21 replies to this topic

[redline]

Hi All,

Some hypothetical questions -

1. What would be the effect on engine stresses of applying brake pressure with the engine held at WOT?

2. Assume an engine is held at peak power rpm (for argument sake 8000rpm) and a braking load is applied to slow it down to 6000rpm whilst maintaining WOT. If this braking load is then instantenaously released - will the subsequent engine acceleration different from 'normal' acceleration between 6000rpm and 8000rpm under WOT?

3. Would such operation qualify as "torque control" or braking? To my mind braking only occurs if the retardation force exceeds the driving torque... In effect, this type of operation modulates the driving torque transmitied to the wheels, and hence could be considered torque control.

Thanks for your input!

[Engineguy]

1. No different than any other load... loading the engine against a dyno, loading the engine against aerodynamic resistance, against the inertial resistance to acceleration, etc.

2. No

3. Ask Max.

[redline]

Hi again...

Assuming i'd like to apply a disk break to the engine output shaft, how would I size it?
Lets take for example an engine developing 130 lbs-ft torque and 120bhp at 5000rpm and 115 lbs-ft torque and 65bhp at 3000rpm (based on real figures of a 2.0L four cylinder engine)...

I'm assuming that the energy i'd have to dissipate to slow down the engine from 5000rpm to 3000rpm wouldn't actually be too great, as the delta_torque is ~15 lbs-ft. As a first guess I was thinking that maybe a motorcycle disc would do in terms of braking torque. The doubt I'm having regards heat build up - say I need to hold at 3000rpm for a few seconds, I'd expect the disc to get pretty damned hot, and more importantly, would it be able to cool down sufficiently quickly for repeated application...

Anyone care to comment?

Cheers,

redline

[JwS]

I would start by figureing out the total energy you have to absorb, basically the horsepower at that rpm times time. Then look at the mass of the disc and the thermal properties of the material it is made from to determine how much the temp. will increase, this will give you an idea if you are in the ball park. Alot of assumptions here, and the heat dissipation is ignored, along with the portion of the heat that will go into the pads, but is kind of a worst case model. For more advanced modeling I would get one of the brake design books out there.
Just taking a guess, I would think a motorcycle disc is going to be a little small, and they are also alot more expensive than something like an automotive disc.
Oh! also, be aware that they are not intended to spin at 3-5000 rpm, you may be risking an explosion there.
JwS

[redline]

Hmmm.... so you think I need to be looking at the total energy (Hp x time) or the difference in energy between the two operating points? If you are correct in saying that i need to base the worst case model on the maximum power, then a motorcycle disc will be small.

About the centrifugal forces, I've done a quick calculation of hoop + radial stress for an automotive sized disc & hub, and we should be well below bursting speed (depending on material clearly...). Really should FEA it though... when i find time!

Can you recomend a good brake design book? It doesn't seem to be a subject which is well covered!

Cheers,

redline

[JwS]

Well, if you want to hold it at 3000 rpm for a period of time you will have to absorb all of the power it is generating at that point, otherwise it will speed up.
As far as books, I was thinking of Puhn's book (spelling?) but now I am not sure how much info was really in there. I don't think it dealt with heat transfer problems that you might be interested in for sustained operation.
I think the first thing I would look at would be the biggest truck disc I could find, after all, there is nothing like overkill! and some forced air cooling to boot. I would still be a little scared of the RPM, especially with a vented rotor, many of them are crude castings, and cracks are not uncommon after a few heating cycles. That is alot of energy stored up.
JwS

[redline]

I see your point... but i don't think I've explained myself correctly! The engine is under load, apart from that applied for the brake, which is why i maintain that you do not need to to brake all the power at that point. The brake supplies an additional load.

Example:- Accelerating in a situation where power to wheels exceeds traction, the engine is at 5000rpm, and i want to slow it down to 3000rpm with the brake. The load applied by the brake is additional to the inertial resistance to acceleration, aero resistance etc... Essentially, the brake would modulate the torque transmitted to the tyres... a form of traction control if you like.

I take your point about the cast rotors - i was actually thinking of machining a disc out of steel. Not sure about the coefficient of friction, but I assume it will be sufficient for a crude test.

I appreciate any further comments/ideas you may have!

redline

[JwS]

Ahhh, I missed that, so the power absorbed would be reduced by the amount being used to drive.
And the disc would be on the driveshaft, which reduces speed alot.....
Is there any way you could use the brakes on the car to do the same thing? Rally cars do this, if you watch you can see the brakes glowing as they drive out of a corner. Of course they are using some sort of computer controlled system. In the old days Eric Carlsson was known for driving the Saab 93 rally car by keeping the gas flat on the floor and left foot braking to control the power. It is also done on motorcycles sometimes, modulating the throttle is slower, and there is a benefit to keeping the throttles open on a CV carburator, it takes a second for the slides to settle into the optimum position.
It is difficult to machine a decent disc, steel will tend to warp very easily due to internal stresses, I would try to find something from a car or whatever.
JwS

[redline]

Hi JwS,

Thanks again for your valuable insights... much appreciated. I had no idea that rally cars used such a system, which i assume is quite complex. Are they using it as a stability control, or to impart a turning momemt to the car? I'm curious how it modulate the brakes, rather than just apply them. What variable do they use in the control loop - force feedback fom the actuator? Deceleration of the wheel in question? Wheel Slip?

You have to think outside the box for the application I'm thinking about, as there are no wheels. Drive is direct from the engine to the .... uhhh... 'thrust generating device', so i am looking at braking the driveline at real engine speed. This may be an advantage in some respect, because there is no torque multiplication by gearbox or final drive. My main concern though is heat build up, given that there would only be 40-60second periods between applications of the brakes.

