Technical Which way around?

Currently reading:
Technical Which way around?

Does this perhaps go back to the fifth gear modification. I think that once tightened, the total length of all the components that are placed on the shaft after the front (ball) bearing will affect how far the pinion gear protrudes. So, if any of mis-sized parts or modifications have added to the length, the effect will be to pull the shaft, and hence the pinion gear towards the front, which will obviously make increase on that 75mm.
 
Last edited:
If you haven't already seen this video by Andrew, a notable member here; (sorry, I can't remember his forum avatar so I can reference him), you may find it very useful.
It reinforces my suspicions about the sizing of your components. On a fifth gear, he discusses machining a small amount from the speedo gear, and the use of different sized washers according to which variation of the gearbox you have. Is it possible that this hasn't be done properly?

 
If you haven't already seen this video by Andrew, a notable member here; (sorry, I can't remember his forum avatar so I can reference him), you may find it very useful.
It reinforces my suspicions about the sizing of your components. On a fifth gear, he discusses machining a small amount from the speedo gear, and the use of different sized washers according to which variation of the gearbox you have. Is it possible that this hasn't be done properly?


I’ve done some measuring and had a think about what does and doesn’t make a difference, be grateful if you could view the video because it was difficult to put this all into words. I’d be grateful gents to let me know your thoughts, thanks.

 
Last edited:
The video was a good idea, and I see what you mean about there being nothing to keep the shaft in place.
I have attached a quick sketch of the principle I'm working from. Once the shaft is assembled and the locknut is tight, and trapping the gears and bushing on the external part of the box, the shaft can't move backwards because it is tight up to the centre race of the ball bearing, and a screwplate locates that bearing. It can't move forwards because of the lip recess machined in the box casing.
Therefore, the as the distance occupied by those gears etc increases, the pinion gear is pulled further away from the crownwheel centreline. As the distance decreases, the pinion is pushed nearer to the centreline. It doesn't alter the fact that with the internal gears being a fixed length, something would have to fit the gap created behind the pinion.
From the factory, the exact dimensions of the components had been measured very precisely, so that there would be only a minor deviation from perfection, which could be taken up by shims. But your box has newly designed components which could easily have upset this balance, and which may not even have been machined to fit.
 

Attachments

  • 20251031_152751.jpg
    20251031_152751.jpg
    920.1 KB · Views: 12
...will do...back soon. :)
The video was a good idea, and I see what you mean about there being nothing to keep the shaft in place.
I have attached a quick sketch of the principle I'm working from. Once the shaft is assembled and the locknut is tight, and trapping the gears and bushing on the external part of the box, the shaft can't move backwards because it is tight up to the centre race of the ball bearing, and a screwplate locates that bearing. It can't move forwards because of the lip recess machined in the box casing.
Therefore, the as the distance occupied by those gears etc increases, the pinion gear is pulled further away from the crownwheel centreline. As the distance decreases, the pinion is pushed nearer to the centreline. It doesn't alter the fact that with the internal gears being a fixed length, something would have to fit the gap created behind the pinion.
From the factory, the exact dimensions of the components had been measured very precisely, so that there would be only a minor deviation from perfection, which could be taken up by shims. But your box has newly designed components which could easily have upset this balance, and which may not even have been machined to fit.
I like your work 😆. Shame I can’t do a certain Aussie joke from our era any more 🙄
Thanks for this stuff, really great as it got me thinking. As the thing is in front of me I can see the nuances, so also with a sketch I’ll try to explain what’s going on as it is a devil to get your head around.
What happens with the components right hand side of the case isn’t relevant. The floating bearing has no lip in the casing so the protrusion of the pinion is dictated by the bearing, shim and cluster as a single block. But the cluster always stays locked against the ball bearing on the right hand side.
By adding a shim between the cluster and the roller bearing, it pushes the roller bearing and hence the pinion further to the left.
 

