Differentials Explained

[doug@taylor-race]

Hi all. I'm new to the forum, but I've been lurking for a while and I noticed there has been a lot of discussion about the different types of differentials available. The following is a link to an article written by Craig Taylor, the President of Taylor Race Engineering. Craig is a Mechanical Engineer, Drivetrain Expert with 40-years of experience, Multi-time National SCCA Driving Champion, Land Speed Record Holder, Designer of the Quaife Viper Differential and my Boss. I learned quite a bit.

http://www.taylor-race.com/pdf/understanding_differentials.pdf

 

[mike & juli]

Your first post and quite the informative one at that, thank you! ~juli

 

[Jeff Monheim]

Thanks Doug. This is something I needed to read!

 

[mjorgensen Woodhouse]

Thanks Doug these are all good things for the community to see and thanks for your help the other day!

 

[Bobpantax]

Hi all. I'm new to the forum, but I've been lurking for a while and I noticed there has been a lot of discussion about the different types of differentials available. The following is a link to an article written by Craig Taylor, the President of Taylor Race Engineering. Craig is a Mechanical Engineer, Drivetrain Expert with 40-years of experience, Multi-time National SCCA Driving Champion, Land Speed Record Holder, Designer of the Quaife Viper Differential and my Boss. I learned quite a bit.

http://www.taylor-race.com/pdf/understanding_differentials.pdf

Thank you for the data. Where does the '08 Viper differential fit into the types discussed in the article? And, what do you think of it?

 

[doug@taylor-race]

Thank you for the data. Where does the '08 Viper differential fit into the types discussed in the article? And, what do you think of it?

Honestly, I'm not familiar with the '08 Diff, so I really can't comment on it. I try to keep-up with changes in all the drivetrain systems, but we provide service to all the Quaife applications as well as most Sportscar and Formula Car series.

 

[Bobpantax]

Here is how the '08 differential is dscribed: GKN Visco-Lok speed-sensing limited-slip differential. And here is more data from GKN's site:

Visco Lok LSD

You must be registered for see images

GKN Driveline’s progressive-locking speed-sensing Visco Lok® LSD (Limited Slip Differential) is proven in front and rear axle applications. Visco Lok® LSD is a traction optimised speed-sensing limited slip differential design







Benefits

• <LI class=l>Tuneable and progressive locking characteristics <LI class=d>Provides maximum torque transfer in excess of skid torque
• Self-actuating with no external controls required

Any comments or observations?

 

[ViperJay]

why bob, its so simple. the visco actuating... torque.... dohicky thing turns the.... locking sensing... thing-a-ma-giggy.... umm.... thing. and the shaft goes into the.... uh.... housing from the front, and..... turns the wheels. its kind of simple when you think about it. therefore, the 08's are better.... i think. call me anytime with your questions.

 

[GTS Dean]

Top notch article! Thanks.

 

[doug@taylor-race]

Bob, The GKN Visco Lok is nothing more than a Viscous differential. Here's an explaination;

"A Viscous Differential looks internally similar to the Salisbury clutch pack type diff, except that the clutch plates and stators are sealed inside a capsule filled with a silicone based fluid. The silicone fluid’s viscosity increases with shear rate, meaning that as the inside wheel starts to spin, the fluid thickens and causes the torque to be applied to the outside wheel.

The hydraulic speed sensing differential uses the differential in wheel speed to actuate a pump, which is ported to a piston compressing the Salisbury-type clutch pack, replacing the need for ramps. Since the speed sensing differential requires wheel spin to actuate, it can only limit wheel spin, not eliminate it. Therefore, very high rates of tire wear on the inside rear will be experienced. Further, the hydraulically actuated clutch pack or viscous pack cannot share the load unless one wheel is spinning, unlike the Salisbury type. This severely limits their value for standing starts and for drag racing.

