weigth transfer measurement

[solex]

hi ..... am working on the braking system.
i wanted to know 1.what sensor could i use to get real time data of weigth being transfered to the front when brakes wud be applied,wud mechanical strain gauges provide accurate reading?.2.we are planning to buy an IR temperature gun,wanted to know which one to buy from whom. we saw a few of them from RAYTECH but those who had bougth faced a problem that the instrument could not pick up temperatures from a shiny surfaces. we are wanting to read out brake disc,caliper,uprigth temperatures.please suggest

[solex]

hi ..... am working on the braking system.
i wanted to know 1.what sensor could i use to get real time data of weigth being transfered to the front when brakes wud be applied,wud mechanical strain gauges provide accurate reading?.2.we are planning to buy an IR temperature gun,wanted to know which one to buy from whom. we saw a few of them from RAYTECH but those who had bougth faced a problem that the instrument could not pick up temperatures from a shiny surfaces. we are wanting to read out brake disc,caliper,uprigth temperatures.please suggest

[thewoundedsoldier]

Why not use accelerometers, front and rear?

Also, I would recommend on-board IR sensors as brake temps really need to be taken on the fly. It would not be hard to mount in the wheel package.

Dynamic data always turns out more interesting than static data.

We have used a temp gun and haven't had a problem with reflection. How shiny are your brakes?

[solex]

first thanks soldier.....could you tell me from which company should i buy the IR sensor(want reliable stuff)..... if i mount them on the wheel package then the sensor would be facing the inside of the disc rigth?????
how could i use accelerometers to get the reading of weigth distribution??????
and yes the ideal thing would be to get temperatures taken on the fly..... have you interfaced the IR sensor with your DAQ or does it have a separate display unit?????

[Mehul Botadra]

Its an efficient way to measure the weight transfer using accelerometers. You have to just place them on the connecting line of the front wheels and rear wheels. In dynamic conditions your accelerometers are going to be subjected to intertial forces, which is exactly due to intertial weight transfer. However the readings dont usually tend to change a lot unless you have extremely soft springs which allow a good degree of squat and dive. Best Luck!

[thewoundedsoldier]

Are you more interested in transient temperature of things like the upright, buttons, etc or more interested in the temperature of the actual rotor?

I recently took this topic head on when running some numbers on oil temperatures of a torsen diff with an inboard brake dumping heat into it. If transient temps are your biggest interest, I would say simple thermocouples available from digikey are a good bet. You could also spend a couple extra bucks and get digital sensors (if you don't have or don't want to worry about a DAQ).

As for the weight transfer calculations, your accelerometer reading multiplied by the static load of the wheel will equal the weight on the wheel at that moment. If you are turning, you will need to do the same with the yaw acceleration as well.

We have not interfaced temp data with DAQ quite yet, especially not rotor temperature data--that is going to take a pretty expensive sensor. Hoping to do something like this:

http://i649.photobucket.com/albums/uu217/FormulaHybridSJSU/Data%20Acquisition/TireTemp.jpg

[Pennyman]

Accelerometers could work, but I think the data would be pretty noisy if you're trying to fish dive and squat measurement from it.

Better results would probably be found using potentiometers on the shocks and then calculating the weight transfer based on the displacement changes.

In terms of IR sensors, there are some on the market for this purpose, and they are pricey. If you have a couple of handy EE's on your team you could use infrared thermopiles wired up to instrumentation amplifiers with the gain set to around 1000 (as the output from the IR sensors is on the order of microvolts), then calibrate the output on some known temperatures.

[MalcolmG]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Pennyman:
Better results would probably be found using potentiometers on the shocks and then calculating the weight transfer based on the displacement changes.
</div></BLOCKQUOTE>

I spent a bit of time looking at this a couple of years ago and you've really got to take account of the damping forces too, they're pretty significant in transients. This of course requires having damper dyno curves, which if you plot in Excel you can create a polynomial trendline to characterise them, then use this expression to calculate damping force based on damper velocity.

