sgraber
|
| posted on 11/4/05 at 03:29 PM |
|
|
Suitable diff for running upside down and backwards?
This is in regards to La Bala, a middy.
Having recently aquired an FJR1300 engine that's shaft driven I am in research phase and thinking about different ways of getting the power to
the wheels.
Since I have very little experience with the different car differential units available, I am hoping that I might get some advice about it here.
Should I assume that a limited slip unit will not run backwards and/or upside down? Am I limited to runing an open diff?
Any suggestions for a common lightweight unit with a very short snout? I am in the USA, but I see a ton of BMW E-series diffs on ebay...
Suggestions?
Mounting: Since my engine bay space is not infinite, I was thinking about making a strong cradle that holds the engine and the differential together
as one solid unit. I want to eliminate the u-joint and mate the output from the bike shaft directly to the input flange of the diff. Bad Idea? Any
suggestions for the shortest possible solution?
Thanks, Graber
Oh yea, I almost forgot. The Yamaha rear diff is a 3.66:1 unit. My cars tire height will be close to the bikes at 24". Should I assume that the
car diff ratio should be close to this number?
[Edited on 4/11/05 by sgraber]
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
|
|
jimgiblett
|
| posted on 11/4/05 at 04:16 PM |
|
|
You might want to check out the Legends race cars which use the FJ1200 motor to see how they get power to the back wheels. Dont know about the FJ1300
though.
Alternatively didnt some early RWD honda diffs run the wrong way?
Cheers
Jim
[Edited on 11/4/05 by jimgiblett]
|
|
|
chrisf
|
| posted on 11/4/05 at 05:55 PM |
|
|
Steve:
Why not run a chain setup? It would be easier and not much more money. You would also worry less about getting the rear ratio right because you could
always change the sprocket.
--Chris
|
|
|
ReMan
|
| posted on 11/4/05 at 06:10 PM |
|
|
I can`t see why, any diff should`nt run upside down.
Provided you stop the oil running out and fill it to the same relative height of oil over the gears, I cant see the difference?
|
|
|
sgraber
|
| posted on 11/4/05 at 06:14 PM |
|
|
quote:
Originally posted by ReMan
I can`t see why, any diff should`nt run upside down.
Provided you stop the oil running out and fill it to the same relative height of oil over the gears, I cant see the difference?
Don't the limited slip units need to run in one specific direction because of the gears in the limited slip unit?
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
sgraber
|
| posted on 11/4/05 at 06:22 PM |
|
|
quote:
Originally posted by chrisf
Steve:
Why not run a chain setup? It would be easier and not much more money. You would also worry less about getting the rear ratio right because you could
always change the sprocket.
--Chris
Aloupol suggested this in the Middy forum, http://locostbuilders.co.uk/viewthread.php?tid=24504&page=1#bottom the main problem with this is the
rotation of the output shaft means that the engine will be on the Left hand side. That's ok for the Brits, but not USA since all the weight of
the car will be all on one side and not balanced at all.
That's why I am leaning towards the car diff at the moment.
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
ChrisGamlin
|
| posted on 11/4/05 at 06:44 PM |
|
|
quote:
Originally posted by ReMan
I can`t see why, any diff should`nt run upside down.
Provided you stop the oil running out and fill it to the same relative height of oil over the gears, I cant see the difference?
From what Ive heard others say that know much more about gears than me, the gears are technically designed to run properly in one direction only, and
running the opposite way can put odd loadings on the bearings etc. ISTR its something to do with the meshing that is designed to pull the two gears
together under load, and reversing that makes them effectively try to push themselves away from each other, but I might be way off on this.