Attachments

  • IMG_2377.jpeg
    IMG_2377.jpeg
    1.1 MB · Views: 11
  • IMG_2381.jpeg
    IMG_2381.jpeg
    587 KB · Views: 9
Last edited:
I like your work 😆. Shame I can’t do a certain Aussie joke from our era any more 🙄
Thanks for this stuff, really great as it got me thinking. As the thing is in front of me I can see the nuances, so also with a sketch I’ll try to explain what’s going on as it is a devil to get your head around.
What happens with the components right hand side of the case isn’t relevant. The floating bearing has no lip in the casing so the protrusion of the pinion is dictated by the bearing, shim and cluster as a single block. But the cluster always stays locked against the ball bearing on the right hand side.
By adding a shim between the cluster and the roller bearing, it pushes the roller bearing and hence the pinion further to the left.
You've obviously been to art school, and I obviously haven't. ;)
So, if you put a (say) 6mm shim behind the roller bearing, the pinion would move out 6mm. But when tightened, the nut would be 6mm further towards the end of the thread, and maybe beyond alignment with the split pin.
Equally, if 6mm was removed from somewhere on the end cluster, and the nut tightened, the plnion would move out by the same amount. But there would then be a gap allowing the roller bearing to move back and forth.
Are you certain about the measurements? If the shaft, casing and main clusters are original, and a matched set it's starting to look impossible that the measurements coold be so far out.
You're possibly going to have dry assemble everything as it is so that you can check using the old method of blueing the gear contact faces.
 
This was on Fiat Forum in 2024 @Bleeding Knuckles mentioned so may be able to advise tre setting up diff?
"Hi Keith, I have a 9/39 Diff in my 126 box having taken a while over the merits of a five speed conversion. In the end I decided to go with a diff gear set from a 126 BIS gearbox. The swap over was simple enough, BUT the setting of the crown wheel backlash and the turning torque setting of the drive shafts took a little time to get right. Providing you are willing to do the job you will see some 60mph at approx 4000rpm, which for me is a steady cruise with a tuned engine."

Is this the people that supplied the five speed or maybe have more details on yours. https://www.dangelomotori.it/en/pro...-fiat-500-f-l-r-and-fiat-126-with-126-engine/

Don't know if any of this relevant.I found it here. https://webshop.fiat500126.com/en/sites/430

Differential​

Final drive ratio
8/39
Backlash
0,08 - 0,13 mm

Tightening torques in mkg​

Nut on bevel gear
5
Crown wheel bolt
4,5
Thanks, useful specs those which I'll need - hopefully soon! My 5 speed conversion, fitted by the previous owner (or even prior) is a Nanni kit made in Bologna. I don't think they're that different though. If I was working from a std box ,for simplicity I'd certainly consider the 9/39. But apart from what is 99.9% a pinion bearing issue, the 5 speed components themselves been faultless (fate-tempting aside) and means I can do a steady cruise at 80mph as it is a VERY long overdrive. I don't boot it in 5th but my motor has very decent torque and will still accelerate in 5th without chucking its guts out for the last 10 years. But now I've said all this, expect another post in the spring with tears :rolleyes:
 
You've obviously been to art school, and I obviously haven't. ;)
So, if you put a (say) 6mm shim behind the roller bearing, the pinion would move out 6mm. But when tightened, the nut would be 6mm further towards the end of the thread, and maybe beyond alignment with the split pin.
Equally, if 6mm was removed from somewhere on the end cluster, and the nut tightened, the plnion would move out by the same amount. But there would then be a gap allowing the roller bearing to move back and forth.
Are you certain about the measurements? If the shaft, casing and main clusters are original, and a matched set it's starting to look impossible that the measurements coold be so far out.
You're possibly going to have dry assemble everything as it is so that you can check using the old method of blueing the gear contact faces.
Correct on the spacer behind the roller bearing, but although the pinion would be where it needs to be in relation to the crown, the bearing housing would be half way out the alloy casing ….. unless I deployed my (crazy?!) idea of putting the spacer between the pinion gear and bearing.
You’re equally right about the split pin, I’d need to drill a new hole as per Andrew’s experience with this spacer either way. Because my split pin is fine where it is it does beg some questions vs Andrew’s install, but my kit is by a different manufacturer.
I’m at the point now where I’m going to put this 75mm figure out of my mind, to that end I’ve bolted the diff onto the box without the rear casing so I can see whether pinion and Crown wheel mesh centrally. Very early days but it doesn’t look 5mm out. After that I can look at the lash etc.
 

