advertisement


Switch Mode Power supply - Help !

E

eddie pugh

pfm Member
I've been trying to get my head around a fairly basic SMPS from a Toshiba DV-R16 DVD/VHS combo recorder player which a friend has asked me to look at and see if its beyond redemption.

Its completely dead so out with the power supply board and the first stage rectification is working. I changed the HOT side reservoir cap which was shorted and now am reading about 330 volts DC on the cap. Next I checked all the caps on the COLD side swapped out a few that were reading a bit low but sadly no joy

I checked the input to the transformer and I was not getting any AC volts so I thought something in the HOT side that was supposed to oscillate and bump the 330v DC up to something really high frequency AC for the transformer to do its stuff was not working. I whipped out Q1001 which is a K3566 Silicon N Channel MOSFET and found it was shorted so I thought worth a try (really great next day service from NIKKO electronics on the bay by the way) More solder fumes but sadly still no cigar.

Now feeling like throwing the whole thing in the trash but can't even go to the tip so its sitting here challenging the old grey cells

I should add I'm not an electronics man but like to dabble and learn and after a bit of research I'm thinking that either the optocoupler IC or one or both of the other transistors on the HOT side must have gone.

I know that SMPS have a high frequency oscillator on the HOT side to function and an optocopler to provide feedback from COLD to the HOT side to regulate the supply.

I thought maybe it was not getting enough load to snap into action but even after connecting it back into the main board and chassis there was no glimmer of response on pressing the on/standby switch

I've added below links to the really poor circuit diag from the Toshiba Service manual and physical board layout and a photo

https://i.imgur.com/KpMfazG.jpg
https://i.imgur.com/5RjTpZz.jpg
https://i.imgur.com/x7YnzeQ.jpg
https://i.imgur.com/1YkaaaC.jpg

Any advice would be really appreciated
eddie
 
SMPS are the repair from hell generally. They have a nasty habit of if you don't find and fix every fault then they tend to go bang (sometimes spectacularly!) on switch on and destroy all/most of the parts you just fitted in your efforts up to that point. This is made many times worse by the fact that until they are working perfectly there is nothing to measure! The oscillator, the drivers, the output switch, the feedback system etc all just sit there dead usually.

If you can buy a new SMPS board for it that's what I'd do if I were you;) It will prob end up cheaper and you will not have pulled as much hair out!
 
That echoes my experience with them too. Another culprit is a small, say, 100uF low voltage electrolytic cap near the controller IC. It handles the startup of the chip or smoothing its supply and the thing won't fire up unless it's exactly in spec. It tends to get the brunt of any peaks on the mains and can go out of tolerance readily.

I fixed a switch mode a couple of weeks ago from a BBC Master micro I had that suddenly stopped working. It has two HV capacitors in series (or parallel for 120v) and needs two bleeder/balancing resistors to keep them at half mains at 240v rectified. One of them was open circuit so one cap had the full mains across it and the other had nothing. The controller was driven from a voltage divider on the cap with no voltage. I only found that because the mains side caps should have discharged when removed from the mains. That tingled.
 
Jez Mike
Thanks for the advice . My first thought was to look for a donor spares machine on the bay but they have been gong for silly money. I've also been unable to source a new replacement board.
I was motivated to try and do a component fix on the Toshiba because I managed to successfully fix two SWMP supplies last week one in a Sony SLV T2000 Hi8 VHS combo and one in a Panasonic DVD/VHS combo Both only required capacitors on the COLD side and that ootstrap starter cap on the HOT side. Fortunately I found some info on YouTube detailing both repairs.

I'm inclined to give up on this Toshiba as I suspect the VHS drive mechanism may also have issues. my friend was a bit cagey about its state

eddie
 
Eddie

I'm only down that hill from you as I recall; and have a spare/expendable old Philips 20Mhz o'scope if that'd help you hunt for live signals. Drop me a pm if it would be any help.
 
Martin

Thanks for the offer.

I've splashed out on one of these cheap as chips component tester devices which if the spec is to be believed can actually measure signal frequency /pulse width etc. Sounds a bit too good to be true but I'm looking forward to trying it out when it arrives. This SWMP might be just the thing to try it out on. I'm going to look for some well insulated probes, 330V DC is not to be trifled with. Do you have any idea what primary AC voltage magnitude these SMPS's transformers generally work at ?

http://rover.ebay.com/rover/1/710-5...0001&campid=5338728743&icep_item=313253742762

What I can't quite understand about this particular supply is there does not appear to be a controller chip unless the MOSFET is able to do both control and switching functions

I clearly have too much time on my hands but we are in lockdown and it has been a bit cold outside

eddie
 
This site contains affiliate links for which pink fish media may be compensated.
I couldn't agree more with your last line!

