PDA

View Full Version : New to the make your own diff game



Sara
04-12-2004, 01:26 PM
Hello everyone!!
I was just curious if anyone had any advice when it came to designing the diff out of aluminum. I'm getting a head start for the 04- 05 car, and was curious to know what you guys had to say about it. Any and all help is greatly appreciated. I'll see you all in Pontiac this year!

Thank you SOOOOOOOO much,
Sara
CU Buffalo Racing

Mi_Ko
04-12-2004, 04:55 PM
Sara, glad to have you here.

I've never heard about a totaly self made diff yet.
Most teams use the gears from the torsen Uni special and making their own housings to save some weight. (maybe you ment this?)

I will not encourage (difficult word http://fsae.com/groupee_common/emoticons/icon_smile.gif)you to make your own, we would do our own too, if we would have the resources, but I doubt if it is worth to do it.
But, hey, it's a learning competition so go on!! I'm looking forward to see it and you telling me I was wrong. http://fsae.com/groupee_common/emoticons/icon_smile.gif

Sara
04-12-2004, 06:23 PM
hahaha making gears would be a feat in it's own! i'm just making the housing for the gears, and stuff. it's pretty cool to learn how all this stuff works, cause i've got to be honest, i don't know much about it.

jack
04-12-2004, 08:53 PM
we made an aluminum diff housing, so do several other schools. i designed the diff, and travis made it, after our first generation diff didnt work. our diff uses "old school" torsen gears, with CV's and just about everything else in that area that we designed and manufactured. the reason the first gen diff didnt work was because the tolerances werent good enough, allthough pretty good. the new diff was made as accurate as possible. the whole diff unit wet, with cv's and mounting brackets, and sprocket weighs 12.3 lbs (5.58kg), which probably saves many pounds from using the original setup. if you make an aluminum housing, make sure that the gears have steel plate to thrust into, instead of the aluminum, and that you make acomidations for the two larger "sun" gears to plunge slightly. other schools also add "breathers" so the diff can expel pressure when hot, however, we reley on leaky seals http://fsae.com/groupee_common/emoticons/icon_wink.gif. with our diff we originally used a nice wide needle bearing on the drive side, but any missalignment in the bearing causes the diff to move slightly which kills the sprocket. now we have a sealed ball bearing for that side, which works, but was very expensive. here is one pic of the diff, i'll try to get more later on.

http://dot.etec.wwu.edu/fsae/SharedPictures/Diff/Diff%20without%20cover.JPG

i have thought about making an entire diff from scratch. on our 4 axis mill here, we could cut gears, althhough it would be quite a challenge. i would make a quaife type, because i think it would be easier to tune, lighter, and you can use spur gears. this would be a serious project though and its way easier to use torsen gears.

oh yeah, it took a four axis mill, and a cnc lathe to make our diff housing

RagingGrandpa
04-13-2004, 11:22 AM
[Lumberg] Hey jack, whaaaats happening. Oh, there it is, I've been looking for my 4-axis CNC mill all over. Yeahhh, I'm just going to go ahead an get that from you.... [/Lumberg, Office Space]

We had to pay to have our housing made this year after a sponsor fell through... over $2k really bit us in the arse.

All Wheeler
04-13-2004, 07:36 PM
We at Newcastle attempted to make our own diff last year. The car didn't run, and a big part of the problem was the amount of time spent on the diff. We tried to make two of our own torque sensing diffs which were parallel mount helicals like a type two torsen. If you are interested in making your own torque sensing diff, I can recommend a starting point or two for you. If you have three sets of element gears, the number of teeth on your side gears needs to be divisible by three. I would recommend that you also look at somewhere between a 2 to 3 module around 30 deg helix to achieve torque bias or around 2:1. This is where things get interesting. The most efficient use of material is when the diameter of your side gears is large. The easiest way to increase your torque bias is to have small diameter side gears. The trade off is up to you. Any higher helix angle makes it impractical to cut on a conventional hob cutter, any lower doesn't provide a very good torque bias. There are alot of places that do hob cutting quite cheaply, I did my work experience at one of these places and cut the gears myself (very painful and not recommended).
Also look at a 4140 or 4130 nitrided. I had a look at alot of materials, and these two invariably came out on top. Aluminium isn't bad for cases, but not a good idea for gears. Hope this helps!
Doug.

