Why is there a 4700 ohm resistor on the this power supply output?

[Miktor]

Some time ago i broke my logitech Z523 speaker system somehow. I tried looking for faults in the various components, no obvious ones to be seen though. I found out the fuse was dead, took the chance and replaced it with a jumpwire. Blew a fuse in my apartment instead, oh well. I then found out that the rectifier diodes (RL207) are all dead. Not sure if that happened before or after i did the jumpwire stunt. Anyway, i've been trying to find the fault since (if it wasn't the diodes), and just learning the pcb and circuit to learn electronics better in general. And i found a resistor of 4700 ohm shorting out the output. That seems really weird to me.

Here's a picture of the power supply and a very crude and incomplete sketch


and here's the bottom of the pcb, showing the 12V output with the 4,7k resistor.


So why short it out like this? To bleed out caps?

Also if anyone has any idea what might've gone wrong let me know. I've checked a lot of caps (no visual signs of damage), and unsoldered a few components to be sure. But can't find anything faulty beyond the rectifier diodes.

[tautech]

Not so much to bleed caps as a minimum load for SMPS stability.
It will only conduct mA's so don't worry over it.

Welcome to the forum.

[Miktor]

Alright, that makes sense! Thank you

[hushaoxin]

SMPS requires a Minimal load

[abraxa]

I hate to be "that guy" but to be honest, I'm a little scared to see someone work on a mains-powered circuit who doesn't seem to be aware of the purpose a fuse serves. Not trying to sound like a dick but you could've set your apartment on fire if your apartment breaker had malfunctioned. Please don't do this again. The diode was already shorting the input section before you put the wire in and it would've been easy to check the diodes with a multimeter before trying to power it up again.

Also, the tried-and-true SMPS debugging method of inserting a lamp into one of the supply lines would've saved you here, too. If you don't want to keep on blowing fuses in case more things are broken, you may want to consider that approach.

[Zero999]

I admit it, I've done that before and to be honest, it wasn't that long ago. It was a stupid thing to do, perhaps more so for me because I should have known better, having worked on mains equipment and been involved with the design of switched mode power supplies in the past.

If you bypass the fuse, it needs to be replaced with something else to limit the current. One thing you can do is connect an old fashioned incandescent lamp (not LED or CFL) in series with the PSU. If the lamp glows, then you know you have a problem.

Did you replace the rectifiers? Another possibility is the switching transistor.

[medical-nerd]

Hiya

So if I am using a multiple output smps as a generic supply providing +5V, +12V, +24V. -5V of which I may be using one or two outputs at a time, or even have it the outputs disconnected for a while - do I need to put resistors across each output to have a minimum load?

Cheers

Steve

[Zero999]

Hiya

So if I am using a multiple output smps as a generic supply providing +5V, +12V, +24V. -5V of which I may be using one or two outputs at a time, or even have it the outputs disconnected for a while - do I need to put resistors across each output to have a minimum load?

Cheers

Steve

It depends on the power supply. Some can handle driving no load at all, whilst others need a minimum load to remain stable.

[medical-nerd]

Hiya

It is a WEIR HSS180 with outputs +5V / -5V / +12V / -12V / +24V

I have attached a datasheet, can you tell me what to look for on this to determine if a minimum load is required?

Cheers

Steve

[Zero999]

Hiya

It is a WEIR HSS180 with outputs +5V / -5V / +12V / -12V / +24V

I have attached a datasheet, can you tell me what to look for on this to determine if a minimum load is required?

Cheers

Steve

There's no mention of a minimum load current. One thing to note, is that the +12V and -12V outputs are only semi-regulated, meaning the voltage will change more than a fully regulated output, given changes in the supply voltage and load, including current drawn from other outputs.

[Miktor]

I hate to be "that guy" but to be honest, I'm a little scared to see someone work on a mains-powered circuit who doesn't seem to be aware of the purpose a fuse serves. Not trying to sound like a dick but you could've set your apartment on fire if your apartment breaker had malfunctioned. Please don't do this again. The diode was already shorting the input section before you put the wire in and it would've been easy to check the diodes with a multimeter before trying to power it up again.

