My E30/R5Turbo track car - Gearbox MkIII

[GeoffBob]

Mega!

Sounds like the old BTCC Sierra Cosworths, other than the rubbing wheel arch noise! Blame that on the heavy passenger...

I love it.

Thanks Theo , you just have to love that noise.

Having studied all the video footage from last Wednesday closely I am convinced that I am braking too hard and too soon (a rookie error) before most of the corners. I put this down to the fact that am too chicken sh!t to commit the car to the corners at the speed I know the car can take them. I have no doubt that I'll improve as I become more comfortable with what the car can do. By the time I'm finished that car will fit me like a glove on a man with five penises.

That wasn't the wheel arch rubbing BTW, that was the prelude to the gearbox packing in

I won't have the box out until next weekend, but I think I have worked out why I lost third gear. I am convinced the 3rd dog ring worked its way off the end of the 3rd gear. As it did so it would have been rubbed against the 3rd/4th selector ring while I was in either 1st or 2nd gear. Next time round I am going to weld the sodding dog-rings to the gears rather than splining them to each other!

[Theo]

That wasn't the wheel arch rubbing BTW, that was the prelude to the gearbox packing in

Oh, it sounds just like a wheel arch nibbling a tyre, uncanny!

I'm sure with practice you will be driving to the limits of the car. It's easy to drive a perfect lap in your mind but in practice, it's much more tricky.

Keep the updates and videos coming, I will try to make a video in my car next time it's on track. It doesn't sound nearly as cool as yours though.

[baptie0]

That looks like excellent fun Geoff, car sounds sweet, shame about the box but I am sure you will have that sorted in no time. keep the progress comming

[gareth]

By the time I'm finished that car will fit me like a glove on a man with five penises.

What a ballistic little chariot! Top marks for getting a decent car and ruining it to an amazing standard. I am in awe and consumed with envy.

I guess with the short wheelbase and large engine providing a lot of engine braking, it'll be easy to get the back end a long way out of shape while downshifting?

[TouringTash]

Every now and then I find a thread for the first time via the home page recent threads but nothing could have prepared me for a yellow Renault
Amazing and inspirational work there fella, if it wasn't 3am I would be off to the garage to dirty my mitts. If you ever feel homesick I'm in Ash next to Brands and could put you up


Just remember your spanners

[MarkT]

Sounds very simmilar to my old track car, must be fantastically satisfying to design and build something such as you gearbox and then be able to use it in anger. Hat off to you, fantastic build.

Sorry to hear about the dog ring issues

[tim_haynes]

Looks Great Geoff, Sounds Lush.

[Mikey_Boy]

Bloomin' Aida! Fan-tas-tic!

Great work Geoff - what an awesome machine!

[appletree]

Great video Geoff, looks like it realy shifts, reminds me of this car with the streight cut gears, just when you dont think it could get any louder it does!

S42B20 300+bhp all N/A!!!!

[youtube][/youtube]

[tim_haynes]

that is flipping fast and stupidly loud that E36!! Love it!

[GeoffBob]

Hi everyone. Thanks for all your comments. I've been away for a few days for work so I have some catching up to do.

You couldn't nearly have enjoyed watching the video as much as I enjoyed making it .

That wasn't the wheel arch rubbing BTW, that was the prelude to the gearbox packing in

Oh, it sounds just like a wheel arch nibbling a tyre, uncanny!

Theo, I had a good poke around under the car, but can’t find any signs of rubbing. I won’t rule out your suggestion though just yet, I have been proven wrong before. And I have to confess the noise only materialized when I had a passenger with me. Curious?

That looks like excellent fun Geoff, car sounds sweet, shame about the box but I am sure you will have that sorted in no time. keep the progress comming

Thanks Baptie. Indeed, I have no worries about fixing the box, I am (after all) the sole supplier of parts I am dead curious though as to what caused the problem.

I guess with the short wheelbase and large engine providing a lot of engine braking, it'll be easy to get the back end a long way out of shape while downshifting?

I would be inclined to agree, but so far have not yet had this experience. Keep in mind that my CR is low (turbo engine) so my engine braking is unlikely as strong as what you think. But I still suspect that you are right. Time will tell (and make a great video).

Every now and then I find a thread for the first time via the home page recent threads but nothing could have prepared me for a yellow Renault
Amazing and inspirational work there fella, if it wasn't 3am I would be off to the garage to dirty my mitts. If you ever feel homesick I'm in Ash next to Brands and could put you up

Just remember your spanners

Sir, you are very kind I’ll add you to the list of zoners I would like to meet when I am next in town.

Sounds very simmilar to my old track car, must be fantastically satisfying to design and build something such as you gearbox and then be able to use it in anger. Hat off to you, fantastic build.

