dolomite sprint bhp

[SprintMWU773V]

We have a dyno rig at work that can handle up to 850bhp so I'd be interested to put a Sprint engine on it to see what kind of figures it produces.

You mean like this?
https://sprint.smugmug.com/Cars/F3-engine/i-pxcZgKS

That is very cool but no ours is a little different http://www.bighealey.co.uk/dynamometer

[soe8m]

I do not have a 20% loss but about 15hp measured by a negative dyno roll out.

The problem with seeing it as a fixed 15 hp loss, rather than as a percentage, is what would be developed at the wheels when the engine is producing less than 15 bhp. It won't go backwards, will it? Neither description of the loss is exactly correct, but when looking at max power the percentage loss is much more useful.

But if you have only 11 percent loss in the transmission (15 of 139), that does seem a remarkably low value. What's the box and diff?

Graham

A well build original box and diff. A bad wheelbearing or uj can cause 5-10 hp loss. The prop centre carrier is a std one but reïnforced less flex.

I do not say the loss of A transmission IS 15hp but mine has.

The loss in hp is measurable with a good dyno. After a full run roll it out and the dyno can measure neg power how many hp is "used" by the transmission. So when i put a big v8 or cosworth in my dolomite having around 500 hp then the loss in my gearbox and diff would raise to 55-60 hp? ( a bit more friction and tordation incalculated ) You cannot guess with percentages. I know you have guess with something but most of the time it's used and even higher percentages to have a pub talk oeh ah more engine hp's.

By the way. A test with or without a (visco) fan at high revs makes a difference of 5 hp.

Jeroen

[GrahamFountain]

I do not say the loss of A transmission IS 15hp but mine has.

The loss in hp is measurable with a good dyno. After a full run roll it out and the dyno can measure neg power how many hp is "used" by the transmission.

Yes, but that measurement of 15 hp loss in your transmission will be specific to the power measured from the engine in hp, i.e 139. Because the machine that worked it out knew what the maximum measurment was. If you took 3 of the plugs out, and re-ran the entire test you'd get two very different values for the measurments of max power (hi Homer) and the loss in the transmission - probably something less than 34.75 and 3.75 hp.

And if it's related to a specific value, it can be expressed as a percentage of that value, with some reasonable expectation that for lower engine outputs the loss in the transmission will be the same percentage of that power output, rather than a fixed loss.

As I said. if it was a fixed loss of 15 hp, i.e. the power at the wheels was always 15 hp less than the engine gives, then when the engine gave, say, 10 hp, there's be -5 hp at the wheels, i.e. 5 hp pulling the car backwards. Since that does not happen, it simply can not be a fixed loss.

Graham

[soe8m]

No, it's from v max to v 0 with the transmission in neutral. Just a measurement of the drivetrain without the engine running. Independent of what engine and in what state it is. Just the transmission and diff of that particular car was 15hp.

Jeroen

[GlenM]

Graham, surely the transmission will use up the same number of horsepower whatever the engine is producing. If the Sprint transmission uses, say, 20 BHP on a standard blueprinted engine, it won't suddenly require the same percentage of BHP from, say, a 1000 BHP V8.

[GrahamFountain]

Okay, if the idea of a transmission with a fixed power loss in hp pulling you backwards with the engine off isn't absurd enough for you, think about it the other way: If it took a fixed power to turn the transmission, then even if you jacked the car up at the back, the wheels wouldn't move till you achieved that power input. And if you think about that, it actually just means they wouldn't ever move, because work is the translation of the point of application of a force, and power is work done in unit time. So if there's no movement, there's no work done and no power used, and so there's still no movement. Or, it has to be moving already before you can make it move, if that takes a fixed power. There may be other reductions to absurdity, but I think two is enough.

It just does not make any sense, any way you look at the physics, that the power loss in the transmission is a fixed value that does not depend on the power going through it. That it's a fixed percentage of the power transmitted will be only approximately true. However, it will be good enough for the purpose of guessing what the difference in engine power has to be for a given a difference in power at the wheels, for the same (or sufficiently similar) transmission. And that's just about the only use it has. So I'm not really sure why so many of you are that bothered about it.

It should, I think, be a reasonably good approximation if all the back forces in the transmission are from friction in the bearing surfaces and between gears. And at a constant speed and transmitted power, that should be largely true. It must, therefore, stop being any kind of approximation when the lubrication between some pair of surfaces breaks down, which will be limited by both the speed the transmission is turned and the maximum power being put through it. But that change only occurs with the loss of lubrication, and that means a big change in wear rate. So it only applies to an overrun or overloaded transmission, and I'm going to ignore it.

The main non-linearity, within the operating ranges of speed and power, will be from the viscous forces with the lubricating fluids. These, unlike frictional forces, are dependent on the square of the velocity of whatever is moving through the fluid. So these forces are related to the speed the transmission is turning and not related to the forces being transmitted: they will change as the wheel speed of the car changes, but not as you apply more or less torque. However, I don't think they are significant in a manual transmission. And they don't tend to make the power loss a fixed value. Rather, they mean it changes in more complicated ways than a fixed percentage loss.

