Backlash in the Transaxle? - Page 12 - 3000GT/Stealth International Message Center

DoctorDex · 10-01-2007, 09:02 AM

Just figured I'd post an update.

I've finally, sort of, fixed the dreaded clunk. This past weekend, I installed a new (used) rear dif (backlash of .004), new carriers, new lobro boot, I got arid of the rubber rear dif. mounts and I had some aluminum bushings pressed in. The car used to make a loud thud from the rear end whenever I'd go from decel to accel. Now it just makes a very light thud from the transaxle whenever the slack is taken up. T

I still has some backlash in my transaxle. I'm going to try and replace my motor mounts next to see if it helps. Then if that doesn't work, I'm going to have to replace the transaxle and probably the motor while I'm at it.

DoctorDex · 10-25-2007, 08:40 PM

I still have a great deal of backlash which is obviously in the transaxle and I'm wondering if anybody knows which part inside the transaxle could be the cause of my problems?

The backlash is roughly about equal in gears 1 - 5 and it seems to be slightly better in reverse. I have an new 300m 18 spline OS, new VCU and new planetary.

Anybody have any ideas?

DoctorDex · 10-30-2007, 08:49 PM

I know I've said this all before, but I figured I'd rehash everything out and then try and propose a fix:

I have an 18 spline transaxle and I've been battling with excessive backlash and clunking ever since I bought the car. I've changed the rear differential (even though the old one was within spec), I've changed the carrier bearings, I've swapped out the rubber rear differential mounts for solids and I've double checked the backlash in the transfer case and everything checks out.

Therefore, I have narrowed down the source of my excessive backlash and the culprit is the transaxle. For the record, I've installed a new 18 spline VCU and a new 300m output shaft. In spite of all of my efforts, I still have a clunk when going from decel to accel (although it has improved slightly) and occcasionally on light decel it will oscillate back and forth which is really annoying. I've driven in a friends VR4, which has a 25 spline VCU and he does not have this problem.

So, then what is causing the problem?

I can't say for sure, but I am pretty sure the design of the 18 spline VCU definitely contributes to the backlash problem.

Here's why: the 18 spline VCU has oblong dowel pin holes and the planetary gear dowel pins fit in loosely inside and there is a large amount of slack space, ~ 1 mm on either side (measured with calipers) of the dowel pin. In contrast, the 25 spline VCU has round dowel pin holes and the planetary dowel pins fit snuggly inside, therefore there is very little backlash from this component in the 25 spline setup.

So what can you do about it? At the moment, the only true fix is to convert to a 25 spline setup. I considered welding over the dowel pins holes and then re-drilling them, but I couldn't find a shop that would touch the VCU with a welder as they were too afraid of wrecking the VCU.

So here is what I'm proposing:

Have new planetary dowel pins made, that are flanged/oblong shaped (on the VCU end), so that they can be press fit into the VCU/planetary.

Now here's my problem, I don't know where to begin. Is this do-able? If not, does anyone have a better idea?

What kind of material would it have to be, it'd obviously have to be something extremely strong and hard and would it be possible to machine something that hard? If not, then how could they be made and would the cost be too high to justify the process?

havok91 · 10-31-2007, 01:23 AM

Doctordex,
you said you bought a new 18 spline vcu? Where and how much? I've done much searching here and can only find that Supercar Engineering is beginning to work on one. I would just like the info so that I can prepare myself.

Also, the stock VCU probably had some, if not all, of its machining done prior to heat treatment. Electric Discharge Machining (EDM--Die sink or wire) can be utilized in machining of hardened steel. But, if it's applicable here, I can't say.

DoctorDex · 10-31-2007, 07:30 AM

It wouldn't be the VCU that would need to be machined.

The planetary gear pins would have to be tapped out of the planetary and new planetary gear pins would have to be made that are flanged/oblong shaped on the end.