Sorry to hear about the dog ring issues

Thanks Mark, satisfying hardly describes it. Even if it’s the end of the gearbox I’ll still have the satisfaction of having once built my own gearbox. Something to tell the grandchildren one day Actually, the gearbox failure really doesn’t bother me. I see it as all part of the learning process. Having to pull the engine out yet one more sodding time, Now that irks me! Having done that job at least a dozen times already I find it to be boring and tedious. Not a difficult job by any means, but at least a whole day of spannering away when I could be doing more constructive things.

Looks Great Geoff, Sounds Lush.

Thanks Tim. Maybe I'll get to bring her down to Kilarney one day.

Bloomin' Aida! Fan-tas-tic!

Great work Geoff - what an awesome machine!

Thanks Mike, I'm honoured.

Great video Geoff, looks like it realy shifts, reminds me of this car with the streight cut gears, just when you dont think it could get any louder it does!

Ta Matt. That E36 is phenomenal. Certainly ate up all the other cars in its way!

Well, the weekend is now upon me. It's p!ss!ng down with rain here at the moment, and predicted to carry on through the weekend. I hope to get a chance to pull the gearbox out, but given the general abundance of sunny weather I see no reason to force myself to work in the rain, there's always the weekend after. I'll see how it goes. Once I get the box apart I'll let you all know what went wrong. Hopefully there won't be any gruesome photo's to follow or I might just give up on the whole gearbox malarkey. At this point in time I expect to find one broken dog-ring , and that’s all. On the other hand it may only be iron-filings left We’ll see.

[GeoffBob]

Well, despite being overcast all day, no rain fell. The engine was thus removed, with the gearbox not far behind (literally), in record time. The gearbox is apart on the workbench and the problem exposed. Seems I managed to break the 3rd dog-ring, which would explain why I could put the stick into 3rd position, but found nothing there. Hmmmm, seems that 400Nm is sufficient to tear these rings apart. I'll have to think this through till I have a solution.



The good news, however, is that the gear teeth and dog-teeth are all spotless. Not a scratch or sign of wear in sight. I am very pleased about this since it indicates that the standard of the local work (particularly the process of hardening the parts) is excellent. I honestly expected to find a few rounded edges, and I honestly found none!

Once I have figured out how to overcome what is clearly a flaw in my design I think I am going to have a top quality product on my hands. And I already have an idea how to solve this one. I won't say anymore just yet.

[gareth]

Ouch! Not all bad though if the rest is all still good.

Looking at that failure, i'd point the finger of blame largely at the lack of radius around the 'ears' causing a nasty stress raiser

Is there sufficient room for a radius in there? Even an undercut one would help if space is tight.

[GeoffBob]

Looking at that failure, i'd point the finger of blame largely at the lack of radius around the 'ears' causing a nasty stress raiser

Good point Gareth! There is a slight undercut (an artefact of the CNC manufacturing process) but not nearly enough. Problem is, the deeper the undercut the less material there is left behind in the ring, and hence the greater the hoop stress. What I really need is for that ring (with its teeth) to be part of the actual gear, not a separate item that splines onto the side of the gear. Problem is, the gear teeth have to be hobbed, and this requires that the cutter overshoot the gear each time, which means the dog teeth are in the way.

I am thinking of keeping them as separate items for the time being, but welding them together before final hardening and grinding. At least then the forces upon the ring (causing it to tear apart) will be more evenly transmitted to the gear, rather than through each of the discrete splines, which are themselves, stress raisers. If that doesn’t work then I have a serious problem.

[gareth]

I see your point regarding the hoop stress, however those corners could cause a stress raiser effectively doubling the stress in that area

there's a little chart i found here which gives some idea of stress concentration factors.
http://www.mae.ncsu.edu/eischen/courses ... ndix_C.pdf

You may well be better off losing 20% of the cross sectional area in favour of a lower stress concentration... Food for thought

[GeoffBob]

there's a little chart i found here which gives some idea of stress concentration factors.
http://www.mae.ncsu.edu/eischen/courses ... ndix_C.pdf

Ta Gareth. Not sure what book that's from, but BTW, those charts are identical to those presented in Appendix A of Mechanical Engineering Design by Shigley, Mischke and Budynas. This is the book that I mostly work out of.

You may well be better off losing 20% of the cross sectional area in favour of a lower stress concentration... Food for thought

Point well made Gareth, I shall certainly take what you say into account. I think I will have all five dog-rings remanufactured (not just the broken one) with an additional radius at the base (one either side) of each tooth.

