My E30/R5Turbo track car - Gearbox MkIII

[GeoffBob]

But the thing that hurts the most is you know that the design WORKS!!! Its just a shame that you have to rely on the shambles of a shower of dickheads to do the work.

Hmmm, yes. You’ve touched a sore spot there Chris. The design does indeed work and don’t I know it I’ve driven the car with the gearbox installed and for a while I tasted victory. Now I find myself abandoning my design when all I set out to do was resolve a few reliability issues . Doesn’t that just suck! Cut the dickheads from the equation though and I figure all will come right in the end.

Chris, thanks for your kind offer mate (very kind of you indeed ), and thanks Alex for the info (PM received, I will respond). I’m going to consider my options for a few days before I rush into anything. There are a fair few reputable workshops left here in SA if I am prepared to have the gears wire-cut instead. I figure that I need to exhaust all options here before I consider having the parts made elsewhere. Sadly, the really good workshops never need to advertise so they are difficult to locate (other than by word of mouth). ARMA gears I found on the internet ”“ I should have known

[GeoffBob]

Currently investigating JH Precision Engineering as a potential solution: http://www.jhprecision.co.za/index.htm

See the list of machines, not bad. I'd be interested to hear comments.

Gears would be wire-cut so I would have to separate the gears on the lay-shaft from the shaft, which is no big deal TBH. That brings forth the option of changing gear ratios if the gears are splined to the shaft. JH has done CNC turning and milling for me before, but never any wire cutting. I am trying to organise a meeting to discuss issues of viability and accuracy. Doesn't help me if he does a top notch job with a wire cutter but it works out cheaper to import an aftermarket race gear-box from Quaife or e-lite, so I'll have to way up the pro's and cons.

[martinpallot]

Geoff, have you considered speaking to beagle engineering about machining your gears, they make the elite boxes and I have heard they do a lot of custom work for the american market at decent prices. That way you could have the benefit of picking a gearbox manufacturers brains about not just the process but areas in which the design can be enhanced?

[GeoffBob]

Thanks Martin, I appreciate the info. I'll definitely look into Beagle. Has the disadvantage that they aren't local to me, but if they are making the e-lite boxes then I doubt I would have quality issues and would not have to worry about sending parts back.

[GeoffBob]

Oooooooh

http://www.beagleengineering.co.uk/

[martinpallot]

Thats the one. I would hope that there would be no quality control issues. I supplied them with very sketchy drawings for my input and output shafts when they were making my gearbox and they used their initiative to turn some sketches I did at work on a piece of paper and emailed through to them into the real thing. Would be interested to see what they can do for you.

[GeoffBob]

Thanks Martin. I've had a few others PM recommendations through to me as well, so it seems I now have a fair number of reputable companies at my disposal, although I have to admit that Beagle do seem like the obvious choice since gears are their business. I cannot deny, however, that I want desperately to keep my manufacturing local if at all possible. I have been approached by a number of individuals locally who are interested in having me modify their own gearboxes for them and all want the parts to be manufactured arms length away. If we have to reach overseas for our parts then we are back in the position where we might as well buy in an outside gearbox (such as an E-lite, Quaife or Sainz). If, ultimately, we cannot have the parts made locally then the best option is to abandon the project altogether. Don’t get me wrong, in order to get my own gearbox up and running I’ll happily have the parts made outside the country and then call it a day, but if my project is to grow into something bigger I will ultimately have to manufacture locally or not at all.

