Issues with Voltage Controlled Current Source

  1. Good afternoon PhysicsForums! I have a design for a voltage controlled current source. I've included the circuit diagram. The basics of the design is that I have an AC voltage and a DC Voltage summed then passed through a 1MOhm resistor to give me a current on the order of 1 microAmp. This of course is determined by the amount of applied DC voltage. The DC voltage is applied via a DAQ board and the MAX4622 analog chip is controlled by the digital DAQ board signals.

    My problem with the circuit is that I have built it on a bread board and everything is fine but after I've assembled it on a pcb I am getting an odd voltage drop on the -15V supply to the opamps. If I put the multimeter across the 15V and ground at the power supply, I get 15V but when I do the same at the -15V it reads 1.4V. I am unsure of what component is causing this issue. Removing the components is much harder on the pcb and like I said it works well on the bread board. I've replaced the AD797 opamps and the MAX4622 chips but I still get the same issue. I also have tried different power supplies and the issue is persistent.

    Resistors, MAX4622 and LT1990 are surface mount components if that helps any.

    I guess what I mostly am looking for are points at where I should try testing the components and what voltages I should expect.

    Sorry for the image link instead of uploading. I was having issues re sizing the image file with ability to read the text.

    Voltage Controlled Current Source (IMGUR)
  2. jcsd
  3. berkeman

    Staff: Mentor

    Are you saying that the -15V output of the power supply is getting pulled up to +1.4V? What is the power supply? Does it have an adjustable current limit perhaps?
  4. Yeah that could be confusing. The power supply is a set +15V, Ground, -15V connection coming straight from the wall outlet of 120VAC. When I dont have the circuit connected to the power supply, I can measure +15V between the +15V and ground. I can also measure -15V between the -15V connector and ground. When I connect the cables going to the circuit and then retest at the same point as before on the power supply. I get +15V for the +15V and Ground, when I test the -15V and ground I read +1.4V. I can disconnect the circuit from the power supply and I read the normal voltages as in the beginning.

    The power supply looks like this but 15V

    Attached Files:

  5. berkeman

    Staff: Mentor

    With the circuits powered off, try measuring the resistance from the -15V input to Ground, and between +15V and -15V on your breadboard and the PCB circuit (power supply disconnected). Do you get similar numbers?
  6. berkeman

    Staff: Mentor

    You can also put small resistors in series with the +15V and -15V lines from the power supply, to measure the current being drawn. Size the resistors to give you about 0.5V drop at the nominal current you expect...
  7. berkeman

    Staff: Mentor

    Are any of the parts on the PCB getting hot to the touch?
  8. That was one of the first things I thought of since a previous version of this board had hot components because I found that the capacitors were lower rated that I thought I had ordered. There are no hot components on the board even after running for more than a few minutes.
  9. berkeman

    Staff: Mentor

    Put in the series resistors to figure out how much current is being drawn from the +/-15V supplies...
  10. I will be doing that first thing in the morning when I get to work. Thank you for your help so far.
  11. you may also try reflowing your soldering joints, as you said the circuit worked fine on the breadboard so either you have a poor joint or you damaged a component during soldering. just two possibilities
  12. berkeman

    Staff: Mentor

    One technique that I learned from a very bright and talented co-worker a while back is to check every node in the circuit with an oscilloscope probe. Look for any clues in the voltages and waveforms that may help you figure out where the fault is...
  13. jim hardy

    jim hardy 5,460
    Science Advisor
    Gold Member
    2014 Award

    If you have a meter that can resolve better than a millivolt,
    you might try reading the voltage drop along your traces to see where the current is going.

    Berkeman's suggestion of 'scoping things is excellent.
    High performance opamps will oscillate from subtle feedbacks due to board layout that would never dawn on us mortals ..
    So use a fast 'scope. Myself, for those troubles I like an analog scope.

    In the AD797 datasheet,
    did you notice their precautions about bypassing in fig 36?
    Are your 10 uf bypass caps good for high frequency,
    and lead lengths <1/4 inch including the track?

    If 'scope shows it's not oscillating probably you're okay on that count.
Know someone interested in this topic? Share this thead via email, Google+, Twitter, or Facebook

Have something to add?

Draft saved Draft deleted