Looking for an inexpensive force gauge

Sort of wrong. Try googling 'force gage,' then come back.

hth, Fred Klingener

Dear jason cooper:

I'm a little confused:

- drop test, &

- breaking strength, &

- fish scale don't seem to go together. I can understand that you might pull up on the fish scale until something breaks... but drop test?

David A. Smith

: I'm a little confused: : - drop test, & : - breaking strength, & : - fish scale : don't seem to go together. I can understand that you might pull up on the : fish scale until something breaks... but drop test?

The fish scale is mounted. I had originally applied force manually at what I felt was a sufficiently constant rate for my purposes, but decided that might not be true after all. I have been applying the force, instead, by attaching one end of the test piece to the scale, and one end to a mass which is dropped from a small height. It works well, but the scale begins to fail after it has been used ~1000 times.

In response to the previous reply... Google was my first choice and, in fact, was where I got the term "force gauge". All of what I've found, however, runs over USD$600. Since I know very little of the field, I was hoping somebody here might be thinking "Oh, he's looking for an 'X' unit...".

Thanks again.

----------------------------------------------------------------- . . . Except when they don't, Because sometimes they won't. - Dr. Seuss

----------------------------------------------------------------- Jason Cooper snipped-for-privacy@acs.ucalgary.ca

Dear jason cooper:

So you are testing the breaking or separation strength of the "thing" that connects the scale to the mass, correct?

I'd just make a simple *strong* hook, place a strain gage on the hook at the closed side, 3:00 position (assuming opening at 9:00-10:00), and monitor the strain gage with a peak detector or just continuously (while a button is held down say). You can calibrate the strain gage/hook system with calibration masses. No worries about losing zero that way, but you can recalibrate the system each day, if you like. Maybe this is already available somewhere...

David A. Smith

Mechanical force gages typically use dial indicators to measure the displacement of a sturdy spring; the indicators are delicate, so those are out.

Load cells that are branded 'legal for trade' are required to survive standardized impact loads. Omega (omega.com) has a fair selection. You can stay within your budget if you homebrew the electronics or have a peak- reading high resolution voltmeter (you still need to provide 'excitation' ; ask Omega for help.). I wouldn't use a 20 lb rated load cell to measure 20 lb loads; I'd go for a 100 lb rating, which will require a more expensive voltmeter but provide a bigger factor of ignorance.

On my budget, I'd use a guided plunger compressing a low-rate spring, and measure the peak deflection by having it compress a column of modeling clay, or scratch a polished copper coupon, or leave a pencil mark on a piece of paper.

But before you invest any money, please help David and the rest of us get un- confused. It sounds like you're dropping something and catching it with the scale, in which case the scale kinetics get involved in the measurement. This is not a good thing.

-Mike-

Dear Mike Halloran:

...

He'd indicated ~3000 cycles to fail the scale. If he truly has anything on this order (or higher), a DAQ system is *required* for proper accounting/capture of the data. I'm afraid the simple mechanical accounting is going to have to be bypassed, and he'll have to go for a wired solution.

Good recall on Omega. I often forget what a resource they can be.

David A. Smith

Thanks for the tips so far. To unconfuse...

The parachute used in BASE jumping (ie, jumping from fixed objects) uses what is known as a "tailgate" to prevent a particular type of malfunction. The tailgate is a simple device

-- a length of 900 lb dacron line is folded over the rearmost 10 lines of the canopy, and an elastic wrapped around the free ends to hold it in place. The idea is that, as the canopy opens, the tailgate delays the opening of the tail until the bulk of the canopy is pressurized.

Two weeks ago I had what is known as a tailgate hangup -- ie, the thing never released. I walked away from it. The effect of each parameter in the closing of the tailgate (what kind of elastic, how it is wrapped, whether it is knotted at one end) has consequently become quite important to me :). I have a background in the physical sciences, so my first response has been to test the thing on a bench under controlled conditions.

What I've rigged is a pair of short lengths of 900 lb dacron line. Each has a loop at one end, so that it can be fixed to something else, with the free ends of the pair wrapped with an elastic in each test as usual. One end is fixed to the scale (which is itself fixed) while the other is attached to a 5 lb weight. In each test, the weight is brought up to a set height of perhaps 8 inches and released. The tailgate always releases

-- the goal of the test is to quantify *where* it releases, relative to other configurations. I'm doing tests of 100 each of each closing method to get the statistics to an acceptable level.

Kinetics is important. The system is not, in practice, brought to failure slowly, and it shouldn't be in the test either. This is why I've chosen to load it as I have. This introduces the questions of (i) whether I've got the speed right, and (ii) how the scale kinetics plays into this, but the most important thing is that the results can be compared amongst themselves in a meaningful way, and I think that's achieved here.

Any input, of course, would be greatly appreciated -- this is not my field of study.

----------------------------------------------------------------- . . . Except when they don't, Because sometimes they won't. - Dr. Seuss

----------------------------------------------------------------- Jason Cooper snipped-for-privacy@acs.ucalgary.ca

Dear jason cooper:

So the scale is hung from the "house", the dacron is hung from the scale, and the weight (on the other end of the dacron) is dropped. Correct? house | v scale | v rope | v weight | v Earth

I'm impressed with the thoughtfulness of your setup. Of course, your life is in the balance...

Mr Halloran had a good suggestion:

you'd need the electronics to convert the signal to a voltage.