The pins I'm referrring to correspond to "3" in the picture:

http://www.stealth316.com/images/cdvcu_pinions.jpg

havok91 · 10-31-2007, 02:55 PM

Those pins hold/align the planets in the planetary assembly. Yes, new pins could be made with a flanged/oblonged head and pressed in. But the oblonged/flanged pins would be a pain to get aligned 100% to the eccentrincity of the oblonged/worn holes in the vcu. Therefore, it would be easier to use a larger diameter pin head (or entire pin with requsite machining on the gears and carrier). This would align much better in a machined vcu that would accept those altered pins. However, I've only seen pictures of the components and I'm speculating. Sounds like some R and D would need to be done (let alone some stress analyses to confirm integrity)--custom stuff.

Again, where did you say you bought a new VCU? You didn't answer that in you last post.

DoctorDex · 10-31-2007, 06:06 PM

Quote:

Originally Posted by havok91

Again, where did you say you bought a new VCU? You didn't answer that in you last post.

I sent you a PM with the guys email address. I can't remember his name off the top of my head.

With regards to the VCU holes, I personally don't believe that wear is that much of an issue with them. The holes of my old VCU where identical in size and shape to the holes of the new VCU. The holes have been heat treated/hardened (they are darker than the surrounding metal), so they seem to be pretty resilient.

I'd be reluctant to machine new holes in the VCU as it is made of very tough metal. I approached a machine shop about this before and they didn't want to touch it. Plus, the new holes would be weaker than the originals because they wouldn't be heat treated.

I wonder if perhaps you could weld a flanged/oblonged head onto the pin and then machine it to shape? If it was machined and then press fit, it wouldn't have to be that strong because there would be no hammering effect because there would be no slack to take up.

As far as fitment, it'd be a near press fit, which you would be able to assemble prior to putting it into the car.

havok91 · 11-01-2007, 07:09 PM

I looked at the pictures, again, from all the previous posts. New machined (not welded flanged--overcomplicated) would be the best bet. Have it machined from a good hardenable steel (likely tool steel, but that all depends on the loads). Dowel pins can are made from a variety of alloys, so, I have no idea what the facory pins in the planetary assembly are. The pin's strength properties could be roughly determined from a hardness test--a tensile test would be better.
The pins will be in single shear regardless of the classification of fit--however,any slack/slop will add dynamic loading in addition to the static loading. Therefore, the pin's shear strength must be at least that of the original's--with more for saftey. The pins would have an interference/press fit in the carrier of the planetary assembly, and a slip/location fit in the vcu.
Pins with larger heads would then be able to register in the vcu (with its oblonged holes--actually like slots), reducing slack/slop. However, the stress distribution would be more of a point type loading in the slots than something that resembles a bearing type load. The long term effects, like fatigue, are unknow unless an analysis is done. Do the slots in the vcu perform a function we are not aware of? My suspision is that the slots allow some sort of miss-alignment of the entire assembly.

When I rebuild my TT's tranny, I'm going to take a closer look at this. Machine things if needed and run some numbers. I'm a big proponent on doing something better than the factory was willing to spend money on.

DoctorDex · 11-01-2007, 11:24 PM

Quote:

Originally Posted by havok91

Do the slots in the vcu perform a function we are not aware of? My suspision is that the slots allow some sort of miss-alignment of the entire assembly.

That is my guess as well. However, they did away with the slots in the 25 spline setup, which means you'd think they wouldn't be necessary

One idea we haven't talked about would be JB weld.

If/when it frags it'd be a good excuse to buy a 25 spline transaxle. I wonder how the compressive strength of JB weld compares to that of hardened steel?

havok91 · 11-02-2007, 01:43 AM

I think JB weld is only rated around 4000 psi. Not much. It also wouldn't be very hard, much softer than a good steel alloy, so, I don't think it would hold up. It probably won't hurt to try, though.
If the 25 spline vcu doesn't have the slots, it may be viable to put in a new bolt hole pattern (exactly like stock but with tighter tolerances), which could be offset evenly with the existing holes. You said the machine shops you talked to wouldn't touch it. But, a good Mold Making shop should have a DIE SINK EDM (repeat) that can machine tough, hard metals. However, this would not be. Is it worth all that work? Maybe.
Props, though, for looking into this idea.
I'll look into this more when I get that far with my car. I got access to EDMs.