Cheers,

redline

PS- I assume Carlsson was keeping his foot flat on the floor to keep the turbo spooled up and hence avoid lag, correct?

[JwS]

I am not sure of the details of the rally car setup, I believe it is used primarily as traction control, not so much as yaw control, and to keep the turbo spooled too, I am hoping someone here has more detail knowlege. Carlsson's Saab 93 was way before turbos, and only had about 50hp, I think the technique is still used by lower hp FWD cars in rallies.
As far as location of the brake, whatever the speed it is turning the same amount of enery is being absorbed, I think you are correct to be concerned about heat build up, think about slipping a clutch for 3-4 seconds at 3-5K rpm, and they have a pretty good size thermal mass (although poor heat dissipation)
JwS

[alexbiker]

Left foot braking into and out of corners is for different reasons in different cars.

In the FWD car, a driver left foot brakes whilst turning in on the power to enhance turn in. This works by (1) causing a forward weight transfer and increasing front end grip, but mostly by (2) effectively forming a controllable handbrake - engine power vs brake force controls the power at the front wheels, whilst the unpowered rear wheels are braking - the whole car pivots around the front and tips into the corner very nicely. Moving to the exit, understeer can be controlled as power is applied in the same way.

In a RWD car, left foot braking can be used for weight transfer, and to control inside-rear spin on the corner exit, thereby acting as a limited-slip diff.

Hope this helps,

Alex

[redline]

"...engine power vs brake force controls the power at the front wheels..."

That is very interesting... now, why not control the power just through the throttle? What advantage is there to modulating power to the wheels through the brakes? I imagine that with electronic control of throttles power can be made to vary almost linearly with throttle opening, so you'd have a much finer control than with the brakes...

The only advantage I could see, which JwS mentioned, could be an improvement in transient response as the throttle bodies remain wide open, but i suspect this would be minimal.

Cheers,

redline

[skiericski]

Originally posted by redline

My main concern though is heat build up, given that there would only be 40-60second periods between applications of the brakes.

At the kind of speeds you're talking about, and with that much time between cycles, a vented disk should have no problem cooling to within an acceptable limit between applications, just make sure it has room to breathe (and a good solid shrapnel shield between it and the driver!) Definately do your homework on those hoop stresses, like JwS said before, cast parts always have the potential for voids and cracks(stress concentrations), so be very conservative with your safety factors!

Can you share details of the application, or is it top secret?

[McGuire]

Originally posted by skiericski
Can you share details of the application, or is it top secret?

Yeah, that's what I was wondering. What does this thing do? Time travel? Male enhancement? I could see how one could make a lot of espresso with it, but beyond that I am at a loss.

[redline]

If it could help with male enhancement, or even produce a decent espresso, I'd be well on the way to making the mega$$$ No, it has more to do with time travel... or shortening the time it takes to get around a corner.

I said you have to think outside the box, because the application is different, in that there are no wheels and tyres to transmit the propulsive force. Thrust is produced via propellers - before you launch into any sarcastic comments, it is not for a land based application! The rest of the driveline is similar to normal automotive applications.

The basic idea is very simple - to alter the propulsive force between left and right sides (whether its a wheel, propeller, jet drive, ionic propulsor etc...). McLaren proved the concept with their 3rd brake system to selectivley brake the inside rear wheel on corners, causing a turning moment about the C.G.

Hope this clarifies things a little - didn't intend to be misterious!!

redline

[redline]

OK, I've done some brief calculations (making many assumptions!) so i'm not too sure about the validity...

For the torque and rotational speeds I'm considering, the heat into (W/sec) the disk due to braking is about 200x greater than the heat dissipated by convective heat transfer to air! I did not take into account forced convection of a ventilated disk.

Even for a low operational duty cycle (~10% is realistic for the application) i think there are progressive heat build up over repeated applications is going to be a problem

redline

[Sir Winston]

The idea of making more power than you need and then trying to remove it is ill-conceived. If you know you need less, you can make less. You have plenty of control over the amout of power you make via spark and throttle control. Spark has loads of bandwidth and throttle has loads of authority.

If you are looking to control left-right torque split then that's exactly what a controllable differential is for.

[blkirk]

Originally posted by redline
OK, I've done some brief calculations (making many assumptions!) so i'm not too sure about the validity...

For the torque and rotational speeds I'm considering, the heat into (W/sec) the disk due to braking is about 200x greater than the heat dissipated by convective heat transfer to air! I did not take into account forced convection of a ventilated disk.

Even for a low operational duty cycle (~10% is realistic for the application) i think there are progressive heat build up over repeated applications is going to be a problem

redline

I must have read this post at least five times, and I just now realized that you ignored forced convection. That will likely be the difference right there. Forced convection is vastly more effective at moving heat than free convection. It is definitely worth having a look at. I'd try to give you a ballpark estimate of how much better it would be in this application, but it's been a good 10 years since I've done any convection calculations. It would take me far too long to get back up to speed before I would trust any answers I came up with.

[skiericski]

He's right, the convection coefficient is the key in this type of problem, and with the insane surface speeds you're gonna be looking at, you can't model it as a still air problem. Ever seen one of those chassis mounted cameras pointed at a nascar's brakes? you can tell when the hit the corner because the rotor instantly turns glowing red, then when they get off the brakes, it almost instantly turns back. Pretty crazy. I assume you included the surface area inside the vents in your calcs? What was the total surface area?

[Greg Locock]

Try SAE paper 200301272 "Measured and predicted temperatures of automotive brakes under heavy or continuous braking"

[golfball]

Forgive my ignorance, but wot's WOT?

[desmo]

WOT = Wide Open Throttle, sometimes also acronymised to "WFO".