In this regard I spent yesterday afternoon with Jan Hugo of JH engineering discussing my current problems and whether we can achieve the desired result with wire cutting. As expected, he is no gear expert (but then neither were ARMA as it turned out, and that was supposed to be their trade) but he is certainly well educated with regard to materials and tolerances. He has made gears before (for the defence industry) and showed me samples. The teeth certainly look more refined than those of ARMA, despite the fact that they are wire cut and not hobbed & ground. Jan does not do grinding in-house, but does outsource the grinding of parts where required. He does this at his own risk. Payment is upon completion of the job (no deposit required) and he sources all materials and takes care of all hardening. Bonus! His QC takes the form of a fully instrumented measuring room as opposed to ARMA’s ”aold fart with a vernier calliper standing in the corner”a. I submitted my drawings to Jan and I expect a quote back next week. Given the machines he has to maintain I doubt he will be cheap, but I have reached the point where I am tired of paying peanuts to buy monkeys. So long as he produces the goods to the specified tolerances cheaper than an imported gearbox I will continue my ambitious plan of producing a locally manufactured race gearbox. If he can’t then I will have to look to somewhere like Beagle just to get my own box up and running and forget the idea of similarly modifying gearboxes for others.

I think that what has to be emphasised here is that JH is sufficiently confident that he can do the job that he is willing to take the work on without me putting a penny in, not even for materials. And if he can’t produce the goods I simply walk away and refuse the parts, so no skin off my nose. For that I am, quite honestly, prepared to pay a bit more. We’ll see how it goes.

[GeoffBob]

Been thinking hard the whole of yesterday and I am currently having strange pornographic ideas WRT my gear ratios. If I am to have my gears wire cut then I am no longer restricted to metric module or other stock tooth profiles as defined by stock cutters. Means I have a lot more flexibility in defining my gear ratios. Hmmm, this could be both fun and productive.

I am also thinking of once again changing my dog-ring design to suit to JH's clearly superior CNC machines. My current design is optimised for the rather old fashioned machinery that ARMA were operating. After Friday's meeting I see solutions where previously they didn't exist. The more I think about it the more I think these new fangled CNC machines can yield far superior results. Goodness, I'm coming over all queer, I may just feel the need to replace my wooden wagon wheels with pneumatic tyres .

[billgatese30]

Horseless carraige Geoff! I call for a good old faashioned witch hunt

Sounds like your taking the positives from this experience at least. Be interesting to see how not being restricted to traditional forms affects possible ratio's

[GeoffBob]

Sounds like your taking the positives from this experience at least.

Not much else I can take away from it Chris. I sure as hell couldn't walk away from ARMA with a working gearbox If I were doing this for a living I'd be absolutely livid, but more than anything else I am doing this for my education. I’m basically paying to find out what works and what doesn’t. So long as I continue to enjoy the learning experience I’ll keep at it. The dick-heads I meet on the way will lose my custom, and that’s their loss, not mine.

Be interesting to see how not being restricted to traditional forms affects possible ratio's

Will still be restricted in the sense that gears must always have integer number of teeth, so ratios are not totally flexible. However, having to conform as close as possible to certain cutter profiles (to avoid excessive undercutting and hence weakening the tooth) meant that the ratios were restricted even further. The main difference with wire cutting (as I have been reading) is that you have a greater deal more freedom with regard to the fillet radius at the root of the tooth (which helps define the strength of each tooth) and the pressure angle can be whatever you decide is best suited to your application. With a metric module cutter you are restricted to a pressure angle of 20’. Having designed the profile of the teeth in CAD you can then examine properties such as contact ratio and interference in order to assess friction and noise. With a hobbed gear you can design in CAD to your hearts content, but are still very much at the mercy of the hobbing machine and the operator as to whether what comes out will conform to your design. The point is, you have to trust the gear-cutter. With a CNC controlled CAD defined profile, operator and machine error seem to be less of an issue, but get your tooth profile wrong and you’re screwed. Your box could scream like a banshee right up until the moment when she strips teeth. After my experience with ARMA hobbing gears for me, and my years of good experience with CAD, I’m keen to give the latter method a try. As with ARMA, it may not work out, but I’m keen to learn anyway.

[GeoffBob]

Has been a while since I posted an update, so here goes.

My experience with ARMA gears left more of a bad taste in my mouth than I realised and so I decided to take a break from the car for a while. This time afforded me the opportunity to tackle a few other unrelated projects that had been lagging behind, as well as spending some of my weekends blissfully undertaking bugger all. And now, having recovered my "mojo" I am back with a burning desire to see this gearbox completed.

