminating another source of run out. That's what I was thinking. Barry Australia
i could of drilled it to match but it would need to be tapped , it was to hard to tap . the back plate was machined in place so was not a source of error , thanks for your input though
Well another great little series, and as you say you pays your money and get what you get, but that is half the fun of having an hobby lathe, most people who are on the internet whether working in metal or wood, seem to spend more time making gadgets and jigs than producing objects, we seem to make things as an aside, you on the other hand are producing things to order, keep it up, I like to watch this kind of thing, just in case I ever come across the need to do it myself, you are doing the spade work for me, many thanks for the upload. Bob
With ER collets, the nut is as important as the rest of the chuck. These are front tightening chucks. They tighten against the front of the collet.If that’s off, everything will be off.
Heya, even good collets are treated as consumables in the machine shop due to the abuse they take. Once you get the run-out in the taper as low as possible, next things to check are the trueness of collet chuck thread and nut. Use a bump tool to align the stock and see if that helps, see my recent video on that and you can get very good radial accuracy with that approach.
If you write to John Mills over at Doubleboost, he might make you a collet spanner with his plasma cutter that fits your adaptor plate. Might be a bit better than a tommy bar.
A few comments. Having done this a number of times for 5c and ER collets on my Myford ML 7 I would suggest your adapter plates holes being the size of your three or four jaw chuck key cross bar. That way you don't need another tool in order to change the collets or stock or remove the chuck from the adapter plate. Try a different collet nut to see if that is causing the runout. That one might be the source of your repeatable 0.002" discrepancy. I found with my Chinese collets, from Banggood, they were filthy with swarf from the slots being cut. A flat utility knife blade and a small triangular file sorted most of my problems. Took some time to do them all but worth it in the end. Is your collet chuck blackface butting up to your adapter plate leaving the inner recess clear of the adapter spigot's face? Any contact there will nullify the machining you did on the Chuck's blackface.
Instead of tommy bars: You have bolt heads hanging out the front of the mounting flange. Stand a post of scrap, standing under a bolt head at 9:00 position (to open) or 3:00 (to close), to the lathe bed. Torque on the nut will be resisted without need for another hand.
I was thinking of building an ER32 collet chuck for my South Bend 9A lathe so have watched a few videos on RUclips and came across your series which was very good but I must admit put me off the idea somewhat. I already have a Bangood ER25 collet set with an MT2 chuck for my Warco WM14 mill so I thought I would try that in the South Bend to see what sort of runout I would get with that. I had to use a MT3 - MT2 sleeve in my lathe spindle but when I measured the runout with a 10mm collet and milling cutter I got 15 microns or 0.00059" which I thought was pretty good. Maybe the lathe collet chuck is back on the cards. Sorry you had problems but very interesting videos.
Hello Ade, I made a chuck to screw directly on the nose of my ml 7, have exactly the same discrepencies with the collets, chased myself up my own ****** before I figured its the collets, be interested to see what you get when you get your sunday best collets!ATB,Cliff
I would have a few more thou clearance in the recess and tap the collet chuck true or the work piece ; and tighten up the three bolts dont take long to do this
hello ades/ what you can try is... in de flens of it... dril ant take 3 or 4 holl,s ?? bolds of m8.. ( inbus ) ( in dutch... stift inbus) ant fine tuning the head.. ???? if it is 0.... 2 zero.. its ok.. ant you cant replease it after you job its don.. (( you dont need a hi speed rpm?? )) the zero mark on the head en center is ewri time the best point of it messer white a dif axels,??? you need that zero !!!!!! tanks for watching gr
YES . IT DEPENDS ON HOW IT CLOCKS UP . oops caps lock on . perhaps protect the collet block with aluminium shims and clock a square one in the four jaw is best . Cheers Ade
Hi Ade, thanks for this series, very interesting. It's now 2 years after you posted, and I've been looking through your later videos to see if you ever improved the setup. I see you sometimes use a MT3 ER chuck, but you also recently used use a bolt-on collet chuck that I think is this one. Did you get a different set of collets? Or did you get a different collet nut? Or are you still living with the 2 thou runout? I suspect you've updated something, but maybe it's in one of your Shed chats and I can't find which one - if you could point me to any video showing any upgrade of the set-up, I'd be grateful.
