ZF failure?

[Russ Noble]

Re: ZF STRENGTH !!!

Great stuff Ron. Just what we're looking for. For comparison with other trannies.

Gear widths

Shaft spacing. That's the distance between centres of the two gear shafts

Ring gear diameter, that's to the outside of the teeth.

Any factory strengthening features eg steel plates to strengthen the aluminium case where the gear shaft bearings are housed,or anything else.

Whilst there are other factors that affect the strength, these are the basic design paramaters that are easily compared.

Ross, sorry for the wind up. :dead: It's just that everyone here seems to send their stuff back to RBT or one of the other ZF specialists, and never look inside themselves. I thought you were our only hope! BTW. Notice I've changed the title of the post, just for you! :chug:

Cheers

 

[wanni]

Watching the picture I can say as follows:
Rupture for shafts deflection.
1) The 4th gear is nearly in the middle of the shafts.
2) The transitory under torque is quiet long in forth gear.
If you look at it in the details, you can appreciate that the teeth brake not at the bottom, as it should be,by higher on the tooth profile. By deflecting so much, the lever as increased a lot and that why this strange failure happend.
Probably the quality of the steel is also not the best, but the main cause is because the driver is offen accelerating hardly in fourth gear.
To avoid this failure, you shall decrese the pressure angle in order to reduce the radial forces and increase drasticly the helix angle in order to reach a cooperation over 2.2.
Regards
Wanni

 

[Russ Noble]

Re: ZF STRENGTH!!!

Wanni, thanks for your expert analysis of the failure.

I'm not sure whether I am understanding you correctly but what you appear to be saying about curing the problem seems to be the opposite of that posted by Randy in post #32 on Richard's thread http://www.gt40s.com/forum/manufacturer-company-announcements/20760-new-trt-transaxle-2.html

Or are you saying the same thing, and I've just misunderstood?

Cheers

Edit:
Having thought about it, am I right in thinking you are looking at this from the point of view of reducing shaft delection, as opposed to making the teeth stronger? Which do you think would be the better way to go?

Racing gearboxes all appear to go for stronger (straight cut) teeth presumably at the expense of greater shaft deflection and of course noise. They have obviously found this the most satisfactory solution for any given gearbox?

 

[Ron McCall]

Re: ZF STRENGTH !!!

Great stuff Ron. Just what we're looking for. For comparison with other trannies.

Gear widths

Input side:
1st-.670" wide 4.475" diameter.
2nd-.710 wide 3.890" diameter.
3rd-.640" wide 3.380" diameter.
4th-.640" wide 3.070" diameter.
5th-.650" wide 2.705" diameter.

Output side:
1st-.810" wide 2.275" diameter
2nd-.775" wide 2.750" diameter
3rd-.725" wide 3.260" diameter
4th-.690" wide 3.550" diameter
5th-.650" wide 3.940" diameter.


Shaft spacing. That's the distance between centres of the two gear shafts

3 1/16"

Ring gear diameter, that's to the outside of the teeth.

8 7/16"

Any factory strengthening features eg steel plates to strengthen the aluminium case where the gear shaft bearings are housed,or anything else.

Whilst there are other factors that affect the strength, these are the basic design paramaters that are easily compared.

Ross, sorry for the wind up. :dead: It's just that everyone here seems to send their stuff back to RBT or one of the other ZF specialists, and never look inside themselves. I thought you were our only hope! BTW. Notice I've changed the title of the post, just for you! :chug:

Cheers

Please note that the transmission I am working on has a non stock 5th gear installed.All other ratios are stock (-2 ZF)

 

[BruceA]

Wanni, either you're coping off me or we are thinking alike?!

Ron, excellent stuff there. Im just having the usual trouble doing a metric conversion!!
Are you able to measure the in/out shafts please and report on their diameter? Probably looking for minimum and maximum diameters of the shafts.
Also, what is the distance between the bearing centres of the in/out shafts?
Oh just one other thing, what is the maximum diameter of the pinion gear, the side closest to the bearing?

Ross, do you run a solid centre or sprung centre clutch plate?

 

[Russ Noble]

Re: ZF STRENGTH !!!

Ron, that was nice and quick. BTW I think you've got your input/output round the wrong way!
Here's how ZF measurements (converted to mm) compare with measurements that we have so far from the other Porsche trans. (the main ZF competitors)

Ring gear
ZF, 215
930, 235
G50/5x, 210
G50/0x, 190

Shaft spacing
ZF, 79
930, 76
G50/5x, 85
g50/0x, 85

Gear widths 3rd/4th gear
ZF, 16
930, 24

That's all we have at present. I'll add to this comparison as more measurements become available.

