5c collet chuck alignment

Hello all,

As some of you may have seen on another thread of mine i have been planning some upgrades to my BH600 lathe. One of which is a 5C collet chuck, i have machined up a back plate and fitted the chuck however i am struggling to get it to fully 0. Initially i machined the back plate to be a shrink fit but it was more like a light press fit to be honest. For a first attempt thought i was happy with the fit. Once i had it snuggly on the backplate i drilled and tapped the 3 holes on my milling machine so i could guarantee a accurate fit. once everything was tight i installed the chuck on my lathe. (For those not familiar with this lathe the back plate screws on and is not on a cam system) Happy with everything i decided to check the run out. I only have imperial collets because at the time i thought it could do both imperial and metric like an ER collet but after reading an interesting thread on hear turns out they can't. Any way i purchased some 1/2" silver steel to use as a test bar as i don't have any pin gauges. Initially the run out was around 0.5mm!! Which for a collet chuck is way to much, i persevered and turned the back plate down so there was some movement and got it down to under 0.1mm run out, better but still not great. After some bashing around with a copper and hyde hammer i had it almost there but something still wasn't right. My battle initially was knowing what part of my equipment to trust? Was my clock out? If the bar bent? Is it my lathe? So today i have been having a play. A few people suggester that the way the DTI was mounted may be causing some fluctuation and the stand wasn't great as it had that adjustable bar in it so a solid bar would be better.

So what have i don't today well i have cleaned the back of the collet chuck, i have cleaned the back plate cleaned the silver steel, cleaned the collets the lot! Then i went about machining up a tool post mounted DTI holder, i ditched my cheap axminster lever DTI and machine a holder up to take my Mititoyo 0.01 plunger DTI all installed and working i made sure i went belt and braces and set the clock to band on centre, clamped the beds up as tight as i dare and locked the DTI on 0 that way i knew everything had no movement no play nothing, what ever the DTI was reading i knew it was on the bar. Might be worth mentioning i am fortunate enough that my workshop is fully insulated and remains at 20deg all the time so i know that the metal is all at room temperature as well. Quick spin on the chuck and i am getting a repeatable 0.03... Happy that its reputable but after all that effort its still out

Sorry for the ramble but i thought i would give you background so you know that i haven't just put a chuck on and expected perfection. My goal tomorrow is to get an imperial pin gauge so i can guarantee the accuracy. I would have thought silver steel is accurate enough? 1/2" Silver steel should be ground to +/- 0.015mm which is much finer tolerance that my DTI can read but thats not to say my piece isn't bent or out of tolerance for some reason? Failing that is the only solution using a hammer and keep chasing the chuck around until it zeros or is 0.03 good enough?

Thank you all for taking the time to read this, hope you can help.

Steve

I can see no problem with the measuring setup shown in your pictures.

What you need to do initially is eliminate the collet and silver steel. Move the DTI plunger so that it contacts the inside conical surface that the collet fits into. With the DTIs shown you will only be able to pick up near the open end of the taper bore but that surface should show no discernable run-out on your 0.01mm indicator.

If that runs true then you need to check each collet

Ian P

.

I agree with what Ian says.

Nick

Is it me or does that collet look like its sitting a bit far out of the chuck, mine all tend to fit back near enough flush with the end of the chuck nose.

Edited By JasonB on 21/01/2016 20:32:12

Cheers for the reply Ian, I do have a lever DTI again 0.01 but its a cheap axminster one, not that theres anything wrong with that but it seems to flicker about a lot, not as smooth as my Mitutoyo clock. I will try it again tomorrow on a more solid Eclipse base and measure the runout in the nose of the chuck. Is there any particular point to measure on as its a taper?

Thats as far back as that particular collet goes Jason, I will have to check the others? I assumed that was normal? Its a Warco collet chuck and the collets were from RDG.... Not sure if that makes a difference.

You may also be chasing something you can't get. A standard Hardinge 5C will run 0.025mm or better so assuming you are using cheaper 5Cs you are not going to get better. Even special accuracy Hardinge will only give 0.005mm so don't expect zero.

With a tapered hole there is no 'one place' that will give the result, you have to test in several places. You will soon know if the collet seating surface is not true.

The next issue is how to correct any error. If the collet fixture can be removed and replaced in the same position every time then my money would be on re-surfacing the cone with a toolpost grinder. (Dremel or whatever and patience.)