Our shop has replaced number of viscous and hydraulic differentials that come in the Viper SRT10 and the Viper Competition Coupe with Quaife differentials. In most cases the stock differential gears were pitted, cracked, or broken because of the overload placed on these gears.

The benefit of the speed sensing differential is that it can transmit drive torque to the “good” wheel when one wheel is in the air. So they are used for drivers that hop the curbs hard enough to throw the inside wheels off the ground. Setup for these differentials is the same as for an open differential, as they do not lock."

I have 25-30 of these in the dumpster right now!

 

[mjorgensen Woodhouse]

"I have 25-30 of these in the dumpster right now!"

Doug,

Steel is high right now so send them my way!

 

[Bobpantax]

Please clarify. You have 25 - 30 GKN Visco Lok differentials in the dumpster right now? The same differential being used by the '08 SRT 10 and ACR? Are you suggesting that the SRT engineers did not properly test the differential before they decided to use it in the '08 SRT 10 and ACR? I ask because the tested track performance for the ACR is phenomenal and no one who has tested the car has mentioned any problem concerning the differential.

 

[j-rho]

Bob, The GKN Visco Lok is nothing more than a Viscous differential. Here's an explaination;

I have 25-30 of these in the dumpster right now!

Doug, the SRT-10 diff is not a regular viscous, but actually more like the hydraulic unit mentioned. It does not have a sealed silicon fluid chamber, but uses the diff lubricant both as the hydraulic pump and shear fluid.

While certainly not an ideal solution, I have found that it can be tuned to work reasonably well with the right fluid makeup.

 

[Bobpantax]

Doug, the SRT-10 diff is not a regular viscous, but actually more like the hydraulic unit mentioned. It does not have a sealed silicon fluid chamber, but uses the diff lubricant both as the hydraulic pump and shear fluid.

While certainly not an ideal solution, I have found that it can be tuned to work reasonably well with the right fluid makeup.

What fluid makeup do you recommend? I have the OEM fluid with the additive and it seems to work quite well.

 

[j-rho]

If you are not racing in a class that mandates the stock differential, I recommend not bothering with different fluids, and going to a proper racing diff like a Quaife or clutch-type unit.

If you must make do (as I must in SCCA stock-class autocross) with the stock components, but have the freedom to change fluid, I recommend using less than the factory specified quantity of friction modifier, and/or changing to a fluid designed to minimize slippage in a limited slip. You'll have to play around with it to suit your taste. I found the factory fluid composition to be way too slippery, allowing far too much inside wheelspin at corner exit.

My altered fluid composition is likely to reduce the lifespan of the factory differential, which should also be taken into consideration should you try same.

I should also note my car is a 2004. I have not driven the 2008 car yet but from what I understand its differential action is much improved.

 

[doug@taylor-race]

Please clarify. You have 25 - 30 GKN Visco Lok differentials in the dumpster right now? The same differential being used by the '08 SRT 10 and ACR? Are you suggesting that the SRT engineers did not properly test the differential before they decided to use it in the '08 SRT 10 and ACR? I ask because the tested track performance for the ACR is phenomenal and no one who has tested the car has mentioned any problem concerning the differential.

Hi Bob, No we don't have the 2008 Model in the (metal recycling) dumpster, but I don't see much functional difference between the two. Although I'm sure the 2008 vesrsion is a clever diff, it still depends on wheelspin to provide the locking pressure. The Quaife only needs a loss of traction to mechanically bias, so long as at least 20% of the drive wheel weight is on the the inside tire, no wheel spin will result.

Regarding your question that I suggested "SRT engineers did not properly test the differential before they decided to use it in the '08 SRT 10 and ACR?" Let me put on my sarcasm hat for a moment.., No, I would never suggest that an engineer made a mistake.

In short term testing, I'm sure that all the differentials work fairly well. However, we have seen dozens if not hundreds of broken stock Viper Diffs. To date, I have never seen a broken Quaife Differential, including those installed in hundreds of race cars of different types.

I guess time will tell.