It is my plan, however, to weld these rings to each gear to keep them in place. I suspect that the odd noise in the video (the one that Theo said sounded like a tyre on a wheel-arch) was the sound of the 3rd dog-ring working it's way off the end of the gear splines. That would have left it rubbing against the edge of the selector ring, which, in any other gear would have been rotating at an altogether different speed. Hence, 3rd gear was effectively trying to engage itself while I was in 1st or 2nd, which is when the noise was present. That would definitely have been enough to destroy the 3rd dog-ring!

FYI, although fairly stiff, I can currently manipulate the position of the remaining four dog-rings (on the end of each gear) by pressing a flat screwdriver up against the edge of the dog-teeth. That tells me that the others were also in the process of working their way loose and that they were awaiting the same fate. What started as an interference fit was working its way towards becoming a slide fit .

[billgatese30]

That video is awesome Geoff. Shame about the box, but as you say, its a great learning exercise.

FWIW I would have thought that the only reason that the dog ring failed was because it came loose on the shaft, therefore by welding it to the gear, would eliminate this risk, meaning that you wouldn't need to remanufacture the other dog rings with larger radii (although it would be a good plan on any new ones being machined) as they should never suffer the same fate.

[gareth]

Looks like you've got it sussed then

Oh i have no idea what that extract is from, i just looked on google for examples, though it's also very similar to a book i have at work

[GeoffBob]

FWIW I would have thought that the only reason that the dog ring failed was because it came loose on the shaft, therefore by welding it to the gear, would eliminate this risk, meaning that you wouldn't need to remanufacture the other dog rings with larger radii (although it would be a good plan on any new ones being machined) as they should never suffer the same fate.

Fair comment Chris, and it is tempting for me to do exactly as you suggest. Problem is, what if I break another one and its down to pure torque on the ring? After that I'll have to have both gears and rings remanufactured due to the weld. It's worth a try though. The worst that can happen is that I have to pull the box out again. I could always pre-empt the next failure by having a set of new design gears and dog-rings made up after the box goes back in (and hope that the new pieces are ready before the box breaks again). These new bits could be made to the new spec (added radius as per Gareth's suggestion and improved splines plus weld etc. In fact, come to think of it, that’s not a bad idea. That way I get to find out (the hard way) whether it was pure torque that broke the ring, or whether it was the ring coming loose that ended its little life.

I am inclined, however, to believe that it was the latter that killed the ring. Following simple Newtonian mechanics, the dog-ring that would be most likely to yield under strain would be the first dog ring, not the third (and it’s not as if I wasn’t hammering 1st gear coming out of the hair-pin). I think it’s likely that 3rd broke simply because it was the first to come lose. Of course, I have no proof of this - it’s all speculation on my part. Only a number of tests on track can confirm either way. If she never breaks again then I’ll know I got it right .

[GeoffBob]

I just got off the phone with Veron Pappas at Arma Gears and he has advised me to rather start over with the gears and dog-rings. We can weld the rings to the gears, but this must be done pre heat treatment if the weld is to last. I have thus decided to remake each of the gears and dog-rings (new design), weld the gear and ring together, and then heat treat them. That should see the box right, imo.

BTW, when I told him it was third that I broke his immediate response (before I could even say anything) was that I would have broken 1st if it was a torque issue. Interesting. He agrees that it broke because it came loose on the splines, and that welding is the solution.

[GeoffBob]

Some more Zwartkops footage for you all, and this time its two very impressive E30's (well, I like them anyway). Enjoy.

[youtube][/youtube]

[GeoffBob]

And another one of the Alpina B3 at Kyalami.

[youtube][/youtube]

[billgatese30]

I just got off the phone with Veron Pappas at Arma Gears and he has advised me to rather start over with the gears and dog-rings. We can weld the rings to the gears, but this must be done pre heat treatment if the weld is to last.

Can't believe I never thought of that before. I guess thats why he earns the big bucks !

Brand new set of gears it is then. At least you'll have some spares so the car can still be moved etc under its own steam should you bust a gear at the track (although you wouldn't want to put them through 100% race pace but you can still drive under its own power.)

[GeoffBob]

Can't believe I never thought of that before. I guess that’s why he earns the big bucks !

I don’t think he was trying to steal you thunder Chris . His point was that the welding process has to occur before heat treatment and final grinding. And unfortunately, all my gears have already been hardened and ground, which basically makes them likely to crack when welded (much like cast iron often does). Believe me, if I could just weld what I already have (as per your suggestion) I honestly would. Then all I’d need is the 3rd dog-ring remanufactured and my car would be back on its feet again. Now I am faced with months of waiting for new gears to be made. I am not happy about this .

[billgatese30]

I never gave much thought to them cracking whilst welding, but it did cross my mind that they would have to be hardened/tempered again otherwise the weld zones would be adversley affected, what I neglected to remember was the fact that they have already been ground to size, so any heating (even by the welding, definately by hardening) would probably warp them ever so slightly and render them useless.