10-01-2007, 09:02 AM	#111 (permalink)
DoctorDex Forum Member Join Date: Jul 2005 Location: Edmonton, AB Drives: '91 VR4 - RIP Trader Rating: (16)	Re: Backlash in the Transaxle? Just figured I'd post an update. I've finally, sort of, fixed the dreaded clunk. This past weekend, I installed a new (used) rear dif (backlash of .004), new carriers, new lobro boot, I got arid of the rubber rear dif. mounts and I had some aluminum bushings pressed in. The car used to make a loud thud from the rear end whenever I'd go from decel to accel. Now it just makes a very light thud from the transaxle whenever the slack is taken up. T I still has some backlash in my transaxle. I'm going to try and replace my motor mounts next to see if it helps. Then if that doesn't work, I'm going to have to replace the transaxle and probably the motor while I'm at it. Last edited by DoctorDex : 12-14-2007 at 09:27 AM.

10-25-2007, 08:40 PM	#112 (permalink)
DoctorDex Forum Member Join Date: Jul 2005 Location: Edmonton, AB Drives: '91 VR4 - RIP Trader Rating: (16)	Re: Backlash in the Transaxle? I still have a great deal of backlash which is obviously in the transaxle and I'm wondering if anybody knows which part inside the transaxle could be the cause of my problems? The backlash is roughly about equal in gears 1 - 5 and it seems to be slightly better in reverse. I have an new 300m 18 spline OS, new VCU and new planetary. Anybody have any ideas?

10-30-2007, 08:49 PM	#113 (permalink)
DoctorDex Forum Member Join Date: Jul 2005 Location: Edmonton, AB Drives: '91 VR4 - RIP Trader Rating: (16)	Re: Backlash in the Transaxle? I know I've said this all before, but I figured I'd rehash everything out and then try and propose a fix: I have an 18 spline transaxle and I've been battling with excessive backlash and clunking ever since I bought the car. I've changed the rear differential (even though the old one was within spec), I've changed the carrier bearings, I've swapped out the rubber rear differential mounts for solids and I've double checked the backlash in the transfer case and everything checks out. Therefore, I have narrowed down the source of my excessive backlash and the culprit is the transaxle. For the record, I've installed a new 18 spline VCU and a new 300m output shaft. In spite of all of my efforts, I still have a clunk when going from decel to accel (although it has improved slightly) and occcasionally on light decel it will oscillate back and forth which is really annoying. I've driven in a friends VR4, which has a 25 spline VCU and he does not have this problem. So, then what is causing the problem? I can't say for sure, but I am pretty sure the design of the 18 spline VCU definitely contributes to the backlash problem. Here's why: the 18 spline VCU has oblong dowel pin holes and the planetary gear dowel pins fit in loosely inside and there is a large amount of slack space, ~ 1 mm on either side (measured with calipers) of the dowel pin. In contrast, the 25 spline VCU has round dowel pin holes and the planetary dowel pins fit snuggly inside, therefore there is very little backlash from this component in the 25 spline setup. So what can you do about it? At the moment, the only true fix is to convert to a 25 spline setup. I considered welding over the dowel pins holes and then re-drilling them, but I couldn't find a shop that would touch the VCU with a welder as they were too afraid of wrecking the VCU. So here is what I'm proposing: Have new planetary dowel pins made, that are flanged/oblong shaped (on the VCU end), so that they can be press fit into the VCU/planetary. Now here's my problem, I don't know where to begin. Is this do-able? If not, does anyone have a better idea? What kind of material would it have to be, it'd obviously have to be something extremely strong and hard and would it be possible to machine something that hard? If not, then how could they be made and would the cost be too high to justify the process?

10-31-2007, 01:23 AM	#114 (permalink)
havok91 Forum Member Join Date: Feb 2007 Location: SLC, UT Drives: 91 RT/TT Trader Rating: (1)	Re: Backlash in the Transaxle? Doctordex, you said you bought a new 18 spline vcu? Where and how much? I've done much searching here and can only find that Supercar Engineering is beginning to work on one. I would just like the info so that I can prepare myself. Also, the stock VCU probably had some, if not all, of its machining done prior to heat treatment. Electric Discharge Machining (EDM--Die sink or wire) can be utilized in machining of hardened steel. But, if it's applicable here, I can't say.