The finger DTI even if not the best will still give you a good ide of the concentricity

The Mitutoyo, which one as you show two?

Ian P

I got a customer reject 5C collet and trued the back and register in a similar way to this

This is all looking more challenging than i first expected. Reading into what you said Jason i think i have got to accept i will never get a true 0 but if i could get it between 0 and 0.01 i would be more than happy. Just hope i haven't got to go as far as boring or refacing

Crazy thought but would a taper reamer work? ARC do taper reamers (MT5) would that be a similar size and angle or is that thought a bit to out of the box?

Hi Steve,

Someone may have to correct me, but I think the taper is a little over 7deg so a MT reamer would not work.

I bought the same collets from RDG and not a single one of them has a runout of more than 0.01, some don't even register on the clock. My chuck however is a Bison one, it came with the back plate fitted, stuck it on the lathe and literally no run out. My collets are pretty much flush with the end of the chuck too.

I would clock everything in sequence. DTI the lathe spindle nose, if ok, Remove the chuck from the backplate, DTI the backplate, if ok, install the chuck, DTI the ground part on inside of chuck, if ok, then it's your collets.

Hi Steve,

I had great troubles getting my 5C collet holder to run true and mine came from the same supplier, cheap but questionable quality, methinks. If you persist you can get it to run true, undersize register and cooking foil shim stock between back plate and chuck helps. Grinding the cone is not the best answer, as it does not address the issues with the other locating surface deep inside, which must not be over size. I understand the collet holders from a certain emporium known for good service with "euro" in their name, sell better ones, bit late now though!

chriStephens

The type of collet chuck that protrudes in front of the spindle nose has the problem both of axial alignment and 'tilt'. If you measure the end of the bar it will show a combination of tilt and the straightness of the bar and how well it is seated in the collet. Repeated measurement with the bar rotated in the collet can show the bend in the bar and average out the seating error. With those compensated you get an indication of the tilt.

I tend to use 1/2" diameter tool steel (HSS Cutting tool ) blanks for checking purposes. Mainly due to having several in 6" lengths so no fear of bent silver steel. I get up to 0.001" run out from my cheap 5C collet chuck which uses a mix of newish standard accuracy collets and ones from my old South Bend lathe of 1942 vintage. In the real world it is about as good as it will get.

I have a 5C collet chuck that I purchased from Chester some years ago and found it to be quite accurate i.e. a bit better than my 3 jaw chucks but not up to the standard I expected. However I noticed that Bison 5C collet chucks have a means of finely adjusting the collet to get absolute concentricity - at a price! It works a bit like a Gritru chuck.

I therefore bought an Adjustable Backplate kit from Hemmingway which gave me the same fine adjustment for only £17.70 and a mornings work. Result - brilliant I can get any collet to run perfectly. If anyone follows this course of action I recommend that 4 fine adjustment screws are used instead of the 3 stated in the instructions as this make adjustment much easier (the same technique as centring things in a normal 4 jaw)

I've had 2 5C chucks from Warco, both bought at one of their open days from the 2nd hand/used/ex demo stock so likely to be questionable in quality/accuracy but at the minimal price paid I thought it worth taking a chance. One mounted up and ran with negligible runout, the other one had noticeable runout (a while back so can't remember how bad it was), so stripped down, cleaned and checked/measured everything and finally concluded that the taper seat had been badly ground. so with it mounted up on its backplate, used a toolpost grinder to take a light skim off the taper and it now repeats perfectly with negligible runout.

It does seem odd that your collets are not seating fully into the chuck, on both of min the collet sits flush with the nose (actually the ground one site slightly below the nose)

I am from the Colonies and have bought the same accessory. After much head scratching some turning and fitting, what I found out was that the unit needs taking apart and cleaned. I had about a 1/2 spoonful of cast iron chips in mine included free of charge. Once cleaned out it is a lot better but it does not come close to rivaling any of the industrial equipment I have ever used. I paid My money, made My choice and would not buy it again.

Hi,

As an alternative option I decided to gamble on a second hand Hardinge sjogren 5CX speed chuck to avoid the potential quality issues with Chinese. The risk with this second hand auction site route was of course wear so I bought one listed as in vgc. I am happy to say it worked out for me. I had to strip it down and clean it and put new grease in it. Run out out is less than a tenth.

Steve