4.2 gearbox countershaft needle roller

I am ready to reassemble the gearbox. Countershaft Needle Rollers look good. Digital micrometer suggests that are 3mm each and no visible damage.

When needle rollers are installed on shaft and have no visible gap between any of the rollers is it safe to say I don’t need to replace them with new ones?

Does it change the answer if I am replacing the laygear cluster with a laygear from another gearbox as the original has damage on reverse/first gear.

I came to these conclusions and questions after reading many great posts on the topic on this forum.

Thank you in advance for any feedback provided!

Kevin

When changing the laygear from another gearbox, measure the diameter of the gears compared to your original. They can vary by a few thou. If you put in one that is smaller, you will have a rattly gearbox. If you try to install a larger one, it will not go together. Needle rollers give very little trouble. They are chrome plated and if still shiny, they are usually fine.

Thank you so much! I’m not a car buff. Long story on how I got into this. Lost at times and having fun.

Thanks again.
Kevin

Hello Dick,
I doubt that many (if any) DIY punters will have the equipment to measure a helical gear’s pitch diameter, the important diameter, particularly when the meshing gears have a fixed centre distance. The only diameter that most can measure is the OD of the gear, which is practically irrelevant except for clearance with the root diameter of the meshing gear.

Regards,

Bill

1 Like

You can tell the difference by measuring the OD of the gears. That is what I was referring to when I mentioned the OD of the gears.

Gears, shafts, pulleys etc. are all made to a tolerance. Important features are typically made to a tight tolerance, not so important feature to a not so tight tolerance.

The involute gear form roll together and the theoretical contact point is at the Pitch Circle Diameter so that the ratio between the mating gears is maintained and therefore the velocity of each, relative to the other is consistent.

The tolerance of a gear’s Addendum is not as tight as the Pitch Circle Diameter; accordingly the OD of a gear typically has a greater tolerance than the important Pitch Circle Diameter. The tolerance of a gear’s OD is double that of the Addendum tolerance.

Casual examination of most gears point to a looser tolerance simply based on the surface finish of the gear’s OD. Further, on many gears where the gear blank before gear cutting is machined in two operations:

  1. Face, OD turn half the length and bore - first op
  2. Face other end to length and chamfer bore - second op

there is often a slight mismatch of the two OD turning operations at the centre. In many instances, a shallow “V” groove is plunged to disguise the mismatch.

Apart from clearance with the Root of the mating gear, within reason, the OD of the gear is unimportant. You could machine a gear 0.010" smaller on the OD than the theoretically correct diameter and if the Pitch Circle Diameter is correct, it would have Zero affect on it’s operation. The OD of the gear never makes contact with anything and therefore, its tolerance can be relatively loose.

Measuring the OD of a gear to determine its size is a flawed method.

1 Like

The question remains, can you swap gears between boxes? Would it be possible to measure backlash to determine if a new combination will work properly? Or will simply any combination work because they’re all machined to the same tolerances on the working surfaces?

Bill says you can. I say you can if the gears are the same. I have run into it before where they will not go together because they are different.

Okay we definitely appreciate the information. Some is too technical for me. But I get the gist… So, is it safe to say I won’t know if swapping the laygear cluster will work until I try to reassemble the gearbox? A tough reality but not a show stopper. I won’t let it deter me from getting this sentimental restoration completed. I’m assuming if it can be reassembled without issue and no binding, then I can assume it is likely going to work? Good news is the needle rollers are probably good to go.
Thank you,
Kevin

Hello Dick,
No, I haven’t said that yet. What I did say is that you can’t rely on measuring the OD of the Gear to determine if its going to be a better fit in a fixed centres arrangement than another gear. Its tantamount to measuring the OD of a thread that’s being screw cut in a lathe, to determine if its going to fit a mating female thread. The most important aspect of the Thread is its pitch diameter, that can be measured via a number of methods; with a Thread Micrometer, or Three Wires of Best Fit, just to name a couple. The OD measurement will only tell you if its within the tolerance set for the Major Diameter of the Thread, but nothing about if its going to fit a standard, mating part or not…

With regards to swapping Lay Gear Clusters, unlike differential Crown Wheel and Pinions sets, the gears in the 4 speed gearbox used in E Types and other models of the time, aren’t matched sets; they have no serial numbers like differential gear sets. They are made to fairly close tolerances and it would have been a case of random selection of components when the gear boxes were being assembled.

The main issue when swapping a new, or alternate used gear into a used gear box, is that the original gear set tends to transfer some of the characteristics of each gear to the other over time. When you introduce a new gear, or one that has slightly different characteristic, then you may have additional noise, but maybe not.

If any of the gears of the original Lay Cluster are in good shape (even just a few teeth), I would be inclined to install the old Lay Cluster in the gearbox with the Main Shaft and measure the backlash between the best teeth of the Lay Cluster and the Main Shaft Gears. Do the same with the replacement Lay Cluster and make the call based on the result. I would be happier with the backlash being less than the original, so long as there is some backlash, rather than more than the original.

Regards,

Bill

1 Like

Thank you for the education. Original Lay Cluster has damage on all teeth of the reverse/first gear. The smallest gear on the cluster.

I will put the gearbox back together and measure backlash for gears 1-4. Then, dissemble, swap cluster, reassemble, and measure backlash.

Wish I would have known to measure backlash before taking apart:-)

I guess worst case scenario is I need a professional shop to make new cluster that meets spec. I know that will be very expensive.

Will try to update when results are in.

Than you,
Kevin

This begs the he question: how did grade the fitting of the gears to different transmission cases?

Hello Paul,
Like many precision parts that make up an assembly in mass production, critical dimensions are controlled by tight tolerances. The more critical the dimension and tight the tolerance, the more rigorous the inspection.

The finish of the gears of your typical All Synchro, four speed Jaguar gearbox indicates a controlled finishing operation after Gear Hobbing. All end and bore surfaces of the gears are ground after heat treatment; accordingly, very tight tolerances are relatively easy to maintain in production. In addition, end float in these gearboxes is regulated by spacers and thrust washers (thrust washers for the Lay Cluster)

When refurbishing the gearbox, one inspection detailed in the workshop manual is the end float of the various gears on the main shaft. In the manual it explains this being done with the main shaft in the gearbox. My method of making this inspection is with the Main Shaft removed from the gearbox housing, with the assembly arranged vertically in a press. The End Float is measured with a dial indicator with gravity assisting in obtaining the reading at one end of the End Float range. I suspect that the original assemblers would have dedicated assembly and inspection jigs to create sub assemblies ready for installation in the gearbox housing.

The diameters and pitch of bores for the Main Shaft Bearings and Lay Shaft are easily controlled in manufacture and as long as the Pitch Circle Diameter of the gears are tightly controlled, random selection of components is readily achievable.

There were many air craft using the Rolls Royce Merlin engine sidelined during WW2, due to many parts not being interchangeable. It wasn’t until Packard started producing them under license did that change. Its my understanding that most parts were redesigned with workable tolerances that allowed mass production of parts that were interchangeable between engines (random selection of parts made possible).

Regards,

Bill

Copy that, and is essentially what I thought, but did not want to assume what I thought I knew was truth.

I also am aware of the use of spacers to set end float, which is a method easily useable in a mass assembly process with less-controllable machining processes, like cast iron cases.