Looking for software based strain guage indicator/ conditoner. Anyone know of one?

[SPL Tech]

I have a 10klbs S shaped four wire load cell that I am going to be using to study dynamic loading on nylon ropes. Basically I am going to drop weights on various ropes and I need to record the maximum force the load cell experienced. I wanted to get a Daytronic 4077, it seems like it would be perfect for my application, however it requires a 120 VAC power source. Being that these tests will be conducted outside, away from buildings and power sources, I am trying to find a software application that I can use on my laptop. The program would have to have the ability to read a minimal of 1,000 samples per second as this is a dynamic loading scenario which will have a peak value lasting but a few milliseconds. Anyone have any ideas?

 

[keithkyll]

Inside the load cell is a Wheatstone bridge. Four resistors. It does not produce a voltage. The bridge requires a tightly regulated power source. Additionally, the output is only a few millivolts per volt of the source. If you give it 10 volts, the maximum output would be 30 millivolts. This is impossible without a conditioner.

 

[SPL Tech]

Inside the load cell is a Wheatstone bridge. Four resistors. It does not produce a voltage. The bridge requires a tightly regulated power source. Additionally, the output is only a few millivolts per volt of the source. If you give it 10 volts, the maximum output would be 30 millivolts. This is impossible without a conditioner.

Ok, so I need a conditioner. Well that brings me to two questions:

1. I can use the Daytronic 4077 in the field if I buy a 120 VAC power inverter and 12V battery, but will that skew the results compared to if I had just plugged it into a house outlet? I can get a true sine wave generator if the output waveform of the inverter matters.

2. Is there a computer program that comes with a conditioner that I can use? I would like to have a graphical representation of what the load cell experiences on my laptop so I can see how changes that I may make will effect the overall loading of the anchor I am measuring. Any bench top type conditioner/ indicator will only give me the maximum and real time loading options, none of them will give me a graph.

 

[OkiePC]

There are many brands of load cell signal conditioners that should work. Most that I have used are aimed at generating a signal that can be read by a relatively slow PLC analog input so I think it would be wise to verify that the speed of any signal converter chosen meets your specification for detecting the tips of the peaks.

A load cell amplifier and separate data acqusition unit for the laptop might be less expensive than a data acq. board made specifically for load cells.

This one jumps out at me as one that should exceed the needs you have shared so far:

http://www.omega.com/ppt/pptsc.asp?ref=LCIC-WIM-BEN&Nav=pred06

And a cheaper battery powered one (not sure how fast it is):
http://www.omega.com/ppt/pptsc.asp?ref=OM-CP-BRIDGE110&Nav=pred06

 

[CharlesM]

I don't think the signal conditioner is the big problem. I think getting something fast enough to collect the data would be the problem. I used a Daytronic one time before that would read the Mv signal in without a signal conditioner. However I don't know if it will collect the data you need. If it did I think it would need a lap top and a RS232 cable to get the data out. I think if it needs to be portable I would look at a portable o-scope or a calibration unit. Some of the Fluke Calibration units have power output and data logging.

http://www.fluke.com/Fluke/usen/Pro...function-Calibrators/Fluke-741B.htm?PID=56122

 

[SPL Tech]

There are many brands of load cell signal conditioners that should work. Most that I have used are aimed at generating a signal that can be read by a relatively slow PLC analog input so I think it would be wise to verify that the speed of any signal converter chosen meets your specification for detecting the tips of the peaks.

When you say speed, what speed are you refering to? The interface speed of the communication bus (most likely USB)? Or the scan rate of the interface that reads the output of the load cell? I agree the scan rate of the indicator needs to be high, at least 500 scans per second.

There are many brands of load cell signal conditioners that should work.

I am mainly looking for a conditioner/ indicator combo unit. I dident even know you can buy separate conditioners, every strain gauge indicator I have been looking at includes a conditioner.

This one jumps out at me as one that should exceed the needs you have shared so far:

http://www.omega.com/ppt/pptsc.asp?ref=LCIC-WIM-BEN&Nav=pred06

And a cheaper battery powered one (not sure how fast it is):
http://www.omega.com/ppt/pptsc.asp?ref=OM-CP-BRIDGE110&Nav=pred06

The first one is nice, but I dont need a scan rate that fast, 15,000 scans a second would be more for something like capturing the force of a bullet striking a steel plate, my application will see forces that, although are fast, are in the milliseconds, not nanoseconds. So a scan rate of 500 Hz. would be fine.

Does anyone know anything about this unit?:http://swann-associates.com/load-cell-amplifiers/strain-gauge-to-usb-convertor-dscusb/

That seems like it would be perfect for my application, its powered by my laptop.