Cheers

 

[wanni]

Hi all,
the center distance between the shafts is 79 mm in the ZF and in most of the Porsche boxes.
This small distance does not allow the shafts to be strong enough.
Russ, you missunderstood. If we considere spur gears ( straight cutting ) we will have only radial ( lateral ) forces. We shall encrese the helix angle in order to reduce them by encreasing the axial forces, if the trust bearings allow this.
Normally the secondary shaft ( pinion ) has no problems because the forces are contrary to the pinion axial forces.
Meaning: the pinion pushes backward and the gears are pushing forward. This allow the fitting of a 4 contacts ball bearing to keep all that in state of to conical bearings as for most of the differentials.
3rd and 4th which are in the middle of the shafts shall have a smaller module in order to push the helix angle over 28-32 degrees. This change will also encrease the cooperation between the to wheels going over 2.0 ( 2 teeth are always ingaged ). Even if the module is smaller, the gear life will triplicate.
Check a TREMEC gearbox, por example a T 56. The teeth are so bended and thin that you could think that those guys are crazy. Not at all. They are probably reaching 3.0 of cooperation with the long addendum cutting and such an angle. Of course the efficiency is killed, but the boxes are strong and quiet. Also the noise level has to be considered.
At RBT they shall learn from TREMEC about gearbox architecture. Especialy for road purposes. Motor racing is a different story. The budjets are allowing to replace the gears at every race. All of you have a different wish about reliability.
Ciao
Wanni

 

[Russ Noble]

Thanks for that explanation Wanni.

 

[Ross Nicol]

Re: ZF supposed failure, in your dreams

I've just got back from Phillip Island Historic meeting and 1109 was there again.Guess what transaxle it has in it? yep an RBT/ZF. It was lapping in 1min 45 to 47sec, similar to my times.I spoke to the chief mechanic and he told me they took the old original box out because they had case cracking trouble but no gear trouble at all.The RBT/ZF has given not an ounce of trouble. The pit board listed 450bhp which woud be about right.My defence case in favour of the ZF's reliability is getting stronger.Anyone care to argue that the race environment is the harshest, gloves on then.Rambo Lambo Bruce Rambo (surely one of them must be your real name) Come on fellow kiwi you can tell us.I have springs in the centre of my scintered Iron clutch plate.Have no fear Russ I don't feel wound up by you and your right I will get inside my ZF if I have any trouble.
Rossepper: Pepper for me please

 

[Sandy]

Ross -

No flame suit needed Ross all good discusson. The problem that I have with the ZF more then anything is that (to me IMO) it looks whimpy (This is not a dig on any folks that use them). Their, I said it aloud. As I said they seem to be a very durable box given all the folks (Including yourself) that use them with some reasonable motors and on the track. My opinion should not take any crediblity from the ZF it is a good box, just not one I want to run.

The sticky tire thing I don't agree with having broken a couple of rear ends in another vehicle only after putting some sticky-er (My technical term) tires. Thought it might just be old drive line (Twisted a pinion in half), then broke the diff a second time. Granted our cars are light so I really don't know where the tire/clutch becomes the fuse or the gear box does.

Your also right the thread name might be better as 'ZF Capabilities' or some such thing. At least we are getting some data on gear width etc.

The funny thing is for sure I'm now jinxed on blowing the 930 box!

Sandy

 

[Ross Nicol]

Re: ZF capabilities "I just love mine"

Yep Sandy the world would be a boring place if we all thought the same.
I can't totally agree with you that the ZF looks whimpy but that is your opinion and points to you for stating it. I would agree if you had said an Audi looks wimpy though.I have seen a failure or two from the Cobra boys with their Jag rear ends.Driver induced of course where too much loud pedal is used on the grass while returning to the track(after an off). As soon as the driven wheels hit the bitumen the output shafts twist like a carrot. A driver with a more mild temperament never has these sort of failures. I could have put that many different ways, that statement. Now whether you have jinxed yourself with the 930 only time will tell but I hope you will be prepared for my comments if the worst happens. Solid driveshafts made from 4130 or better, heat treated to a Rockwell standard with plunging c/v at each end will sort driveshaft weaknesses.The head mechanic for 1109 identified the rubber doughnuts in the driveshafts as the reason for reliability but I say again I have had no trouble with mine.
I like the term fuse Sandy as it describes the weak point that can act to save other parts. Well done. I think your previous driveline trouble may have been self induced, now will you be honest and fess up to this? I agree all good discussion, only thing is now I really want that 930 to blow up and Russ will probably hate me for that.
Flame retardent Ross

 

[Russ Noble]

Is Ross' ZF going to blow before Sandy or Russ' 930?

