Differentials

[BMH]

Hello everyone! I have a few questions regarding differentials. We are not a first year team, but this is the first year we are truly engineering the car and starting from scratch. I was designated the drivetrain lead. The main questions I have is where do I need to start as far as trying to select the differential, by that I mean are there certain things I can model that will help me decide on what type of differential to select, etc.

As of now, I am leaning towards a spool setup due to low availability of funds along with it being a lightweight setup. What teams are currently using spool setups? I would like to contact them regarding more information on the spool setups.

Also, can anyone recommend some good literature on this subject?

Thanks!

[BMH]

Hello everyone! I have a few questions regarding differentials. We are not a first year team, but this is the first year we are truly engineering the car and starting from scratch. I was designated the drivetrain lead. The main questions I have is where do I need to start as far as trying to select the differential, by that I mean are there certain things I can model that will help me decide on what type of differential to select, etc.

As of now, I am leaning towards a spool setup due to low availability of funds along with it being a lightweight setup. What teams are currently using spool setups? I would like to contact them regarding more information on the spool setups.

Also, can anyone recommend some good literature on this subject?

Thanks!

[MalcolmG]

I would suggest your first step should be to discuss what characteristics you're looking for with whoever is in charge of vehicle dynamics/suspension in your team. If a spool is chosen then it makes life easy on your end, but the kinematic and unsprung designs will have to be built around that decision too.

[Welfares]

We've run a spool for a few years now, on the happy side, its cheap and easy, just remember you'll have to tune your suspension around it a bit. There's been a fair bit of discussion on it and other diffs, the find button is a good place to start.

[Mike Cook]

We have played with a spool a lot. I would recommend not using it for a fsae car. The car will be a lot less predictable, your setup will constantly need tuning for different parking lots/weather. It might sound like a good idea, but it isn't.

The taylor diff is really nice but not very cheap:

http://fsae.com/eve/forums/a/t...12600868/m/473109543 (http://fsae.com/eve/forums/a/tpc/f/412600868/m/473109543)

[BMH]

Mike: I would really like to use the Taylor diff, Its just more than we will probably be able to afford this year.

I understand that the selection of a differential is largely dependent on the suspension and geometry setup. Supposing that we end up going with a spool set-up, what are some things in particular that would need to be done with the suspension and all?

I have previously done a search on this website on various diffs. At this point I am looking more towards Diff modeling and simulation of some of our choices in diffs. I would just like to get some numbers to support whatever diff we end up choosing.

[Conor]

ATV differentials are a good option if you're working with a tight budget. Our team ran a cam and pawl style limited slip out of a Honda Rubricon with a fair amount of success and I know other teams have too. They're not the best, but they're light, cheap, simple, and they get the job done.

[exFSAE]

A popular, recurring subject.

The Taylor Race Engineering website has a good document going over the various types of differentials. They do play a big part in handling. In terms of modeling.. it should be easy to model an Open and Spool diff.

In terms of setup though, that spool will cause heaps of understeer and I'd imagine a pretty abrupt transition to snap oversteer when you break the rear end free with drive torque.

Until you're lifting the inside rear wheel you'd probably need a heap of rear roll stiffness to get the thing to rotate.

Can't imagine it'd be very good through a slalom.. and the FSAE tracks at MIS and Fontana are long and slalomy..

[BMH]

exFSAE: Can you recommend some books that touch on modeling diffs, or at least provide equations?

I'm not exactly sure how to go about modeling the diff. I assume that you would use the output torque from the engine and somehow distribute the torque to the wheels and then figure out torque distribution or wheel speeds under different conditions (wet roads, turning, wheel spin etc.)

[ben]

Open diff = equal torque to both driven wheels
Spool = equal rotational speed of both driven wheels

I'm guessing the best way to model a plate-type limited slip diff is to find the torque required to traction limit the inside tyre and then multiply this torque by the bias ratio and apply it to the outer tyre.

Ben

[Mike Cook]

BMH,

exfsae is pretty much correct. If you have a spool, it tries to drive both rear tires at equal speeds. In a corner, one wheel is going to need to spin faster than the other but since this isn't possible with a spool, the rear tires will create longitudinal forces which will try to push the car out of the corner (understeer). The best way to combat this is to take all the weight off the inside rear tire so it can't generate any longitudinal forces in the corner. This require really high rear roll stiffness.

For a steady state corner, this is probably do-able, but for things like slaloms, the transitions between having the inside rear tire on the ground and picking it up leads to a tight /loose thing which is tough to drive.

If the taylor is out of the budget, then I agree, some of the ATVs have good limited slip style diffs.

[carlosm]

We use the honda ATV diff and have had great success with it. In that it hasn't broken and works. If money is tight and you still want to run a diff seriously consider the honda ATV, they are 130-150 brand new from honda. If you search the forum you will find a thread going into depth about the ATV diff options, specifically from honda. Also if you want some solidworks models of the cam and pawl and how we mount it shoot me an email and i would be happy to send you what we have. The main issue with these diffs is getting the output shafts made, unless of course you have a company that can do splines for you. In that case you're good to go and it should be easy.

[Chris Allbee]

For splines we used to have T-slot keyway cutters reground by a tool shop to match our spline specifications. Then it was just a matter of patience and an indexing table on the mill.

[Ian_N]

Spool diffs aren't bad but will give you a fair wack of understeer to deal with.
As for the ATV Diffs, they're alright from what we've looked at here but as yet I've not been able to trackdown a maximum torque rating for them (Kinda crutial if you don't fancy seeing your diff explode on the acceleration event!)

Clutch pack (Sailsburys) is what we've used traditionally here at UH to good sucess (Regular top 20 finishes in FSUK for the past 10 years, 3 class wins in class 1-200 till 2007, 5th and 2nd in FSG for the past 2 years), though the one we've used was on the heavy side but for the most part reliable and robust.