I think that is what is known as a School Boy Error

Shame about the wait, especially as it currently your summer. If you were over here it wouldn't have been half as bad.

[gareth]

As a cheaper option, how about remaking the dog rings then trying some form of mechanical retention to keep them in place. roll pin, circlip etc.

What sort of loads are these likely to see?

[GeoffBob]

Shame about the wait, especially as it currently your summer. If you were over here it wouldn't have been half as bad.

I am mulling over whether to put back a stock W55 box for the time being. I still have the one that I started out with on the project. Probably not a bad idea, although it'll mean shifting the engine in-out again when the custom box is repaired. The alternative is that I leave my car standing for at least 3-months (collecting surface rust on the brakes again) while I wait for the new gears to be manufactured. Of course I won't have my lovely custom ratios (which I have to say make such a difference on the track) but it's better than having no car at all.

As a cheaper option, how about remaking the dog rings then trying some form of mechanical retention to keep them in place. roll pin, circlip etc.?

No room I'm afraid Gareth. The joys of trying to build a set of straight-cut synchroless gears into the space of a stock gear casing.

What sort of loads are these likely to see?

The full load transmitted from the driven (pinion) gear (driven by the lay/counter shaft) via the dog-ring to the selector ring to the selector hub to the output-shaft.

FYI, The only reason that the dog and selector rings can be manufactured so much smaller than the actual gear-pairs is that gears transmit the shaft torque via forces exerted upon any two mating teeth that happen to be in contact at the time, while the dog-rings and selector rings use multiple teeth that engage simultaneously. If its absolute figures you are after I have the calculations in my file.

[gareth]

Nope, that gives me the idea for the loads. I should have been more specific though

What i was thinking was: If the dog ring working lose and getting chewed up is what broke it, could a few small roll pins retain it in position on the shaft. They'd only have to withstand axial loads. only one new dog ring would need to be made and the others could be modified. Quick, cheap, relatively easy but does add another stress raiser to the dog ring. maybe it should go through the thickest part?

It may be a stupid idea but if enough stupid ideas are thrown around, a good one could come out somewhere!

[GeoffBob]

Great idea Gareth, and I welcome your input. Sadly, there simply is no room to fit any sort of mechanical retaining mechanism. And making room would weaken the ring even further.

You are right, though, about the retainer only having to withstand the axial load. I am considering silver-soldering the rings to the gears as I think this would provide adequate strength and could be done at relatively low temperature. Some strategically placed flux and I think the silver would suck right into the gaps between the splines. Just an idea at this time, not one that I plan to act on just yet.

[keri-WMS]

What temperatures might the gears and oil reach do you think?

[Ziggy]

I am considering silver-soldering the rings to the gears as I think this would provide adequate strength and could be done at relatively low temperature. Some strategically placed flux and I think the silver would suck right into the gaps between the splines. Just an idea at this time, not one that I plan to act on just yet.

Brazing (or even epoxy) was my first thought when it transpired that you couldn't weld...

[keri-WMS]

I asked about temperatures as I was wondering about Loctite 648, in testing it stays at 100% strength when aged in motor oil at 125 degrees. Constant 150 degrees looks ok, disassamby is at 250 degrees.

http://www.loctite.sg/sea/content_data/ ... mpound.pdf

Again probably a daft idea, low risk to the existing bits if it works though!

[GeoffBob]

Thanks gents.
No idea is a bad idea, every little bit helps.

Ziggy, I don't think epoxy would work as it will likely shear (even if not affected by the oil), but brazing would definitely work .

Keri, you're a genius! From that datasheet you posted a link to: "Typical applications include holding gears and sprockets onto gearbox shafts and rotors on electric motor shafts". Even if it doesn't work it's still positive info as I'll know for next time that something stronger is required.

My idea at the moment:

Get new gears and rings (welded to each other) made in the long term (spares on the shelf ready for the next hiccup).

Get new dog-rings with tighter splines made up in the short term, and press these onto the existing gears with loctite 648, as per Keri's suggestion.

As an experiment I'll see about brazing some of the original (stock) rings and gears together to see if it works. If the brazing rod doesn't take or it requires too high a temperature I'll try silver solder instead. An hydraulic press will work to evaluate the strength of the resulting bits .

[keri-WMS]

Keri, you're a genius!

Ah shucks!!!

(...just hope it works out well now! )

If the bits are not a tight enough fit then there's another type with particles in it that works well (606 or 601 rings a bell) but I assume you won't need that?

[GeoffBob]

I'm sure it will work fine Keri

Three photo's, courtesy of Robin Miles, taken at Zwartkops. 'Had a passenger in the car with me. Fortunately he ain't heavy, he's my .....