• Spring 2024 meetup in Calgary - tentative date Saturday, April 20/2024. Other regions are also discussing meet ups. If you want one in your area get going on organizing it! discussion
  • We are having email/registration problems again. Diagnosis is underway. New users sorry if you are having trouble getting registered. We are exploring different options to get registered. Contact the forum via another member or on facebook if you're stuck. Update -> we think it is fixed. Let us know if not.
  • Spring meet up in Ontario, April 6/2024. NEW LOCATION See Post #31 Discussion NEW LOCATION

Shaft Repair and Gear Bushing Replacment (Colchester Master 2500 Lathe)

RobinHood

Ultra Member
Premium Member
This shaft is out of the headstock from my Cholchester Master 2500 lathe. They call it the "second shaft". It is the only one that has a sliding gear assembly with a plain bearing that is not splined to its shaft. This is because it can run faster/slower than the shaft it sits on, depending on gear selection. This gear set is also the most heavily loaded one since it either drives the low speed or the high speed bull gear on the spindle.
This is the second shaft with the sliding compound gear
IMG_0208.JPG

Here is the overview of the headstock. You can see the second shaft just above the spindle
IMG_1098.GIF

This is the problem
IMG_0209.JPG
The gear bushing has started to score the shaft. In particular the hardened bushing retaining rings.

These guys...
IMG_0211.JPG IMG_0212.JPG
The one on the left is pretty much unscathed but is very close to also start digging into the shaft because the bushing is worn down. The right one shows clear evidence (the shiny inside surface) of it contacting and marring the shaft.

Here is why (i believe)...
IMG_0210.JPG
The inside of the compound gear is lined with a bronze bush. It does have two oil grooves in it, but somehow it still managed to start chewing itself to bits. The side facing us engages the low speed bull gear => low speed, high torque, high load on one side of the gear. With possible poor fit from new and poor lubrication, this side started wearing down faster, eventually allowing the bushing retaining ring to make contact with the shaft, creating the groove seen on the shaft in the pic above.

The repair:
First new centers needed to be cut into the end of the shaft as the existing ones were not concentric with the bearing journals ( i have no idea why; the shaft was cylindrically ground at manufacture so i first assumed the centers would be good, but not so). I don't have a cylindrical grinder or a cylindrical grinding attachment for the surface grinder or T&C grinder big enough for a shaft this long. But i have a tool post grinder. So that's what i used to fix the shaft. Final dimension is not important in this case as i will be making the gear bushing ID match the shaft. Key requirements: cylindrical and no taper.

IMG_0213.JPG
Here we see the shaft after the first pass. The areas most worn are clearly visible (dull coloured). This pass was also a taper verification pass. I had to move the tail stock just a tad to correct.

And the final product...
IMG_0214.JPG
We finished with about 0.0002" taper over the 4.25" length. I ended up playing around with the tail stock lock tension (it moves the quill just a bit), to get there during the final passes.
No lathe dogs were used to drive the shaft. Just the friction on the center in the spindle and witness marks to confirm no slipping. The grinding passes were such light DOC (partly because it is such a small TPG), that it was not necessary. The live center in the tail stock has back to back spring washers to take up the expansion of turned parts. It has about 2 thou axial movement. I used about 1 thou of that as preload between the centers. Works great.

Overall i took off 0.0036" after the initial pass to clean up the shaft.
IMG_0215.JPG

Next will be the removal of the old bushing in the gear and making the new one...
 

PeterT

Ultra Member
Premium Member
That's pretty delicate setup work, bravo. Was .0036" the resultant diameter difference from contact? About how much DOC per pass was the grinder happy with? What grit is the stone?
 

RobinHood

Ultra Member
Premium Member
0.0036” was the amount i took off after the first pass. The first pass was about 0.0008 to 0.001”deep. So the total removed from the original OD was just shy of 0.005”.

DOC of about 0.001 to 0.0015” are possible. I stuck with 0.0004 for the most part as i did not want the wheel to start digging in and make matters worse (ask me how i know.... it gets ugly in a hurry). The TPG is only 1/4 HP. And the 3” wheel is the max size allowed. In this case i used a 46 grit wheel.
 

RobinHood

Ultra Member
Premium Member
Some progress....

Pushed out the bushings in the gear. Here is what we found:
IMG_0216.JPG
Should have known that they would not just use plane Jane bronze bushings... They are special GLYCODUR F super low friction, steel on the outside, sinter tin bronze in the middle with a PTFE (Teflon) impregnated inner layer. (www.glycodur.de).

