Forums

MS Differential?

MS Differential?

I was told there is a reduction gear between the motor and the driven wheels, but I was asked how the difference in rotation between the driven wheels is handled and realized I had no idea. Embarresed! Help!

mrrjm | September 21, 2013

I assume yes. Unlike Chevy and the Volt Tesla is not very forthcoming with technical information. I also want to know is the reduction done? Reduction gears with a chain or just gears? I wish someone would take the car apart and post videos on youtube.

LarsT | September 21, 2013

It is simple: Reduction gear is a fixed gear with 9.73:1 reduction ratio and of course there is a differential to make up the difference in rotation in curves build into the drive unit (no car would work without it).

reitmanr | September 21, 2013

LarsT
Thanks. Makes sense.

jat | September 21, 2013

There is an open differential between the axles, just like on any car that doesn't have hub motors. Tesla achieves limited slip by selectively applying the rear breaks to transfer torque to the the non-slipping wheel (the same as Porsche and some others do). This is tied into stability management which also needs to brake the rear wheels independently, and is probably why SM can't be turned off.

Another case of replacing complex (and non-adjustable) hardware with simple hardware and complex software.

jackhub | September 21, 2013

Help! Something I don't understand. Reading this, the first images that come to mind are Ben Hur racing his chariot, the conestoga wagons crossing the prairie, and motorcycles with sidecars- all turning sharply without differentials- not to mention that little red wagon I once pulled around. How and Why?

jat | September 21, 2013

@jackhub - those aren't powered wheels, so they turn freely. If you have a straight axle where the inside wheel turns the same as the outside wheel, one or both are going to slip. Some vehicles can get away with that like go-carts or skid-steer construction equipment, but you don't want that in a car on paved surfaces.

WhisperingCJ | September 21, 2013

@jackhub. Your examples are where the two wheel are not connected in any way. Add a motor between them and there needs to be a way to ensure the motor is not turning the two wheels at the same rate when turning because the one on the outer edge would need to be turned faster

jackhub | September 21, 2013

Hmm. But if both wheels are connected to the same axle, and the axle turns both wheels when the axle turns, why does it matter what turns the axle- a motor or a horse?

jat | September 21, 2013

@jackhub - it is really hard to believe you aren't a troll, but for now I will assume you aren't.

The horse doesn't turn the wheel -- the horse pulls the cart, and the wheel turns because of that motion. Each wheel is free to turn at its own rate.

If you really are having difficulty understanding this, go read wikipedia.

jbunn | September 21, 2013

Jack,

On a wagon or a chariot, both wheels are NOT connected to the same axle. It's more like a pine derby racer, or a red wagon. The axle does not spin. Each wheel spins at the end of the axle.

This has nothing to do with Tesla. Just red wagons, roman chariots and ships of the prairie.

DickB | September 22, 2013

@jat@jaet.org
“Tesla achieves limited slip by selectively applying the rear breaks to transfer torque to the the non-slipping wheel “

I do not think this is what Tesla does. I seen you post this a few times and I also wonder how TM does control the slip.

I have a 2012 MS 85kw with over 10,000 miles. I put the car up on a lift and ran it. I notice the brakes do not come on at all. What I notice is the power is cut way down and the wheels turn very slowly. You also see this on the KW meter on your dash and the TC warning is on also. Not sure if the power is cut down to both wheels or just the one that is slipping. Where did you get the info that TM uses the brakes for TC? From my testing it looks like it’s all electrical and no mechanical brakes involve at all.

mrrjm | September 22, 2013

Almost all manufacturers use the brakes and control throttle to limit wheel spin. I am sure Tesla is the same. Someone who has the car pull off to the side of the road where one wheel is on the pavement the other is in the dirt and punch it and see what happens.

cwmenne | September 22, 2013

For traction control, the immediate response is independent braking as needed, then power is reduced to limit additional wheel spin, which is much smoother than jamming the brakes on over and over again against 400+ HP. This is the same on most new cars, including the Tesla.

chrisdl | September 22, 2013

DickB:
+1 It would great to get a source on this.

jat:
Those accusations are uncalled for. Not everybody is as technical as some of us are.

jat | September 22, 2013

@DickB - when I was having the car pull to the right under acceleration (and to the left under regen), I was talking with the service manager at the ATL service center. When we were having trouble identifying the problem, I asked if it could be something in the limited slip differential, and he said the diff on the Tesla was an open diff and the limited slip functionality is provided by braking the spinning wheel. I haven't disassembled it or done any further investigation, but I assume he would know what he is talking about. The problem I was having turned out to be some bolts that needed to be tightened on the suspension.

pebell | September 22, 2013

@jat, that was probably the most uncalled for "troll accusation" that I have ever seen. The chariot question was very valid, because at face value it appears to be two wheels attached to a single axle. Jbunn's answer (not yours) nailed it - the axle doesn't rotate on a chariot, the wheels rotate independently around the fixed axle.

jat | September 22, 2013

@pebell - I had no problem with the first question, but persisting after two explanations rather than just reading up on it seemed hard to believe he was really having trouble understanding it. Despite that, I still assumed the question was genuine and explained further.