I don't use it much but i changed the collet nut and it was far better. I am going to change it to fit my rotary table on the mill where I will probably use it more often . Cheers Ade.
What were you measuring it with? I’m assuming you’re metric. 0.02mm indicator, 0.01? Sometimes you luck out too. That’s the problem with these unknown brand tools. They’re very variable.
Able, Why use a backing plate could you not bolt straight to the spindle face, I have a 7 x 14 Real Bull mini laity with a 100mm spindle the collet chuck bolts direct to the spindle. This removes the risk of the backing plate adding any inaccuracies? Just a thought. Barry Australia
This collet chuck was made to go on a rotary table & the hole pattern is completely different as well to the lathe mounting spindle, that is why he had to use a backing plate.
@@relentlessbandit I bought a ball bearing nut Er40 in US, I wan to try, I spend with import tax and transport more then 60 euro, but if will make the difference I will be happy, I will let you know guys
Thanks, Ade, your video was most inspiring. I did things very slightly different to you, and my lathe is a WM290V. Knowing that the taper on a collet is 8 degrees, I very accurately set my compound to that, and set about cutting the taper on a billet of aluminium, as you had done. I also secured the collet chuck with a tapped hole in the end of the taper. And I also found that the rear face of the chuck was running out, which I trued.
When I assembled everything, I placed the aluminium taper into the chuck, and clocked it on the exposed taper. No runout at all. Then I tried a collet on a 20mm mill cutter. 2 hundreds runout.
But then, I oiled the bore, and tightened the collet a bit harder. No runout at all...!!! (And my collets are from the same stable as the lathe). I must admit, I was surprised, and happy! So is the secret using 8 degrees, rather than "following the bore taper", and for the collets, oil and tight? Who knows, I only completed this today, but have a lot more checks to make.
I should have mentioned... The backing plate was running out as well, (which I trued). Altogether, not great quality parts, although I was pleased to note that after 6 years, my lathe (taper roller head bearings) did not need adjusting, and it was all running true.
I know you did yours some time back, is it still running good?
Cheers!
I am an amateur self taught model engineer. A few years ago I bought an old ML7 of 1949 vintage. The taper inside the headstock although far from perfect, was running within 0.015mm even after all thes years. I spent a fortune trying to find an ER 25 collet to run accurately on that lathe. I tried chuck fitting, 2MT, Back plate etc. The best I got was just over 0.06mm . I think that most of these Chinese and Indian stuff are just rubbish. They once used to be cheap so one wouldn't mind so much if they turned out to be crap but have you seen the prices of them now a days? Thanks for a very entertainig video . You have more than enough skill to produce your own collet chuck from scratch knowing that it will run true. The quality of the collets also play a major role. I would say give Veterx or Bison a try but they are far from cheap. ATB.
Interesting series Ade, the top and bottom is you cannot get everything perfect even on the clock. As you demontrate pushing on the lathe headstock gives runout, when cutting this will do exactly the same so even if the collet bore ran perfect as soon as you start cutting the preasure created would create runout due to the bearings and on many of these smaller lathes they are not tapered so very little you can do to preload them like in a bigger lathe. Maybe changing the bearings would help but your chasing a fannies hair(to quote john mills, doubleboost)
All that work and you get shafted by a collet. Still you got that collet chuck running very nice.
Keep up the good work
An interesting series of videos. Unfortunately I doubt you'll find any ER collets that come with some type of guarantee of decent accuracy for less than 30 quid. Rego-Fix, Bison etc are expensive for a reason. My full set of Bison ER 40'S and an R8 collet chuck cost me close to $900. Even the best unless it's pure luck will show at least some measurable run out. It should be very small and around or under .0005" for the good one's but it will be there. High accuracy ER collets guaranteeing .0002" maximum runout or less are available, but there even more expensive. You did real well by getting the low run out numbers for the lathe spindle mounted chuck that you did.