 

[SPL Tech]

I don't think the signal conditioner is the big problem. I think getting something fast enough to collect the data would be the problem. I used a Daytronic one time before that would read the Mv signal in without a signal conditioner. However I don't know if it will collect the data you need. If it did I think it would need a lap top and a RS232 cable to get the data out. I think if it needs to be portable I would look at a portable o-scope or a calibration unit. Some of the Fluke Calibration units have power output and data logging.

http://www.fluke.com/Fluke/usen/Pro...function-Calibrators/Fluke-741B.htm?PID=56122

By "fast enough" do you mean the scan rate of the indicator? And using a laptop is fine, I want a unit that can give me a graphical representation of the load the load cell experienced, so a laptop will be required anyway.

 

[James Mcquade]

Have you considered the shock effect on the load cell and the damage it will create on the load cell?

A sudden load on a load cell will create a shock - an initial massive load condition of 6 to 10 times the actual size of the weight actually being placed on the load cell. this effect will cause the internals wheat stone bridge to fail prematurely.

we always had to over sizt the load cells at a former oem location i worked for.

regards,
james

 

[CharlesM]

Most of the signal conditioners that I have worked with are rated in % output over time. A typical setup would be 90% output in 100ms which would be to slow for your application. So if you did need to convert the signal from a mV signal to 0-10VDC you would want to make sure the signal conditioner was fast enough to get all the changes.

I think USB and RS232 will be to slow to gather the data directly. So your device will need to buffer the data in a table that you can upload to the lap top. It looks like the link you posted will do this. The only bad thing I see with what you posted is it only has +5V to power the bridge. This will give you half the resolution you would get with 10Volts. It might be fine for your application.

 

[SPL Tech]

Have you considered the shock effect on the load cell and the damage it will create on the load cell?

Of course, the peak loading is specifically what I am trying to measure. But we are talking about a 150lb weight being dropped onto a 10,000 lb load cell with a dynamic nylon rope between. The absolute maximum load the cell will experience under the dynamic conditions I will subject it to is 5000 lbs. as the rope has a breaking strength of a bit under 5k.

But in order for the load cell to sustain any damage, wouldent I have to exceed its rated SWL of 15,000 lbs?

 

[SPL Tech]

Most of the signal conditioners that I have worked with are rated in % output over time. A typical setup would be 90% output in 100ms which would be to slow for your application. So if you did need to convert the signal from a mV signal to 0-10VDC you would want to make sure the signal conditioner was fast enough to get all the changes.

I think USB and RS232 will be to slow to gather the data directly. So your device will need to buffer the data in a table that you can upload to the lap top. It looks like the link you posted will do this. The only bad thing I see with what you posted is it only has +5V to power the bridge. This will give you half the resolution you would get with 10Volts. It might be fine for your application.

What do you mean convert mV to 0-10VDC? As I understand it, the conditioner puts out 5VDC, the 5VDC gets reduced through the load cell at a rate of 3mV/V and the output of the load cell that goes to the indicator is measure in mV. So if the input is 5VDC, the maximum output would be 15mV.

What do you mean when you say "too slow". Are you talking about the interface between the computer and the conditioner (USB) or are you talking about the resolution/ scan frequency of the receiver (measured in Hz. or scans per second)?

 

[CharlesM]

What do you mean convert mV to 0-10VDC? As I understand it, the conditioner puts out 5VDC, the 5VDC gets reduced through the load cell at a rate of 3mV/V and the output of the load cell that goes to the indicator is measure in mV. So if the input is 5VDC, the maximum output would be 15mV.

Most would use 10VDC to get 30mV out of the load cell. This will increase your resolution.

What do you mean when you say "too slow". Are you talking about the interface between the computer and the conditioner (USB) or are you talking about the resolution/ scan frequency of the receiver (measured in Hz. or scans per second)?

Not all units buffer the input into a table. They will only give you one reading per update of the com interface. I've had problems when the device would update at 500htz but the data was only updating on the lap top 10 times a second. 10 times a second is fine if you are getting 50 values every update but if you only get one you could be in trouble.

 

[califflash]

you might check out a battery operated scale head with a serial output. I have some of these systems in my plant but I am not too familiar with their inner workings as the systems they are coupled with are proprietary. GSE and Gainco both make scale heads that would suit the application and you should be able to pull the values via rs232 and modbus.

 

[rdrast]

There is a 9-36VDC version of this Rice Lake 920i, but somehow I think that would be an expensive solution for you.

 

[danw]

Vendors like Natiional Instrument have both PC hardware and software for 1000Hz sampling. If you need the max load, then your probably only need a 'snapshot' of X samples, not continuous data collection at 1,000 points/second, so a 5K buffer should cover data collection.

Consider a trigger to initiate data collection, if event timing is of any interest, otherwise triggering tends to be human initiated.

There's no commercial Modbus (client) master I'm aware of can handle 1000Hz polling. And Modbus typically doesn't fetch files. Be wary of Modbus as a protocol for this type of data acquisition.