KORAD programmable PSU Voltage kludge

[MikeQ]

I recently bought a KORAD USB controllable PSU with the intention of implementing a lithium ion battery charger in software. Unfortunately the PSU reports inaccurate voltages when within close range of the setpoint. When the measured voltage gets within 0.2 volt from setpoint, it reports/displays the setpoint instead of the measured voltage..! I assume this is deliberate by the firmware, but neither the vendor nor KORAD are too happy admitting this. And of course, it might be only my unit defective, however not very likely.

I wrote down all the detailes here; http://whiteboard.ping.se/HackValue/KoradPsu

Much appreciated if someone could verify my claim that this is a problem with the series, and not only my device (or as the vendor told me, my multimeter ).

Thanks.

[HalFET]

What did you use to measure these values?

[MikeQ]

What did you use to measure these values?

"Reported voltage" and "current reported" from the PSU itself via the serial API, consistent with the values on the PSU's display. "Current limit" the limit set via the serial API. "True voltage" calculated from "current reported" over the 20 ohm load via ohms law. This value is verified against an external voltmeter as well.

[Per Hansson]

Yes this is a documented issue with this PSU, see here:
https://www.eevblog.com/forum/repair/korad-ka3005p-power-supply-calibration/msg1105223/#msg1105223

The user "Tony Mach" has even started work on a custom firmware for it if you want to get your hands really dirty further down in the same thread!

And then if you are really down to it why not just throw everything out and design your own logic controller?
https://www.eevblog.com/forum/projects/korad-ka3003d-redesign-and-upgrade/

[MikeQ]

Many thanks, Per!

Now I know this really is by design and can continue my dialog with KORAD based on this. Hacking the firmware would have been an idea, but with IP protection enabled in the MCU requiring writing new firmware from scratch ... no thanks. Patching away the condition toggling from true value to setpoint probably quite easy, but requires the original firmware.

What a shame.

[glarsson]

When repurposing a bench power supply to charge batteries it is important to find out if the power supply can survive when the battery tries to fees power back into it. Many power supplies will not survive.

[alm]

To prevent current flowing from the battery into the power supply (e.g. when power to the power supply gets interrupted, or some protection triggers), a series diode between power supply and battery is often recommended. Obviously this will degrade voltage regulation at the battery terminals.

[glarsson]

The diode makes it hard to stop charging LiIon batteries at the correct tension.

[Kalvin]

How about using reverse current protection:

http://www.ti.com/lit/an/slva730/slva730.pdf

Using dedicated reverse current protection circuits will reduce the voltage drop compared to the diode solution.

One may also want to simulate and investigate this simple "ideal diode" circuit using a P-MOSFET and two [quite well matched] PNP-transistors:

http://jiggerjuice.info/electronics/projects/power/ideal-diode.html#type3