Funny this thread should come up on the front page again.... Not to take anything away from the OP, but I just finished building my own pull testing rig and was wondering what to do with it.

Tsuli- What kind of equipment do you have for testing / what kind of tests do you have in mind? I did mine on the cheap for a project I'm working on, so only a 5" stroke x 10 ton hydraulic pullback ram for static testing. But rather than using rated cord as a fuse to ballpark breaking strengths I got my hands on a 10,000 pound s-beam load cell and a Daytronic 4077 strain gage indicator, which should allow me to measure both static and dynamic forces up to ~44kN. The 4077 has an actual analog channel, so capturing true peak load values in dynamic tests shouldn't be a problem (unlike with most other indicators where you're limited by the scan frequency). But I only dabble in these sorts of things, so I'm curious what your take on this sort of this is since you probably know much more about it than I do. At the moment I've just been playing with pull testing cams on the test jig, but if I can figure an easy, packable AC power source for the 4077 (couldn't find the DC version cheap) packing it to the crag for real world testing of most anything is certainly possible. Feel free to PM me and we'll compare notes.

-a.

...2500 PSI working pressure with a 1-1/2" shaft works out to ~20kN. Did I mention it has 36" of stroke?...

Aric,

I have pondered the issue of dynamic load testing a little(and I even built a dynamic load test meter and gage for Metolius), and my conclusion is that conventional strain gages are probably not reliable for accurate dynamic load testing. Capcitance gages based on a high frequency (~100 Mhz) carrier should be used instead.

I assume that the testing jig is small enough that wave transit times can be safely neglected. I also assume that the manufacturers of quality strain gages use materials not subject to significant piezoelectric effects(electric field coupling to mechanical strain) and that this dynamic problem can be safely ignored. In any event, it is likely that the RC time constants of such effects would be sub microsecond and therefore not of importance to millisecond measurements that we are interested in.

The problem I am concerned with has to do with the temperature change of materials on expansion. The strain gage block is in thermodynamic equilibrium between the applied mechanical stresses and the internal thermal pressure of the gage block's lattice. Due to the expansion or contraction of the gage block, mechanical work is done on the thermal pressure. Therefore, the temperature changes. Now, the temperature change is relatively slight, but the electronic strain gage is measuring a similarly slight change in resistance using a bridge circuit to effectively amplify the slight resistance change. A well designed bridge circuit is designed to be temperature compensated so as to cancel the effect of changing temperature. Unfortunately, this temperature compensation is only going to work to the extent that the entire gage bridge circuit changes in overall temperature. A gradient in temperature across the gage will render the compensation useless.

Now, when you consider that the gage material is different than the strain block and that the stresses in the block have some kind of complicated shape dependence, we see that the sudden application of strain will set in motion rapid fluxes of heat in the vicinity of the gage itself. The thermal equilibration time for a plastic film against aluminum is probably in the low millisecond to high microsecond range. Meanwhile the thermal equilibration in an aluminum gage block probably has many different decay constants for the different excited thermal modes varying from hundredths of a second to perhaps a second or so. Therefore, during these fast intervals, unless you know and have carefully studied the design of the gage block and the internal stresses and thermal fluxes, I do not think you can count on thermal compensation to cancel the (induced) temperature effect errors. As I mentioned above, since the termperature changes are of similar fractional scale as the resistance changes(both being proportional to V/V0) it appears to me that dynamic measurements are subject to a potentially first order error in scale factor - and therefore of uncertain accuracy. From a practical perspective, an error of even +/-20% would render measurements against specification standards(like biner strengths) of relatively little value.

Reviving an old thread, again…

I read through this thread and unfortunately a lot of the links are to EBay auctions that have ended. I’m looking to build a cheap tester that I can compare different designs with. I’m not too concerned with the accuracy that Aric’s tester gets but I would like a good idea of what each item will hold. I think you covered the jig and puller well enough in this thread (besides I can figure that part out for myself) but the load testers seemed vague without the links working. It seems to me that there are three or four different types (what I gathered from this thread) strain gage indicator, a load cell, a dynamometer, and digital crane scale. When I do searches for these things it comes up with all kinds of stuff. To help narrow the search down is there different key words you use or do you just fish through each time. Also what is the minimum poundage you would suggest? And just to see how much info I can pump out of you guys, what options would be necessary? I imagine something that tells the peak holding power but is there anything else?


Randall

Can someone familiar with this thread offer me a little advice on a reasonable setup to aim for?

I am a high school teacher, and I would like to do some dynamic load testing of different systems with students. For example, I would like to hang a load cell from the ceiling, and drop weights from the cell. I would like students to examine the effect of dropping distance on the peak force, and ideally our setup would generate so many force readings per second, so we could examine how the energy is absorbed by the system. We do NOT want to test to failure, so we don't need a high-capacity setup. Dropping a 5# weight from various heights, and examining the ratio of forces generated to the weight dropped, would do what we need.

This kind of setup would also be useful for some static testing, such as examining how an equalized anchor shares a load among individual anchors.

I have a background in physics, but I am getting lost in all the transducers, s-beam and canister load cells, and readouts I have been looking at. Can someone suggest a simple load cell to look at, and suggest what else I need to capture the data that the load cell would output?

Thank you.

Eric

Can someone familiar with this thread offer me a little advice on a reasonable setup to aim for?

I am a high school teacher, and I would like to do some dynamic load testing of different systems with students. For example, I would like to hang a load cell from the ceiling, and drop weights from the cell. I would like students to examine the effect of dropping distance on the peak force, and ideally our setup would generate so many force readings per second, so we could examine how the energy is absorbed by the system. We do NOT want to test to failure, so we don't need a high-capacity setup. Dropping a 5# weight from various heights, and examining the ratio of forces generated to the weight dropped, would do what we need.

This kind of setup would also be useful for some static testing, such as examining how an equalized anchor shares a load among individual anchors.

I have a background in physics, but I am getting lost in all the transducers, s-beam and canister load cells, and readouts I have been looking at. Can someone suggest a simple load cell to look at, and suggest what else I need to capture the data that the load cell would output?

Thank you.

Eric