Attachments

  • IMG_2375.jpeg
    IMG_2375.jpeg
    831.8 KB · Views: 11
  • IMG_2374.jpeg
    IMG_2374.jpeg
    1.4 MB · Views: 11
  • IMG_2372.jpeg
    IMG_2372.jpeg
    1.6 MB · Views: 13
  • IMG_2369.jpeg
    IMG_2369.jpeg
    1.3 MB · Views: 12
  • IMG_2376.png
    IMG_2376.png
    5.3 MB · Views: 11
Hat's off to you guys for persevering, nothing simple in this life.
I loved the Aussie video and the bottles of special gear oil required to do the job.;););)
Yeah I'm pushing on through thanks to you and Peter.
Yeah, that guy Andrew, what a legend, 've never seen so many empties in a workshop o_O
So, as mentioned earlier I've given up on the 75mm measurement as its not practical to achieve :rolleyes:
Besides which, at the end of the day on any pinion/crownwheel the end-goal is to achieve the correct contact patterns for power and freewheeling and then the backlash. That's the theory ;) I'm using this guide as a start as its easy to follow
 
I know you often can’t trust chatGBT but filtering a couple of things out like pinion bearing preload this sounds quite a sensible guide!

🧰 Tools & Materials Needed​

  • Bearing puller / press
  • Dial gauge with magnetic stand (for backlash measurement)
  • Torque wrench
  • Feeler gauges
  • Engineers blue / marking compound
  • Shim sets (for pinion depth and side bearings)
  • Workshop manual or pinion depth specification (see below)
  • Clean oil, rags, and patience

⚙️ Step-by-Step Setup Guide​

1. Inspect all components

Make sure:

  • Crown wheel and pinion are a matched pair (they’re lapped together at the factory and usually stamped with a matching number).
  • Bearings and races are in good condition or replaced as a set.
  • The diff housing is clean and free of damage.

2. Install the pinion (without final preload yet)

  • Fit the pinion bearing races and install the pinion with the correct depth shim behind the inner bearing race.
  • The pinion depth is usually specified relative to the centreline of the crown wheel (for Fiat 126, typically around +0.10 to +0.25 mm marked on the pinion head, relative to a nominal setup distance).
  • You’ll find a number stamped on the pinion (e.g. +05 or -03) — this tells you how much to adjust from the nominal depth given in the workshop manual.

3. Set pinion bearing preload

  • Fit the pinion bearings and crush sleeve (if used; some 126 units use shims instead).
  • Tighten the nut gradually while measuring the rotational torque on the pinion (without the crown wheel fitted).
  • Aim for 0.9–1.2 Nm (8–10 in-lb) of bearing preload on new bearings.
    (It should rotate smoothly with noticeable resistance, not loose or notchy.)

4. Fit the differential with crown wheel

  • Bolt the crown wheel securely to the diff carrier — make sure bolts are clean and use thread locker (these are critical).
  • Fit the differential assembly with its side bearings into the housing.
  • Insert the side shims (one on each side) — these control both backlash and bearing preload.

5. Adjust backlash

  • With a dial gauge on a crown wheel tooth, rock the crown wheel back and forth to measure backlash.
  • Fiat 126 spec: 0.10–0.15 mm (0.004–0.006 in).
  • Adjust by moving shims from one side to the other:
    • Move crown wheel closer to pinion → reduces backlash.
    • Move it away from pinion → increases backlash.
Keep total side shim thickness constant to maintain bearing preload.


6. Check tooth contact pattern

  • Apply engineer’s blue or marking compound to 3–4 teeth on the crown wheel.
  • Rotate the pinion in both directions under light load.
  • Inspect the contact pattern:
    • Too close to toe/root → pinion too deep.
    • Too close to heel/top → pinion too shallow.
    • Adjust pinion depth shim accordingly, re-check backlash afterwards.

7. Recheck all settings

  • Backlash after adjustment: 0.10–0.15 mm
  • Contact pattern: centered and even across tooth face
  • Pinion bearing preload: smooth and within spec
  • Side bearing preload: firm (no end float)

8. Final assembly

  • Once you’re satisfied with settings:
    • Fit new seals.
    • Torque all bolts to spec (crown wheel bolts ≈ 65 Nm; check your manual).
    • Fill with correct gear oil (usually 80W-90 GL-4).

