3 Wire Load Cell question

[MrSeaton]

Hi,

I'm messing around with some load cells I pulled out of a bathroom scale and I noticed that these load cells have only 3 wires unlike the 4 wire load cells I have been reading about. Does anyone have any information on these types of load cells and maybe what wires are for what?

Thanks
John

 

[Sceadwian]

https://www.societyofrobots.com/sensors_forcetorque.shtml
and
https://en.wikipedia.org/wiki/Load_cell

might help.

 

[MrSeaton]

thank you for the reply

 

[MrSeaton]

It was a good read but unfortunately there wasn't anything on 3 wire load cells.

 

[JimB]

Assuming that this load cell uses a strain gauge, you then need to put the SG into a wheatstone bridge circuit to be able to read the changes due to load.

Have a look here:
https://www.vishay.com/brands/measurements_group/guide/tt/tt612/612c.htm

JimB

 

[j.p.bill]

You have only 3 wires because it's a half bridge. Only 2 strain gauges present. It may be used with two fixed resistors to make a full bridge.

 

[MrSeaton]

So if a wheatstone bridge consits of 4 resistors your saying that the 3 wire load cell takes the place of two of the resistors? And this is because it has two strain guages?

Going with that, How then do i connect the other two resistors and how do I find out what the resistance is of the strain guages? Without taking it appart how can I map out the wires?

I have a fluke 77III, can I use that to tell me what the resistance is and how do I translate the readout to a resistor I can buy at radio shack?

Thanks
John

 

[MrSeaton]

Ok, That was a stupid post... A little reading and I got my answer.

for this explanation I will call the wires wire1 wire2 and wire3
When testing the load cell I noticed that touching wire1 and wire2 gave me a reading and then touching wire2 and wire3 gave me the same reading but touching wire1 and wire3 gave me a reading twice the that of the first two radings.

This gave me an idea. I knew I had some resistors laying around so I found 2 that were the same and I tested each individually and then I twisted two of the ends of the resistors together and measured them from the two untwisted ends and it gave me the same result, twice the resistance. Ok, that makes sense.

Now I just have to figure out this diagram I found on a site and see if I can picture this perticular setup. I'll modify the image and post it here to see if I'm right.

 

[j.p.bill]

The individual rsistances were probably close to 350 ohms. If you use two fixed resistors to complete the bridge, matching the resistors closely is more important than them being exactly 350 ohms. Anything from 360 ohms up to 1000 will probably do fine. Check the Omega Engineering site for info on strain gauges and arrangements of them.

 

[MrSeaton]

Ok, here are the two images

 

[MrSeaton]

Ok, here is the image I found on the internet.

Here is the one that I modified to what I think I will have when I complete the circut with the other two resistors.

Ok, I think I need to make one more image to clerify things a bit and then I think I've got it.

Thanks
john

 

[MrSeaton]

Ok, here is the other image.

Thanks for letting me know that they didn't have to be exact. That was another question I had. you must have read my mind.

Off to go do some more reading and some more testing...

 

[MrSeaton]

Oops, I forgot my question lol. Are the v- v+ right in the pic?

 

[MrSeaton]

OH, and the resistance of the load cell was 748 each and 1498 combined

 

[MrSeaton]

Ok, I found the correct configuration for a half bridge circuit. I modified my image I found to show the correct configuration.

 

[MrSeaton]

I made the circuit and it works. The only thing is that the output is in millivolts. I will have to build an amplifier circuit to be able to bring the voltage up to 2.5 - 5v. I'll post more on that later.

Here are the pics of the design so far.

 

[JimB]

Nice pics, in focus too, makes a nice change from some of the recent pics on here.

To amplify your bridge output, it would be easiest if you use an op-amp, a 741 if you have one, but they are a bit of an old design, a TL072 would probably be better.

If you moved your bridge completion resistors to one end of the board, there would be room to mount the op-amp and associated components.

JimB

 

[MrSeaton]

Thanks for the kind words on the pics.

I'll do some looking around and see if I can find some info on the TL072 and a diagram on building the circuit.

Thanks again for the info.

 

[MrSeaton]

Well, I went to radioshack today and got some parts for my amp. I ended up using LM324 Quad Op Amp and I built a 3 Op Amp Instrumentation Amplifier.

I've got it on the bread board for right now just to make sure I could get it to work. To my surprise it works like a champ. I'll attach some pics of what I've got so far.

My next step is to design a pcb board and maybe try my hand at etching one out. I should be able to make one nice and compact that way.

Anyway here are the pics.

 

[Leftyretro]

Does your instrumentation amp have provisions for zero & span adjustments for calibration trimming purposes? Typical if you increase the gain resistor to more then you normally need then connect a 10 or 20 turn trimmer pot to the output of the amp to ground and take your measurement output from the pot's wiper. Zero adjustment is usually a trimmable voltage reference feeding the input to the amp.

Not sure of the weight range of your cell but several years ago I build a 0-5 gram scale using a very small Omega Co. strain gage. I used new US quarters as my validation weights. I forget what a quarter weighs but they are pretty tight tolerence and simple to validate and calibrate with.

PS a Tare function is also very useful but that is more easily done in the measurement system then in the sensor circuit.

Good luck