Quaiffe ATB transmission losses.

[Blatman]

I have a Quaiffe ATB equipped English (salisbury) diff in the back of my IRS Westfield sprint car (but don't let that put you off!)

Took the car for a run up on the rollers at Northampton Motorsport, and after a quick run up (it was a "shootout" type session), I got my printout...................

I was pleased with the power output, but the transmission losses were over 50bhp. So, do I have a problem, or do ATB's suck power 🤔

The road operator mentioned that the car seemed to want to squirm sideways out of the rollers, and suggested that there may be a diff problem, or rear set up problem. I know that the rear end is properly adjusted for camber (1 degree neg), toe (out, just a smidge), damping, tyre pressures (18psi, 205/60/13 Falken 502's), etc. The car was strapped down quite tightly, but movement was still apparent.

Any ideas, anyone (apart from buying a Caterham.........).

TIA.

[AMMO]

Blatman

 

What was the wheel horsepower/overall power or percentage of transmission loss? Did other vehicles have similar losses?

 

At the Dyno/Curry 1 Mike Bees' car made 229 bhp with 52 bhp/22.7% transmission loss. Is your car around about the same power?

 

AMMO

[Blatman]

Yes it is pretty similar (on these rollers....... It made 248 at Pete Baldwins place a while back, with no strapping down of the car, just front wheels chocked, and adjustable rollers). It made 223bhp, with a corrected (for what?) figure of 217bhp. Transmission losses were calculated at 53bhp, an overall loss of just under 25% of the 223bhp figure (I can't be arsed to do the maths right now...). I was disappointed in that, BUT it occurs to me (after reading the recent debate sparked by the Puma Racing coastdown/power at the wheels controversy) that I have a bog standard flywheel, a heavy duty road clutch plate, tyre diameter on the small side, a straight cut synchro box, and "standard" rear brakes. Also, as mentioned, the car was strapped down pretty firmly, and the rollers are not adjustable. Even so, I'm a little concerned, but you seem to suggest that there may not be too much to worry about.

 

Edited by - Blatman on 27 Oct 2002 13:10:10

[AMMO]

Blatman

 

Was that Peter Baldwin's rolling road when he was at Marshall's in Cambridge? If so, that dyno was known to read around 20 bhp more than most. Not that it was a problem if you used it as a comparator. It was worth going there just for Peter's tuning skills.

 

I never worry about dyno figures too much as the damn things all read different. That's why I'm quite keen to get a few cars on the same dyno on the same day to compare. My simple rule is that when you are on the back straight at Snetterton and people are going past you, that's when you need more power. When we first started racing motorcycles in the mid '80's my bike dynoed at around 85 bhp on a Shenke engine brake. Other competitors were quoting 100 bhp +. We used to go past them in a straight line.

 

Rule of thumb for transmission losses is: on bikes 7- 9 % and on cars 18- 25%. I don't know (or care) if this is particularly accurate. It's easier to use this than worry / argue about it. Then you can get on with other things like sorting out the package so you can win races or whatever else it is you are trying to acheive.

 

The correction was probably for 20 degrees centigrade at sea level. So on a hot or cold day you can take into account how much the engine would of made if the temperature was 20 degrees at sea level. Then there are other corrections like SAE and DIN (that might not be a correction, just a measuring method, can't remember). You can cheat a bit by using DIN as it gives power figures that are a bit higher than SAE (I think that's the right way around). Another snippet of information that has not been retained in my head.

 

I'll give you some examples:

 

On one occasion I took my bike to two different dynos on the same day and got a reading of 139 bhp on one and 132 bhp on the other. TTS bought a new dynojet dyno same model and spec. as a previous one and had figures that were consistently 7% different.

 

I have some nutty German friends near Hockenheim who built their own dyno which could be converted from a a rolling road to an engine brake. They would only use flywheel figures and had calculations to do transmission losses which took into account things like flywheel and wheel weights. All the bike magazines in Germany use them as it's supposed to be the most accurate, TUV approved etc. Our 139 bhp (at rear wheel) Guzzi used to go past their 150 bhp (at crankshaft) Guzzi on the back straight at Assen.

 

I don't know what the answer is to acheiving accurate power figures is, but I don't lose any sleep over it. I suggest you don't either.

 

You should, however, seriously consider getting a Caterham. 😬 😬 😬

 

AMMO

[Paul D Jones]

Lap times, and silver pots on top of the fire place.

 

All dynos and rolling roads read differant

Air temp

Air pressure

Tyre size and distortion on the rollers compared to real world road conditions

Ram air effect of an air box etc.

The choice of gearing makes a differance

The way the ramp up loading of the rolling road set up will give a differant reading

And lots of other effects.

Use the same rolling road to compare BHP befroe and after mods etc.

 

 

 

Paul.

See My Car Here

[Blatman]

Hi Paul.

Yeah, I know all about the variables of the various operators, and the differences in readings that can occur. I'm not bothered by the power output difference, as I was expecting it. I was more concerned with the operator stating that the car was trying to go sideways, and the only cars that did that in his experience, had rear suspension or diff problems.

AMMO.

I was at Petes place back in April or May this year, and I though it was in Hertfordshire, although it wasn't that far from Cambridge. It definitely wasn't in] Cambridge, although it may have been Cambridgeshire......I'm confused.............

Whatever, I'm not losing any sleep over the power figures being different, but the transmission losses are higher than I expected, so I wanted an opinion on those. Seems up to 25% isn't unusual, but I'd still like to see if I can address them, and get them down a bit.

[Blatman]

I'm not confused. Pete Baldwins rollers were at Wilshers Garage in Royston. If it's the same rollers and set up that you know about, AMMO, then it's reasonable to assume that they'll still be reading about 20bhp high. I'm still not concerned with the output though, just the transmission losses.

