Question about collets

[Tin Falcon]

here is photo of mine in use
I was doing some light milling in the lathe.
Tin

 

[Swifty]

We must not forget that collets, like 5C style, are designed to hold a specific size of part, vary by a small amount on the diameter being held and the holding pressure will be impacted.

Paul.

 

[Tin Falcon]

true a full imperial set is in 1/64 .015" increments.
Tin

 

[kd0afk]

3C collets are three times the money of 5C. I will probably go with a 3C system though. Thanks for all the help.

 

[rodw]

I thought it might be worth posting up some observations I made of runout on my 6 month old Chinese lathe. The other day I looked at spindle taper and could not see any runout. Today I measured 0.01mm. So here is what I measured. (Difference between high and low readings quoted)

Spindle internal taper 0.01 mm (0.0004")
3 jaw on 3/4" end mill 0.06mm (0.0023")
5C Spindle Collet adapter with 3/4" end mill 0.05mm (0.00197")
4 jaw holding 3/4" end mill 0.01mm (0.0004") same as spindle taper

I have been meaning to remake a collar I need at the back of the spindle adapter. It is possible I will get the collet adapter running a bit better than this when i do. I have also got a Swiss made DTI accurate to 0.002mm on its way so I may revisit these figures.

Maybe I could have got the 4 jaw better with a bit more practice. But I was able to dial it in to the accuracy of the spindle itself so I am happy with that.

So the message in this for me is the 4 jaw is the most accurate holding device I have. I was hoping the collets would have fared better so more homework required, maybe they will improve. All of these devices exceed my abilities as a machinist.

 

[platelayer]

I thought it might be worth posting up some observations I made of runout on my 6 month old Chinese lathe. The other day I looked at spindle taper and could not see any runout. Today I measured 0.01mm. So here is what I measured. (Difference between high and low readings quoted)

Spindle internal taper 0.01 mm (0.0004")
3 jaw on 3/4" end mill 0.06mm (0.0023")
5C Spindle Collet adapter with 3/4" end mill 0.05mm (0.00197")
4 jaw holding 3/4" end mill 0.01mm (0.0004") same as spindle taper

I have been meaning to remake a collar I need at the back of the spindle adapter. It is possible I will get the collet adapter running a bit better than this when i do. I have also got a Swiss made DTI accurate to 0.002mm on its way so I may revisit these figures.

Maybe I could have got the 4 jaw better with a bit more practice. But I was able to dial it in to the accuracy of the spindle itself so I am happy with that.

So the message in this for me is the 4 jaw is the most accurate holding device I have. I was hoping the collets would have fared better so more homework required, maybe they will improve. All of these devices exceed my abilities as a machinist.

Yes the only way one can improve on those it is to acquire a higher quality lathe. Many high precision lathes have the spindle bore and taper finish ground in situ so as to eradicate the manufacturing errors. Watch maker and instrument maker lathes are done that way so spindle runout is zero with the collets held to very close tolerances, and are often better than the spec. sheet would seem to indicate when new. - I have seen Swiss Schaublin lathes with virtually no runout at the workpiece, but there are always ways around these issues without spending pots of money.

 

[platelayer]

true a full imperial set is in 1/64 .015" increments.
Tin

It used to be the case that one can order any size of collet within the range from the manufacturer for a specific requirement probably at a premium though. Watchmaker collets are often available in increments of 0.1mm (0.004" approx.)

 

[platelayer]

Where do get your information from?

I should say that I was thinking in terms of collets drawn in by a drawbar, however looking at the the collets chucks such as those by Bison the advantage might not be so clear cut as these chucks do have gearing and the thread to wind in the collet giving an improved MA.

 

[rodw]

I should say that I was thinking in terms of collets drawn in by a drawbar, however looking at the the collets chucks such as those by Bison the advantage might not be so clear cut as these chucks do have gearing and the thread to wind in the collet giving an improved MA.

Better MA but more runout on the chuck maybe?:hDe:

Still, I think collets are very handy and I have started to breed a bunch of 5C ones in addition to my ER32's. The collet block is very handy. Also look at some of the stuff ArcEuro trade do in the UK. They have ER32 collet blocks and ER32 holders that fit in a 5C collet holder.

Maybe Little Machine Shop has a reasonable priced 3C collet.

