MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Transmission efficiency

As we all know a lot of power is absorbed in the transmission. In a standard MGB only 65 bhp actually reaches the rear wheels, according to Peter Burgessīs book.

This winter I will replace the 3 sync non o/d gearbox in my ī63 MGB with a 5 speed Sierra box. Everyone I know who has put a Sierra box in their MG comments on how much quieter it is. That has led me to think that it may be more efficient, too.

Noise and heat are symptoms that energy is used, right? Will a quiet and cool gearbox like the Sierra 5 speed let more bhp through to the wheels?

Tore
Tore

Tore, gearboxes typically do not lose much power. If they did you would need massive cooling. Industrial gearing loses typically 1-2% per mesh. All the transmission losses are associated with tires.
Art Pearse

Hi Tore

Art is correct. We see about 1 to 2ish bhp gain at the wheels with a 5 speed Ford box. The tyres are the big drain of power. We are seeing transmission losses around 22 bhp for road going Bs. This rises to around 27 for race spec tyres and speeds.

Peter
Peter Burgess Tuning

Very interesting. So the Ford box really is a little more efficient - but only a little. I was not aware that the tyres absorb that much power. I guess there is not much to be gained by looking for tyres with low rolling resistance for our cars.

Tore
Tore

Don't sacrifice grip though, remember the old Michelin X tyres? :)

Peter
Peter Burgess Tuning

Peter, how do you measure the "at the flywheel" bhp?
Art Pearse

I've seen it done by measuring the drag by driving the rear wheels in gear with the clutch down, and subtracting that from the 'measured at the wheels' BHP, as part of the same run.
PaulH Solihull

Hi Art, we measure power most repeatably with inertia. We accelerate the two rollers, the weights and inertia are known, The acceleration is sampled 100 times per revolution of the rollers. acceleration in a known time gives bhp. We test in 4th gear or direct top. We then run a coastdown test where we either dip the clutch or knock it into neutral and the pc times the negative acceleration. We press a button and it tells us what the 'flywheel' power is. It lets us compare cars like for like so we can see if the transmission losses are too hich for that type of car. Over nearly three years we have seen lots of combinations so we get to see what saps power.....wide tyres, high speed, independent rear setup with millions of UJs!

Peter
Peter Burgess Tuning

Interesting. I always wondered how. Do you think your round rollers deform the tires more than would a flat road, and so over-measure the tire drag?
Art Pearse

Hi Art

I know where you are going but deformation not enough to show up we feel, at least not with the upto 400 bhp we have seen as a max so far.

Many folk reckon to use a single roller dyno because of tyre deformation and other reasons. We opted for twin roller because we feel they are inherently safer if say a tyre fails.

We had a TR4 engined Peerless to setup on our rollers, it subsequently, many months later, ended up on a Dynocom large (flatter contact like a road)single roller dyno, identical figures at the wheels and transmissionlosses up around the 190 at the flywheel. We took this to show the tyre deformation on twin rollers was not a problem.

Does this help with your question?

We use the cosatdown losses to give us an overall picture so we can compare engine/transmission performance MGB to MGB, Mini to Mini etc. It gives a 'flywheel' figure but I for one would not treat this as gospel even though the computed flywheel figures closely match those we have been given from engine dynos, but that, in itself doesn't mean much either.

Peter
Peter Burgess Tuning

Thanks for the explanations Peter. It all makes sense now.
Art
Art Pearse

Lots of new information for me here! While we are on the subject of efficiency: I remember to have read somewhere that an overdrive unit absorbs something in the region of 3-5 % of the power trasmitted when it is used. That is in addition to what is lost in the gearbox itself.

Can this be correct?

Tore
Tore

It has its own gearing so has to absorb some power, plus the oil pump developing some 400psi.
PaulH Solihull

Hi Paul

You are spot on I think we saw about 4 bhp losses in OD top !

Peter
Peter Burgess Tuning

So in addition to complexity, weight and cost, that explains why the concept of overdrive units was so easily killed by the simple fifth gear...

Though I suppose a little power is lost even in an indirect fifth gear, compared to direct fourth.

Tore
Tore

Got it in one Tore, we are told to run the cars in direct top so we get the losses as close as we can to 'reality'. Most MG racers junk the OD just to save weight.