Sara
04-14-2004, 01:00 PM
Hey guys!
Thanks so much for all of your help, i have actually gotten an old 2 piece torsen diff out of an 86' audi quattro, i didn't realize all those gears needed to be exactly the way they were when i started taking everything apart, and putting it back together, but messed up the whole timing thing (like i have said, i'm just starting to really learn about how this thing really works and stuff). anyway to put the gears back they way they should be? I look forward to talking with you all in detroit, and i wish you all the best of luck!
Thanks!
Sara

Sara
04-14-2004, 01:02 PM
oh yeah one more thing, do you know of any good manuals or books that i could get my hands on to read up on this subject?
thanks

Mi_Ko
04-14-2004, 01:13 PM
Sara, we have the nearly the same diff from a lancia delta and we did make the same mistake as you. http://fsae.com/groupee_common/emoticons/icon_smile.gif

Try this:
There are small marks on the element gears. (A letter or a number. I dont remember) All these marks have to bee on the same side.

Disco
04-14-2004, 07:54 PM
Sara,
Jump on the Torsen website to find the timing information. It a bit tricky to work out, but remeber that 0 degrees is classed at vertically up when the torsen is on it's side, hence the element gears at horizontal. Send us your email and I'll send you a few pics of the diff we used.

Cheers
Steve
sr_price@hotmail.com

jack
04-15-2004, 12:01 AM
i have done a little searching for literature on diffs, but every book seems to have the same thing, says how they work but thats it (carrol smith books, milliken sp?, transmition books).

when you time your diff it must be correct, otherwise from what i understand, it will grenade. here is how you time the old school torsen: the element gears are marked with either an "e" or "s" in the root of the gear, make sure the same 3 are touching the same "sun" gear (all e's or s's on a gear). label each set of 2 element gears 1,2,3. now each element gear has a spot, when you stand the diff up on its side. i.e: top, second window. when you put the element gears in, make sure that "e" or "s" is on the left side. now, you must index each gear a certain amount of teeth out towards you. here is what each element gear needs:

Window #1: top: 0, bottom: 0
Window #2: top: 3, bottom: 3
Window #3: top: 2.5, bottom: .5

that should do it. you should check out the torsen site too, incase i made a typo, personally, i found the torsen directions difficult and confusing to figure out, so this should help. good luck.

Daves
04-15-2004, 12:58 AM
After spending about $2500 and many hours of labor last year, we decided to purchase a Taylor Race differential for $2360. We have not done much driving with it yet. I did accidentally drop it off a table before ever using it, and it still works. I believe the fluid never has to be changed. An interesting note is that the whole housing spins with the gears. The only things that don't spin are the outside bearings. A drawback is that there are some problems with the Taylor drawings on their website, but nothing that has hurt us tremendously.

Sara
04-19-2004, 05:20 PM
Thanks Jack for your help! i got the gears put back in order, and moving with no resistance. YAY!!!

Sara

drivetrainUW-Platt
06-05-2004, 08:11 PM
Sara,
i designed our diff for our 2004 car, have lots of solidworks drawings, would love to discuss it with you if ur still interested in the design, send me an email at duwem@uwplatt.edu
Mike Duwe
UW-P Drivetrain Leader 2005

-
06-06-2004, 02:52 PM
Sara,

A note on the old style Torsen diffs. I have heard straight from one of their tech reps that there is only one way for the gears to go back together. On occasion we had discussions around our lab as to whether it was back together correctly even though it all fit back in, but the gears didn't spin by hand. As it turned out if you just dab a little bit of oil on the gears, it allows you to easy turn the gears and lets you know everything is working ok.

