Is there a trans guru in the house?
 

Is transmission work close enuff?
Are these bolts, ring gear to carrier on a 901, single time use or reusable? ......
http://forums.pelicanparts.com/uploa...1202097755.jpg

Got some trick sh..ah....stuff goin' in. ....http://forums.pelicanparts.com/uploa...1202097969.jpg

Hayden used new bolts on mine but if I were doing this myself I would've used the old ones. The only true stretch bolts that I know of are the rod bolts inside the engine. If they feel good when you torque them then they are fine.

-Andy

Sometimes I use them twice in non-critical applications where a street type clutch is used and the power is modest.

Where twin or triple plate clutches are installed and/or a larger displacement engine is used, especially with race tires, I replace the bolts as cheap insurance.

I hate cutting corners on things,....:)

More data for the curious. 993 info.


http://forums.pelicanparts.com/uploa...1202259825.jpg


http://forums.pelicanparts.com/uploa...1202259837.jpg


Steve,

What's been your experience with setting up the new diff (taper bearing preload) if installing just the new diff and no other changes to the transaxle?

Generally it's ok to reuse those bolts for most applications though as Steve mentioned, on purpsose built racecar why cut corners? However, under no circumstances would I advise reusing the lock clips that go between the bolt and the differential.

The other thing to watch for is that with Guard's LSD (versus his torque biasing differential which is pictured) the stock open diff bolts aren't long enough to get proper thread depth and you need to get longer bolts. Shouldn't be an issue for this project but on the 915's in particular this is something to consider...

So I'm rootin' thru my 9eleben shop manual looking for the proper way to check the R & P backlash in a 901. There's a nice picture showing the special tool used for this....course I haven't gots one of those. Does anyone know the radial distance from the center of the differential (transversely... as in out put flanges) that this measurement is taken?

JP

For measuring, use the ring gear radius. Easy to make up something to do the measurement. You can only come close anyway, as shims don't come in an infinite variety of thicknesses.

http://forums.pelicanparts.com/uploa...1202625876.jpg

When first I installed a Quaife in a 915 I discovered that I needed longer bolts for full thread engagement. At the time a 928 diff ring gear bolt had the extra length and the funny shape to hold the keepers. But about 901s this deponent sayeth naught.

And I have gotten away with reusing these bolts. Is this a 12mm 12.9 bolt? Those will take 150 lbs/ft of torque without failing (think 6 bolt flywheel bolts). Torque here is a lot less, is it not? I've known otherwise meticulous shop owners even to reuse the flywheel bolts for street applications. But, unlike Steve's customers, if something breaks I have only myself to blame. And I can't suck as much power out of my engines as he does for his customers.

Walt

More gratuitous advice
 

More:

JP - you ever set the preload and side to side location of a ring gear before?

I'm guessing not.

I'm guessing you are replacing the carrier bearings, though.

And I'm guessing you have yet to remove the old carrier bearings, because you aren't reusing the old diff housing.

So: take your old ones off as best and as carefully as you can (assuming they are not visibly and obviously scored and way messed up) and hone them until they will just slip over what they fit over on the nice new housing. But more to the point, so you can pull them off with your fingers. Strong fingers, but fingers.

Use these old bearings (as long as of the same brand as the new ones) for your setup, and only when done install the new ones. Which are a press fit. Not a heavy press, but enough of a press that they are hell to remove without damaging (as you will have discovered). There is a really neat tool you can use to pull them off, but you don't have it (and neither do I).

Another trick you can easily pull is to set the preload (I assume the procedure is about the same for the 901 as the 915) before you install the ring gear on your new diff. That gear can only get in the way for preload setting, which should only involve the bearings and the total shim stack thickness. Then you can install the gear and go about swapping shims from side to side to set the lash.

You can do the honing with a drill (or drill press) and a large two or three stone hone of the sort you can purchase at a good hardware store or NAPA or otherwise that says it will work on the diameter involved.

Walt

Hay, Walt, good to hear from you again.

Basically the trans was set up by G Ledbetter some time back and the guts look lovely...including the R & P gears which are smooth as a baby's ass. I figure the best I can do is return the ring gear to exactly where is was with the old carrier.
I figure I am .003 off on the short side and may need a shim there, but want to check the preload. Assuming that it good, I'll want to check the backlash to see if it is in tolerance.
Your tool looks like sumthin' I'd build:D......but the radial distance to your indicator is still unknown.....that is what I need as the arm length is gonna change the number. Would the radial distance to the center of the ring gear be what I need? The ring gear on a 914-901 doesn't get in the way cause you gotta pull the gear stack to get the sucker out. It does give you incentive to get things right before reassembly, tho.:D

You're right, I've never been here before.....but I'm used to that. I got some advice from H Burrville (sp?) and am proceeding apace. ......
http://forums.pelicanparts.com/uploa...1202662508.jpg

Measuring radius
 

JP - not for the first time in my life I said too much and obscured what I said about what someone wanted most to know.

