[SOLVED] LTC3780 Current limit

[luckym]

Hello, I have a DC/DC modul with LTC3780 (It is modul from ebay). The connection is almost the same like in DS (page 12)
http://cds.linear.com/docs/en/datasheet/3780ff.pdf.

Is it possible to add current limit regulation to this modul? I need to have common GND of Input and output.

Thanks for any advices....

michal

 

[mtwieg]

It already implements a current limit in the controller. If you want to make it a current regulator (as opposed to a voltage regulator with a current limit) then it might be possible, but will require modifications to the control loop, and output voltage limiting protection.

 

[luckym]

Thanks for reply, The current control in the IC is part of inner control loop (if i understand that...) and don't know how to utilize this for current limit. I want to keep the voltage regulator and add current limit. I want to use it for charging Pb-ACCU, so need voltage regulator with current limit.

I've tried this wiring (look at the figure) but unsuccessful - it oscillates. I had only LM358, may be wrong OA... Is this right way? or should I try something different (schematic)?

thanks for help..

 

[D.A.(Tony)Stewart]

Thanks for reply, The current control in the IC is part of inner control loop (if i understand that...) and don't know how to utilize this for current limit. I want to keep the voltage regulator and add current limit. I want to use it for charging Pb-ACCU, so need voltage regulator with current limit.

I've tried this wiring (look at the figure) but unsuccessful - it oscillates. I had only LM358, may be wrong OA... Is this right way? or should I try something different (schematic)?

thanks for help..

View attachment 112027

Vref comparator has far too much gain and no BW limits. Do you know control theory?

What you want is

1. slow response rise to CC limit,
2. then CV limit,
3. then when I= 5% of CC limit , foldback to CV float for Pb batteries and latch.
This topology does not do that. Search site for solutions.

 

[mtwieg]

Thanks for reply, The current control in the IC is part of inner control loop (if i understand that...) and don't know how to utilize this for current limit.

Yes, but the inner current mode control loop has an upper limit past which it won't allow peak inductor current to increase no matter what (see page 22 of the datasheet). But it's only a peak inductor limit so it does not accurately regulate the actual output current over a wide input voltage range. Also it seems to have built in foldback limiting, which may be a problem for implementing external current limiting.

I want to keep the voltage regulator and add current limit. I want to use it for charging Pb-ACCU, so need voltage regulator with current limit.

I've tried this wiring (look at the figure) but unsuccessful - it oscillates. I had only LM358, may be wrong OA... Is this right way? or should I try something different (schematic)?

thanks for help..

View attachment 112027

The schematic looks reasonable, but it's not surprising that it oscillates. Injecting current into the Ith node (actually, pulling current out of Ith) may provide better results than injecting into Vfb. Are you actually working with the real circuit, or are you simulating in LTspice?

 

[luckym]

Vref comparator has far too much gain and no BW limits. Do you know control theory?

No... I know very little about this. and this is me problem... Can you recommend something for reading?

- - - Updated - - -

Also it seems to have built in foldback limiting, which may be a problem for implementing external current limiting.

ok, maybe this explains the strange behavior when I connected this circuit to VFB pin.

The schematic looks reasonable, but it's not surprising that it oscillates. Injecting current into the Ith node (actually, pulling current out of Ith) may provide better results than injecting into Vfb. Are you actually working with the real circuit, or are you simulating in LTspice?

[/QUOTE]

I'm working on real circuit....

Now I'm thinking about this: I don't need fast response in current limiting... There is a potentiometer - top resistor in voltage divider. If I replace it with digital potentiometer will be the stability of the control loop influenced? I can sense current with MCU and set voltage via this pot.
It would work like an outer slow control loop. At the start I'll begin with voltage setting low, and gradually increasing until current limit. Is this a way?

 

[FvM]

Your circuit oscillates because the current control path has effectively infinite gain by using an OP in comparator configuration. You can turn the circuit into a PI controller by adding a RC circuit. As you said, the current control loop must not be fast, so there shouldn't be a problem to find a stable configuration.

 

[luckym]

Hello... I had a time to play with it and this schematic works fine for me...


It has a drawback with low current limiting. The Output of OA doesn't go totally to ground and there is voltage drop on the diode. But for me the limiting current is above this value, so it is ok....