Reverse engineered boost converter

A project log for RPN Nixie Calculator

A programmable RPN calculator using way too many Nixie tubes

robert.c.baruchrobert.c.baruch 05/19/2016 at 02:5312 Comments

So I got the Chinese DC/DC converter, and it seems to work. Powered with 12VDC and set for 170VDC loaded with 150mA, efficiency is 75%.

The chip, as well as the board, is labelled with the Chinese for "Strong Electronic Technology". Note the two heatsinked MOSFETs. The switching diode is in the lower right.

I didn't turn it on, I took it apart first.

The schematic (click to embiggen):

(Updated to show a transformer T1 instead of a single inductor. It probably has a turns ratio of 1:14, but I won't be sure unless I sacrifice the transformer and unwind it. Also corrected the current sense circuit.)

The bright minds over at the eevblog forum identified the chip from its topology as the UC3843 SMPS controller.

The only weird thing here is that input MOSFET. It appears to be for reverse input polarity protection. A MOSFET and two resistors when a single diode would do. I must not be seeing something.

In any case, now that I designed my own, with fewer parts, maybe I can get better efficiency?


Paul Kocyla wrote 03/29/2017 at 18:20 point

I wonder why one pin of the secondary coil connects to the primary (mosfet drain) instead to ground. Maybe it´s for routing-convenience?
If it works with ground connection it should be easy to make an isolated or symmetric supply of this module.
I´ll try this out in a second... yes, it works.
The mosfet circuit on the left seems to be the undervoltage lockout, can be skipped.

  Are you sure? yes | no

robert.c.baruch wrote 09/06/2016 at 19:10 point

Yes! Mine came with a piece of paper showing what the actual terminals were, and I had to spend a few seconds making sure that no, the paper wasn't a misprint, and that it was pointing out that the markings on the unit were reversed.

  Are you sure? yes | no

Masterice wrote 09/06/2016 at 15:42 point

It may be of interest to anyone presently thinking of using these HV psu units from China that all the ones I have seen recently have the English 'IN' and 'OUT' printing on the PCB by the terminal block IN THE WRONG POSITION.

The IN should be the out and the OUT should be the in.

I'm using these (and other) HV psu units to power retro simple valve (tube) transmitters. So far - so good.

  Are you sure? yes | no

robert.c.baruch wrote 05/21/2016 at 18:30 point

That's actually disturbing -- thanks for noticing that. I wonder what happens if I pop it out and replace it with an IRF630.

  Are you sure? yes | no

K.C. Lee wrote 05/21/2016 at 18:54 point

You should get a bit of margin for the voltage rating of the MOSFET e.g. 20% or so.

  Are you sure? yes | no

pepi wrote 05/21/2016 at 22:10 point

You may need one with a higher current rating. That HY1707 is rated at 80A, and the IRF630 for <9A. So maybe something like the IRFP260N would be more appropriate (depending on how important it is in this design).

  Are you sure? yes | no

robert.c.baruch wrote 05/21/2016 at 23:57 point

So I tried the IRF630 (heat sink, thermal paste and everything) and it worked for about 5 seconds, then the voltage started to sag. I put the HY1707 back in (heat sink, thermal paste and everything) and it works fine. The IRF630 also got hot, while the HY1707 did not.

So that's really strange. I'm pretty sure I'm not drawing more than 9A from the power supply -- it's capable only of 7A.

It may be somewhat important -- I'm suspicious that having a VDS across the HY1707 that exceeds its specs is asking for trouble. Like, premature power supply failure.

  Are you sure? yes | no

pepi wrote 05/22/2016 at 09:45 point

The IRF630 has an Rds on that's quite a bit higher than the HY1707, so this could explain that. You can try the IRFP260N or another mosfet that can take 200V and appropriate current and which has a lower Rds on.

  Are you sure? yes | no

pepi wrote 05/22/2016 at 11:53 point

And by the way, what matters is that the inductor, when the mosfet is conducting, can draw quite a lot of current. So you don't need to be drawing 9A at the output for the inductor to draw that much or more.

  Are you sure? yes | no

pepi wrote 05/21/2016 at 16:55 point

Looks like that HY1707 is only rated for 70V. So they seem to be using it out of spec, at least if you want to output e.g. 200V. 

  Are you sure? yes | no

robert.c.baruch wrote 05/25/2016 at 03:34 point

I went back and looked at the inductor. It's actually a transformer, so the VDSS rating makes much more sense. I've updated the schematic to reflect the new component.

  Are you sure? yes | no

pepi wrote 05/25/2016 at 08:44 point

Great! That actually makes a lot more sense. I was already a bit surprised how they were able to get this performance from a simple boost, because the duty cycle would be quite high, and it's harder to get good efficiency that way. So this explains it, and then the HY1707 is most probably fine. You can put a scope on it to see what kind of voltages it reaches when the device is outputting 200V. See also the following app note which discusses this design:

  Are you sure? yes | no