The SMPS transformer will see input chopped from full-wave rectified mains - so about 340VDC round here. For lower voltages, the duty cycle will generally increase to achieve target v DC output. That makes for a negative input impedance w.r.t Mains AC (which is why they cook-out under low-voltage conditions)

As you say - not to be trifled with. ATB.
 
SMPS...

Clever to reduce the mains transformer into
a choke or similar.

I had a lovely JVC Nicam VCR that died on me,
turned out a psu Zener Diode had gone o/c.

All about PIV or peak inverse voltage...

https://en.wikipedia.org/wiki/Peak_inverse_voltage

Yep. you need a scope and a psu cct diagram
to succeed quickly, once done you will chuckle
and take on anything.

Good Luck.
 
There are some 10 diodes on the HOT side in addition to the chunkymain 4 rectifying diodes which I've checked and are known good hence the 330 volts DC at the cap. In the parts list some of the diodes are listed as "rectifying diodes" while others are listed as "switching diodes" There are also one or two refered to as zener or reference diodes

I was wandering what the difference is between a rectfying diode and a switching diode.?

I think I will open this up again in the next day or so and try to check all the diodes with my multimeter. Hopefully I can verify whether they are open circuit while they are still on the board.

eddie
 
The rectifying diodes on the secondary (plus sometimes some others as flywheel diodes etc) have to be fast or they will burn up trying to switch on and off fast enough (more the off actually) and so they are usually Schottky rectifiers or for higher voltages fast-soft recovery types.
 
eddie pugh said:
There are some 10 diodes on the HOT side in addition to the chunkymain 4 rectifying diodes which I've checked and are known good hence the 330 volts DC at the cap. In the parts list some of the diodes are listed as "rectifying diodes" while others are listed as "switching diodes" There are also one or two refered to as zener or reference diodes
Click to expand...

eddie pugh said:
I was wandering what the difference is between a rectfying diode and a switching diode.?
Click to expand...

Ahh, a little research journey for you. ;)

eddie pugh said:
I think I will open this up again in the next day or so and try to check all the diodes with my multimeter. Hopefully I can verify whether they are open circuit while they are still on the board.

eddie
Click to expand...

You will have to lift one end of them diodes off the board for a truth measurement. IIRC, ;)

At least you don't need to refer to the Truth Table and their boolean functions. :(
 
Darmok said:
Ahh, a little research journey for you. ;)



You will have to lift one end of them diodes off the board for a truth measurement. IIRC, ;)

At least you don't need to refer to the Truth Table and their boolean functions. :(
Click to expand...

Usually not in fact... which is one small mercy! Most will have a capacitor to ground after them.
Beware that Schottky's measure about 0.3-4V as opposed to 0.6 ish for conventional ones and that they have significant reverse leakage compared to normal ones.
 
Arkless Electronics said:
Usually not in fact... which is one small mercy! Most will have a capacitor to ground after them.

Beware that Schottky's measure about 0.3-4V as opposed to 0.6 ish for conventional ones and that they have significant reverse leakage compared to normal ones.
Click to expand...

Oh yeah, well since I retired from the industry, I spend little time on refreshing my memory.

Those caps need to be in good order else your measurements could be misleading?

Have you ever been a CRT TV tech Arkless?

If yes, do you remember those Triplers or in other words Diode Ladders?
 
Darmok said:
Oh yeah, well since I retired from the industry, I spend little time on refreshing my memory.

Those caps need to be in good order else your measurements could be misleading?

Have you ever been a CRT TV tech Arkless?

If yes, do you remember those Triplers or in other words Diode Ladders?
Click to expand...

I've certainly repaired plenty back in the day... and CRT monitors. I still have some triplers and LOPT's lying around in fact.
 
Thanks to you all for all the above. A few things to investigate there and today I received my new GM328 Tester so I will do some more testing .
First off will be the transformer primary. If I understand correctly this toy can inject approx 5v AC at variable frequency up to 2 MHz into the primary so I will try and measure some secondary voltages to see if the transformer is still intact.

I'll update in a day or two

eddie
 
You must log in or register to reply here.


advertisement


Back
Top