[SOLVED] Opamp Oscillation Problem - Help Required

[goldsmith]

Hi again hshah
I don't know why you used that capacitor in parallel with BE junctions ( it seems unnecessary ) . the out put stage is a simple complementary stage at class AB . isn't it ? and there is an op amp , to drive them . with this arrangement you can use cheaper opamps ( usual opamps ) . and i think perhaps you know how that feedback loop in that opamp does work ? and i suppose that you know how a freewheeling diode does work too ?
Best Wishes
Goldsmith

 

[hshah8970]

@Goldsmith:
Yes, I understand the feedback loop as well as the purpose of the diodes.

I don't know the purpose of the capacitor (this is just some DC motor driver circuit I found online).

What I want to know is: what maximum current output will I get here if the maximum current output for each transistor is 3A?

Moreover, for my application that has a much higher output current requirement (20+ A), what transistor models should I use?

Humza.

Update 1: I would also highly appreciate people giving me alternative circuits to amplify the opamp current for the motor's inrush current. Thank you.

Update 2: Guys, the local electronics market here doesn't have high output current BJTs available. Any idea how I could use capacitors to provide my motor with the required high inrush current?

 

[goldsmith]

Hi Humza
What you are referring to by 20A ? are you refer to the inrush current ? or perhaps more powerful motors ?
Anywhere , what kind of power transistors are available in your town ? is 2n3055 and Mj2955 available in your area ? if yes , use some of them in parallel together and don't forget to add balance resistors ( in series with emitter , and as low as 0.27 or 0.1 ohms ) .
Best Wishes
Goldsmith

 

[hshah8970]

@goldsmith:
Yes, I was referring to the inrush current, which I can only guess. I will be checking the availability of the resistors you mentioned tomorrow. And the resistors you mentioned, should their power rating be high? Because normal resistors would burn out on such currents. Is there a special name for high power rated resistors?

 

[godfreyl]

As soon as input voltage is applied, the motor starts off fine but quickly slows down to a halt and the output voltage at the opamp wanes. The opamp heats up.

What sort of heatsink is the opamp mounted on?

It sounds like the opamp is shutting down because it is overheating. Overheating is due to the power dissipated by the opamp.

Example:
Let's say you are using a +-16V supply, and the opamp is supplying +6V to the motor. Let's also assume that the motor only draws 3 Amps.

In this case, the power used by the motor = 6V * 3A = 18W. However the power dissipated as heat by the opamp = 10V * 3A = 30W. Unless the opamp is mounted on a good heatsink, it will overheat and shut down very quickly even though the current is only 3A.

 

[goldsmith]

Hi again
You can calculate the curren through each resistor and then their power . ( select them with enough wattage ) .

 

[hshah8970]

Alright so I'm pretty sure now that the opamp is oscillating. Firstly, I checked the inrush current and it was nowhere near the 10A limit of the opamp.

Secondly, I left my circuit sit for a while and found my motor's rotor rotating in either direction i.e. it would turn right for a second and the left for a second and then stop. This would repeat for a while. When this happened, I also saw my input supply voltage increase and then fall to zero.

This confirms oscillation, right? Because the motor was rotating both ways.

 

[FvM]

This confirms oscillation, right? Because the motor was rotating both ways.

Same problem as before, without a detail measurement, this confirms nothing in my opinion. (I don't say, I does not oscillate).

How can the input supply voltage change? This only "confirms" that the supply can't output the required current under all conditions, which may be a basic problem cause as well.

 

[godfreyl]

When this happened, I also saw my input supply voltage increase and then fall to zero.

This confirms oscillation, right?

No, it conforms the power supply isn't doing it's job. The opamp can't work properly if the supply is fluctuating like that.

btw:

What sort of heatsink is the opamp mounted on?

 

[hshah8970]

@fvm:
The oscilloscope in my lab is unavailable (I'm a student) otherwise I would've taken a picture of the output waveform, first thing. I removed that fluctuation by adding bypass capacitors with each supply. That seemed to do the trick. But now I'm stuck with the earlier problem: the motor speeds up for a while and then comes to a halt. And I don't know if this is of importance, but the supplies my lab has are unregulated. Moreover, the supplies are perfectly able to provide the required output current. I tested them directly with the motor and they worked just fine.

@godfreyl:
The supply stopped fluctuating once I added bypass capacitors to the circuit; I've also attached a freewheeling diode across the motor in reverse parallel. The problem in my last post has been solved. However the original problem remains and my motor starts up fine but then it comes to a halt and the output voltage of the followers drops to zero as well. And yes, the opamp heats up. But I suppose that is normal. Here is a picture of the heatsink I just took.


I'm thoroughly perplexed about this issue. I wonder what I'm doing wrong. Should I include the motor in the closed feedback loop of the voltage follower? If yes, then how? Is the motor's back EMF causing this anomaly? As always, help will be appreciated. Thank you.

Again: the opamp as well as the power supply are perfectly able to provide the inrush current. The normal operating current of the motor is around 1.5A and the inrush current is a little less than 3A.

 

[FvM]

Here is a picture of the heatsink I just took.

Did you mix up the pictures or is there no heat sink at all? In this case the observed behaviour would be pretty understandable.

I also don't believe that effective bypassing of a power OP can be achieved on the shown test board.

But now I'm stuck with the earlier problem: the motor speeds up for a while and then comes to a halt.

Thermal shutdown due to insufficient heat sink is the most likely explanation.

 

[hshah8970]

@fvm:
I see! I thought that the silver thing on top of the opamp was a heatsink. Now I've found that it isn't. I also understand now that bypassing requires circuit boards and shortened wires plus surface mount capacitors (correct me if I'm wrong).

Does one make heatsinks themselves or are they purchased?

 

[FvM]

An aluminium sheet of sufficient thickness (2-4mm) and size (> 10 x 10 cm) would already help. A commercially availiable heatsink is a better performing and more compact solution.

 

[godfreyl]

I also don't believe that effective bypassing of a power OP can be achieved on the shown test board.

You mean supply bypass capacitors? IMHO, that shouldn't be a problem if the capacitors are on the board close to the opamp, and due care is taken with the rest if the wiring (especially "grounds").

@Humza: Here's an example of a heatsink (and bypass capacitors):

 

[hshah8970]

Thanks to all who replied and helped me detect the problem. It's quite funny how I got to learn all about opamp oscillations, loop stability, and snubber circuits simply because my opamp didn't have a heat sink on it.

Again, thank you.

 

[godfreyl]

It's quite funny how I got to learn all about opamp oscillations, loop stability, and snubber circuits simply because my opamp didn't have a heat sink on it.

As my boss once said, many years ago:

Finding the right answer to a question is fairly easy.

The hard part is finding the right question.

:wink: