Brake parasitic drag

[aceofbase]

Does anybody know a method of measuring parasitic drag in the disc brake (pneumatically operated)? It's a kind of parasitic torque that presents in all modern brakes. After applying the brake and when it's released there is still some local contact between the lining and the disc. But after a few rotations of the wheel this torque disappears. I need to find how big is this parasitic torque (it's considered to be pretty small) and is the possibility that this torque can appear not only applying the brake but when driving off-road. Currently I'm thinking of some kind of a sensor that could measure the force between the pad and disc - but I didn't find any appropriate type of sensor... I could try to use the brake dynamometer – but it will be the last thing I would try, cause I want to measure this torque right driving the truck depending on the road conditions.

 

[GregLocock]

Strain gauge the spindle in a locality close to the calliper mounting bolts. calibrate it by locking the brakes and applying a known torque to the wheel. Check for cross talk by taking it for a long drive over you test surface without using the brakes. If you get no significant signal then you are good to go.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[aceofbase]

Yeah, but still it's a problem to get information about the brake torque. E.g. I want to have a laptop in the cab that will show me the brake force (so as to measure a torque) in the real-time, during the driving... Actually most of the time (during the ride) the brake force should be 0 N, but in some cases, as I wrote before, it can exist.

 

[SomeYahoo]

I think that Greg's repsponse would help you here. When you lock the brakes and apply a series of known torques, you can get a calibration between micro-strain and torque. Program this into your data acquisition software on the laptop and you should get real-time torque.

 

[aceofbase]

Hm... I was thinking about it and think that in this case with a strain gauges near the mounting bolts I'll measure the torque not on the brake disc, but the torque on the wheel, cause all the forces from the road will be transmited to that strain gauge. I think maybe to put a strain gauge right on the discs hub (cause it's not like the new type of splined discs, but the disc with a kind of hub on it) and measure torque on it. But still I'm not sure about the wheel forces that can affect such measurment. But still idea with a strain gauges is the best one. Thank you!

 

[MikeHalloran]

Instrumenting the wheel or the brake rotor hub eats up your budget pretty quickly, because you have to condition and transmit the signal from a rotating object to a translating object and provide power for the transducer and the transmitter.

What Greg said was to put strain gages near the BRAKE CALIPER mounting bolts, not the wheel mounting bolts. The bracket that carries the caliper is stressed only by brake torque and by the weight of the caliper as magnified by bumps.



Mike Halloran
Pembroke Pines, FL, USA

 

[aceofbase]

Oohh. Thank you. I confused. But still I don’t know how I will measure the torque: the calliper is connected to the brake carrier by means of two bolts (or even sometimes 4) and where am I supposed to stick the sensor and what force will it measure… Here is the pic of the brake

 

[GregLocock]

The calliper bolts to the spindle via those two bolts. You need to choose a place on the spindle near those mounting points which, ideally, will 'see' braking forces but not other forces from the wheel. To establish where that is needs experiment, experience or analysis.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Fabrico]

Driving off road may cause friction from deflection, but that action may also cause the pads to further retract.

Except for arguably the off-road part, would the subjective numbers tell much more than a simple but carefully controlled coast down test? Or something even simpler like turning the wheel by hand?

 

[aceofbase]

Yeah, turning the wheel with a digital (so as to put different measurements in the memory) torque wrench was the very first idea and I will do so: but just to compare the results with a sensors. The problem is that I can detect that parasitic moment even with the wrench - but it will be just a moment in the first moment after releasing the brakes and I will not know for sure if it will take place anywhere during the ride. And YES – the bumps will also cause retracting and the chance that the pads will have a contact with a disc is very very little. But that’s the thing I should prove: is that really so little that we don’t even have to care about it or it really exist and we have to use some device in the brakes to force the pads move out of the disc.

Now I'm trying to figure out where to put the sensors: I think that's easier to put them on the brake disc. But somewhere in that part that Then I will try the idea to put them on the spindle near mounting points...

