Colchester Bantam Mk1 - Eccentric Spindle, damaged rear spindle bearing

SnakeyJ

Registered
Registered
Joined
Oct 27, 2018
Messages
95
I've spent the last 3-4 evenings trying to level and align my lathe properly, but have had some marked periodic eccentricity in the spindle, that was running out and showing significant play (lateral and axial) at the spindle nose. The lathe is a 48 years old and was auctioned off by a local college last year, but have hardly used it until the last month.

I dusted off the manual and did some research on line and decided to tighten up the spindle preload (done via nut on spindle at rear of headstock) - quick diagram and description from the lathe manual.

Rear spindle bearing and preload adjustment:
rear spindle gamet bearing.jpg

Sectional drawing of spindle from Colchester manual:

236085-3aef2bb97e79db747be74fb51cf6caa9.jpg


Destructions from user manual:
236086-5aca0511ca36e21684391299cf2f6953.jpg


I only had a cheap Chinese 'fake Mitutoyo' dial indicator, which is very twitchy and gives little confidence, so I cautiously increased the preload until I was able to reduce the play and get around 2 thou periodic runout at the spindle nose. The adjustment was not smooth, in that I could turn the nut by hand to tighten, and this would make the spindle tight to turn but it loosened up again after a couple of turns of the spindle and I could then nut was loose - repeated 4-5 times before stopping.


Finishes were greatly improved, but quickly deteriorate as I work away from the headstock and at 6" the eccentricity is markedly visibly (>15 thou) . I did various checks with cam lock chucks, centres and a taper mounted collet, but these all confirmed the excessive run out.

I ordered in a 15" MT4 ground test bar and a proper 0.0001 Mitutoyu indicator, so I could properly measure and test this before making any further adjustment or investigation.

However, these have merely confirmed the errors and I had to strip back the head and have a proper look to see what's going on. On indicating both ends of the spindle I found that the error was more pronounced at the rear of the spindle and that this was the source of the eccentricity (not a bent spindle). I removed the preload nut and had a proper look - the rollers all seem to rotate correctly and looks okay externally, but looking at the inside of the bearing with an inspection camera I think I have a visible crack in the cone..... The front bearing appears ok.


There are some other marks on the cone visible in the video, but as far as I can tell these appear to be scratches.

18-11-01 230254.png



Having said all this I have reapplied the preload and the eccentricity/run out looks a lot better. I'm also not seeing any movement or expansion of this crack, even when applying some lateral force to the spindle.

I'm just waiting for a pin hooked wrench so that I can try and properly adjust the preload down, but would be keen to hear any thoughts and advice on this.

These bearings are expensive and I'm guessing there's little point in replacing the back bearing on its own. I do have Colchester part numbers; 03-912 Front and 03-913 Rear, but can't see any sizes or equivalent Gamet part numbers.

Has anyone here pulled the spindle and replaced bearings? It looks like I can probably take the spindle shaft out of the front of the headstock, If I can only loosen remove the rear bearing, but concerned about causing any more expensive damage. I have looked around online, but no clear procedure for this.

spindle layout.png

Spindlebearingadjustmentenlarged_zpsa898cf58.jpg
 
Just got a quote back on the bearings from a UK supplier (on the original Colchester part no's) @ £ 650 including carriage/tax (USD 845) - slightly better than I had feared, though if I go this route I want to double check that the spindle is true and there are no other horrors lurking in the headstock.

Alternatively I did see a couple of Mk1 headstocks being sold by breakers at around half this price - the bearings would be a bit of a punt but offers a supply of donor parts.
 
It's rare for a rear spindle bearing to fail first; I would probably get another used headstock and swap parts to save $$
you have to wonder what caused it- lack of lubrication or particulate contamination, maybe infant mortality?
mark
 
Being operated probably for years with the bearings loose did them no favors, either. I would be inclined to suspect that whatever is wrong with the bearing has been wrong since it was new. In which case, someone in QC at the factory screwed up. The problem with buying a used headstock is that you never know before-hand whether or not you are buying someone else's problems. Before doing either that or buying a new spindle tail bearing, I would pull the spindle and take it to someone else to mount it between centers and confirm that there is nothing wrong with your spindle. If there isn't, I think that I would buy the new bearing (assuming that all of the other parts on the spindle, including the head bearing, are still good). If there is something wrong with the spindle, then buy a used headstock.
 
Thanks Mark and Wa5cab - good points from both. I figured I can check the spindle for true by mounting it up in v blocks and indicate it out on a surface rotating to confirm any twist or bend. If there's any damage to the spindle then I will definitely take a punt on another head and salvage the best parts.

I'm also thinking I should tighten it all up first and test to see if I've managed to re-seat the bearing properly - depends on how the bearing looks when it all comes out.
 
Yikes! Did you say $850 United States greenback dollar bills for 2 tapered roller or ball bearings? And then the discussion turned to
maybe infant mortality?
:eek:
As in, the factory that made the bearings or the factory that assembled the lathe, had a quality control failure that incurred this problem?
someone in QC at the factory screwed up.
:faint:
That is the sort of thing I would expect from the Chinese mainland, but not Old World manufacture. I hope this turns out to be the previous owner's operator error, or neglect and abuse. That, I could live with.:big grin:
In any case, I wish you the best of luck, and I do surely appreciate your dedication to QC!
 
I meant that I couldn't think of anything that could have been done to the fully assembled machine that would have cracked the inner race. So absent believable proof to the contrary, I think that the machine must have shipped new with the problem. My first step if I were trying to investigate the cause would be to determine whether the factory had Labour problems around the time that the lathe was built. Deliberate sabotage would appear to be the most believable cause.
 
Is there something about Colchester bearings that is unique? There are millions of bearings in the world and I find it hard to believe that a high quality replacement cannot be found at a more reasonable price. With that said, it is hard to figure out how a bearing race would crack while in place unless it was cracked on installation in a bore that is not fitted well for it.

By the way, SnakeyJ, you did a really nice job of presenting your issues. Clear, concise and well thought out.
 
Westerner: I think that price was for one bearing- double yikes
 
Back
Top