Regardless of the torque of the motor the limiting factor of how much torque is put through the gearbox is invariably the amount of grip of the tyres. Light them up off the line and there is very little torque being transmitted, because at that point there is little grip. That is obvious when you see someone slow off the line because they had too much wheelspin.The wheels and hence the gearbox are not transmitting enough torque to move the car forward. The maximum torque of the motor is transmitted through the driveline when there is no wheelspin and the drivers boot is hard in it at revs corresponding to the max torque of the motor. So usually in 4th or 5th. On and off the throttle in that situation would put additional shock loadings in as well I would think. I'm sure the rubber donuts on the originals will virtually eliminate any shock loading thus immensely helping longevity.

My choice of trans given the available suitable options of 930, G50/5x or ZF and given their approximate equality came down to ECONOMICS. $1500 for a 930 that I might have to slip some bearings into and maybe file up the syncros a bit, as against the sort of prices the 5 speed trannies were going for. Had a G50/5X or a ZF come up for that sort of money instead of the 930, then that's what I would have gone with. It was a no brainer and like Sandy I was also comfortable with the 930 capabitities. Time will tell. I am traditionally hard on gearboxes, so mine may well feature on the "930's failure" thread before anyone elses.No cushioning in mine either, solid centre triple plate and CV's and 14" rears and reasonable torque. In saying that no-one has yet pointed to ANY 930 failures and I'm sure some of them must be driven by a selection of the brainless leadfoot peddlars that Ross has alluded to. I hope that's not a description of Sandy, is it?:lol:

And no Ross, I won't hate you if a 930 blows up. It's all in the interests of science!

Really, I don't think there's all that much between any of them. The ZF has got to get the vote because it was original fitment but I'm sure any of the Porsche turbos will at least match it strengthwise for less dollars. IMHO of course!

Cheers

 

[Sandy]

Sadly to admit, both drive line failures were from Speed shifting a 377" small block powered International Scout. Never had a problem until I got some big off road tires that at the time seemed like a good idea. Same bad driving and at the time I would say both breakages were as you put it mindless, and I'll be the first to admit it

The failures were interesting, twisted the pinion clean apart in the center, then the second broke the spider gears. Was a Dana 44, so not all that tough, but toss in some big tires, and a good sized motor with loads of TQ and boom.

I have spares on the 930's, but if it blows I'll fess up!

Sandy

 

[BruceA]

Russ, here are some more 'box details for your comparison chart, the main details of a Renault UN1-13 gearbox, the unit from the 21 turbo.

Crownwheel dia 190mm
Pinion size dia 67 down to 52mm

Shaft spacing 76mm

Distance between shaft bearings 232mm

Gear size:
1st 20mm (can be made up to 25mm, mine is a custom made gear)
2nd 20mm
3rd/4th 22mm
5th 18mm (I will be making a new gear for this. New width will be 21mm)

The original input shaft has undercuts and wastings with a minimum size of only 17mm!! My new 1-peice shaft has a minimum of 27mm and no undercuts.

So when you look at the other gearbox sizes, one thing sticks out in terms of strength. First is the crownwheel size and what may be 2nd, is shaft diameter. From peoples use of the various gearboxes, it does seem like, for current power usage, shaft spacing is not a great factor in strength, nor is gear size, given the perhaps smaller size of the ZF gears (width)

 

[Russ Noble]

So when you look at the other gearbox sizes, one thing sticks out in terms of strength. First is the crownwheel size and what may be 2nd, is shaft diameter. From peoples use of the various gearboxes, it does seem like, for current power usage, shaft spacing is not a great factor in strength, nor is gear size, given the perhaps smaller size of the ZF gears (width)

Not sure that I can entirely agree with that analysis of gear size relevance Lambo. In reality it is the gears that are shown here failing in the ZF. If they were considerably wider, then surely they would be quite a bit stonger and maybe able cope with the extra stress induced by shaft deflection?

Shaft spacing approximately dictates shaft diameter. Wanni states that shaft spacing of 79 mm on the ZF does not allow the shafts to be built strong enough. Interestingly the 930 shaft spacing is 76mm but the 3rd/4th gear width is 24mm ie 50% wider than the ZF. No-one has so far been able to document a 930 failure whatsoever. Maybe we should rename THAT thread! :dead:

Thanks for those UN1-13 measurements, I'll try and add them in later.

Cheers

 

[BruceA]

In reality it is the gears that are shown here failing in the ZF. If they were considerably wider, then surely they would be quite a bit stonger and maybe able cope with the extra stress induced by shaft deflection?

Yes, that is what I said above, wider gears are not failing. Perhaps in the case of the ZF, where 4th is in the middle of the shaft support bearings where shaft deflection is greatest.

Shaft spacing approximately dictates shaft diameter. Wanni states that shaft spacing of 79 mm on the ZF does not allow the shafts to be built strong enough. Interestingly the 930 shaft spacing is 76mm but the 3rd/4th gear width is 24mm ie 50% wider than the ZF. No-one has so far been able to document a 930 failure whatsoever. Maybe we should rename THAT thread!