 

[j-rho]

The Quaife only needs a loss of traction to mechanically bias, so long as at least 20% of the drive wheel weight is on the the inside tire, no wheel spin will result.

While I don't mean to hinder the efforts of those selling their products, a couple things should be noted-

A Quaife relies on *some* tractive torque from the inside driven wheel. In a car that dances over curbs, this can be a problem when the inside rear wheel is in the air, offering no resistance, and receives 100% of the input torque. In some cases if the driver is not careful, that inside tire can spin up considerably, and cause big problems when it lands with a large speed differential vs. the road.

The other issue is in situations where there is enough power available to overwhelm the inside tire even with it on the ground. As you begin to get inside wheel spin, the Quaife behaves the opposite of the Viper's stock diff - instead of the increasing wheel speed differential causing the differential to lock up more tightly, the now-spinning inside rear tire offers decreasing amounts of resistance as its longitudinal slip % increases, thus increasing the amount of input torque it receives, worsening the situation, until the driver backs out of the throttle, or the car settles down out of the corner and returns towards even weight over each rear wheel.

The former case applies to road racers, the latter generally only applies to autocrossers. To combat the latter draawback to this style diff, Torsen made a really nice piece, the T2R, which behaves like a helical (Quaife/Torsen) diff but has some adjustable preload characteristics of a clutch-type. I don't think they make one for the Viper though.

Party on,

 

[GR8_ASP]

OK, I will bite.

It is clear you are on here trying to sell some Quaife differentials. No question they are a good piece and capable of enhancing the performance of our cars. I remember evaluating the Torsen type diff back 25 years ago and understood its benefits and of course costs. Worked great on high horsepower front wheel drive at the time. So no question as to how they work, their strengths and weaknesses and such.

But I think you need to do some more research into the Viper diffs before you start coming to conclusions. The only diff that matched your hydro-lock definition is the 2003-2006 SRT-10. Not Gen I/II. Not comp coupe and not 2008. And so far the visco-lok diff for 2008 has had glowing reviews. Time will tell from a extreme performce perspective but I feel it is very inappropriate for you to scare people with 2008 SRT's based on scant direct evidence with them.

If you read through the archives here you will find plenty of negative comments regarding the 2003-2006 diff. From performance driveability to durability. Some of us have even replaced the cross pin with a more durable pin due to early failures (that I have 600 RWTQ helped me make that decision).

Gear durability concerns are somewhat different. You must first explain which gears are considered weak. Generalisms will not work in the gear arena. Pinion, ring, side or output, spider or planetary. They are all different. Obviously the side and planetary gears are of a unique type between traditional differentials and the Quaife or Torsen type. But the ring and pinion are not.

 

[doug@taylor-race]

Doug asked me to step in here with a few comments. My name is Craig Taylor. I am the engineer that designed, in conjunction with Quaife, the Automatic Torque Biasing Differential produced by Quaife for the Dodge Viper. Hopefully I can clarify a few issues, and stimulate further discussion.

First, full disclosure: We at Taylor Race Engineering, Inc. really do want to sell Quaife differentials for Viper cars. We believe in their performance, strength and reliability. It is the only differential that we know of that has a limited lifetime warranty, even in racing applications. However, we want to ensure that the Viper community is well informed of the method of operation, advantages and disadvantages of each type of differential. Therefore in response to your questions and comments:

1. We do have a recycling bin full of the Hydro-Lock differentials (2003-2007 Viper). We have accumulated a number of the Visco-Loc differentials as we are replacing a number of them with the Quaife unit, but not nearly as many as the Hydro Lock.