Having said that Ive also seen many people say that in the real world most diffs cope with it fine. 
|
|
|
Hellfire
|
| posted on 11/4/05 at 06:52 PM |
|
|
Just give it a try. Flip the diff, plug the breather, relocate and see what happens running it upside down. If it's knackered after a few
hundred miles, it might become costly. If it's knackered after a few thousand miles, treat it as a consumable and get another one. They
aren't too expensive after all. You could pick one up for the same cost as one tyre. Then we'd all know once and for all whether
it's a viable option or not. Go for it...............
|
|
|
sgraber
|
| posted on 11/4/05 at 06:54 PM |
|
|
I'm not too worried about the ring and pinion. I used to be very hung up on the issue, but I'm over that. I am more interested in the
effects of limited slip internals being run backwards...
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
Aloupol
|
| posted on 11/4/05 at 09:27 PM |
|
|
The most common LSD has ramp pieces, which angle gives the amount of locking effect, and they are sometimes different in accellerating and in engine
brake situations. So maybe your diff will work strangely if run backwards. Need to unmount it and check the ramp angles to be sure.
About the layout I suggested: it's easy to add an intermediate axle, behind the diff, with two chains. On this axle you can put an electric
motor to sort the reverse gear problem. By the way it becomes easy to place the engine in the body's centerline.
I will make a sketch because I feel that my writing is confused.
|
|
|
Aloupol
|
| posted on 11/4/05 at 10:12 PM |
|
|
Hope this pic is a bit more clear..
 
Rescued attachment FJR1300-.JPG
|
|
|
sgraber
|
| posted on 11/4/05 at 10:43 PM |
|
|
quote:
Originally posted by Aloupol
Hope this pic is a bit more clear..
Yes! This is along the lines of what I was envisioning. Chris Fiaccone drew up something similar and also suggested that the jackshaft might be
replaced with a reversing box. However, an electric reverse could easily be designed and for probably a lot less money.
Placing the engine inline with the centerline of the car is an added bonus. Packaging the transmission off to one side aids in the compactness of the
driveline, one needs hardly more than 24" longitudinal in front of the half-shafts! Neat.
I have to wonder how much driveline loss this setup will have? Would using 2 chains in the driveline compare favorably with a standard automotive
diff?
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
soggy
|
| posted on 11/4/05 at 10:50 PM |
|
|
have you heard of a phantom grip diff its USA made it fits where the sun gears go and gives a limited slip action for half the price.search the web
for info.
if at first u dont succeed use a bigger hammer!
|
|
|
Aloupol
|
| posted on 11/4/05 at 11:21 PM |
|
|
The reverse gear: I was thinking of that too but I wonder if it's strong enough. It's designed to be driven by the gearbox torque, and the
sub axle has a little bit more.
Efficiency: I don't know the exact numbers but I don't think it's a lot. If it was some cooling system should have been developed by
bikes tuners ;-)
Maybe by replacing the chains by straps (like the Harley Davidson bikes) it's possible to save something.
The Phantom diff: is it the stuff I saw, using the axial reaction of the diff's gears to press clutch discs, without ramps? Is it worth? Which
car uses it?
Compactness: I plan to use this kind of jack axle in my car, with a standard (transverse) bike engine, in order to minimize the wheelbase, and to have
half axles of equal lenghts.
LSD: The guy who tested the R1ot said it's not that important to have one in a middy BEC, due to the good weight balance and reasonable power. I
won't sell my children to buy one, if an open diff is good for that great kit car it will suit for mine.
|
|
|
sgraber
|
| posted on 11/4/05 at 11:45 PM |
|
|
quote:
Originally posted by Aloupol
Compactness: I plan to use this kind of jack axle in my car
Where will you fiind your jack-shaft components? I am thinking about 1" go-kart bearings and carriers, but the hollow shafted cart axles are
probably too fragile for the torque figures we are reaching. How to attach the sprockets to the shaft? I don't think a keyway will be strong
enough. Can you point me in the right direction? This a likely a part that will be best made by a qualified machine shop. Maybe a halfshaft modified
with splines on each end to accept standard bike sprocket?
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
Peteff
|
| posted on 11/4/05 at 11:53 PM |
|
|
I'm a bit confused.