 

[swall]

The seals in the wheel bearings would cause some amount of drag when turning the wheel. Don't know how much, but when motorcycle wheels are statically balance by rotating them to find the heavy spot, you need to use a rig that allows the axle to rotate, not the wheel bearings, as seal drag will render the wheel balancing attempt hopeless.

 

[Rob45]

Interesting point, Swall:
How about removing the wheel bearing seals for the duration of this test?
Then measure the rotational inertia of the tire/wheel/hub/rotor assembly (any good mechanical engineer knows how to do that, right? <grin>) and do a spin test with seals installed, giving you a measure of seal drag.
It could even be that seal drag exceeds brake drag under most conditions.

And Ace: I notice you're using air-applied brakes: would this be an Eaton study?
What *IS* the mechanism for retracting the pads when you don't have an hydraulic seal to do the work for you?

 

[BobM3]

Use a full bridge with 4 active elements (2 gages give a positive output with positive torque and adjacent gages in the bridge give negative output with positive torque). You'll probably also need a stress concentration to get the stress high under the gage. Drill a hole in the the part in the load path and locate the gages near the hole. You better spend a lot of time making sure the location you pick isn't subject to off-axis (non-brake torque) loads. I don't know the magnitude of the drag torque but my guess is that the signal will be swamped out by other loads (off-axis, vibration, etc.) Also small strains can be sensitive to thermal strains.

 

[Rob45]

Why not free-float the caliper and put the strain gage(s) on a torque arm?

 

[aceofbase]

2BobM3: You just guessed what I’m doing right now! We calculated that we can not just put the strain gauges – the strain is about 0.5microstrains and nobody knows any type of sensor to measure such a strain: it’s just below the normal error of such a measurement. So the minimum strain that can be detected is about 5 and I found out the new diameter of the brake disk hub and we will chisel it out to put the strain gauges there on the weakened surface. About the temperature: yes – that’s a problem. But we know the temperature distribution thru the brake disk and during the short test it won’t affect the results.
2Rob45: About the brake: this one on the test is made by Knorr-Bremse – this is the biggest brake company in truck industry. They also own Bendix. But I didn’t hear about Eaton’s air brakes. I know that Haldex and Marathon are making air disc brakes as well – but it’s pretty the same mechanism. Retracting mechanism is a spring. But normally it doesn’t help This spring moves out calliper but the pads are still in contact with a disk. They are not pressed but still are making some parasitic torque. During the next movement they are retracted by means of vibrations but still in some case might be pressed to the disc. This right what I’m measuring in this test About the seal drag: I believe it is constant – so will figure out how to find it and then just keep in mind.
About the last offer: in this case I think we won’t be able to measure this brake disc torque during the movement.
Any new ideas ?

 

[Rob45]

Ace:
Thanks for the explanation.

 

[Fabrico]

You could add a small spring to the end of each pad for complete retraction. This is something that would be much more touchy to do with hydraulic disc brakes. It would not take long for mechanics to start throwing the springs away. But you still get to advertise the worlds only drag free disc brakes.

 

[Rob45]

Hydraulic disc brakes don't need return springs; that's what seal rollback is for.

 

[Fabrico]

Interesting. I'd bet a dollar to a donut there are OEM's that don't observe that. And how effective is it on the billions of sliding calipers?

On the mechanical, if the components retract, you would not need to retract the pads unless they are a tight fit for rattling or some other reason. If they are dragging, something else is holding them there.

You could move the air quick release to each chamber and use the puff of air to move the pads back, cool them, and dust them off while they're at it.

 

[patprimmer]

My experience is that if you knock the pads back even a few thou to far (like by strut and stub flex while cornering) you can loose so much pedal that on first application the pedal hits the floor.

The flex in the piston to calliper body seals is normally enough to give suitable clearance, but only pulls the piston, not the pads back.

Single spot callipers make it more difficult as there is no device to centralise the calliper over the disk, so one side may drag.

I have seen pads where the pad clips into the piston and would be pulled back with the piston.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.