The measuring radius is the same as the ring gear radius. Don't know what that is, but you just measure your ring gear diameter and, etc.

And if someone chimes in and says the radius is to the center of the ring gear teeth I won't be offended. I'm going off of memory of what a rebuilder told me. But I think it is to the outer edge of the teeth. Too bad the factory manual doesn't say, but they presume you have the factory tools, which take that part out of the equation.

Never throw a piece of steel away. You never know when you will need part of the latch system from a long since replaced garage door to make a measuring instrument. The bigger steel bits became the steel frame for my tire storage shed.

Walt

Or aluminum.....I have some angle that used to be square tubing that will make a fancy tool that I'll never use again, stylish even.:D

Thanks for the info.

Checked the preload today. It's .007 (or so) the book says .006...close enuff?

Diffident differential differences - continued
 

More, this time on the general subject.

1) Maybe someone ought to hunt down a moderator. Don't 911 tranny discussions usually end up on the 911 rather than the engine building forum? I don't care a whole lot, but someone may.

2) For want of the proper terms, and inability to draw pictures on pictures, I'll invent some terms.

The carrier bearings slide onto the diff and come to a stop when they bump into a spot where things are machined wider. We'll call these the stops. The shims fit between the bearing and the stop, to be more precise. But the non-adjustable baseline is the stop.

If you can measure precisely (to the 0.001", maybe) the distance between stops on each side on your stock diff, and on the replacement diff, you could use that information to set the preload. You'd also need to know what the old shim stack total thickness was (which of course you do), plus knowing that the preload was within spec on the old diff (and cover plate and case), and you trusted (as I am inclined to do) that the carrier bearings are dimensionally the same (if you just reused the old ones, that would be a given).

So if the stop to stop distances are the same with stock vs Guard, the total shim pack thickness ought to be the same. And, if there is a difference, you could just calculate what your total shim pack should be - thicker if the Guard is narrower, and vice versa.

3) Invented term 2 is the ring gear mounting plane on the differential carrier. If the distance from stop to stop is the same, and the distance from either stop to the ring gear mounting plane is the same, you can divide up the shim pack the same way it was and the lash ought to come out the same. And if it is different, you ought to be able to figure that out also, and decide which side's shims need to be thicker, and which thinner. And, even if the stop to stop distance (see 1) above) is different, you can take that into account.

So, if you have a nice granite plate with a dial indicator measurer and whatnot, I suppose you could do the measuring with the requisite precision. Maybe even a wide enough digital or dial caliper. In which case you ought not even to need to measure the lash, since if the numbers work out right and it was good before, it should be good after.

I don't think anyone quite trusts things that much. Sure you and I both trust that the pinion depth doesn't need to be rechecked. But I think most will check preload (at least using the cover plate gap method) and lash. Just to confirm things. Lash setting is kind of a cut and try process, at least for me. Especially on a tranny for a 911.

But is this, gasp, for a 914? Were it for a 911 I'd say just plop what you have in there and measure to see what you have. But you'd have to take it out again to install the lock plates if for a 914, no?

Walt

(A friend who was a 914 guy actually forgot the ring gear flip when rebuilding a buddies 911 tranny. So the 5 speed reverse is not an urban myth.)

Which is about what I did, but I don't have a surface plate/height gauge (I'd have used a CMM at work but THEY frown on that). I used a depth mic & parallel. So, I used the cover plate method to check the preload and got .007 rather than the prescribed .006 and am gonna call it close enuff. BTW, you did it again, Walt.:D

Seems like the measuring radius measurement would be to the ID of the ring gear? See the image in Grady's third post (#11)

http://forums.pelicanparts.com/porsche-911-technical-forum/248508-915-repair-what-tools-needed-change-carrier-bearings.html

Kevin

I see why you might think that. But that diagram is part of explaining how to set pinion depth. Which involves knowing the distance from the flat end of the pinion shaft to the centerline of the differential (which connects the center of the right carrier bearing or its race or the hole the race sits in with the center of the left). doesn't have anything to do with lash, other than that the lash is stamped on the OD of the ring gear in case you forget what it should be. Though, unlike the R numbers which vary some with each individual R&P set, I think the lash number is not based on measurement of the specific set. It is just a standard.

