Buck being converted to battery charger, 1kW

[cupoftea]

Hi,
A customer has sent us a 1kW power supply and wants us to adjust it so that it clamps its output current to 34A maximum. However, the customer has requested, that if we can, can we make it clamp its output current to an increasing value if the output voltage is lower….. Ie, at 28V output, clamp the output current to 34A, and at 18V output, clamp it to 38A….and the clamp current level being linear between those two points.

The attached (PDF schem and LTspice sim is what we have so far. The ringed current clamp is the bit that we have added in.) BTW, i have to confess that the BOM they supplied us with does not have a value for R35....So we have guesstimated what it probably is. (its a DCDC module and we cant break it open)

This one is a long , messy story….customer originally supplied the PSU not realising it was supposed to be a battery charger….so now wants to modify it for battery charging.

The only way we can control this PSU is via the TRIM pin…..thats the only thing we can access.
Can you agree that , given the attached shown circuitry, there is no way that the current clamp level can be made to increase as the output voltage reduces?

--- Updated Feb 13, 2022 ---

May you also agree, the following?...

The customer thinks that changing U8 from an integrating error amplifier to an error amplifier with a DC feedback path (ie a high value resistor from opamp output to inv, as well as a capacitor from opamp output to inv) will mean more chances of avoiding instability.

Do you agree that this doesn’t add up?

 

[BradtheRad]

You're walking into risky territory. This story (long and messy by your account) is in a category with the car owner who goes to a shop carrying a carburetor he bought for a 'bargain price', and asks the mechanic to install it.

Tradesmen usually insist on obtaining parts themselves, because they're expected to guarantee their work.

So after you fiddle with your customer's power supply he may want to hold you responsible for anything that goes wrong later. 1kW suggests the system is expensive. It may have an existing problem in the battery (or whatever). You'd need to do extensive testing on the system to make sure you're not being set up.

 

[cupoftea]

Thanks, but its no problem....this customer is giving us a go.

Essentially, can you confirm 2 things...
1...Inv amp error amps dont give you fold out current limiting? (clamp current increasing with lower battery voltage)
2...Inv amp error amplifiers dont give you better stability margins than integrating type error amplifiers?

 

[Taihung]

Hi
Can you share what kind of PSU ?
I had converted a PSU from voltage mode to current mode with 12V60A DSP-750, with 800W and run with PFC+LLC controller.
I use it as battery charger or electrolyzer. More than 40pcs are still working.
Before that I also though as you did, designed a external buck with 60A adjustable output. Yes, it run but make higher BOM cost and reduce efficiency. I use MCU to control outpput voltage & current.
This PSU/DSP LLC controller is not prefer for variable output so much. After modify it can change output voltage from 7-13.5V @60A (65A max). It enough for charging 12V lead acid or 03 cells Li-ion battery pack.
Luckily the DSP support I2C to read out voltage, current, temperature, fan speed. The rest to do is only display to LCD.
In each PSU/DSP have power controller & monitor controller. In this kind PSU they use analog control for PFC + LLC and Microchip 16F/18F/24F for monitor&protetion.
The trick have to do is:

1. Remove devider feedback voltage resistor by 10V zenner (offset 2.5V from reference), so the much output voltage is 10V + 2.5V = 12.5V (over 13.5V, the Over Voltage circuit or MCU will cut off).
2. Remove Under Voltage circuit (normaly under 10.8V will cut off) by cutting net on PCB and route it to 12V sub power. MCU still monitor output voltage, if setting output voltage < 6.5V out of start-up time (few second), MCU still turn off main power. I also not suggest to mod this controller operate at low output voltage, because switching frequency going high and control loop is not stable.
3. Route output of current sense circuit (output of opam) to the mid point of devider feedback voltage resistors. Add a resistor to adjust offset. MCU also read current via ADC. Adjust offset voltage to this point also adjust the output current.
4. The regulating current under 9A is not linear, beacause SHUNT measure circuit disadvantage when input voltage too small.

PSU still have Over current protection, if output current higher 65A, it will shut off.