I also said in the chariot example each wheel is free to turn at its own rate, so I am not sure why you think that was insufficient anyway.

DickB | September 22, 2013

@jat@jaet
I also talk to service mechanic but he wasn’t clear how the TC work. He also told me that the MS has a parking pawl which I don’t think it has. The roadster has a parking pawl which I seen in one of the videos. I haven’t seen a picture of the MS differential apart yet. Elon did say a while back that the TC was all electrical since the mechanical was to slow to react. That’s what got my interest on how this differential work. So far I haven’t found the answer.

I will put my MS back on the lift next time I get chance to make sure the brakes are not going on. Last time the car was on the lift you could hear the slight drag from the pads as the rotor was turning. Gave it more throttle and no change in sound from the pads, but the TC came on preventing the wheel to go faster. I still heard the sound and felt no pressure in the rubber brake hose. Putting slight pressure on the brake pedal, then you feel the pressure in the brake hose and the sound from the pads will change.

pebell | September 22, 2013

@jat, i didn't say your answer was wrong, I said Jbunn's answer _nailed it_ because instead of calling troll or refering to Wikipedia, he took the time to try and understand what caused the confusion, and addressed and resolved it.

I know you went on to (correctly) answer the question, but what's the point of calling troll on someone _you_ feel takes a little bit to long to "get it"?? If you don't feel l like answering any more, just go to the next thread and leave it to someone with more patience..

pebell | September 22, 2013

Typo: ... If you don't feel like ..

chrisdl | September 22, 2013

Pebell: Agree. Negativity reflects badly upon this forum. Friendly does it.

reitmanr | September 22, 2013

Beyond all the varied replies, I am hearing that of course there is a differential in the rear end. I hadn't thought to ask about limited slip. Sometimes we get more than we bargained for.
Thanks for the fire hydrant of explanations. Bottom line for me is, of course there is a differential. We are just not entirely clear how all functionality is provided. Seems to work well on my MS.

Bikezion | September 22, 2013

Here's the blog about the roadster, I imagine it is very similar. It sounds like it electronically limits power. It seemed to work extremely well in the test drive I did, in the model s.
http://www.teslamotors.com/blog/slip-sliding-away

I'm not much of a fan of the brake based systems, but they do work, although counter intuitively. Why fight against the energy usage with the brakes?
I have a trutrack in my range rover and it works very well (not that the rover could ever spin the tires on pavement!) until the wheel comes off the ground, then it basically behaves like an open diff. Right when you need it most!
I think tesla nailed it (as with the rest of the car!), with room for improvement with a torque vectoring system
http://www.rimac-automobili.com/concept_one/torque-vectoring-7

jat | September 22, 2013

@Bikezion - to do torque vectoring, you need multiple motors, and probably hub motors which have their own negatives (unsprung weight, primarily). If you are building a money-is-no object performance car, then that is fine (see the SLS eDrive and the aforementioned Rimac), but it isn't really practical for what the Model S is aimed at.

Bikezion | September 22, 2013

I think the model S is a far better car than either of those 2, although the torque vectoring has some huge advantages, with some downsides, that may not be worth the trade offs. As I said I think Tesla nailed it perfectly with the S. Its not a perfect car, but a perfect first attempt!

jackhub | September 22, 2013

@jbunn. Tks. got it!

ye | September 23, 2013

DickB, when you ran your car on the lift, did you hold one of the rear wheels to try to prevent it from turning? That's the only time the car would need to apply the brakes (to the other wheel). If both rear wheels are able to turn easily, the car can simply reduce power to the motor, to prevent them from turning much faster than the front wheels.

DickB | September 23, 2013

@ye – What I’m trying to find out is if the TC uses the brakes. I don’t think MS does. It seems to be all electrical.

With the car up on the lift, TC off, push down on the throttle and both wheels were going about 80 mph. The speedometer does not work since the front wheels are not turning. Let off the throttle and the regen goes up to 60kw. Push down on brake and the wheels stop and you feel the pressure in the brake hose going to the caliper.