One reason I'd think backlash would be measured at the OD rather than the ID is that with a larger radius your measurements are more precise. If it moves a thousandth at the ID, it might move two thousandths at the OD, which would be easier to measure accurately (this exaggerates, I guess - I didn't calculate, but you get the point). Improves the resolution?

JP

Well, I did it again, but by accident. I started that epistle, then went for a walk, came back, saw I hadn't sent it, and sent it. In the mean time you had asked a simple question. But this time what I did was not answer the question, not answer it but obscure it with blather.

And I agree with you on the question. I don't think you can be super confident that you can measure preload using the gap method to the thousandth of an inch anyway, even though that side plate sits amazingly squarely on there, aided by all those studs.

Now Grady says this plate gap measurement is just to get you in the ball park, and you should use your handy inch/ounce or metric equivalent beam torque wrench to measure drag. And of course he is right. But I don't have one of those, and don't know anyone who does short of a guy who rebuilds trannies for a living. And while I can purchase a foot pound beam wrench for relative peanuts, for some reason smaller means much more expensive here.

I did try another approach. Remember that fancy tool I made for measuring lash? Well, I had the tranny on the stand in its normal orientation with the diff in but the pinion shaft not installed. I fixed the output shafts so the diff acted like it was locked, as you need to do to measure preload as I recall. And I turned things until my lever arm was horizontal. At a calculated distance out on that arm I hung a weight. I figured that I could calculate inch/ounces (or anything else) by how much weight it took to make things turn. Of course I had to deal with static vs moving friction. And the reduction in effective lever length as the lever arm moved away from being parallel with the ground (or, if you wish, out of tangency with the center of the earth). And so on. I was not at all satisfied with the results. Seemed like a good theory, though.

Were I in your shoes, and after measuring everything and concluding that the shims should replicate what used to be, and then found I was only a thousandth off of nominal using the rough method, I'd do what you are doing - call it good and move on to dealing with the backlash figure.

Walt

I talked to Mr. Ledbetter today and got all sorts of insights. He checks lash to the center of the ring gear but says it doesn't make much difference as long as you're on it. He also sets the preload much tighter ....larger gap and explained his methodology.....so I redid it and got .007.....he then goes with the turning force check. I do have a cheepa$$ HF inch pound torque wrench that I bought to tighten rocker shafts....wow, a multi-use tool.:D

I checked something he said after we got off the phone and, as I understood him (stuff was flying over my head so I was a couple sentences behind), didn't like the result so I'll have to bother him again. It's only a matter of time till Mr. Cramer throws out this thread nohow.

JP

I've got an inch/lb torque wrench. Use it all the time. Not expensive at all.

But for the preload you want an inch OUNCE wrench.

Or has it been that long since I got out my shop manual?

Walt

15.6 to 20.8 inch pounds.....had the manual open to the page in the shop.
Gettin' old is a *****, neither one of us were sure.:rolleyes:

Walt,

Thanks for the correction. The way I understand the diagram is R is the theoretical design distance from the center of the ring gear to the end of the pinion. This number appears to be based upon the "blueprint" design specs of how the ring gear and pinion gear would ideally go together? I see that this R varies from transmission to transmission. For instance:

915 trans R = 66.30mm
930 trans R = 82.29mm
Early G50 trans R = 61.00mm

For ring-pinion sets prior to 1985, the dimension N on the face of the pinion was the deviation from the design distance R, determined by specialized testing of the ring-pinion set before installation into a transmission. So upon installation into a transmission, the amount R + N is the pinion depth dimension to be achieved with the necessary bearing retainer plate shims, measured by the special VW 385 tool. I think I got it.

Thanks to you and JP sharing your ideas. I'm close to getting into this myself. Got the diff out of my G50 and am devising how i'm going to get the carrier roller bearing races out of the case so I can install new bearing assemblies upon a replacement LSD. Also plan to use the old bearings for use in setting up preload and setting backlash. Lots of reading and understanding to absorb!

This should about do it for me. Got the backlash checked and it's .006-.007" vs
.16-.20mm (rounded) from the book. A bit towards the bottom of the tolerance.
My tooling is a kind'a makeydo, but functional. .......http://forums.pelicanparts.com/uploa...1203270483.jpg

can you take any close up pics to post on your set up & what specifically you were checking for us beginers?
Great job from the looks!! Thank you! Bob

JP

Piece of cake?

Did you have to change any of the carrier bearing shims?

I like the adjustable dial indicator holder. All I have are those rod and magnet base ones. What is yours?

KTL - I've only seen the inside of a G50 once, and that was watching a mechanic change to steel cones or some such synchro stuff out in the open at a race track. Though various tranny fundamentals doubtless are much the same.