And there are op-amps that can be configured to provide a peak-hold function for the produced voltage. (No link for this one, some Fluke DVM's have this feature built in.)

And Mr. Halloran is also correct in his insistence on a 100 lb rating. You may have only a 5 pound weight, but the impulse of bringing it to a stop could be much higher. The Dacron rope will provide *some* dampening.

Also, the surface characterisitcs of the rope will be altered by repeated cycling. I would suggest that new rope sections be used periodically, and substituted in the *middle* of a run, so that possible effects will be highlighted.

Now it is time to pay the piper. How many jumps are performed on a single "line" before it is replaced? Just some enlightenment sought...

David A. Smith

You have the basic set-up very nearly right. I'm not sure if you omitted this only for clarity, but the "rope" part is in fact two pieces of rope bound together with an elastic and one end. The free ends go where the free ends of a rope would go in the setup you showed.

We've been re-using the elastics, but watching very closely for problems from cycling (as you noted). Interestingly, that doesn't seem to be a problem over a ~100 drop test run.

: Now it is time to pay the piper. How many jumps are performed : on a single "line" before it is replaced? Just some : enlightenment sought...

The line itself in a tailgate is replaced only when it's lost (every 50 jumps, maybe). The elastics are replaced every few jumps. The lines on the canopy are generally replaced when the canopy is. While a skydiving canopy can be used as many as

(1) When one jumps a skydiving canopy, one jumps with a reserve parachute. If you were wrong about how many jumps your canopy could take, it'd mean a reserve ride. No big deal. BASE rigs don't generally include a reserve, since there generally wouldn't be time to use one anyway. You're more careful when you don't have a second chance :). (2) The opening of a skydiving canopy is controlled by a reefing system which results in generally slow, comfortable deployments. When one jumps from ~200 feet, though, the most comfortable opening is generally the fastest one (~15 g's can be reached on a hard-ish BASE opening, as I understand it), so the reefing system is omitted. Wear and tear on the canopy fabric increases accordingly.

Thanks again for the tips. I may be back with questions about the details :).

----------------------------------------------------------------- . . . Except when they don't, Because sometimes they won't. - Dr. Seuss

----------------------------------------------------------------- Jason Cooper snipped-for-privacy@acs.ucalgary.ca

I just want to see if I've got this right... A load cell takes in some voltage and puts out some other voltage, the latter depending on how much force is on the cell. The setup you're recommending is something like this:

cell -> op-amp -> meter

so that I can read a maximum that only occurs briefly on a meter which samples (as the ones you mentioned do) every 1/3 second...

If that's correct, and if I've already got an ADC on hand, could I rig the following:

cell -> ADC -> computer

and put in a little time calibrating to obtain the same result?

----------------------------------------------------------------- . . . Except when they don't, Because sometimes they won't. - Dr. Seuss

----------------------------------------------------------------- Jason Cooper snipped-for-privacy@acs.ucalgary.ca

Dear jason cooper:

Absolutely. I just saw that I had "spent" your budget, so I was looking to go cheap (which is usually a mistake).

The strain gage circuit is a form of a "wheatstone bridge", so you'd need a power supply and some form of amplifier. Radio Shack will sell you a little Basic Op Amp guide and the parts to do a quick *analog* cicruit that could output to your ADC. Your ADC may be able to do strain gages directly, which I think a lot of them do.

Just be sure your ADC sample rate is high enough that you capture the spike. Or just analyze the "impulse" waveform for area, shape of rise, shape of falloff, and so on.

You can calibrate with the OpAmp circuit or with the ADC and software. Static weights will make the job easy.

Good luck!

David A. Smith

The advice directed towards your force measurement has been good so far. A stock arrangement for force measurement is a cantilever strip of spring steel with a strain gage bonded on a surface. The arrangement suggested for you was rather beefier.

The area where you need to take account, is the signal magnitude. Aluminum loading is kept under a millistrain, and steel under several millistrain. A bridge strain gage has a gage factor that is a small multiplier on the resistance change in one of its legs - so a force for 1 millistrain might provide a signal that is (say) 2X the change in resistance. If the bridge is fed with 5 volts, the two outputs might provide a signal of 5 volts X 1/1000 strain X 2 gage factor (hypothetical numbers only). You see this would amount to only

As others mentioned, you can rig a bridge DC supply, and an op amp to an ADC even using half-remembered data (such as mine) and still get satisfactory results by calibrating with a known weight.

Brian W

One other question... I found this:

----------------------------------------------------------------- . . . Except when they don't, Because sometimes they won't. - Dr. Seuss

----------------------------------------------------------------- Jason Cooper snipped-for-privacy@acs.ucalgary.ca

Dear jason cooper:

I am unfamiliar with the specific configuration. Find out before you buy it how they intend you to use it, because thier specification sheet is unclear. Find out how the force is supposed to be applied in relation to the component geometry. Your choices are:

I don't know how to read these types of sensors just by looking at them... maybe one of the other posters can.

David A. Smith

This is a precision, rugged, waterproof load cell, excite at 10 volt, sensitivity 3mV/V full scale. As far as I know, it is intended for push pull loading, it is resistant to side loads....

Brian W

My company makes a similar style load cell. You are correct that it is for tension or compression loading. The specs might be misleading. It states it will handle up to 50% side load. That means a 50% side load won't damage the cell. However, it will significantly alter the cells output producing an error. It is important that the side load component is reduced to a minimum. Use of rod end bearings, shperical washers etc. can help.