That explains the black inside on two out of the three. The middle one was about half and half - some black coating and the rest bronze and the most worn one was just bronze. Oh, yes, they come split like that from the factory. See the two extreme ones below.
IMG_0217.JPG

I was still puzzled how come we have this uneven wear and the one side retaining ring chewing the shaft. Here is what i think happened, having further examined the compound gear: the gear teeth are hardened, the rest is relatively soft. Somehow there is a hard area inside the bore of the gear on one side -> heat treat error? See the pic below, note the discoloured area about the size of a thumb print
IMG_0219.JPG IMG_0220.JPG IMG_0221.JPG
The final bore size and snap ring groove was machined AFTER the gear teeth were made (and heat treated). The grooving tool severely deflected as it was going through the hard area and the depth of the groove is much less than required in that area (clearly visible by naked eye; as well as the tool chatter that cut the bore). QC did not catch the mistake. The gear was installed on the shaft with the bushing retaining ring just barely clearing the shaft OD. As time went on - i am guessing within hours of first operating the machine - the Teflon in the bushes settled (the run-in period; the manufacturer mentions that). The snap ring started marring the shaft and left a rough surface finish. As the gear is shifted from Hi to Lo, the PTFE coated bushings had the Teflon abraded and thus wore to the state we see them today.

I'll use a Dremel tool with a narrow disc to grind the snap ring groove to size.

Because of the wear on the shaft and the subsequent refinishing and now smaller OD, i can't replace the bushings with GLYCODUR F ones since they only come in certain standard sizes.

So i chose to use Olite Bronze bushing material. I started by turning the OD for a press fit into the gear bore
IMG_0218.JPG
Yup, i used an adjustable reamer as a mandrel (i have no expanding mandrels). Worked great. Then i pressed them into the gear and bored them to size. That all went well, until....

The bore turned out a real nice sliding fit on the newly ground shaft, but the gear set has a massive run out on the formerly damaged side. Turns out that even though i used the "good" side of the gear bore to indicate it into the 4-jaw, there is a problem with the other end. I will have to press out the bushings and start again - this time ignoring the gear bore altogether and work with the gear teeth as a radial and axial reference. Because i need the teeth to mesh properly with their mates. The bushings inside the bore will be whatever with non-concentric IDs and ODs. But that's fine, as long as the gear teeth run true.

That is coming up....
 

PeterT

Ultra Member
Premium Member
Hard to know but one possibility is may have been turned between centers to nominal oversize dimensions, then heat treated, then centerless ground to final dimension which doesn't reference the centers anyway. Maybe they just assumed shafts would be replaced & not re-dressed.
 

RobinHood

Ultra Member
Premium Member
@historicalarms and @PeterT excellent observation and great answer! I totally forgot about that, even though i mention having to recut the “bad” centers in the shaft before i could regrind it.....
 

RobinHood

Ultra Member
Premium Member
Here is how i solved the gear bore run out problem: i took some aluminum and made a special collet for the 3-jaw that was bored insitu like Joe Pie shows in this video.
.

A ring was cut out of a slug
IMG_0224.JPG

The rough bore was machined before the two relief slots and the one through slot were cut on the mill using a slitting saw (i don't have a vertical band saw)
IMG_0231.JPG

The ring was re-inserted into the 3-jaw and finish bored. The shim piece then removed and the gear clamped in the collet. That caused the outside to run perfectly true. The oversized oilite bushings had already been pressed in. Now all i had to do was bore the ID to size for a good sliding fit on the reground shaft.
IMG_0232.JPG

That turned out very well and there is no appreciable runout of the gear set any more.

Another "problem" with the same gear was the fact that it was not meshing well with its mates as far as axial location was concerned. The pics below show the offset when in high range - both to the driven gear and the driver gear
IMG_0202.JPG IMG_0203.JPG

When in low range, the gear set was making contact with the side of the spindle gear, as seen here
IMG_0205.JPG

The shift linkages are not adjustable on this lathe. I decided to make the shift fork with a 0.100" offset pin and replace the factory centered pin fork. The 100 thou was arrived by placing a gauge block between the gears in low range and checking for alignment like so
IMG_0206.JPG

The new fork was made from some round stock bearing bronze i had on hand. I first machined an oversize pin (3/4") onto the stock and then parted off to overall length of the shift fork plus some extra as i wanted the pin to end up longer than factory for a better fit in its shift arm. The part was then transferred to the mill with the rapid indexer in the vertical position and held in a 5C collet
IMG_0239.JPG

The round stock was machined square and the center was cut out to accept the gear with a close sliding fit. I used the rapid indexer because that way i had to set the tool offset only once and then just spin the part to end up with a square. The next operation was carried out in the vise with the boring head reducing the pin diameter from 3/4" to 3/8". The vise held the square part of the fork with the spindle initially centered on the pin. Next i moved over 100 thou in X to form the offset pin.
IMG_0241.JPG

And the final product. The original is on the right. The fork engages on the big part of the compound gear.
IMG_0243.JPG

First test runs indicate that the modification was a success. No more contact in low range and much better alignment in general.
 
Last edited:

PeterT

Ultra Member
Premium Member
Well done on multiple fronts.
Investigating / improving / remedying vintage machines with issues like this is a skill set unto itself IMO.
 

wharris1

william
Premium Member
I just came across this
thank you for the information I may have something similar to look for on my colchester triumph
 

RobinHood

Ultra Member
Premium Member
You are welcome.

Let us know what you find and how you resolve the issue. Always interested in new ways of investigating/repairing problems.
 
Top