📘 Tips​

  • Always replace bearings in matched pairs and re-shim — don’t reuse old shims blindly.
  • Write down your shim measurements so you can backtrack if needed.
  • If you hear whining after assembly:
    • Whine on acceleration → pinion too deep.
    • Whine on deceleration → pinion too shallow.
 
I know you often can’t trust chatGBT but filtering a couple of things out like pinion bearing preload this sounds quite a sensible guide!

🧰 Tools & Materials Needed​

  • Bearing puller / press
  • Dial gauge with magnetic stand (for backlash measurement)
  • Torque wrench
  • Feeler gauges
  • Engineers blue / marking compound
  • Shim sets (for pinion depth and side bearings)
  • Workshop manual or pinion depth specification (see below)
  • Clean oil, rags, and patience

⚙️ Step-by-Step Setup Guide​

1. Inspect all components

Make sure:

  • Crown wheel and pinion are a matched pair (they’re lapped together at the factory and usually stamped with a matching number).
  • Bearings and races are in good condition or replaced as a set.
  • The diff housing is clean and free of damage.

2. Install the pinion (without final preload yet)

  • Fit the pinion bearing races and install the pinion with the correct depth shim behind the inner bearing race.
  • The pinion depth is usually specified relative to the centreline of the crown wheel (for Fiat 126, typically around +0.10 to +0.25 mm marked on the pinion head, relative to a nominal setup distance).
  • You’ll find a number stamped on the pinion (e.g. +05 or -03) — this tells you how much to adjust from the nominal depth given in the workshop manual.

3. Set pinion bearing preload

  • Fit the pinion bearings and crush sleeve (if used; some 126 units use shims instead).
  • Tighten the nut gradually while measuring the rotational torque on the pinion (without the crown wheel fitted).
  • Aim for 0.9–1.2 Nm (8–10 in-lb) of bearing preload on new bearings.
    (It should rotate smoothly with noticeable resistance, not loose or notchy.)

4. Fit the differential with crown wheel

  • Bolt the crown wheel securely to the diff carrier — make sure bolts are clean and use thread locker (these are critical).
  • Fit the differential assembly with its side bearings into the housing.
  • Insert the side shims (one on each side) — these control both backlash and bearing preload.

5. Adjust backlash

  • With a dial gauge on a crown wheel tooth, rock the crown wheel back and forth to measure backlash.
  • Fiat 126 spec: 0.10–0.15 mm (0.004–0.006 in).
  • Adjust by moving shims from one side to the other:
    • Move crown wheel closer to pinion → reduces backlash.
    • Move it away from pinion → increases backlash.
Keep total side shim thickness constant to maintain bearing preload.


6. Check tooth contact pattern

  • Apply engineer’s blue or marking compound to 3–4 teeth on the crown wheel.
  • Rotate the pinion in both directions under light load.
  • Inspect the contact pattern:
    • Too close to toe/root → pinion too deep.
    • Too close to heel/top → pinion too shallow.
    • Adjust pinion depth shim accordingly, re-check backlash afterwards.

7. Recheck all settings

  • Backlash after adjustment: 0.10–0.15 mm
  • Contact pattern: centered and even across tooth face
  • Pinion bearing preload: smooth and within spec
  • Side bearing preload: firm (no end float)

8. Final assembly

  • Once you’re satisfied with settings:
    • Fit new seals.
    • Torque all bolts to spec (crown wheel bolts ≈ 65 Nm; check your manual).
    • Fill with correct gear oil (usually 80W-90 GL-4).

📘 Tips​

  • Always replace bearings in matched pairs and re-shim — don’t reuse old shims blindly.
  • Write down your shim measurements so you can backtrack if needed.
  • If you hear whining after assembly:
    • Whine on acceleration → pinion too deep.
    • Whine on deceleration → pinion too shallow.
Correct principles in all respects, but not all the details match the gearbox of the 500/126.; I would use the Haynes or similar.
If the positions of the rings that adjust the mesh of the crownwheel/pinion have not been moved, and if you trust them having been correct at the start, I would leave it alone. Setting that up is very fiddly,and you do need a dial gauge. and probably a metal frame or jig to hold the box.
The principle is simple, but operation hard. You wll be using the castellated rings by tightening one,and loosening the other by the same amount repeatedly until the mesh is correct. Then you will be tightening or loosening both of them equally until the correct preload has been achieved.
It took me hours to set up, hours to achieve, but was very satisfying once done.
 