Peter
Peter Burgess Tuning

Peter, one more question!
If your machine does the HP calculation based on inertia x acceleration or deceleration, do you include the inertia of the car components - wheels, driveshaft, engine? They all need to be accelerated along with the rollers.
Art Pearse

Hi Art

We use the system purely as acceleration on the rollers. You would have to strip the car and work out every inertia value for every component otherwise. The way an inertia dyno works, if you took all the readings for a car then lightened the flywheel you would record a slightly higher bhp at the wheels figure, with no differences in transmission losses. At first sight this seems wrong but you have to remember that if you lighten the flywheel the car will accelerate faster, this means more bhp must be available at the wheels. To me it is like measuring potential. As a person wrote on another thread about flywheel lightening, it is removing the burden of accelerating the flywheel.

We also use the eddy brake facility for holding constant speeds for testing or constant loads which is different from inertia only testing.

Only inertia testing gives time after time repeatable results whereby changing something like a plug gap shows up!

I hope this answers your questions.

Peter
Peter Burgess Tuning

Hi Art

I mentioned your question to my friend Martin. Let me introduce Martin. When he was 17 he wrote to me asking how to build a flow bench for his Physics exams! He has never looked back and we have remained in contact and become mfirm friends ever since, he has pursued a professional career in engineering leaning to the electrical/electronic side. He is now 42 and I am 55! To go off at a Tangent(inherited from my Mum) with so much controlled by electronics I am so pleased I have Dave Gollan (our technical Insultant), Martin and my Son Simon who all understand these things and keep me on the straight and narrow with regard to electrically things!

Martin wrote....

Quite right, but their effect is reduced by the drive train gear ratio, and then further diluted to a negligible factor by the ratio of weight: inertial roler to combine inertia of the vehicle drive train. The Dynocom kit takes measurement of the acceleration of the roller of known mass, and hence the input energy imparted at that interface. Post vehicle transmission losses and inertia, what ever is left is measured (the energy available to accelerate the vehicle), algorithm applied to calculate and report in the wheel figure. The same scenario and physics applies when the car is driven down the road.

The coast down test on the Dynocom measures the combined coast down inertia of both the inertial roll and the vehicle dynamics. Interestingly, by using this method and the relatively slow coast down given the large mass of the roll, the normal error factor of windage loss of the roll and tyres is also captured, and hence a reduction in the final error budget.

The industries accepted bible on this is Engine Testing written by A.J Martyr and M.A Plint and this points to the fact that the roll inertia only needs to be 5x that of the drive train to give accurate results, and in the case of the Dynocom setup it exceeds any road vehicle, other than may be a heavy goods vehicle, and I don't think we'll be testing one of those particularly soon :-)

Regards,
Martin

PS I could not find the specific roll inertia number, will ask Paul who designed them at Dynocom if anyone needs the number. The Dynocom we run can deal with 5000lbsft torque and 1250 bhp. Peter
Peter Burgess Tuning

Peter
that's intersting about the overdrive losses. I always figure that too, but on the road there is a definite and consistent gain using overdrive 3rd over direct top, 1/2 sec in a 2km hillclimb. And I have the STN80 gear set so the difference between overdrive 3rd and direct top is only 3.5%. This was quite contrary to what I'd expected having similar results to you on the rolling road. But it was so consistently different I'd never junk the overdrive now, particularly for rallying when many corners are 3rd gear.
Paul Walbran

Hi Paul

I think the torque multiplication available in OD third is greater than that in direct top so it will accelerate quicker. I will have a play next time an OD car comes in and see if there is much difference in indicated power third to OD third, this will be a little seat of the pants as I will be visually logging raw (uncorrected)peak power figures . You can't beat 6 different gear ratios can you!

Peter
Peter Burgess Tuning

Hi Peter
Yes, I think it's more a case of the rev band being suited to the road configuration, but the difference did surprise me when I first found it. I have since found it so useful in rallying becuase the 3rd gear corners are usually followed by shortish straights that don't get past overdrive 3rd. On top of that, on gravel there is a nice little tweak to the tail when it flicked out of overdrive at the entry to the next corner, setting it up with a nice bit of sideways at just the right time :-)

Though to be honest, it all caught up with me on the last marathon rally here. It seems that the linings will take about 2.8 of these events, each with 700 miles of special stages, so say 2000 miles of this treatment before disintegrating dramatically. Another thing to put on the work shcedule before every event in the future!
Paul Walbran

This thread was discussed between 17/12/2012 and 21/12/2012

MG MGB Technical index

This thread is from the archives. Join the live MG MGB Technical BBS now