Power OP four quadrant bench supply

[Amper]

Hi!
As usual i hope i didnt overlook some older post disgussing the same topic.

Im the fortunate owner of two APEX PA03 Power OPs. These devices can run up to +-75V and 30A, have thermal protection and a nice salting of tasty beryllium oxide ceramics.
Since my recent projects require battery simulation and a general high power for quadrant supply/amplifier would be an amazing addition to my lab id like to give them some housing work.

My stuff usually only requires OPs in the basic use cases and i never had to come up with some design myself so i failed miserably at finding a solution my self.

What i want to do is to use a second low power low voltage OP to measure the current over a shunt to ground like in a usual bench supply. The Voltage should easily be done by just amplifying a +-10V reference using the PA03 directly. But how do i get both at the same time? Since i want the current limit only to kick in at a specific value i suspect i need a diode for non linearity pulling "down" the voltage reference but what will happen in the negative case? I thought about anti parallel zeners but something doesn´t feel right.

I hope someone understands my problem by the vague description given. It would be amazing if you could push me in the right direction.

cheers

[Kleinstein]

The cross over from voltage control to current control is indeed nonlinear. Using diodes is kind of a natural way.

As far as I see it one would have too possible cases:
1) have a single set voltage and in theory independent positive and negative current limits.
2) have a set current and separate positive and negative voltage limits - this probably not the way it is wanted.

So for case 1) it would be 2 separate circuits for the positive and negative current limit with a diode each to pull the voltage down or up.
Quite often one might choose a symmetric limit, but would still have the 2 limiters.

[David Hess]

You have it right; something more than a single diode connected current control loop is required.  The examples I quickly found use a class-B or class-AB voltage output stage like your APEX PA03 and enclose it inside *two* diode connected current control loops with one acting on positive current and the other acting on negative current.

This forum discussion shows some examples:

https://www.eevblog.com/forum/projects/simple-4-quadrant-psu-with-remote-sense-please-recommend-topologyschematic/

The functional diagram of the LT1970 power operational amplifier also shows this configuration.

[Amper]

Thanks!
Yes, i played a bit in lt spice yesterday and when i went to bed i noticed my mistake. Two current setpoints is just not enought even when you want a symmetric output limit. The anti parallel zeners worked but then other quadrants didnt do what they were supposed to.

If i find the time ill play around a bit more and give you an update what i will come up with.

[duak]

Amper, you may also wish to look at the hp 6825A to 6827A bipolar power supply amplifiiers.  There is a write up in the hp Journal from July 1974.
www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1974-07.pdf

Kepco made similar instruments, the BOP - Bipolar Output Power supplies.  Most of the manuals available on line show the block diagram and give an idea of how they work.

Cheers,