Strain gauge amplification problems

Discussion in 'The Projects Forum' started by jlangho, Oct 22, 2011.

  1. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    Hey all, just found this place, looks like a sweet little forum.

    anyway, I built an amplification circuit for my wheatstone bridge thats in 1/4 bridge setup. I tested the whole thing in lab with a function generator and voltage supply. I'm using a LM741 for the +/- 6V supply and a LM348N for the amplification circuit. I realize these aren't the best to be using but they're all i had at the moment.

    In the lab, it worked well. amplified correctly and everything seemed kosher. Hooked it up on the car and I'm experiencing some issues. It will work....sometimes....lol....The only time i can get it to work, is if i touch my pinkie on the + terminal of the battery and my thumb on the - terminal of the battery. Then if i push on the strain gauge with my other thumb, i get a reading through our data acc system. However if i don't "bridge" the terminals with my hand I get no output.

    tried higher and lower capacitor values from about 1μ all the way up to 1M and it didn't seem to have any effect.

    I'm guessing that there's a ground issue...or a capacitance issue somewhere. I'm also betting to guess that my problem is in my voltage supply circuit since it worked fine in lab but not in the field.

    attached is an image of the circuit. thanks!
     
  2. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Can you provide more details concerning what your strain gauge application is for?

    What strain gauge are you using? part number?

    hgmjr
     
  3. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    application is for FEA (finite elemental analysis) analysis of structural members primarily in a-arms and frame elements. Simulation results can only give you so much and its essential to have real data. After we get a working circuit this should help us with frame triangulation and tube sizing for light weighting the entire car.

    I realize you guys have a "no automotive" policy but this is for FSAE, which is a student design competition to engineer and compete with a Formula style car. We design and build the whole thing so I assumed this would be a little bit different than the usual.

    its a micro-vishay 350Ω strain gauge. We're using .01% tollerance resistors for the bridge completion.

    here's the digikey link Strain gauge
     
    Last edited: Oct 22, 2011
  4. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    Note that the strain gauge circuit (not shown), must have some DC path back to the virtual ground. Otherwise, the LM348 will be unable to have defined bias voltages.

    Your story that it starts working only when touched is suggestive of a problem of this kind.
     
  5. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    that is one thing that I failed to mention. Currently I'm using a 5V sensor supply voltage from our ECU/data logger to excite the bridge. I'm going to change that and use the + 6V supply I made and connect the grounds as you mentioned.


    this idea was mulling around in my head but I just failed to mention it :p
     
  6. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    I just looked at my "intermediate" schematic that i drew up in lab. When i was applying the wave source i did have the grounds common...going to the lab now to test it
     
  7. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    tried it....I'm guessing the 5V line from the ECU is an elevated 5V...nothing ever works well if you try to tie its ground to something else.

    I tried hooking the power for the bridge into the +6/vgnd line coming from my power regulator and it doens't have enough power dissipation to work well. The line droops way down to about 1.1 volts.

    I also tried to power the wheatstone bridge with just a voltage divider and it works but now i'm not sure about a few things.

    1) the ground of the v-div for the wheatstone bridge is also the -6V line that gets fed into the op amp as well.

    2) it works if these two lines are left disconnected or tied together but then they give different values

    I'm guessing my problem is the voltage sources for the circuit...making sure that everything is happy. Im assuming that if i could get a good source for the op-amp power as well as the power for the bridge I'll be set.

    Would using something like a LDO (something like an LM317) to regulate my 12V line work better? Also if someone has a better circuit for the +/- 6v line I'm all ears as well
     
  8. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    Made a voltage regulation circuit with a LM317. The issue seems to be that our ecu/data logger is measuring the voltage differential between the ouput of the op amp and the - of the battery....which is used as the -6V line.

    I'm going to try and make a voltage inverter with a 555 and see if it works or not
     
  9. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    Made the voltage inverter with the 555 but it seems like something in the circuit went bad. Time to go back to the lab! any ideas for help?
     
  10. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    I'm attaching a circuit of what i tried to do today with the 555 as a negative voltage generator.

    Also, I found this post

    http://forum.allaboutcircuits.com/showpost.php?p=84027&postcount=2

    with the design by sgt wookie (fig 10)

    does the circuit perform well under load? If at all possible I'm looking for a solution that will drive my bridge as well as the op amps. Is this a viable solution at all?

    thanks
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Your connections to the 555 timer are wrong.
    You have power (Vcc) connected to pin 7; that won't work at all.
    You have pins 8 (Vcc) and 4 (Reset) connected to Vcc via a 4.7k resistor (R13); that won't work either.

    Replace R13 with a piece of wire.
    Change R12 from 33k to 1.2k.
    Add a 4.7k resistor between R12 and the junction of the wires going to trig/thrs and the top of C1.
    Connect pin 7 to the junction of this new resistor and R12.
    C8 is too small; increase it to 100nF
    C9 is MUCH too small; increase it to 10uF.

    But, even that probably won't be enough to drive your bridge, nor will it be regulated. You really need a well-regulated very low noise supply for that bridge.
    I don't know how much voltage you are planning on placing across the bridge, either.

    If you're speaking of the LM675 virtual ground - it's similar to the datasheets' version but actually the version you saw in that thread would not be stable, as the LM675 is not unity-gain stable. It WOULD work with a power opamp such as a L272, L2722, etc.

    But as I mentioned already, you need a very stable supply for the bridge.
     
    Last edited: Oct 24, 2011
    jlangho likes this.
  12. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0
    The schematic i found for the 555 timer was one off the internet. Thanks for the heads up on that. I'll try some things out and see how it works. The 555 circuit won't be used for the bridge. That voltage supply will come from the LM317 regulator circuit that i included in there as well.

    I appreciate all the help :) its one thing to learn all the theory behind this crap and its another to apply it lol :cool:
     
  13. mcasale

    Member

    Jul 18, 2011
    210
    12
    Hi. It seems you are going to great lengths to use dual power supplies. You can look at using a single supply op amp, and just regulate your +12v to give a nice clean voltage.

    I have used the AD8571/AD8572/AD8574 op amp in a PRT amplifier, which is not too different from a bridge circuit. I think these are SMT only, but there might be something similar in through-hole (if that's what you're using).

    If you bias the bridge at the mid-point of the supply, you should have no problem with the common-mode. The instrumentation amplifier will also need to work with positive signals, so you have to verify what direction your bridge signal goes.
     
    jlangho likes this.
  14. jlangho

    Thread Starter New Member

    Oct 22, 2011
    15
    0

    The only reason why is because i had all of this stuff laying around and we're on a bit of a tight timeline. I'll get a quick schematic drawn up for a similar amp of what you're talking about just for reference so we know we're on the same page....lol...I was also thinking about this but by the time digikey shipped and blah blah blah...it would put us about a week behind timeline. Might end up being that far behind anyway :p


    Edit:

    looks like this guy might work http://search.digikey.com/us/en/products/NJU7004D/NJU7004D-ND/638814
     
    Last edited: Oct 24, 2011
Loading...