SG3524 COMP Pin Usage

That is a really old part.  I bet the designer of that is long ago retired.  Typically, you would use the COMP pin to compensate the circuit for stability.  Appropriate RC network you can be used to add an additional pole and zero to the small signal response.  You would feed back a portion of the output to the inverting input of the error amp and a portion of the internal reference to the non inverting input to form the control loop.  I do not think there is any provision for implementing voltage feed forward for the input voltage.  Additionally you can clamp or limit the output of the error amplifier.  Internally this is for shutdown and current limit.  You could also add an additional control externally.  If you are trying to do as you suggest, it will get quire complicated as your external circuit would have to source or sink whatever the output stage of the internal amplifier was trying to produce.  The other current limit and shutdown circuits, just clamp teh output to ground during fault or shutdown conditions.

All this begs the question: "why are you considering this device?".  I'm sure there are much more modern alternatives with all the features you require.

Thanks for the quick response John. I understand this is an old product, admittedly, that is a bonus in one regard. There are several documents found on IEEE Xplore that suggest the SG1524, and UC1524 COTS parts have been tested to >20 kRad Total Ionizing Dose radiation tests and survived. Obviously there is no guarantee since these are not Space rated parts but it is a good starting point for implementing COTS parts in Space applications and TI wouldn't ever be able to guarantee any results on this line of parts due to that. I understand that the production process could have changed since the tests were last performed on these COTS parts. Any design I end up going with will be radiation tested in the end but having reasonable assurance of the part is something that is highly considered.

The other option I can think of which is a little less ideal than the voltage feed forward (thanks, I didn't realize this was the correct term to use here) is to try to implement the error amplifier and current limit inputs to both monitor voltages. I'm just starting to look into this chip (or likely the SG2524 due to temp ratings, same functionality as far as I can tell). The error amplifier should be relatively straight forward since it's original use is to monitor the output voltage. I would presume that if I kept the input voltage within specifications on the current limit amplifier I could also use it to monitor a voltage. The idea is that the circuit would force the solar panel to keep a certain voltage and the output voltage would be limited, If the voltage on the output rose too high it overrides the input voltage amplifier and forces the PWM to reduce the output voltage. This would really end up being an issue of voltage scaling if this were to be a viable option. I was hoping to be able to easily connect to the COMP pin and provide my own op-amp conditioned signals.

Thanks for the reply, I look forward to any more direction you may be able to point me in.

JohnTucker said:

Probably what you want to do is implement all your external circuitry, then provide that as an input to the to the error amp.  you could configure the error amp as just a unity gain buffer.

This crossed my mind earlier. I'll certainly look into it more thoroughly. Thank you for the help.

Hey just a quick sanity check here. Even though the functional block diagram on page 2 shows the current limit inputs as an op-amp. Page 9 suggests that this is not the case but then again I could be looking at a specific use case. I'm not entirely sure why one page would suggest an op-amp while the other does not. What is your take on page 2 versus page 9?

It is essentially a comparator with a threshold of 200 mV.