You mentioned you thought the chuck was case hardened, if it was you would have seen soft steel below the case hardening somewhere between .005" - .030" deep so I strongly suspect it was induction hardened since that's the usual method today. Given the cross section with most lathe faceplates it's also common for them to warp a bit after the finish machining at the factory with cast iron face plates. As far as I know it's pretty much expected to re-machine them on the lathe there to be used on just as you did. Unless you did so and forgot to mention it I think I would have tightened the spindle bearing preload a touch before doing anything else. It may or may not have made any difference in the finished results, but one of the first tests done when putting a new lathe into service and before it machines it's first part is checking the bearing clearances and adjusting if needed. That very sharp aluminum specific carbide tip worked to end your chatter problem. But a lot of times wrapping the outer rim of the part with a large elastic band, rubber strap, or even multiple wraps of wire solder can fix the chatter when nothing else does. It's a very old machinist trick that changes the harmonics out of the range it's chattering at. Reducing the rpm and / or upping the feed rate can also help sometimes.
the whole thing turned out to be a bit of an adventure . i did sort the spindle bearings out and i like the ideas for vibration suppression , cheers Ade.
A great set of videos demonstrating the worth of buying decent gear in the first place!
I appreciate why he used a back plate, as he had to drill it to fit the chuck, why not drill the collet chuck, with the appropriate drills
to fit the lathe spindle, thus eliminating another source of run out. That's what I was thinking.
Barry
Australia
I suppose knowing the chuck was hardened & didn't know if it was drillable at that stage.
i could of drilled it to match but it would need to be tapped , it was to hard to tap . the back plate was machined in place so was not a source of error , thanks for your input though
Well another great little series, and as you say you pays your money and get what you get, but that is half the fun of having an hobby lathe, most people who are on the internet whether working in metal or wood, seem to spend more time making gadgets and jigs than producing objects, we seem to make things as an aside, you on the other hand are producing things to order, keep it up, I like to watch this kind of thing, just in case I ever come across the need to do it myself, you are doing the spade work for me, many thanks for the upload. Bob
more than welcome . Ade.
With ER collets, the nut is as important as the rest of the chuck. These are front tightening chucks. They tighten against the front of the collet.If that’s off, everything will be off.
Heya, even good collets are treated as consumables in the machine shop due to the abuse they take. Once you get the run-out in the taper as low as possible, next things to check are the trueness of collet chuck thread and nut. Use a bump tool to align the stock and see if that helps, see my recent video on that and you can get very good radial accuracy with that approach.
another nice video ade, thanks for sharing, look forward to the next
regards...keith in swindon
Thanks for the vidios, enjoyed them. I've just bought a Warco WMV 240 and a WM 14 mill i've got all this to come!
Great video again, look forward to the next one.
If you write to John Mills over at Doubleboost, he might make you a collet spanner with his plasma cutter that fits your adaptor plate. Might be a bit better than a tommy bar.
A few comments.
Having done this a number of times for 5c and ER collets on my Myford ML 7 I would suggest your adapter plates holes being the size of your three or four jaw chuck key cross bar. That way you don't need another tool in order to change the collets or stock or remove the chuck from the adapter plate.
Try a different collet nut to see if that is causing the runout. That one might be the source of your repeatable 0.002" discrepancy.
I found with my Chinese collets, from Banggood, they were filthy with swarf from the slots being cut. A flat utility knife blade and a small triangular file sorted most of my problems. Took some time to do them all but worth it in the end.
Is your collet chuck blackface butting up to your adapter plate leaving the inner recess clear of the adapter spigot's face? Any contact there will nullify the machining you did on the Chuck's blackface.
Instead of tommy bars:
You have bolt heads hanging out the front of the mounting flange. Stand a post of scrap, standing under a bolt head at 9:00 position (to open) or 3:00 (to close), to the lathe bed. Torque on the nut will be resisted without need for another hand.
It never gets old
Nice one Ade, looking forward to the next one!
I was thinking of building an ER32 collet chuck for my South Bend 9A lathe so have watched a few videos on RUclips and came across your series which was very good but I must admit put me off the idea somewhat. I already have a Bangood ER25 collet set with an MT2 chuck for my Warco WM14 mill so I thought I would try that in the South Bend to see what sort of runout I would get with that.
I had to use a MT3 - MT2 sleeve in my lathe spindle but when I measured the runout with a 10mm collet and milling cutter I got 15 microns or 0.00059" which I thought was pretty good. Maybe the lathe collet chuck is back on the cards. Sorry you had problems but very interesting videos.