As a function of my interaction with Jan Hugo Engineering I have thus far redesigned my gearbox to take advantage of his wire-cutting machines. This was a more involved process than I initially realised and has culminated in the manual design the gear teeth in order to generate the DXF files from which the gears will be cut. The process of designing the teeth is a little more involved than simply choosing the pressure angle and then plotting the profile of each tooth (which follows an involute curve) and requires specific attention be paid to the root of each tooth in order to maximise tooth strength. By sizing gear width and tooth profile correctly I have managed to limit the maximum shear stress applied to each tooth to below 250MPa, which is well within budget for a hardened carbon-steel gear. So no issues there.

Shown below are the details of the five gears and five pinions that make up the proposed gear set.



Ratios are thus as follow:

1st = 2.040:1
2nd = 1.533:1
3rd = 1.211:1
4th = 1.000:1
5th = 0.859:1

The fact that the ratio of each gear and pinion was not limited by the imposed use of a metric-module cutter on a hobbing machine meant that I had greater flexibility in the choice of possible ratios.

The result of these ratios is shown below. The final drive ratio is the same 41:10 as before.

[e301988325i]

That's hopw I set my gears up in gran turismo, the further up the gears the closer they get. . . keep it coming Geoff!

[GeoffBob]

That's how I set my gears up in gran turismo, the further up the gears the closer they get.

And pretty much how it should be done as I understand it. Equates, in my case, to a delta of roughly 35km/h per gear. Although my previous gear setup was similar, I was restricted in my choice of the ratios due to the imposed use of metric modulus gear cutters.

1st gear is, as before, good for 100km/h, which makes for an interesting time driving the car onto the trailer (due to the absence of torque at anything less than track speeds). I am thinking of investing in a winch at some point.

Shown below is the new design dog to selector-ring profile. Compared to the previous design I have reduced the number of dog teeth from twelve to six, but substantially increased the dimensions, and thus strength, of each tooth. With the added room (as a result of throwing out the old splined dogs) I have included a substantial radius at the root and apex of each tooth. This again increases the strength of each tooth and fully takes advantage of the wire cutter.

keep it coming Geoff!

Cheers, will do

[Mikey_Boy]

Hey Geoff - I didn't realise so much had happened!! Sounds like you have had some tough luck, but keep rooting at it - there are a lot of folks here willing you to succeed!

Just as a (small) aside - back in the day whilst at Ilmor (what was McLaren / Mercedes F1 engine source, now Mercedes High Performance Engines) based in the UK, we sourced all our gears from Switzerland, despite some of the best suppliers in the world nearby...

Glad your mojo is back with you!

Mike

[GeoffBob]

I'm guessing your gearboxes ran like clockwork Mike ?

Ilmor eh? Must have been a dream job that. Lucky you.

[Mikey_Boy]

Ha ha! Very good...!

Not always - gears were problematic at times, but these were used for cranks gears and intermediate gears up to the cams (race engines don't use chains or belts as you know) so in the never ending struggle to limit lossed throught these gears, there were always failures...

What I have experienced with gear design (root of tooth absolutely critical, final finishing of gear also critical) shows what a good job you are doing so far...

By the way, you may remember my installation of the M3 g/box to my (soon not to be!) M20 engine'd track car? Best thing I have done outside the Z3 steering rack for quality of drive etc. Shows that the other bits of the driveline are just as important as the engine...

Keep it up Geoff - keeping fingers crossed you come up trumps with a good supplier!