As for getting the old diff out of the old Audi, how much of a pain was it? We never got to the point of having to get one out of there, but from what I heard they are pretty hard to get out since they are basicly in the very heart of the transmission/transfer case.

-Mark

Sara
06-09-2004, 07:09 PM
I was lucky to have a friend who just had a spare rear axel laying around, it was a total pain in the ass to crack open!! i just wanted to see what the diff looked like and how it worked and mess around with it... and now i know how the gears go on, and that they are in a specific order! thanks for all of your help everyone!

Dr Claw
06-10-2004, 11:36 AM
I think LTU has written a few papers on aluminum diff's from back in the day ('99, '00?)...i think we might have been the first to do it, but i am not entirely sure about it.

we did an aluminum casing this year too. it ended up cracking the day of comp though because we used 3 through bolts to bolt it all together, and then 3 hardened steel dowel pins to help spread the load. just remember this:

hardened steel pins < aluminum

that is not the 'less than' symbol from 4th grade...it's the 'fish symbol' from 3rd grade when you didnt really understand what it did other than eat whatever was on the other side of it. hardened steel dowel pins eat aluminum for breakfast, on race day. i should post pics http://fsae.com/groupee_common/emoticons/icon_smile.gif

luckily our school is "20 minutes away from everything", including pontiac, so we chained our disheartened diff guy to the mill and had him machine the stock casing to fit his carrier and everything ended up happy, ableit heavier.

have fun though

Broderick
09-09-2004, 10:22 PM
I know I'm draggin up an old topic, but to make the housing pictured above, (WWU's) what type mill was needed? 3 or 4+ axis? Also, did you turn it or machine the gear housings first?

Ryan
RFR '05

jack
09-10-2004, 12:03 AM
well, travis made it, but i can tell you we used our haas VF4 mill (3 axis) with a 4th axis that bolts onto the table, and plugs into the mill (its not one of those manual things), oh yeah, we turned the "blank" on our haas VF2 lathe first, which was then chucked up in the 4th axis. and i had the personal pleasure of taking the original 50lbs block of al, and making it round on a manual machine, so round stock could be used in the cnc lathe.

PS, the cnc lathe code was written on the machine, and the VF4 code was all done in catia. travis wrote the code for one of the 3 "sides", stuck a rotate code in (4th axis, 120 degrees), and repeated.

johnnySV
09-14-2004, 06:24 AM
does anyone have experiance integrating diffs from utility quads like the Honda Foreman ??

and i assume the reason for such customization of the rear drive differential stems from the axial input (chain / sprocket drive) / axial output configuration?

this thread has also been helpful: http://fsae.com/eve/forums?a=tpc&s=763607348&f=125607348&m=4126049904&r=32610054611#32610054611

Courtney Waters
09-15-2004, 10:18 AM
A 4-axis CNC is NOT needed to make an aluminum diff housing. For that matter you don't need a CNC at all, but it sure makes life easier. The diff housing I designed for our 04 car was turned on a CNC lathe to save time and due to the detailed profile. No reason why you couldn't do it on a manual lathe though. All of the "windows" for the element gears and holes for their pivot shafts were done on a Bridgeport (yeah, with a 2-axis setup on it, but it could have all been done manually). The key was a fixture I built to accurately position the housing at 120-degree intervals and two phases (since the axes are perpendicular to the windows). I can go into more detail, but a picutre is worth a thousand words as they say:

http://www.mirafiori.com/~courtney/misc/fsae/04car/04_diff_fixture.jpg

Charlie
09-15-2004, 12:40 PM
Our drivetrain guy in 2003 made our aluminum housing completely manually, with a 3 axis manual mill with indexing head, and a manual lathe. PITA yes, but you work with what you have available.