As an aside, the effects of viscous forces are significant in automatic transmissions, because of the torque converters. But that's complicated because they depend on the difference in the input and output speeds, not the absolute speed the converter is turning. I'm not clear what the nonlinear relationship between loss and the power going though the torque converter will be. But these viscous forces will, in effect, massively tighten the coupling between the input and the output of the torque converter as the speed difference, and thus the transmitted power, increases. And that's a negative feedback effect, which will tend, very strongly, to try to make the relationship linear. So even the complications are complicated. But, fortunately, it's irrelevant in the discussion of a manual transmission.

Graham

[GrahamFountain]

Graham, surely the transmission will use up the same number of horsepower whatever the engine is producing. If the Sprint transmission uses, say, 20 BHP on a standard blueprinted engine, it won't suddenly require the same percentage of BHP from, say, a 1000 BHP V8.

It should, if you'd kept every part of the transmission the same, including the gearbox and tires, and if the transmission is a perfectly linear system, and the 1000 hp didn't bugger it to bits, take something like 200 hp of 1000 if it took 20 of 100 (I did pick easy numbers there). But if the transmission suitable for a 1000 hp V8 took a fixed 200 hp from whatever is input, a 100 hp engine would never be able to start the car moving.

Is possible that a 100 hp engine wouldn't do well moving a car meant to take 1000 hp at full speed (200+ mph); that 100 hp would not be able to move it at all is obvious bollocks. Sorry, but it is.

However, as implied in the above post, there're questions about exactly how linear a system the transmission of a car is. And going from 1000 to 100 hp (even 155) might be more than a bit of a stretch – I'm certain that a transmission built for 1000 hp wouldn't take near the same percentage as one built for 100, or 155, or 135, or whatever. But when comparing 90 hp at the back wheels with 139, and working out roughly what that means at the engine, it should be near enough – given that the transmissions, including box and tires, are sufficiently similar.

Given only that they're both manual transmissions, the transmission losses could easily vary quite a lot; Mad Mart gives about 20 percent for a doly and I give about 25 for a TR7. Jeroen's 15 hp loss giving 139 at wheels, or 10 percent loss (15/(139 + 15)*100), seems a very small value in my, admittedly limited, experience. But the 154 hp that value gives seems like a very credible value for a blueprinted engine. So 10 percent, or 15 of 154, loss – however you care to express it – just appears to reflect an exceptionally well put together transmission.

Graham

[Carledo]

What seems to be evident from all this, is that the ONLY way to get an accurate engine BHP figure is to test it at the flywheel on an engine test rig! End of story!
I can't be bothered with all this and am pretty happy with the Rolls Royce idea, either the horsepower is "sufficient" - or it isn't, in which case, something needs to be done!

Steve

[GrahamFountain]

Well it's nice to know what the power at the flywheel might be, but the car is moved by what you put on the road.

Graham

[cliftyhanger]

Possibly a more useful reference figure would be the standing 1/4 time?
That would then ignore engines that have big horespower but no torque until very high RPM and so on, and a more realistic idea of what the car is like to drive.
Otherwise I understand the engine dyno is the "gold standard" if you wish to compare power outputs. A normal rolling road is OK, but so many factors can affect the results that the same car on different days is likely to get different results. Go to another RR and it will be even more random.

[soe8m]

Possibly a more useful reference figure would be the standing 1/4 time?
That would then ignore engines that have big horespower but no torque until very high RPM and so on, and a more realistic idea of what the car is like to drive.
Otherwise I understand the engine dyno is the "gold standard" if you wish to compare power outputs. A normal rolling road is OK, but so many factors can affect the results that the same car on different days is likely to get different results. Go to another RR and it will be even more random.

Rolling roads are sometimes misused and that gives the differences in outputs. There's also a factor CW value parameter you have to use and that is a bit guesswork. When you have CW low then your power will be less and CW high you have a more positive readout. Engine power is only measurable as stated on an engine dyno. The way you measure the power loss is also calculated with factor X. The roll out is measured the other way round, like a brake measurer roller at a MOT station. Gearbox in neutral but the friction is less than have power on the gearbox. That factor X is a std value what a dyno uses to calculate. That way it is a fixed value of a particular car no matter what engine. A more powerfull engine will give slightly more loss but not in percentages. I think my gearbox casing will melt loosing 200hp transferred into heat.

Jeroen

[GrahamFountain]

A more powerfull engine will give slightly more loss but not in percentages. I think my gearbox casing will melt loosing 200hp transferred into heat.

Well, obviously, a Sprint transmission won't handle 1000 hp, which is what I already said. And a transmission that will handle that much power will probably lose a rather different amount to the Sprint's – don't know if it's more or less, just different. However, if the 1000 hp transmission was linear and it did lose 200 hp of a thousand, it would lose 20 hp of one hundred. Actually, because it won't be perfectly linear, I think it would lose much less than 20 hp of 100, maybe less than 5 or 10. So the loss from a much bigger engine will be a greater, not a lesser percentage. But I'll explain why that should be, later.

