Help figuring out part of a switching PS schematic

Hi All,

Below is a link to a portion of the power supply for an Alesis M1 Active MKII monitor. This is my first time trying to figure out a switcher schematic and there's a part of the circuit that isn't quite as intuitive as the rest of it. What function does Q5/Q2 and associated components serve here?

Thanks!

www.bicyclerecordings.com/various/Picture1.png

Bump?

That's very strange, and possibly very-very tricky. I can't explain it.

It's probably a form of ZVS active clamp for a flyback converter.

Usually there is one blown 25 cent cap in those things!  That's what keeps repair shops open these days.  There was a good TAB book on switching supplies, but even that was oblique.
Does it sit there and tick?  That is the usual cap symptom.
Mike

I'm no switcher expert, but I can guess as well as anybody.

It looks like the unregulated DC is connected to the top of a center tapped transformer. The center tap is pulled to ground by a common source power fet Q1, autoformer action drives the other end of the winding negative, cutting off Q2, When that bottom end of the winding (3) swings positive relative to the center tap it turns on Q2..

Transformer T2's floating winding is driving the gate of Q2 out of phase with Q1, so it specifically turns off when q1 is turned on.

Troubleshooting doesn't require perfect knowledge of exactly what everything is doing. Start ohming out the active devices to see if one has failed, or look for a bad passive component.

or not...

JR

At the moment I'm more just learning about the circuit for general edification. I haven't gotten around to troubleshooting these power supplies yet, but I will sometime before too long.

Glaime, can you explain a bit further what ZVS means? I also don't quite know what you mean by flyback converter either...

bobschwenkler said:

Glaime, can you explain a bit further what ZVS means? I also don't quite know what you mean by flyback converter either...

Click to expand...

http://en.wikipedia.org/wiki/Flyback_converter

About zero voltage switching, any good power converters book explains that. Also with Google you can find good articles.

Ok, if you just want to figure out the circuit, I would suggest getting the data sheet/app notes for that specific regulator IC. They are mostly similar but not identical so it will be helpful to see precisely what each pin is doing with typical voltages and waveforms.

You may get lucky and find that actual circuit or something close in the app notes.

JR

I did check out the IC data sheet with the intention of at least figuring out what the pins on the schematic are...

For anyone who's interested, I took another look at these power supplies today. It was a pretty easy fix, C8 on both units was wiped out. I measured one at 80nF (supposed to be 220uF) and the other even lower. On the PCB C8 is right next to R3/R4 which appear to be running quite hot; the PCB is starting to carbonize around the solder pads or these resistors. I think this is just how the circuit runs as designed. I ran across another similar comment or two while doing some searching about these power supplies.

I must admit this really got me. I was hoping that the masters would bring some explanation with their analytical skills but I can't blame them as switching psu design is not everybody's cup of tea.

I have spent some time in looking around and, while I could not locate particularly an exact example on the net and in the two books I have (Power Supply Cookbook, Marty Brown-Power Supplies, Switching Regulators, Inverters and Converters, Irving M Gotlieb), the closest I could get was similar networks that increase the eficiency on the switching Mosfet. Not producing a very obvious result on the output, but to help the Mosfet perform better. A bit like acting as a heatsink but electrically.

However, went a bit further and asked someone I know whom works for a well known company and specialises in switching design.

It is apparently a clamping network to prevent the voltage rise during switching on the drain of Q1 due to C and L (transformer) within the circuit. And as expected above, to protect Q1. He says that the main reason is really the economy, otherwise the maximum voltage rating of Q1 would have to be much higher, hence an expensive part. His view is that it can be done simpler but although this is an expensive way of doing it, it is really a much higher performance circuit.

Unfortunately this was on late Friday afternoon and he obviously did not have the time to reply in full, but  promised me to mail further info which I will post here.

Cool, thanks so much for the info! I'm interested to hear what further your friend says about it.