I am trying to get a weight measurment using 4 strain gauges in a wheatstone bridge configuration (one bridge for each gauge like this http://sub.allaboutcircuits.com/images/00205.png). What is the best way to do this? I was thinking of summing up the differential voltages from the sensors via a summing op amp, then amplifying that voltage with a gain amplifier (INA125). I cannot figure out how to do this as the differential voltage refrences the constant side of the bridge, not ground. Do i even need a summing op amp.
When i measure the differential voltage from my bridge, i get 3.3mv deflection with my weight on a single strain gauge. When i put 4 gauges in parallel and adjust the bridge, then distribute my weight to all 4 gauges, i only get .8mv, or 1/4 of the single gauge configuration with all of the outputs connected together. How do i use a summing opamp to add the .8mv from each bridge to get back my 3.3mv?
So what is the best way to do this???
I think you first want to use the instrumentation amplifier to amplify each bridge output, and THEN use a summing op-amp, or better, feed each in-amp output to an A/D input and sum their inputs in software.
Putting op-amps before the in-amp will just add noise and inaccuracy to your measurements (unless you design them for high precision, which is what the INA125 is designed to do in the first place so why reinvent the wheel?)
Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html
My system runs on 3.3 volts and the amplified output needs to be referenced to ground, not the bridges lower potential branch. i also did not want to use an amplifier for each bridge output. do i need to do that so that i get an output that is refrenced to ground? i know i can use the ina125 to amplify each bridge output and sum the result (i dont have 4 adc inputs available to sum it in code) with an opamp but that seems like too many opamps are needed. cost is a major factor and the ina125's are single amp chips at ~$5. i was hpoing to get away with a single summing amp for the 4 bridges and amplify the summed output (in mV) using a single ina125.
thanks for the input though! opamps are an area of difficulty for me.
Ok,i will try to be more specific.. I am just trying to make a digital scale using a 3.3 volt arduino. I have found this: http://www.youtube.com/watch?v=fPzUtzFJFus
and it seems simple enough. However, their implementation only uses one strain gauge. I wish to use 4 gauges (one for each leg). The ones that i have are half of the bridge, requiring the other half to be breadboarded. The 3 wire gauges are 750 Ohms with no load and look like this:
This was my first idea(please excuse the crudeness, the drawings are for functionality puropses. It may not be wired correctly), thinking that the current from each leg would be summed at the node going into the op amp. My DMM only read a .8mV deflection with my weight equally distributed on all 4 sensors. However, If i measure the output voltage of EACH bridge (disconnecting the varable output from eachother) and place my weight on a single sensor, i get a 3.3mV deflection. Makes sense since that one sensor now sees 4 times the load. I could just take this design and amplify the output with a gain of 1650, but it seems that this configuration hinders resolution.
Idealy, I would like to do something like this:
Where each .8mV output of each bridge is added to get my 3.3mV which is then amplified for the arduino. This seems more correct, but again, im an op amp noob. One thing that is really confusing me is getting a signal refrenced to ground and not the constant voltage between the two 188 Ohm resisters of the bridge.
Do you have lots of other things going on in your project? Why not just send each of the four strain gauge outputs to a separate analog input and sum them inside the Arduino by software? Are the other analog inputs needed for other things? Sorry to be so boring. But at least I didn't start down the path of why have multiple strain gauges?...
Why not just send each of the four strain gauge outputs to a separate analog input and sum them inside the Arduino by software?
The other analog inputs are being used by other sensors. The reason for needing to use 4 gauges is that i dont need a mechanical mechanism to direct all the force to a single point. All the newer digital bathroom scales have a strain gauge in each foot and each read a distributed load, then sum the result for the internal converter to get the users weight. I am basicaly trying to reproduce that but for an arduino analog input.
This is what i want:
Single power supply with an amplified differential output that is refrenced to ground (for the arduino ADC). However, i just need to know what the best way is to add 3 more differential voltages from 3 more bridge circuits to the instrumental amp (anyone know of a cheaper instrumental amp that will do this, im all for it!). :)
Im not complaining about anything. Cheaper is always better if i can get the same functionallity. The INA125 was recommended to me and so far it seems to be the cheapest "insturmental" amplifier and easiest to hookup. Thats all. The last schematic i posted is exactly what i want, but with 3 more bridges. The question is how does one sum the differential voltages of 4 bridges and amplify that sum via a device like the INA125 or AD620 to MCU levels, which needs to be refrenced to 0volts, not VCC/2.
This is simply for a digital scale. nothing more. The other ADC ports are for other scales. Thus i need to make multiple summing/amplifier circuits and so cheaper devices are better. Accuracy should be close.
Throw me your ideas. Dont be afraid.
Currently i have a setup with all 4 gauges in parallel. I get a deflection of .8mv for a weight of 165lbs. I can simply use the previous INA125 amp to get MCU voltage levels and in theory, it should work. I would prefer however to add each .8mv differential voltage together so that my weight is represented as 3.3mv deflection instead of .8mv. That way, by my calculations, i only have to amplify the output of the sum by 1000 instead of 4000, reducing noise, etc.