With TC on, the wheels will only go about 10 mph. The power is being cut down and the brakes are NOT coming on. This TC does not use the brakes like some other cars.

Last winter I stop on an icy road and step down on the throttle and the car just crawl down the road. You could see that the TC was working but very little power going to the wheels. I think TM did a very good job on TC and the regen.

ye | September 23, 2013

I thought we were talking about whether the Model S achieves the same effect as a limited-slip differential, and if so, how it does it.

A regular differential lets the rear wheels turn at different speeds. If one wheel is much easier to turn than the other, e.g., because it's on ice and the other one is on pavement, the wheel that's on ice will turn twice as fast as usual and the wheel that's on pavement won't turn at all. And you won't get anywhere.

A limited-slip differential doesn't absolutely force the two wheels to turn at the same speed, but it makes it somewhat difficult for them to turn at different speeds. That way, the wheels will turn at different speeds when you drive around curves, but both wheels will still get power even if one is on ice.

A car with a regular differential can simulate a limited-slip one by applying the brakes gently on the faster wheel if it notices that the two wheels are turning at very different speeds. This will cause the slower wheel to speed up.

jat@jaet.org said that this is what he thinks the Model S does. You (DickB) cited the behavior of your car on the lift as evidence against his opinion. But it isn't really, because even if he were right, your car would still have behaved exactly as it did behave.

DickB | September 23, 2013

@ye – Sorry if I didn’t explain this correctly, but I own a transmission shop for over 40 years. We do repair the complete driveline so I do know how a differential works. My question was on how the TC (traction control) works. Does it use the brakes to stop the spinning wheel or is it all electrical? After running my car on the lift it seems like it’s electrical and the brakes are NOT use.

One of the main reason I bought the Model S was how the driveline works. Love driving this car.

jat | September 23, 2013

@DickB - given the thread title and context (you even quoted my comment talking about limited slip in the differential), I assumed we were talking about differentials, not TC.

Yes, TC is simply limiting the torque requested of the motor to prevent slipping. I wouldn't be surprised at all if it looked at the speed sensors of the front wheel to decide how much power to let go to the rear wheel. On dry pavement, you actually get the highest coefficient of friction at around 10% slip, so it would still let them turn. That has nothing to do with the functionality of a limited slip differential.

ye | September 24, 2013

Sorry, DickB, it's hard to tell who know what sometimes.

Do other cars' traction control systems apply the brakes even when both driven wheels are turning the same speed? Why would they do that, when they could just reduce power to the engine?

Brian H | September 24, 2013

What about SC? Does it somehow differently drive the wheels?

ye | September 25, 2013

I don't see how it could. There's only one motor, whose torque is split evenly between the two rear wheels by the differential. If the car wants a different split, I'm pretty sure it needs to apply the brakes on one wheel.

therealseanarcher | February 24, 2014

Here's a wonderfully good explanation with animation of how an open differential like the one in the Model S works. http://auto.howstuffworks.com/differential2.htm . What the most important part to realize is that an open differential always provides the same amount of torque to both wheels. This means that the amount of torque is limited to the greatest amount that will not cause a wheel to slip. So if one wheel is on ice and loses traction, the amount of torque drops so low that the other wheel has no torque either.To overcome this, the Model S uses the ABS signal of a slipping wheel to apply enough braking to that wheel only and emulates traction, which in turn allows more torque to be applied to both wheels. This gives the wheel with real traction the opportunity to put some of it to the ground.

Brucelli | February 25, 2014

In the absence of actual facts, wouldn't it be a neat solution to have each of the motors ( there are two in the S) drive each of the rear wheels independently via two separate reduction gears? Then there is no need for brakes or a limited slip diff. Provide both motors with the same frequency power ( for the same nominal speed) from the speed controlling inverter. The wheel requiring less speed ( inside corner) pulls less power from its motor than the outside wheel motor by the nature of induction motors. If one wheel loses traction, it will draw less power from its motor, and speeds up slightly. The wheel with traction runs at normal speed and draws normal power from its motor. Again, by the nature of induction motors.

For the cost of a second reduction gear, and the saving of the differential gears, a solution that does not rely on the "parasitic" braking strategy!

sagebrushnw | February 25, 2014

@ Brucelli

Are you saying "...each of the motors ( there are two in the S)" that there are TWO MOTORS in the Model S?

The Model S has only one motor.

EQC | February 25, 2014

@ Brucelli, sagebrushnw:

I've been wondering if the Model S has 2 motors. Visiting a Tesla store, you can see:

1) a semi-dissected motor, usually mounted to the wall. About 1-foot in diameter and probably a bit over a foot long.