Is the engine the wrong way round in the drawing? The drive comes off the side of the gearbox. If you've taken the shaft drive off it will it
still be driving the wrong way?
yours, Pete
I went into the RSPCA office the other day. It was so small you could hardly swing a cat in there.
|
|
|
madman280
|
| posted on 12/4/05 at 12:11 AM |
|
|
No problem. They are on the streets everywhere. The front diff in a small four wheel drive truck rotates backwards. If you use a clutch type limited
slip you should be able to build-up what you need by using one from a rear unit of the same make and size. For example. One could use an Ford 8.8
front diff and use the LSD carrier from a 8.8 rear diff. The difference is in the gears and case. Not 100% positive what fits what in smaller sizes
but this might give you a place to start looking. Now finding the right gearing may be a whole other story, but this excel spread sheet might be of
some help there: http://members.lycos.co.uk/alister667/speedcalc.xls
[Edited on 12/4/05 by madman280]
|
|
|
madman280
|
| posted on 12/4/05 at 12:16 AM |
|
|
Had another thought. If you were to use a racing type quick change you can rotate them backwards since they are bevel type gears not spiral bevel like
most diffs. I've seen Winters and Halibrand midget diffs used this way. If you do a bit of a web search on midget or speedway sprint cars you
should find a couple sources. They make IRS type diffs as well as the live axle variety. Not usually listed in a catalog. Call and telling them what
your doing. I'm sure they can offer a couple sugestions.
[Edited on 12/4/05 by madman280]
|
|
|
chrisf
|
| posted on 12/4/05 at 01:03 AM |
|
|
Aloupol:
My idea was to use an inline reverse box as the jackshaft. The whole unit could be slide fore/aft to tighten the chain.
Actually, I suggested to Steve that he use a standard size inline reverse
box (scroll down), he could decide later if he wanted to go with a Quaife unit or use a standard jackshaft. Alternatively, he could put a
toothed gear in the middle of the jackshaft and run an electric reverse off that. All three solutions could be designed to act interchangibly.
Moreover, using the jackshaft means an easier final gear ratio. At any rate, the chain drive seems much easier than having to flip heavy
differentials. Just my ideas...
--Chris
|
|
|
The Shootist
|
| posted on 12/4/05 at 03:08 AM |
|
|
Production LSDs are clutch or viscus
The clutch type has spring loaded clutches which will slip when a certain torque differential occur between the two wheels. This will function in
either direction.
Viscus uses silicone fluid in place of the springs. When the difference in speed between the two axles heats the silicone it expands and causes the
clutches to bind, locking the axles. This type will also work in either direction.
The ring and pinion gears should be fine turning backwards otherwise downshifting and engine braking would kill them on a regular basis.
Looking in my "Machinists Handbook" 1941 edition gears typically have a symetrical profile, therefore the direction of rotation is
mute.
Now one thing we haven't examined is lubrication. Now I can say from personal experience that a rearend, turned on it's side, with one
axle pointed straight-up, has run lightly loaded in a mill in our plant for decades. In fact we did just change it out due to a failure, but it has
amassed over 2,000 hours of operational time before it failed.
I believe, firmly, that a rearend will work just fine in a small car in an upside down and running backwards orientation. The windage effect should
give plenty of lubrication to the bearings.
|
|
|
Aloupol
|
| posted on 12/4/05 at 05:27 AM |
|
|
quote:
Originally posted by sgraber
Where will you fiind your jack-shaft components? ... This a likely a part that will be best made by a qualified machine shop.
Yes I think so. Didn't tought at this for the moment but it doesn't seem impossible to make from scratch. Possible to weld parts (lathed
flanges) on a tube or a bar, then lathe the assembly.
quote:
Originally posted by Peteff
Is the engine the wrong way round in the drawing? The drive comes off the side of the gearbox. If you've taken the shaft drive off it will it
still be driving the wrong way?
It was the other way in my first drawing but I understood from what Steve said that it had to be like this. It makes sense, thinking of the
configuration of the "diff" on the bike.
quote:
Originally posted by chrisf
Aloupol:
My idea was to use an inline reverse box as the jackshaft. The whole unit could be slide fore/aft to tighten the chain.