For earlier Porsche transmissions there was a Porsche tool which fit the 915 (maybe a couple of others) and was pretty easy to use (because I borrowed one and could use it). And the VW tool which had various adapters so it could be used on more trannies, and is more expensive.

Some folks can set up a tranny just with blueing, but I think it is just too hard to get confident with something like that unless you can play around under controlled circumstances. Spend a day or more assembling things so they are known to be correct and seeing what the blueing says, and then purposely making them wrong in a known direction, etc. There are pictures around of what the pattern should look like, but when I've tried it I have just never felt confident. The guys who really know their stuff that I know often use blueing as a final check.

My understanding is that pinion depth (initially set as you say - to a standard specification) is set acoustically. The ring and pinion are installed in something, and while being rotated the one or the other is moved fore and aft until the desired frequency is reached. Then that offset is inscribed on the set. At least that is the 915 system.

An engineer/mechanic/shop owner friend says he sets up the trannies by measuring, but he doesn't use the factory tools. He must have figured out some other way accurately to locate the centerline of the differential. If you could measure from there to the end of the pinion you'd have it. Or measure to the case by the pinion, and then back with the pinion in to its end, you'd be there also.

What might be fun would be to figure out how to use an old differential housing as the basis for a home brew pinion depth measurer. Old diffs are a dime a dozen, because folks like me and JP have replaced them with somehting else. The factory tools would be a steal, I think, at maybe $1,000. Tough to justify for something you might do once every couple of years, if that often.

Walt

Hay, like a bumper sticker that a buddy used to have said:"don't follow me, I'm lost".

This is my first time mucking about with a ring & pinion brought on by being a CSOB.....and wanting to learn how to do it. I spent enough monies on the LSD to ruin what I jokingly refer to as my budget. Tack on an engine rebuild with this...:(

I'm checking the backlash between the ring & pinion.

The widget on the end of the pinion shaft is to hold it firmly (a poor imitation of the factory tool) while moving the ring gear....stop to stop as it were. It's a 30mm socket over the end of a big effin' bolt back there whose name escapes me, welded to a piece of strapping then bolted to the intermediate plate. I'm not happy with the "firmly" part of this and am gonna try something else.

The aluminum angle face is centered on the axis of ring gear rotation....via eyeball and marked to approximate the radial center of the ring gear & bolted to the axle flange (there is a factory tool for this also).....which is where this mess started.

Were I not a dummy, I woulda checked this as set-up before taking the thing apart. I'd feel much more confident at this point knowing the original lash.

No changes to the shims, Walt. Central Tools Inc. is on the box.....no other marking except "Visegrip". It takes some foolin' around to get the flexible arm.....er...inflexible/stable.

When measuring the backlash of a gear you want to measure it at the outside edge. Here's a picture of this I found on the Mopar Muscle website:
http://forums.pelicanparts.com/uploa...1204391621.jpg


Obviously, in the case of the 901/915/930 ring gear, the case is in the way so you can't put the indicator on the edge of the gear directly. However, you still need to place the indicator at same distance i.e. the radius of the ring gear. It could just be the perspective, but it looks like your distance is too small.
http://forums.pelicanparts.com/uploa...1204391385.jpg


Just tip when checking the backlash on the ring gear: Make sure there isn't any slop in the flange to diff union (which would skew your readings). One transmission I worked (a friend's 930) on had some slop in the output shaft spines. My friend had to make a tool to bolt directly to the differential instead of bolting something to the output flange.
-Chris

Jezze Chris, you're a little late for this conversation. The engine/trans went in the car today.

As Mr. Ledbetter explained it, the gap (backlash) is the same (or should be) throughout the length of the contact area twixt the R & P. The radius of the ring grows as does the radius of the pinion as you move away from the center of rotation.....makes sense to me.

Your Mopar dude has a real bad cosine error with the angle of the indicator to the face of the ring tooth.:D

The new owner of my 914 has carried thru with a trans mod "plan".
10-12 lbs of excess rotating has ben eliminated. It's rather specialized...I figure not many here are rushing to the exits to duplicate this.:D .....
http://forums.pelicanparts.com/uploa...1262191001.jpg

Oh you autocrossers. Never need 4th or 5th, do you.

Not often do AXers go over 80 mph.....maybe PCA at Qualcom. Decent 901s are cheep.

A builder still can use ones imagination in the Prepared & Mod classes at Autocross....more so in PCA than SCCA but there is still room at the latter.
You want a turboed twin rotor in a Sprite.....you'll find a class to run in at autocross.