Meanwell PSU modification (remote control)

[js671]

Hi folks

Excuse my fumbleness- come across something I’m trying to figure out.

I have a piece of equipment containing a frameless Meanwell power supply which appears to have been modified to be remotely controlled from a PIC16F. The supply is powering an RF PA stage (exclusively) which is controlled by a display interface run from the PIC which displays power adjustment options, current output power (RF) etc.

Input voltage at full power to the RF PA is 48v at some amps.

The PIC & other bits are run off an aux 18v supply (regd downstream to 12), this works grand and everything except the PA runs great.

I’ve been able to get the PA running from a bench supply as well, with the 18v aux running the control circuits. Adjusting bench supply current makes the carrier power adjust and alls sweet.

The PSU itself is toast - salt air, perished insulation on the mains input and a large arc later, the board is blown to buggery beyond repair. Now I could just fire a PSU in there and run the thing at full power but adjustment would be handy. Which is where you guys come in

The control line from the PIC to the PSU is a single wire, connected to an unused pad on the Meanwell. This APPEARS to trace to pin 3 of the UC3902 on the Meanwell. Scoping reveals a fairly dirty (but could be the RF getting into my scope leads) “PWM ish” signal which jumps around all over (again could be the RF!)

I think it’s somehow modulating the load share duty cycle on the PSU but must admit that it’s a new one on me & the manufacturer doesn’t produce schematics of their products (although I do have the Meanwell drawing, hence working out where it goes. I’ve got no experience with these load share chips.

Meanwell have discontinued the PSU!

Question therefore is….whilst not obviously officially supported by Meanwell, is there the potential of making a similar mod to another Meanwell PSU that the gang here knows of & may be able to help with?

Happy to post up a few pictures of the goosed supply later if it helps

Thank you!!

[js671]

Turns out the “PWM ish” signal is a 1-12v signal.

[MathWizard]

So there's only 1 PSU, so no load sharing, but with the UC3902, and the PIC, was the power/voltage/current being limited to the RF power amplifier ? Or was it just set for some max setting for no load sharing ?

IDK these chips at all, or a load sharing circuit, but pin 3 of the UC3902 is just some open collector, so some current from the PIC, gets passed through that BJT, but then there must be some external circuits that take that current and does something with it, that I guess some other pin on the same chip, senses.

I don't see any sample circuit in the datasheet, did you map any of the circuit ?
https://www.ti.com/lit/ds/symlink/uc3902.pdf?ts=1721630027139&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FUC3902%253Fbm-verify%253DAAQAAAAJ_____4dajIFJHgIFMtm-7HBrQAM5udC3X4ED8RjOakr2oXVVtuaX4svmVy1jgniyTC-Zl0SClzdOIBOFu6zO5C60AWChPJkEAXH2z9Ywp6643hyi1FDTF_bJQTVckdYzV9LxKduCA0MYzSglRZWNwR4kKiRJLtpYgtR4Besc_nIUkbCjBMq2-ZeA7mlQFWj2m2UBxeoHztA9RCGQzg_7DieqiHjF1LiXGqKcRDd0w4Sr_VDjJAiQX3vtr3bH7RTfwYuB-TqDanWlUXXpXcYpcruBwNrMGF-_cWNwPgd2Bn9AZ4WbZP8