What's new
What's new

How are ring and pinions setups assembled in a production environment?

Cole2534

Diamond
Joined
Sep 10, 2010
Location
Oklahoma City, OK
I was watching some guys set up a R&P the other day and it seems like a lot of fiddling around with trial and error. Shim selection, torque checks, gear printing, crush washers? SLOW.

So how was this done on an assembly line back in the day? I gotta assume they used quantitative methods rather than T&E, but I could be wrong.
 
I expect the case and pinion have been gauged and the computer spits out the correct shim pack. The ring gear is set to a specified clearance to the pinion, I have no clue hoe that could be automated but it probably is.
 
That's what I anticipated. But if that's the case, why aren't those required values published in the service manuals and then followed by install techs/etc?
 
Pinion depth is usually etched on the head of the pinion gear either as a hard dimension from the ring gear centerline or a deviation as +/- from a nominal ring gear centerline to pinion head dimension. This dimension is usually included in the service manual for the rear end. Preload on the pinion bearings is normally set with a crush sleeve and pinion nut torque. Setting for ring gear position is set by measuring backlash and adjusting ring gear position accordingly- either by shifting shims from one side to another or adjusting collars. Obviously, all this is not for factory assembled units, but bottom line is setting up a ring and pinion without having the pinion depth tool and/or the case spreader tools can mean a lot of trial and error to get a good mesh. Didn't answer your original question, but. . . . .
 
I’ll preface this with “it’s been a lot of years ago and a lot of beers ago”. I can fill in some of the blanks on factory production builds of class 8 truck axles. These are the big boys not pickups.

First of all the ring and pinion gears are lapped together as sets. This is for wear characteristics and to minimize noise. Both are phosphatized to facilitate proper break in. Pinions are mounted to a pinion cage separately. Pinion bearing preload is set with a solid hardened washer. The stack of components are measured to get an approximation of the required spacer thickness. This is assembled and bearing rolling torque is measured. Rinse and repeat using spacers that are at station in 0.001” increments till rolling torque spec is met.

The mating ring gear is mounted on to the differential and set in the carrier. The gear assembly can be adjusted left or right using the differential bearing adjusting nuts.

Now the pinion cage is assembled to the carrier assembly. It is adjusted in and out by shims under the mounting flange. The ring gear tooth faces are painted and moved into proper contact position with the adjusting nuts.

A moderate amount of torque is transmitted through the assembly and both gear positions are adjusted until a specified contact pattern is achieved on the gear tooth faces along with the backlash setting. Differential bearing preload is another long story.

Lots of trials to computerize and automate with little success.
 
The factory will have a complete set of shims and setting tools and will be working on runs of the same axle. The gear housings/third members are machined to very close tolerances,so each gear set will have its pinion depth marked so any deviation from the design depth will be easy to get the correct shims. Pinion depth is the first step. I'm pretty sure they have charts or programs that tell what side shims go with various depths. Of course if it uses adjustable ring nuts then its going to be a manual job reading the pattern. Don't know if robots can do it now or not. In a factory setting it would go pretty quick compared to doing it in the field but would still be labor intensive. Assembling is quick but you still have to read the pattern and see if you have to change anything. I would also be courious how they do hundreds of thousands of axles.

By the time I graduated from hs I could set up any kind of gear set.I have probably set up at least one of every kind of pre 70's axle made(passenger cars/light trucks).That was easy money.I made my own setting tools and case spreaders.A lot of weeks I made more in the garage than I made at my day job by building rear ends,transmissions and engines.
 
there is/was a long time member here, location "ann arbor", can't recall the name.
Has written about it maybe 12 years ago.
 
I asked a similar question of a fella from ( Richmond gear?) years ago. His answer was something to the effect of- the assembly line gears are all machined to the same dimensions, thus, the "Master Dimension" listed for each TYPE of gear. The only variable is the finished dimension after lapping, which is when the backlash and PHD are marked on the gearsets, and the corresponding shims are mated with the sets before they go to the floor. This explains why there are rarely a bunch of smaller shims in factory assembled diffs. They usually only need one or two added to the "masters". That, and the fact that the factories allow a much wider tolerance than us picky drag race guys did...
 
Now or in the past?
In the old days piston bores where labeled and the assembly guy put the correct sized piston in each hole.
Now the holes are all made the same size and no more fitting.
Auto production machining tolerances have come a long ways in the past 30 years. This eliminates fits ,"colors", or classifying in many cases although still done in some driveline parts.
Remember that in production you have a lot of housings and gear sets on hand.
Back in the day a set and housing would be measured when made and color coded for size. You put a "blue" set into a "blue" housing.
If the torque test failed it went to repair to be hand matched and shimmed.
Processes varied somewhat at different makers.
Aftermarket or repair does not have the luxury of many sizes to pair up so the tedious process.
Bob
 
You guys that think setting up a diff is slow need to see my friend Dan do it. It would blow your mind. 30 years he's made a damn good living doing just diff work for 1/2 the price of any other shop around.

If there was an automated diff setting robot/cell/machine on an assembly line somewhere I'd put my money on Dan being better and faster.
 
Another one with a distant (mid 70's) memory of planetary gear setups in a UAW plant. We did measure some parts and prescribe selective fits (e.g. control body valves) or call out various gear body shims etc. But I can't ever recall UAW workers being given freedom to play around with selective fits on their own. Too much distrust all around.
 
..... But I can't ever recall UAW workers being given freedom to play around with selective fits on their own.....

No put the blue with blue or green with green.
But is possible so understand. A friend got the job of putting in camshafts.
There are 8 cams and each motor has a code to tell which.
For two weeks long he put Corvette cams into every motor built on his shift as these were the easiest to grab.
Selective fit normal back in the days. What are those stamped letters or punch marks on the oil pan rail and/or different paint color dots?
I can assure that this exists today in some build.
Sucks as a supervisor that you have lots greens and most the mating stock are blues and yellow. What to do then?
File that under fun days.
Bob
 
Last edited:
I have no idea how it is/ was done at the factory, but I remember back in the early 80’s those Ford integral carriers were really troublesome. You’d get it to quite down at one speed while accelerating only to have the customer come back a week later complaining about a whine while decelerating at another speed. Even remember I’d have to change many housings to finally cure the noises. And I mean some of them absolutely screamed. You couldn’t turn the radio loud enough .
That said, I’m sure they’ve come a long long way because of the advances in cnc manufacturing.
On another note.... I absolutely loved the Ford 9” removable carriers. Painted the gears....set backlash while in your vise at your bench, and then stick it in the housing. Those are the best designs I’ve ever worked on as far as the tech is concerned. But the Dana’s were the true brutes as far as my recollection goes. They are tough. If I were to buy a real pickup, I’d absolutely want a Dana rear end.
 
They use a machine on the Ford line made by a company called Marposs. For the shimmed dimensions it measures the gear positions and preloads and then lists what shims to install. It measures the gear set post-shimming and indicates whether further shimming is needed. The pinion gear preload is set with a crushable shim, and the machine measures the running friction while tightening the pinion shaft nut. I worked at a company that ended up using a variant of this machine to shim missile seeker assemblies for alignment and preload and got to see the machine in operation. It took about 30 seconds to select the shims, 30 more to measure the shimmed unit, and about 15 seconds to set the crushable pinion shim. Very impressive.


Sent from my iPad using Tapatalk
 
I just retired from a company that built the all the equipment and integrated it in to assembly line loops It is all done with gauging of housing , pinions with bearings , carriers with bearings and and offset codes for each ring and pinion pairs black magic and lots of math
 








 
Back
Top