 

[mu38&Bg#]

When I added a 5C chuck to my Clausing 4900 lathe with a threaded spindle nose I made a backplate for it and the 5C taper came out at .0002" TIR. I checked it a year or two later and runout had not changed. Modern production lathes use bolted on collet chucks. The method is not the problem it's the quality of the parts. I use collets anytime the material fits.

Greg

 

[kd0afk]

The whole reason for me to get a collet chuck is to have tighter control over the concentricity. Most of what I turn is 1" brass and aluminum. I don't want to use a pot chuck for that so I will stick to my original plan. The bearing play on my spindle is about .005" last time I checked and the collet will be machined in situ so I'm not worried about run out.
The 5C collets are less expensive and the collet chuck itself is less expensive than a 3C setup and for me right now affordability is a deal breaker. I will still have the 3 jaw chuck and the 6" 4 jaw chuck.
I also want to support the kid at MLA, he is doing excellent work and I like the looks of his chuck.
Platelayer- I wasn't trying to infer anything negative about your abilities or experience.

 

[platelayer]

The whole reason for me to get a collet chuck is to have tighter control over the concentricity. Most of what I turn is 1" brass and aluminum. I don't want to use a pot chuck for that so I will stick to my original plan. The bearing play on my spindle is about .005" last time I checked and the collet will be machined in situ so I'm not worried about run out.
The 5C collets are less expensive and the collet chuck itself is less expensive than a 3C setup and for me right now affordability is a deal breaker. I will still have the 3 jaw chuck and the 6" 4 jaw chuck.
I also want to support the kid at MLA, he is doing excellent work and I like the looks of his chuck.
Platelayer- I wasn't trying to infer anything negative about your abilities or experience.

Are you sure you mean .005" and not .0005". If you have .005" (5thou) then you need to do something about it as the bearings are badly worn, and would be worthwhile doing something about for much improved accuracy. If concentricity concerns you then sort the bearings out first before testing the runout of the spindle.

 

[kd0afk]

Are you sure you mean .005" and not .0005". If you have .005" (5thou) then you need to do something about it as the bearings are badly worn, and would be worthwhile doing something about for much improved accuracy. If concentricity concerns you then sort the bearings out first before testing the runout of the spindle.

Sorry, you are correct. .0005". I'm talking about the ammount of deflection I get when I flex the spindle with a bar. I am a follower of Halligan142 on youtube and I'm referring to his video on adjusting the spindle clearance.

 

[kd0afk]

I've made a decision. When I finally go with a collet system I am going with a 5C.

1. It has the capacity I need
2. They are more affordable
3. The chuck will be machined on the lathe so there will be no runout
4. I have a 4.5' bed so loss of length isn't an issue
5. I'm mainly turning brass and aluminum so I don't think that'll be an issue.
6. It looks awesome!!

Thanks for the valuable input as usual guys. I'll let you know how it turns out.

 

[platelayer]

I've made a decision. When I finally go with a collet system I am going with a 5C.

1. It has the capacity I need
2. They are more affordable
3. The chuck will be machined on the lathe so there will be no runout
4. I have a 4.5' bed so loss of length isn't an issue
5. I'm mainly turning brass and aluminum so I don't think that'll be an issue.
6. It looks awesome!!

Thanks for the valuable input as usual guys. I'll let you know how it turns out.

Yes well one makes ones choices and pays ones money I suppose, however I don't think the collet chuck is the necessarily the best solution and certainly in the UK typical ones such as the Bison one are not cheap, but I can see it might be an interim solution. Probably the best solution would be to acquire something like a Hardinge lathe in good condition with a collet closer on the headstock. This way the spindle taper and bore should be perfectly concentric, and the collet closer makes for rapid material changes. Of course obtaining a Hardinge lathe in appropriate condition is not easy and never cheap. - I suppose they are more common in the USA, but in the UK good ones are always at a premium. Anyway don't let me dissuade you because I digress. The 5C collets are made by all and sundry, but Hardinge ones almost certainly will be the most accurate providing they are in good condition.

 

[kd0afk]

Here is a follow up. I bought the kit from MLA. Unmachined it weighs a little more than my 4" three jaw chuck and way less than my 6" four jaw chuck, so no change in the deflection department. It gets machined in-situ so it will be concentric to the spindle.
I'm sorry if I disappointed anyone by going with the kit that costs $135 rather than selling my lathe putting several hundred dollars with it and buying a Hardinge. No offence but I think that suggestion was ridiculous.
I have started a thread for documenting the build so any follow-up on this subject will be there.
Thanks for the tips guys.