10-31-2007, 07:30 AM	#115 (permalink)
DoctorDex Forum Member Join Date: Jul 2005 Location: Edmonton, AB Drives: '91 VR4 - RIP Trader Rating: (16)	Re: Backlash in the Transaxle? It wouldn't be the VCU that would need to be machined. The planetary gear pins would have to be tapped out of the planetary and new planetary gear pins would have to be made that are flanged/oblong shaped on the end. The pins I'm referrring to correspond to "3" in the picture: http://www.stealth316.com/images/cdvcu_pinions.jpg

Sponsored Links
3000GT/Stealth International Message Center Advertisement

10-31-2007, 02:55 PM	#116 (permalink)
havok91 Forum Member Join Date: Feb 2007 Location: SLC, UT Drives: 91 RT/TT Trader Rating: (1)	Re: Backlash in the Transaxle? Those pins hold/align the planets in the planetary assembly. Yes, new pins could be made with a flanged/oblonged head and pressed in. But the oblonged/flanged pins would be a pain to get aligned 100% to the eccentrincity of the oblonged/worn holes in the vcu. Therefore, it would be easier to use a larger diameter pin head (or entire pin with requsite machining on the gears and carrier). This would align much better in a machined vcu that would accept those altered pins. However, I've only seen pictures of the components and I'm speculating. Sounds like some R and D would need to be done (let alone some stress analyses to confirm integrity)--custom stuff. Again, where did you say you bought a new VCU? You didn't answer that in you last post.

11-01-2007, 07:09 PM	#118 (permalink)
havok91 Forum Member Join Date: Feb 2007 Location: SLC, UT Drives: 91 RT/TT Trader Rating: (1)	Re: Backlash in the Transaxle? I looked at the pictures, again, from all the previous posts. New machined (not welded flanged--overcomplicated) would be the best bet. Have it machined from a good hardenable steel (likely tool steel, but that all depends on the loads). Dowel pins can are made from a variety of alloys, so, I have no idea what the facory pins in the planetary assembly are. The pin's strength properties could be roughly determined from a hardness test--a tensile test would be better. The pins will be in single shear regardless of the classification of fit--however,any slack/slop will add dynamic loading in addition to the static loading. Therefore, the pin's shear strength must be at least that of the original's--with more for saftey. The pins would have an interference/press fit in the carrier of the planetary assembly, and a slip/location fit in the vcu. Pins with larger heads would then be able to register in the vcu (with its oblonged holes--actually like slots), reducing slack/slop. However, the stress distribution would be more of a point type loading in the slots than something that resembles a bearing type load. The long term effects, like fatigue, are unknow unless an analysis is done. Do the slots in the vcu perform a function we are not aware of? My suspision is that the slots allow some sort of miss-alignment of the entire assembly. When I rebuild my TT's tranny, I'm going to take a closer look at this. Machine things if needed and run some numbers. I'm a big proponent on doing something better than the factory was willing to spend money on.

11-02-2007, 01:43 AM	#120 (permalink)
havok91 Forum Member Join Date: Feb 2007 Location: SLC, UT Drives: 91 RT/TT Trader Rating: (1)	Re: Backlash in the Transaxle? I think JB weld is only rated around 4000 psi. Not much. It also wouldn't be very hard, much softer than a good steel alloy, so, I don't think it would hold up. It probably won't hurt to try, though. If the 25 spline vcu doesn't have the slots, it may be viable to put in a new bolt hole pattern (exactly like stock but with tighter tolerances), which could be offset evenly with the existing holes. You said the machine shops you talked to wouldn't touch it. But, a good Mold Making shop should have a DIE SINK EDM (repeat) that can machine tough, hard metals. However, this would not be. Is it worth all that work? Maybe. Props, though, for looking into this idea. I'll look into this more when I get that far with my car. I got access to EDMs.