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.

OPA541: Voltage Sucked Down on OPA541 5VAC Circuit Output

Part Number: OPA541
Other Parts Discussed in Thread: OPA548, OPA547

I am currently working on a design to generate a 5VAC (rms) @ 400Hz supply capable of 5A continuous. The design contains a non-inverting amplifier that takes a 213mV (rms) @ 400Hz voltage waveform (generated by a COTS board) and boosts it up to 5VAC (rms) @ 400Hz. The circuit is shown below:

This circuit works just fine, effectively generating the 5VAC (rms) output that I desire... that is... until I put it under load... When I connect this thing to our expected load (special incandescent aircraft light-bulb design, the voltage on the output waveform gets sucked down to 4.6VAC (rms) and draws 1.14 Amps AC (rms), and stays at 4.6VAC (rms) until I remove the load. When I remove the load, the amplifier output returns to 5VAC (rms) @ 400Hz. What could be causing the voltage to get sucked down like this?

Some additional information (not sure whether it will help anyone looking at this, but figured I'd share as much information as I actually understand/know). I've tried a specialized industrial supply capable of 5VAC @ 400Hz (at very high current) and the voltage does not get sucked down at all. Under normal circumstances the bulb draws 1.15 Amps AC (rms) from the industrial supply when 5VAC 400Hz is applied, not impacting the voltage whatsoever. I am convinced there is something wrong with my amplifier circuit...

Additionally, I am heatsinking this with a huge chunk of aluminum I found back in the shop. A thermal gun indicates that the actual part (not the heatsink) is at roughly 30 degrees C under these conditions.

Additionally the load is a string of incandescent bulbs in parallel, which are antiquated and thus I cannot provide specifications for. I estimate the load (based on measurements previously mentioned) size to be 4.4 Ohms.

Let me know if there are any other details that would help if I provided, or am missing something.

  • Hello Steven,

    I wouldn't expect a normally functioning OPA541 to exhibit a drop in the peak-to-peak output voltage when the load is applied. Something unexpected is coming into play in the circuit. 

    I've ran some simulations on your OPA541 circuit and peak power dissipation is under 30 W so the setup is well within the safe operating area (SOA), providing the heat sink arrangement is removing the dissipated heat. You indicate the OPA541 package temperature is about 30 C so it appears things are working thermally.

    I did a calculation of the gain error based on the limited open-loop gain (Aol) of the OPA541 at 400 Hz. Figure 2. in the datasheet, the Open-Loop Gain and Phase vs Frequency graph, indicates an Aol of about 65 dB (1778 V/V) at 400 Hz. The calculation results in a gain error of about 50 mV so instead of 5 VRMS, the output would be 4.95 VRMS. Clearly, that isn't the cause of the much lower 4.6 VRMS output you are observing.

    The output behavior tends to suggest that there is a resistance in series with the OPA541 output going to the load. It would only take ~500 milliohms to reduce the peak output volage from 7.1 Vpk (5 VRMS) to 6.5 Vpk (4.6 VRMS). 

    I suggest observing the OPA541 power supply pins, input pin, output pin and at the load point with a DSO as the load is connected and disconnected. Look for any unexpected changes in any of the levels besides that observed at the output. Also, if you have another OPA541 see if it behaves the same manner as the one you are currently evaluating.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • Thanks for your reply. I guess I don't quite understand when you say:

    The output behavior tends to suggest that there is a resistance in series with the OPA541 output going to the load. It would only take ~500 milliohms to reduce the peak output volage from 7.1 Vpk (5 VRMS) to 6.5 Vpk (4.6 VRMS).

    Is there something wrong with having a resistance in series with the output? Isn't that already what it means to have a load attached?

  • Hi Steve,

    What I meant is that there may be some unexpected resistance appearing in series with the intended load. Current flowing to the load would cause a voltage drop to be developed across the unexpected resistance, resulting in less voltage across the intended load. That unexpected resistance might reside with the wires that connect the OPA541 output to the load, circuit board traces, or a resistance developed at some connection point(s) within the output circuit loop.

    The OPA541 will drive a resistive load, and complex impedances (Z) such as motors and transducers that can have both resistance (R) and reactive (+/-jX) characteristics.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • Tom,

    I see what you're saying. No that's not what's happening. The input voltage is also being dragged down as well... It's behaving more like it can't supply enough current to the load.

    The voltage at the actual output (pins 5 and 7) is being dragged down directly.

    Thanks,

    Steve

  • Hi Steve,

    You mention the input voltage is dropping along with output voltage dropping. If the voltage change at the OPA541 output divided by the voltage change at the input equals the correct closed-loop gain Av = 1 + (3.38 M/ 0.15 M) = 22.5 V/V, then the op amp is functioining correctly. That indicates the problem is elsewhere in the circuit.

    For starters, make sure that the power supplies you are powering the OPA541 circuit with a supply that can provide the required current. Take a look at the OPA541 supply pins with and without the load applied and monitor the voltages with a DSO. They should remain relatively constant with and without the load.

    An input voltage change that occurs when the load is applied to the output suggests a current loop issue with the PC board. If the input signal return and the output signal return share a common ground return the output current may be affecting the input current. You noted that the gain was correct without the load, but drops with the load connected. The output current in the latter case is going to be amperes. Try reducing the load (increasing the resistance) and see if the voltage gain comes closer to the expected value. That could provide some evidence of a ground loop issue.

    I would expect that you are using continuous planes for the positive, negative and ground planes. Can you describe your PC board construction used for the OPA541 circuit?

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • All of the supplies look okay. My PC board construction is actually just a perf-board with a bunch of soldered connections and fly wires going every which way. The soldering is pretty good though, but a ground loop could possibly be the issue.

    What can I do to address this issue quickly and efficiently?

  • Hi Steve,

    Unfortunately, prototype construction may not result in the needed performance. I do think you will be able to achieve this if you have a properly constructed PC board designed for high circuit current levels, that minimized ground impedance. This involves designing boards having ground/power planes, or at least one having a significant ground pour and wide, short traces for the circuits that carry high current.

    The OPA541 is one of our older power op amps. It doesn't have an evaluation module (EVM). However, there is a somewhat newer power op amp the OPA547/OPA548 and we designed a new EVM for it a couple of years ago. The OPA547/OPA548 EVM documentation shows its PC board design in Figures 2 and 3. It is a relatively simple, two-layer board. The top layer is mostly ground plane, while the bottom layer is primarily the signal routing plane. This EVM readily demonstrates the full performance capabilitites of the OPA547 and OPA548 power op amps.

    You can find the OPA547/OPA548 EVM documentation here:

    www.ti.com/.../sbou132.pdf

    If you follow and apply the layout techniques used for the OPA547/OPA548, I think you will be able to achieve a similar level of performance from the OPA541.

    Regards, Thomas
    Precision Amplifiers Applications Engineering