2) The naked "skateboard" chassis/power train of the Model S. On this, you can see the motor as well -- but it is over 2 feet long, with the reduction gear/differential assembly coming out of the middle of it. Definitely seems like there are 2 of the "motor units" like what they have mounted on the wall.

I could be mistaken in interpreting what I've seen, or there might be some technical reason why the two halves with the reduction gear in the middle don't count as two motors, but it certainly seems like Brucelli's idea has some merit.

chrisdl | February 26, 2014

One side is the AC motor, the other side is the inverter.

It look like this:

Wheel == Drive Shaft == [[ Differential ]] == Drive Shaft == Wheel
||
[[ AC Motor == Reduction Gears — Two-way Inverter ]]

The parts between [[ ]] are what is in the enclosure that you saw.

It works like this:

When putting power into the drive unit (= accelerating / driving):
1. DC Current flows from the battery into the Inverter (discharging)
2. Inverter creates AC to feed the Motor
3. Motor drives the Reduction Gears
4. Reduction Gears drive the Differential
5. Each side of the diff drives a different Drive Shaft (left and right)
6. The drive shafts make the rear wheels turn

When using regenerative braking:
1. The wheels turn the drive shafts
2. Drive shafts turn the differential
3. Differential drives the reduction gears (which act as multiplier gears now)
4. The reduction gears make the AC motor turn (which act as a generator now)
5. The Inverter creates DC to feed the battery
6. DC current flows from the Inverter into the battery (charging)

Brian H | February 26, 2014

To make it dead simple, the inverter makes the current wiggle back and forth to drive the motor. In regen, it removes the wiggles to feed the battery.

therealseanarcher | February 28, 2014

Yes, there is beauty in this design. Tesla packages the power electronics module (PEM) within the motor housing to keep the whole drive train in a nice compact enclosure, protected, liquid cooled, easy to maintain and with a low center of gravity. Brilliant! The only time you'll see a second motor will be in the AWD version and the Model X, but that will be between the front wheels. Why build a drive shaft and complex center differentials if you can just plop another motor up front?!

lolachampcar | February 28, 2014

and then there is the SLSE with four independent motors. Horribly impractical but wow!

chrisdl | March 1, 2014

lolachamp:
Yeah, that's a sweet ride. And it has a decent battery for once. (Although it doesn't last long, I guess.)

4-independent motors theoretically sounds like the best way to drive an EV. One motor per wheel. I wouldn't be surprised if future (VERY future) Tesla's would use the same solution, but then perhaps first with 2-independent motors for rear drive only. No need for an old-fashioned differential any longer. Could be good question to ask Elon at the next event. What would his opinion be of such a solution for the future?

Brian H | March 1, 2014

What happens if one motor quits?

Zebuf | March 1, 2014

If one motor quits, the cars systems would recognize this, and disengage it. Then you essentially drive a pretty big go-cart!
Not very good for hard accelleration, but perfectly able to take you safely wherever you want to, under normal road-conditions.
Much preferrable to having only one engine if it should fail...
;-)

lolachampcar | March 1, 2014

I do not think Elon would ever go for something so complicated for the small benefit it brings. It is simply too much trouble for anything other than a super car (I'm thinking Roadster II here).

jordanrichard | March 1, 2014

Correct me if I am wrong, but when an AC motor fails, in becomes a generator but not in a good way. I mean, when you get on the accelerator for the MS, the power goes from the motor to the wheels. When you let off, the power stops flowing to the wheels and the wheels drive the motor, giving you massive resistance, hence slowing down. So if one of the wheels motors in a 4 motor EV fails, then it would create massive drag on that one wheel. That would be one helluva surprise if it were one of the front wheels.

DickB | March 1, 2014

@therealseanarch...

As far as I can tell the MS does not use the brakes for a wheel that is slipping. It will limit the amount of power to the motor. Had my MS up on a lift and ran it. There were no brakes applied, just the power was cut down. You also see this on the dashboard. The front wheels were not turning since the MS was up on a lift. Not sure if that makes a difference.

It looks like the MS limited differential does not use the brakes and is control by power to the motor.

Any one else try this on their MS when car was up on a lift?

Brian H | March 1, 2014

I meant on a 4-motor in-wheel car. Swerve city?

LMB | March 1, 2014

(LMB spouse)

I assume with four motors you would need 4 sets of reduction gears as opposed to the single set in the Model S. This seems expensive and less robust.

Pages