[GeoffBob]

root of tooth absolutely critical, final finishing of gear also critical

You can say that again Mike. I can hardly believe that the design of my gearbox has become so involved that I have spent so much time designing something as totally obscure and unknown as the root of a gear tooth. Gears are so much more than just finding the correct number of teeth to achieve the desired gear ratio ”“ contact ratio (the average number of teeth in contact at any given moment in time) and interference (whether a tooth must be undercut in order to allow two teeth to roll over each other without obstruction) are also incredibly important aspects of gear design that must be considered in order to ensure a reliable design with minimal noise and vibration. Keeping vibration down helps prevent failure due to fatigue. FYI, my gears (1st through 5th) have contact ratios of 1.64, 1.66, 1.69, 1.71 and 1.73 respectively (each was aimed at a value of 1.7). A contact ratio of 1.2 is considered the dead minimum allowed, better than 1.5 is required if the gearbox is to be reliable.

I think that the effort I have put into this gearbox is a testament to my own stubbornness and arrogance rather than my engineering skill. In any other universe you’d have to be mad to take on a task like this. Will it ever pay dividends once built? The answer is no it will not, but the process of getting to the finished item will have been worth every penny ”“ if only to say that I did it. Lets not forget that I have already had a taste of driving a car with my own design gearbox onboard. Having tasted blood once I want more.

By the way, you may remember my installation of the M3 g/box to my (soon not to be!) M20 engine'd track car? Best thing I have done outside the Z3 steering rack for quality of drive etc. Shows that the other bits of the driveline are just as important as the engine...

If I remember correctly it was me that steered you towards that Getrag 262 Mike. I am familiar with its internals and it is indeed a thing of beauty. Very few people understand or appreciate how that dogleg gearbox came about and why it is that 1st gear on a dogleg box is really 5th gear.

The other bits are INDEED as important as the engine Mike. I am glad that you will be fitting a Getrag 262 to your S14 engine ”“ it’s the way God intended it to be .

[Mikey_Boy]

That's just awesome Geoff! Bloody mindedness and determination... Loving it!

And of course - it never pays dividends other than the satisfaction when it all comes together - (as I recall, Keith Duckworth spent a tonne of time sorting out gear vibration issues on the DFV...)

From what you have stated above, I think (from the very little I know about gear design for gearboxes), you are very very nearly there!

Oh yes - the Getrag will indeed be heading onto the back of the S14!! Your advice choosing that box was invaluable...

[GeoffBob]

Glad I was of some use Mike, I seldom am these days

FYI, I delivered the new drawings to Jan Hugo Engineering this afternoon. We went through the drawings one by one and he seemed happy. He can do everything except the lay-shaft and spigot shaft, which I will have to have done elsewhere due to the necessity to shape (not hob or mill) the splines.

I now await his quotation. Given the quality of his work it won't be cheap. But then I have tried the cheap route once already and it didn't exactly work out.

[Connor]

Just read from start to finish, and wow what a read and what a build.
Some very good engineering has gone into this and is a real credit to you, shame about the messer company but its just bad luck, onwards and upwards

[GeoffBob]

Hi connor. Thanks for reading through and thanks for your comment. Apologies for the slow reply but I haven't been around much lately. Have had a few other projects keeping me busy.

With regards to the latest design of the gearbox, I haven't even been able to get a quote out of Jan Hugo Engineering. This is a symtom of a bigger problem that designing engineers like me face on a daily basis: The reputable companies are too busy doing big jobs to take any interest in small one off jobs like mine, and the companies who are prepared to take on my work have no other work due to their reputation for poor quality. As per the reccomendations posted by others, I am rapidly gravitating towards having the work done back in the UK. I seem to recall that there was one company in particular that took my fancy at that time, Beagle I think it was? I may make look them up in the new year once Christmas is over.

[appletree]

Any work being done over Christmas Geoff?

[GeoffBob]

Not likely Matt, too busy relaxing Been a hard and busy year for me at work and my car has had to take something of a back seat. On the other hand I have enjoyed the Christmas break doing very little even vaguely constructive.

And you Matt? Not too much work I hope. Tis the season and all that ...

[appletree]

Just swopped a few bits round on the car, playing with the missus car abit trying to get beeter MPG out of it as its our daily. 06 Mondeo TDCI other than that just chilling out...

[GeoffBob]

Enjoy!