2. Both the Visco-Loc and Hydro-Lock differentials are activated by a difference in wheel speed which activates a pump, pressurizing and compressing the clutch pack. The compression of the clutch pack is proportional to differential in wheel speed (wheelspin). The pump in the Visco-Loc is valveless creating the necessary pressure in proportion to the shear rate of the viscous fluid. The pump in the Hydro-Lock is a positive displacement gearoter pump (like an engine oil pump) that is activated by a difference in wheel speed. Unlike the Visco-Loc, it draws its working fluid from the differential lubricant. The pump works in both right and left wheel spin by a clever reversing of the outer rotor when transitioning from a right wheel spin to a left wheel spin. In fact the Visco-Loc is a vast improvement over the Hydro-lock because it is much quicker acting. This is because the working fluid is contained in the differential, not drawn up from the sump of the differential. Additionally, it does not have to take the time to “reverse” the pump as in the Hydro-Lock.

3. The advantages of the Visco-Lock and Hydro-Lock are: 1) if the spinning wheel is off the ground, it still can lock, delivering torque to the wheel on the ground. 2) It does not create an understeer influence which is the major problem of the Torque-Lock differential of the 1992-2002 Vipers.

4. The disadvantages of the Hydro-Lock and Visco-Loc are 1) The dependence upon actual wheel spin to initiate locking of the differential and 2) the clutch pack does not share the drive load unless wheel spin is present. The lack of sharing of the load in straight ahead acceleration (drag racing) is the Achilles heel of these differentials. During the launch, the spider gears, side gears, and crosspin carry the entire load, resulting in frequent failures. Most of the Hydro-Lock and Visco-lock differentials that we have removed are failed, or have cracked spider gears. (Note: The Torque-Lock differential does share the launching load, as the clutches are activated by input torque)

5. The operation of an Automatic Torque Biasing differential is quite simple. When the differential senses an impending loss of traction on the weaker wheel, it shifts torque to the wheel with greater traction. It varies in a linear fashion from 50%-50% to 80% - 20% bias. In other words it can bias up to 80% of the available drive torque to the better wheel.

6. The advantages of the ATB differential are: 1) Power on understeer is greatly reduced when compared with the Torque-Lock differential. The result is improved lap times in road racing. 2) Six planet gears drive each output (side) gears, as compared to just two in the Torque-Lock, Hydro-lock, and Visco-Loc. The immense strength of the Quaife ATB makes it very popular with drag racers and aftermarket tuners. Applications of up to 1200 HP have been easily handled by the Quaife ATB.

7. The only disadvantage of the ATB differential is that it requires a reference torque from the inside wheel. This means that at least 20% of the drive wheels’ weight must be on the inside wheel for maximum acceleration out of a corner. If the driver is a curb hopper, the inside wheel will spin freely while it is off the ground, and cannot bias any torque to the outside wheel. However, with a T-1 Viper driven by National Champion Bobby Archer, we documented a 2 to 3 second improvement in lap times at Texas Motor Sport Ranch near Fort Worth.

 

[ViperJay]

see? this is exactly what i was saying in post #8! simple!

 

[jmsrx]

Here are some more pictures and some animations for us verbally challenged people.

HowStuffWorks "How Differentials Work"

 

[viperbilliam]

No wonder I can't get rid of my stock differential! I'll just keep it for collector value. As a satisfied consumer, these Quaifes work. It's the first mod I'd do.

 

[mjorgensen Woodhouse]

A big part of the advantage when jumping the curbs is that the torque is NOT all transfered to the outside wheel all at once. This is NOT a good thing because you can "overload" the traction ability of that tire and get into a very loose situation (needing to pedal off the throttle and slowing down in the process). This is why Bobby and many others are faster with a Quaife. The overall "comfort level" it gives let's the driver push harder into each corner.

In an autocross situation though this is not the best situation and on my particular car I ended up doing the opposite that you would think to make the car faster. When the car was setup for optimum "cornering power" I could not get on the gas and exit the fastest because the inside rear was off the ground to much. The "wrong thing" ended up being the right thing for this though and I increased my front spring and bar rates by almost 200% to get the rear of the car to stay flat and give me the ability to "steer" out of the corner on the gas early. On the road course though my car is a real handful but I LOVE it's ability to point and accelerate at will.