[AMMO]

The Peter Baldwin I'm thinking of is a championship winning Mini racer. He was originally at Marshall's in Cambridge but has since moved on. I tested my Cooper S there. He did a really good job setting up the carburation.

 

Seems that transmission losses rise with power output. Don't know if cooling the transmission with a pump and rad would have any effect. Oil might make a small difference. You could pay attention to all the bearings. I used to build all my race gearboxes with worn (but not knackered) bearings. The gains are probably very little, but it all helps. 125 racers used to run in all their bearings, including wheel bearings, on a lathe.

 

 

AMMO

[Blatman]

Yup, same fella.

Cooling may help, but as it's a sprint car, it'll never get any hotter than when it's on the rollers. C'est la vie......

I hadn't considered tight bearings, but I'll be having a look, and asking some questions of that nice Phil Stewart bloke.......

Thanks for the help.

[Graham Perry]

Ammo/Blatman, At Dyno/Curry 1 my car showed mechanical losses of 30 BHP on the coastdown. It has the AP Torque biasing diff which although of a different design works on similar principles to the ATB . The at the wheels number was 151 BHP. I did however have my diff set up slack deliberately to reduce losses though. Thats why its so noisy. I was surprised how low this number was. I would suggest Blatman that getting it set up different might make a difference although you have to consider the gearbox as well.

[Mike Bees]

Biggest contributor of losses on the rollers are the tyres.

 

Graham - you seem to be saying that there is some adjustability in the AP diff? How/what/why?

 

Mike

[Fast Westie]

Blatman

I don't know about the losses in the ATB, but I do know that mine used to get very hot in a 10 min race (150C case temp). Hence my fitting of a diff cooler. I have had it rebuilt with very little bearing preload (Mike - this is what Graham is talking about) and it doesn't go much above 95C.

 

What you may find is that the transmission was still cold when you did the run and the oil drag had quite an effect. The Quaife ATB has lots of gears in it to slosh around.

 

The car in front is a Westfie1d

[Peter Carmichael]

Blatters,

 

Roller losses are mostly goverened by roller speed. Which gear did you have it in? AMMO's rule of thumb is a bit of a red herring because a 250 bhp car run up in fifth gear will show the same coastdown losses as a 100bhp car... IF you have the same number of people sitting on the back, the same tyre pressure, etc. etc. I have more than 95 spreadsheets of various rolling road runs from Emerald and they show losses that do not mean yours are in any way out of the ordinary.

 

The key bit with rolling roads is the repeatability. If you are ever in doubt, ask the operator to do another run and see if it lines up. If you look at the "at the wheels" figures they will be all over the place, but a properly performed coastdown test will give the same figures within a horespower or two.

[Blatman]

Peter.

It was 4th gear (1.1), coast down was from 60mph, IIRC. Repeatibility won't be a problem 'cos they seemed to know what they were doing, the only variable being how tightly they tension the straps when trying to keep the car in one place...........In this instance it was pretty tightly strapped down, and they seemed to do two or three coast downs, although they may have just been trying to settle the car. It wasn't really the place to start getting into too much detail, as it was a shootout-cum-pissing contest for us Westfield boys (I came second to a chipped Cossie turbo 😬)

After the Puma Racing discussion a while back, I'm convinced that looking at the at the wheels figures is nonsense, and I'll only use rollers for a broad look at before and after figures when I've been tinkering.

I'm a little happier that the rolling losses aren't considered to be as surprising as I found them, but the comment from the operator that the car was trying to crab sideways out of the rollers, and his suggestion that this behaviour is often indicitave of a rear end problem of some sort, has me thinking that an examination of the diff/rear end may lead to some improvements being found.

Thanks for the input, fellas.

[Graham Perry]

Mike, As Fast Westy says, you can vary the diff pre-load so that it goes from impossible to turn by hand to very easy to turn by hand BUT you get more backlash,noise etc as a result and its not as nice to drive. Mine is very slack and this probably contributed to the lowest losses I have ever seen on the rollers for my car. Like yours it has been 50BHP in the past. Also my gearbox is also quite loose as its been rebuilt so many times.

[mr_ed]

Blatman,

 

Surely when you're going in a straight line, the Quaife ATB would absorb no more power than any other diff, open or LSD...

 

It's only when one wheel is going quicker than t'other, that the magical workings on the diff actually do anything.

 

So, when on the rollers, I'd not expect anything to be happening in there. It's just the crown wheel and pinion doing any work.

 

Cheers,

 

Ed

 

 

 

[J.R.]

At last someone -ED has stated the obvious.

 

Are both back tyres the same rolling diameter or is one worn?.

 

This would create squirm and extra power loss.

[Blatman]

Yes, I had realised it's open when being used on the rollers, but I thought the the internal workings are all still rotating, and they have quite a lot of mass to them. Whether it's more or less mass than a plate diff I don't know, and it's certainly more than a boggo open diff. Even so, I don't think mass is causing the losses. Can the worm gears get stuck, or be slightly engaged when moving in a straight line, ie, can they be prevented from disengaging fully, for whatever reason (wear/other internal problem) 🤔

Both rear tyres appear to be evenly worn, and have equal amounts of tread on them. I'll be checking the wheels for roundness/buckling too, then, I guess, although they looked fine from where I was standing (very close).

 

[AMMO]

The percentage losses were calculated from a CCC article of a dyno shootout from a few years ago. As I said, Idon't know if accurate or not. The important rule of thumb is the back straight of Snetterton one.

 

At T&M all the cars were run up twice and the graphs overlapped perfectly. The transmission losses on my car were in the region of 22 % as was Mike Bees'. Mine was 38 bhp and Mike's 52 bhp.

 

AMMO