Last edited:
Correct principles in all respects, but not all the details match the gearbox of the 500/126.; I would use the Haynes or similar.
If the positions of the rings that adjust the mesh of the crownwheel/pinion have not been moved, and if you trust them having been correct at the start, I would leave it alone. Setting that up is very fiddly,and you do need a dial gauge. and probably a metal frame or jig to hold the box.
The principle is simple, but operation hard. You wll be using the castellated rings by tightening one,and loosening the other by the same amount repeatedly until the mesh is correct. Then you will be tightening or loosening both of them equally until the correct preload has been achieved.
It took me hours to set up, hours to achieve, but was very satisfying once done.
I echo your thoughts on using the Haynes book as the reference but I must admit I do like the easy steps on the AI 😄
As for the task ahead I don’t trust any of the settings so I’ll be adjusting the castellated rings for the mesh and lash. I had a little practice and yeah the principle is fine but it’ll be fiddly. I’ve got a dial gauge but I’ll need to find a way to hold it and the box reliably.
The crownwheel has suffered a bit but this wasn’t just by the lockup this has been over a bit of time (see pic). The pinion doesn’t have any corresponding markings though.
I’ve determined what pinion shims I need, with none inserted the pinion needs to come out by 1.25mm to have a central contact position with the crownwheel so I’ve ordered plenty of shims to add to the few I’ve got as I expect there to be some fine tuning to that figure.
Lots a fun winter nights ahead …. by the end of it all I’ll have a drink problem too like our Aussie cousin 😏
 

Attachments

  • IMG_2389.jpeg
    IMG_2389.jpeg
    1.3 MB · Views: 9
I echo your thoughts on using the Haynes book as the reference but I must admit I do like the easy steps on the AI 😄
As for the task ahead I don’t trust any of the settings so I’ll be adjusting the castellated rings for the mesh and lash. I had a little practice and yeah the principle is fine but it’ll be fiddly. I’ve got a dial gauge but I’ll need to find a way to hold it and the box reliably.
The crownwheel has suffered a bit but this wasn’t just by the lockup this has been over a bit of time (see pic). The pinion doesn’t have any corresponding markings though.
I’ve determined what pinion shims I need, with none inserted the pinion needs to come out by 1.25mm to have a central contact position with the crownwheel so I’ve ordered plenty of shims to add to the few I’ve got as I expect there to be some fine tuning to that figure.
Lots a fun winter nights ahead …. by the end of it all I’ll have a drink problem too like our Aussie cousin 😏
It sounds like you've got it all together in your mind now, which takes you half way there. I had brain overload for a week when I did this. Here's the link to my fragmented writings about it. Towards the end of the thread it has an image of the frame I made from angle-iron, which really helped,
There is also an image of a simple tool I cobbled up to make precise turns of the castellated rings more easily achieved.
 
It sounds like you've got it all together in your mind now, which takes you half way there. I had brain overload for a week when I did this. Here's the link to my fragmented writings about it. Towards the end of the thread it has an image of the frame I made from angle-iron, which really helped,
There is also an image of a simple tool I cobbled up to make precise turns of the castellated rings more easily achieved.
What a terrific thread you did and wow, what a faff indeed. Great practical step by step insights from someone who went through the pain barrier and back – several times 😩. I’ve read through it twice but I’ll be back to it again. Thanks for sharing.
But there’s no avoiding it so I’m glad I’m not planning to use the car much if at all before the spring and although I’ve got the engine out / head off to do some tidying up I should have enough time🤔
 
Last edited:
My You Tube is going crazy, and firing all sorts of interesting videos at me. See this one, which should start at the point where a big spacer is placed behind the pinion and before the bearing.

Blimey that’s a turn up isn’t it, a spacer behind the pinion?! I thought it was a bit matto because I hadn’t seen it done before but there you go, thanks for sharing. If I do end up further inwards with the pinion that it currently appears one thing I noticed with my test washer was it was larger on the OD and would restrict oil flow, so if I do go down that route it needs to be the std 32mm OD. Wish I had a lathe, so many little projects on the go where I need to turn parts, oh well, no space or funds🥺
But hey, all good for now, shims should turn up on Wednesday and I will take it from there. Any further insights from you and the crew before then much appreciated😎
 
Back
Top