Ade, what a palava, wouldn't you have been better off, making your own faceplate/ER32 collet chuck?
with hindsight yes . but it was fun fighting it . Cheers Ade.
Very nicely done ! Thanks for sharing
Hi Ade…did you try turning the MT3 chuck to various positions in the sleeve to see if the readings changed?..🤔
Hello Ade, I made a chuck to screw directly on the nose of my ml 7, have exactly the same discrepencies with the collets, chased myself up my own ****** before I figured its the collets, be interested to see what you get when you get your sunday best collets!ATB,Cliff
I would have a few more thou clearance in the recess and tap the collet chuck true or the work piece ; and tighten up the three bolts dont take long to do this
keep up the good work,and thanks for the info,
hello ades/ what you can try is... in de flens of it... dril ant take 3 or 4 holl,s ?? bolds of m8.. ( inbus ) ( in dutch... stift inbus) ant fine tuning the head.. ????
if it is 0.... 2 zero.. its ok.. ant you cant replease it after you job its don..
(( you dont need a hi speed rpm?? ))
the zero mark on the head en center is ewri time the best point of it
messer white a dif axels,???
you need that zero !!!!!!
tanks for watching
gr
Great video ade I’ve got some ok German collets . Use on the mill and lathe
silly question....could you use a collet block held in a 3 or 4 jaw chuck ...to get the collet advantage of holding small round parts?
YES . IT DEPENDS ON HOW IT CLOCKS UP . oops caps lock on . perhaps protect the collet block with aluminium shims and clock a square one in the four jaw is best . Cheers Ade
@@ade63dug thanks Ade, i measure the runout and report back. but its a quick way of holding a small brass part
Hi Ade, thanks for this series, very interesting. It's now 2 years after you posted, and I've been looking through your later videos to see if you ever improved the setup. I see you sometimes use a MT3 ER chuck, but you also recently used use a bolt-on collet chuck that I think is this one. Did you get a different set of collets? Or did you get a different collet nut? Or are you still living with the 2 thou runout? I suspect you've updated something, but maybe it's in one of your Shed chats and I can't find which one - if you could point me to any video showing any upgrade of the set-up, I'd be grateful.
I don't use it much but i changed the collet nut and it was far better. I am going to change it to fit my rotary table on the mill where I will probably use it more often . Cheers Ade.
Nice work !
is there any way to check the bearing surface on the inside of the collet nut? this could distort the direction of pull when tightened.
very good question , i will have to look into it , cheers Ade.
Thanks for the video.
Shop-apt look at the web sit they do er32 collets, i buy carbide end mills from them and they are great
APT are all chinese import also
Congratulations
After watching your video I had to go and check my exact same 3 MT collet with a 3/4" end mill. virtually zero run-out. You have another issue.
What were you measuring it with? I’m assuming you’re metric. 0.02mm indicator, 0.01? Sometimes you luck out too. That’s the problem with these unknown brand tools. They’re very variable.
Able,
Why use a backing plate could you not bolt straight to the spindle face, I have a 7 x 14 Real Bull mini laity with a 100mm spindle
the collet chuck bolts direct to the spindle. This removes the risk of the backing plate adding any inaccuracies? Just a thought.
Barry
Australia
This collet chuck was made to go on a rotary table & the hole pattern is completely different as well to the lathe mounting spindle, that is why he had to use a backing plate.
Cheap equipment is like cheap wine. At times quality may be very good but inconsistency can cause you to gag.
I found the nuts contribute to the runout. My two chinese nut vary a lot in the run out.
Buy a good quality nut.
I have to agree. I bought an "ball bearing nut" and it almost took readout to zero.
@@relentlessbandit I bought a ball bearing nut Er40 in US, I wan to try, I spend with import tax and transport more then 60 euro, but if will make the difference I will be happy, I will let you know guys
If work I will do also for my mill er32
Also
ruclips.net/video/KkUzH1rU848/видео.html
ruclips.net/video/BWqr43rUIC4/видео.html
2:54 .showing 400 thou?? you serious?? .400 thou is a tons!!
Abe talks in metric/imperial 400th of a mm