• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 49 subscribers
  • Views 669 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OPA381 power supply question

Warren Pinkard
Prodigy 20 points
Other Parts Discussed in Thread: OPA381

Hello,

I am using the OPA381 to build a transimpedance amplifier for a calibrated photodiode from Thorlabs. The signal I am measuring is currently 1 kHz, but I may need to measure something as fast as 10 kHz, and in the extreme 50 kHz. The amp can be found here, and the photodiode here. I am trying to build a circuit just like the ones in Figure 1, 3, or 4 of the datasheet found here. I will need to do some experimentation to find out if I need a pull down resistor and/or a reverse bias, so I'm not sure which one of those circuits I'll use in the end.

I was able to use another TI document titled "Compensate Transimpedance Amplifiers Intuitively" to select a feedback resistor and compensation capacitor, but I'm pretty lost on picking a power supply. I'm a chemical engineer by training, so I don't know a lot about circuit design or electronics in general.

I was thinking about using this power supply because it is dual output +/- 5V, and I could experiment with a pull down resistor if necessary. Then I read that switch mode power supplies are very noisy, and I started looking at a linear +5V, +12V, -5V found here. I was looking at the HTAA-16W-AG in particular. Since it is triple output, I could run the op amp, bias the photodiode, and use a pull down resistor all from one power supply. Since it's linear instead of switching it should result in less noise as well. It's more expensive, though.

At this point I read something about by-pass capacitors in this document, and I got lost. I don't know what bypass capacitors are, or how to properly select them, or when they are necessary.

So here are my questions:

Is there anywhere I can read about properly selecting bypass capacitors for a given power supply?

In Figure 1 it shows a 1 uF capacitor on the positive side of the op amp power supply, is this a bypass capacitor? Should I just assume that I will always need to put a 1 uF capacitor there even though it isn't shown in Figures 3 or 4?

Would either of the power supplies that I have linked above work? Would I need bypass capacitors for them?

Is there another power supply that TI recommends that would be better?

Finally, and this is on a topic other than power supplies. If I were to use a big feedback resistor, say 10 MOhm, would I have any difficulty measuring the voltage output? Most multimeter leads are 10MOhm impedance, which is on the same order of magnitude. Would I end up pushing a lot of current through the leads instead of through the circuit? Should I try to find 1 GOhm leads? How should I go about measuring the voltage on the output of the op amp if I use a large value feedback resistor?

I would eventually like to hook up a BNC plug to this circuit and measure the voltage using a DAQ card on a computer. Would I run into difficulty with a high value resistor?

I realize that this is probably below the technical level of most of the people on this forum, but I really appreciate any help you can offer. I'm a bit lost and don't know where else to turn.

Thanks a lot,

-Warren

  • 0
    Expert 1220 points

    Hello Warren,

    A bypass capacitor serves a couple of purposes. Generally, you will see 10uF to 100uF individually or in parallel near the power supply itself. These will generally be aluminum electrolytic or tantalum capacitors. You will also generally see  1nF to 1uF near the actual device Vdd pin (roughly 2mm away). The capacitors provide a 'charge bucket' when the device requires transients in current. When the device requires additional supply current (perhaps it is driving a load, or features of the device turn on momentarily) the supply voltage may dip and this may lead to poor device performance. The capacitors are there to help alleviate this. These capacitors also minimize electrical noise on the supply pin by providing a low impedance path to circuit common. As the capacitor value is made to be smaller then the higher the frequency content is shunted to circuit common and thus away from the devices power supply. You should *always* place a bypass cap in your circuit. Again, larger values near the power supply module itself and the smaller values as close to the device pin as possible. When it comes to "what value?" well, you will often see 1uF. Why? Probably because that is what the applications engineer had in their drawer at the time when writing the datasheet or application note. There are additional papers written on bypass capacitors to the point where the series resistance and inductance of the package are considered. Sometimes this level of analysis is required depending on what frequencies may interfere with your design and what the load is.

    If you want to measure the *output* with a standard, let's say Fluke, you will not have a problem. You would be loading the output of the amplifier with a 10M ohm resistor and considering first order effects, I would say you would be A-OK. I would suspect that if you do run into problems with such a large load it would be due to the large resistance values and parasitic capacitance with the meter inputs and cables.

    -Ken

  • 0 in reply to Ken Dillinger
    Prodigy 20 points
    Hey Ken,

    Thanks for taking the time to respond to my questions. The information you provided has been really useful, and I've done some more reading on bypass capacitors so I think I've figured it out. I think I'm really struggling with the intimidation factor of approaching a whole new field so the advice you've given me has provided the confidence I need to move forward, and I appreciate it.

    -Warren