--Chris
Yes, but The only thing to check is the torque this unit can stand. It's designed to be mounted between the gearbox and the diff's
reduction gear so with a smaller torque than in "our" jack axle.
quote:
Originally posted by The Shootist
The clutch type has spring loaded clutches which will slip when a certain torque differential occur between the two wheels. This will function in
either direction.
As far as I know the clutch type LSD has, in addition to the spring, a jacking system that uses the motor torque to press the clutch, and these parts
have different shapes for acceleration and deceleration. Maybe it's not difficult at all to switch the parts, I don't know.
[Edited on 12/4/05 by Aloupol]
|
|
|
Rorty
|
| posted on 12/4/05 at 06:00 AM |
|
|
Steve, what is the estimated weight of La Bala? I doubt if the type of electric reverse Locosters are familliar with would cope with the weight of
your car. If you ran a more powerful motor, then you'd be getting into power-sapping alternators and heavy duty batteries.
I would look into either a diff with a built in reverse, or possibly a canabalised transaxle to provide a mechanical reverse. Basically, what you need
for a compact reverse is a set of planetry gears.
As The Shootist says, any diff will perform OK in any orientation as long as it's suitably lubricated.
With the engine orientated as in Aloupol's drawing of the Busa, an industrial reverse mechanism in line with the engine's output shaft may
be worth considering. Some are made with aluminium housings which would help keep the weight down.
What about one of those reverse transaxles I posted recently? I know they're a bit pricey, but I believe it may be a cheaper solution in the
long run.
Cheers, Rorty.
"Faster than a speeding Pullet".
PLEASE DON'T U2U ME IF YOU WANT A QUICK RESPONSE. TRY EMAILING ME INSTEAD!
|
|
|
sgraber
|
| posted on 12/4/05 at 01:50 PM |
|
|
La bala is fairly light. The car weighed 1,126Lbs with the 1.6l 4AGE. and that was in driveable condition, but without bodywork, windshield, etc. My
finished estimate was 1,400Lbs (or around 650Kg). The new drivetrain is (est) 150Lbs lighter than the old. So at 1,250Lbs (560Kg) la bala is still
heavier than the typical locost BEC, but not extremely so. FWIW - My weighing was done using 4 bathroom scales so there is probably a margin of error,
but not significant to the SWAG calculations we need to do here.
Why not eliminate the jackshaft completely but still use a chain driven differential, an intermediate shaft to the longer side, and two equal length
short half-shafts like the newer front wheel drive cars use to eliminate torque steer? I might be able to use the entire drive assembly from a Honda,
starting from the differential outward to the hubs.
Rorty, I haven't completely discounted the mini-transaxle, but want to pursue (discuss) all other non-comercialized ideas to their logical
conclusions.
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
sgraber
|
| posted on 12/4/05 at 03:38 PM |
|
|
Rorty et al,
This is what I have in mind. Just a quick sketch in Max.
 
Chain Driven diff render 1
Note that the red blob in the middle with the little gear is an automotive style gear reduction starter that would be a reverser unit. The large green
blob in the back is the engine, the purple blocks would be the bearing blocks and the blue sticks are portions of the car frame....
I intend to take the entire half-shafts, intermediate shaft and differential from a Honda FWD transaxle since that centers the engine mass in the
correct location relative to the centerline of the car.
Steve Graber
http://www.grabercars.com/
"Quickness through lightness"
|
|
|
chrisf
|
| posted on 12/4/05 at 05:26 PM |
|
|
In my opinion, that is the best idea yet. But just a thought...If the driveshafts are dissimilar lengths, will they really cause torque steer? 1.)
You're not steering in the rear with a deDion and 2.) you won't have that much torque.
At any rate, your idea would be the easiest to fabricate/machine and lends itself well to your deDion rear. This is something you can sort out between
GRP work. Might break things up a bit.
--Chris
|
|
|