But first, what you're saying, in effect, is that if you were to connect a 15 hp engine to your transmission, it would not even begin to move the car. That is, until the input power exceeds that 15 hp threshold, the output won't even turn with the wheels off the ground. So where is that power going, and how does it get there if nothing turns? Power, being the movement of force at speed (1 Watt = 1 Newton moving 1 Meter in 1 second), can only get into the transmission if it is turning. If it's not turning, all that goes in is torque. And as power is directly proportional to torque times rpm, the power at zero rpm is zero, never mind what the value of torque might be.

There will, even with the wheels off the floor, be some threshold for the torque, to overcome the stiction in the bearing surfaces and meshed gears, etc. But that's not power, and once this torque threshold is overcome the transmission will turn, albeit slowly, for very little power input. I would be very surprised (gobsmacked, even) if you couldn't turn your transmission, weight off wheels, from the nut at the front of the engine, in gear, with the plugs out, using only a big torque wrench (and can thus measure the torque threshold empirically). But to generate 15 hp that way, you'd have to be Heracles incarnate.

Moreover, how do you explain how this "black hole" of a transmission, which is magically sucking 15 hp from anything put in, does not try to pull the car backwards when it's out of gear, or have the gearbox/clutch spinning wildly all the time you're parked? There must be something stopping that; beyond the fact that perpetual motion machines don't exist.

I apologise if you take any offence, but you must certainly answer these questions before anyone with any grounding in physics will take any note of this theory of yours.

My opinion, based on (what I consider) fairly simple physics, is that your transmission would definitely take something not very far off 1.5 hp from that 15, and the car would be propelled with about 13.5 or 14 hp. Therefore, I'd expect that 15 hp to get the car to maybe about 35 mph, rather than be entirely unable to move it. If it was just aerodynamic forces, assuming a top speed from 139 hp of 120 mph (don't be insulted if it's more; it's just an example), 13.5 hp would move it at about 55 mph ((139/13.5)^1/3 * 120 mph). But allowing for rolling resistances as well as drag, it will be much less. I know a Sprint engine with 2 cylinders not firing (so a bit less than half power) gets a TR7 (not mine, honest) up to about 60. Whereas, assuming only aerodynamic drag, and a top speed of 120 mph for full power, half the power should get it to about 95 mph (0.5^1/3 * 120).

That's generally the case with linear systems, i.e. their loss or gain is expressed as a ratio that does not depend on the power level. It's usually expressed in decibels, but can be expressed as a percentage. So a linear system with 3 dB attenuation would lose 50 percent of the power put into it, i.e. half it; and one with gain of 3 dB would add 100 percent, i.e. double it.

As to the reason why I expect the loss to be a smaller fraction of a smaller power: As already said, a car transmission is not going to be an exactly linear because of, e.g., the viscous losses to the fluids, which will vary depending on the speed you drive the system. Generally, those viscous forces increase with the square of the speed (doubling the speed takes four times the force; trebling takes nine times; quadrupling takes sixteen times; and so on); whereas frictional losses go linearly with the force, and are (very largely) independent of speed. So, since it's more than likely a 15 hp engine won't turn the transmission as fast as a 150 hp engine, while the losses due to friction with the 15 hp should be one tenth of those from 150 hp, the viscous losses are liable to be very much less than one tenth. As a result the transmission will lose significantly less than 1.5 hp – though I'd not like to guess at the actual figure. It certainly won't lose anything like 15 hp of that 15 hp – that would just be silly.

As an aside, these viscous losses are further complicated by cavitation: they don't just keep increasing and increasing with increasing speed – at some point the shear stresses are too much for the fluid, and, generally, it forms bubbles: cavitates. It also tends to make a lot of noise at the same time, which is something of a problem with propellers on submarines and keeping them quite. It's also behind what happened in 1894 when Charles Parsons first launched Turbinia and connected a simple propeller directly to the output shaft of a steam turbine, put it in the water and turned it on – a hell of a lot of banging and bugger all propulsion. Still, he was getting 35 knots out of her with three props on each shaft by 97.

Yet further aside, Parsons was still using direct drive when HMS Dreadnought was launched: which postponed the First World War for 8 years, (according to Jackie Fisher) till the Kiel Canal was widened for the Kaiserliche Marine's own Dreadnought type, all big gun, battleships (Nassau, Helgoland, Kaiser, and Koenig classes). In 1908 Jackie Fisher predicted the start date for WWI to within 2 months (October 1914), based on when the widened Kiel Canal was to open.

The point being that, sometimes, theoretical prediction can have practical importance – the very existence of "French's contemptible little Army" (the BEF) at the First Battle of Mons was vital to the defence of Paris against the Schlieffen Plan, despite its being totally outclassed by von Kluck's German 1st Army. And had Paris fallen, at the very least, Alsatians wouldn't speak French, and I wouldn't have to eat Quiche Lorrain – unless it was sausage in a flan. So, maybe not all bad then.

And it just has to be said that General Alexander Heinrich Rudolph von Kluck was no chicken.

Graham