If you haven't looked at this before, it's quite interesting, I recommend it.

[thewoundedsoldier]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Pennyman:
Better results would probably be found using potentiometers on the shocks and then calculating the weight transfer based on the displacement changes.
</div></BLOCKQUOTE>

Then you are missing out on sidewall deflection, damping force, and any compliance yield in the system. I'm not so convinced that the weight on the spring is the same as the weight the rotor is expected to stop.

That being said, you are probably right that it will give you better 4-corner feel than a couple accelerometers.

[thewoundedsoldier]

I suppose since the interest is in the weight transfer and not the weight on the corner, you're right. I was wrong.

But,...

[rkraft]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by MalcolmG:
I spent a bit of time looking at this a couple of years ago and you've really got to take account of the damping forces too, they're pretty significant in transients. This of course requires having damper dyno curves, which if you plot in Excel you can create a polynomial trendline to characterise them, then use this expression to calculate damping force based on damper velocity.

If you haven't looked at this before, it's quite interesting, I recommend it. </div></BLOCKQUOTE>

While it may be interesting, Malcolm, your approach leaves out so much information that I doubt its validity. Your damper dyno data is most likely taken at one speed and one frequency (or even worse, it is a compilation of peak forces at peak velocities). It is not a good representation of what the damper is actually doing on the track.

The next time you are on the dyno, try doing sweeps at different amplitudes and frequencies. I'm sure you will discover some interesting information that will lead you to different ways of measuring weight transfer.

[thewoundedsoldier]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by rkraft:
While it may be interesting, Malcolm, your approach leaves out so much information that I doubt its validity. Your damper dyno data is most likely taken at one speed and one frequency (or even worse, it is a compilation of peak forces at peak velocities). It is not a good representation of what the damper is actually doing on the track.

The next time you are on the dyno, try doing sweeps at different amplitudes and frequencies. I'm sure you will discover some interesting information that will lead you to different ways of measuring weight transfer. </div></BLOCKQUOTE>

Wasn't his point that damping force is significant when considering weight cornering? I don't understand why that would change at different frequencies or loads. The dampening force should still serve to skew the spring force away from perfectly representing the transfer.

This is getting murky, so I guess I'll bug out now. Oooooh--dampening curves!!

[rkraft]

It's extremely significant. Damping force changes with velocity, acceleration, frequency, and stroke amplitude. Taking a numerical differential of shock displacement and combining that with a pkF-pkV curve does not take most of that into account.

[exFSAE]

Before we all go bananas here...

Question #1 - How accurate does it have to be? Or as a corollary - how inaccurate can it be?

[MalcolmG]

Ryan, I should point out that I didn't expect complete accuracy from this method (I was using it more to get a feel for how significant the damping force was than to get a precise value for weight transfer). I do realise that dampers are displacement sensitive, but sometimes there's a lot to be gained from spending a bit of time getting numbers that might be within 10-20% of the true values and then moving on to something else, unless of course you intend to use it for something where absolute accuracy is important

[rkraft]

exFSAE asked the right question here. But if you're using the data to look at how fast you are transferring weight and its effect on balance, 20% will be tough to get using that method. All of your inelastic and non-suspended weight transfer is ignored, and your estimated damping force will still be a bit off.

[Pennyman]

Sounds like strain gauges on the pushrods might provide time-dependent weight transfer data independent of your damper settings.

Although, setting it up would probably be a pain, and I'm not sure about the time delay on a strain gauge...

[Vittorio]

Don't forget that inelastic weight transfer can't be seen from pushrod strain gauges.
Probably in this case accelerometer data is the most simple and useful source of information.

[thewoundedsoldier]

Oh hell let's just say 1.5g decel. Draw the turning force out of a hat, or something.

[Zac]

Clearly tire force hubs are they way to go.