True true. I would really like to know what the shaft sizes are on a 930 and ZF - as this shaft deflection syndrome! is looking more and more like a weak spot.

Just as a matter of interest, what was the issue with all the Colotti gearboxes on the orginal 40's? What kept failing in them?


Cheers
Bruce

 

[Rick Muck- Mark IV]

Just as a matter of interest, what was the issue with all the Colotti gearboxes on the orginal 40's? What kept failing in them?


Cheers
Bruce

The fact they were made out of compressed pasta? The ring and pinions failed often, Ford even made up R&P sets from the steel used in the good ole' 9" axle to improve reliability. The newly available ZF with its' FIVE speeds made the Colletti obsolete....why continue to fettle if when there was a ready to go, off the shelf gearbox.

Rick

 

[Jim Cowden]

Wanni

A technical question
Is there a relationship between gear width and deflextion of the shaft.

Example.

2 gearboxes both have a main shaft of 25mm od.
Both the same material.
box 1 has 16mm wide gears.
box 2 25mm gears.
Both the same bearing spacings.
Disregarding gear strength.


Does the surface area of the 25mm cause less deflection due to the load being spread over a larger area.
And with the narrower 16mm gears do they cause more deflection due to the same load on a narrower area.

The narrower area being the bearing width on each gear.
Or does it just come down to shaft diameter.

Jim

 

[Wanni Albertini]

The parameters which are influancing the shafts deflection are the helix angle and the cutting pressure angle.
Of course if the hubs of the wheels are larger, than the shafts are deflecting less by rigidity add on the external.
Ciao
wanni

 

[DCraig]

My first post here. I’ve worked on Porsches since 1968 but I still love the project cars. Perhaps I can help by adding my two cents worth on the ZF boxes used in Porsches.

Unless there are some specific questions concerning the 901 boxes, I will not include them since they don’t get along well with higher horsepower inputs. Their normal mode of failure is to twist the rear of the main input shaft off. All Porsche ZF boxes use helical cut gears.

The 915 box is probably the all time favorite transaxle to use in mid engine cars because they are reasonably plentiful; you can flip the ring gear for correct rotation; parts are still available (but getting pricy); limited slip as well as Quaife differentials are available; and unless you abuse them, they will last a long time and give trouble free service. More on abuse later. The aluminum cases are a bit stronger than the magnesium cases but case failures, except for pushing out the differential side cover, are only found when internal parts get broken and fed through the gear train. The factory found that it was cheaper for them to produce the aluminum cases, the strength factor was a bonus. The early 915 boxes had skinny little ears on the guide sleeve that rides inside the slider. They were prone to failure. These were increased in size in about 1976 and are used in all later boxes. They are upgradeable to the early boxes.

The 930 boxes have gears that are 30% wider than the 915 gears. The G-50 gears are also 30% wider than the 915 gears. This was done because under racing conditions and big time horsepower, the gears would fail occasionally. I have never seen a gear failure on a 930/935 box that wasn’t caused by debris from another failed part.

The 934/935 boxes have internal oil pumps (915 race car boxes did also); a spray bar directing oil against the ring gear were it meshes with the pinion gear; input and pinion shafts have drilled oil slots; and the input and pinion shafts are nitrided. To my knowledge these are the strongest boxes made.

Now lets talk about abuse. We use friction synchros up until the G-50 box which uses GM style synchros. Friction synchros DON’T react well to speed shifting or short ratio shift towers. It overpowers the synchro and lets the slider teeth bang into the gear engagement teeth. That’s the grinding noise you hear since the gears on the input shaft and pinion shaft stay in constant mesh. If anti-friction gear oil additives are used the synchros stop working.

The number one killer of transaxles is when the driver downshifts, especially into first gear, and the expanders over speed and the anchor block breaks. This spreads the synchro ring and feeds it through the internals. Not pretty. If the driver shifted into first over 20MPH on a 928, even if the clutch was never engaged, the anchor block said bye-bye. The number two killer is starting the engine without oil in the transaxle. The needle bearing last less than 60 seconds. Number three killer is using regular gear oil in a limited slip. Limited slip gear oil STINKS badly. I have used Swepco oils for many years.

Under big torque loads open differentials break their spider gears; LSDs can warp where the ring gear meets the carrier; any of them can push the cover plate into deformation, which opens the ring and pinion clearance, and can then clean all the teeth off the ring gear; I’ve never seen a titanium spool hub fail.

I believe we are dealing with hard driven street cars here instead of all out race cars that take full power abuse for long periods of time.

If I can figure out how to do it, I’ll post pictures.