Also, the tried-and-true SMPS debugging method of inserting a lamp into one of the supply lines would've saved you here, too. If you don't want to keep on blowing fuses in case more things are broken, you may want to consider that approach.

No honestly i understand your concerns. I knew what I was doing, and it was really dumb must admit. But at least I won't ever do that again.

I think I'll go with just buying more fuses, incandescent lamps can't easily be bought here in the EU anymore. Seems like you can get a ton of fuses from China for cheap anyway.

[Miktor]

I admit it, I've done that before and to be honest, it wasn't that long ago. It was a stupid thing to do, perhaps more so for me because I should have known better, having worked on mains equipment and been involved with the design of switched mode power supplies in the past.

If you bypass the fuse, it needs to be replaced with something else to limit the current. One thing you can do is connect an old fashioned incandescent lamp (not LED or CFL) in series with the PSU. If the lamp glows, then you know you have a problem.

Did you replace the rectifiers? Another possibility is the switching transistor.

Yeah my friend suggested some Op-amp current limiter, but it seemed to be limited to low amounts of current. Also i was impatient and dumb.

Haven't been able to replace the rectifier diodes yet, don't have any around. Transistor seems fine? It's not shorted at least.

[james_s]

Surely you can get special purpose or decorative incandescent lamps? Also look on ebay, I've bought 240V incandescent lamps there before.

*Never* bypass a blown fuse! If a fuse is blown you might try replacing it once but if it blows again something is shorted. I would start by checking the diodes and also the chopper transistor, it's normally the only power transistor on the primary side. One or both is/was likely shorted.

[Miktor]

Surely you can get special purpose or decorative incandescent lamps? Also look on ebay, I've bought 240V incandescent lamps there before.

*Never* bypass a blown fuse! If a fuse is blown you might try replacing it once but if it blows again something is shorted. I would start by checking the diodes and also the chopper transistor, it's normally the only power transistor on the primary side. One or both is/was likely shorted.

I suppose i could. I'll look into it. Need to find some rectifier diode replacements though. Also took another look at the chopper transistor, and it has 2 between source and drain. Might've been the culprit in the first place!

[james_s]

That certainly sounds bad, and it's a common failure point on switchmode power supplies so I think it's reasonable to assume that the transistor was the problem. Rectifier diodes are easy enough to find.

[Miktor]

That certainly sounds bad, and it's a common failure point on switchmode power supplies so I think it's reasonable to assume that the transistor was the problem. Rectifier diodes are easy enough to find.

unsoldered it, and now it reads just fine though, must've been some of the rectifier diodes still left on the board messing up my readings. Unless it went open from the heat of unsoldering, but find that unlikely.

[Zero999]

Surely you can get special purpose or decorative incandescent lamps? Also look on ebay, I've bought 240V incandescent lamps there before.

Yes, incandescent lamps for ovens and some electric fires (to make them glow a nice red) are still widely available in the EU, since LEDs and CFLs won't work at high temperatures.

That certainly sounds bad, and it's a common failure point on switchmode power supplies so I think it's reasonable to assume that the transistor was the problem. Rectifier diodes are easy enough to find.

Unfortunately the transistor normally fails short circuit, hence the blown fuse, and it can take out the transformer too, which can be very difficult to rewind.

[james_s]

I have not yet encountered a bad transformer in a SMPS but that's not to say it couldn't happen.

[Damianos]

The weird thing is not the specific resistor but the question about it!

Except of the fuse bypass procedure ...
What is a "dead diode"?
How was the transistor tested? What kind (BJT/MOSFET) is it?
How many rectifier diodes does this circuit have? It was referred that "some of them remain on the board"!
...

I have no intention of discouraging anyone, but it is strange to ask for the last component ... while "the world is burn"!

[Dave]

I have not yet encountered a bad transformer in a SMPS but that's not to say it couldn't happen.

Generally the fuse should blow before the winding inside the transformer gets a chance to overheat and burn the enamel coating... unless you bypass the fuse, of course.

The incandescent lightbulb truly is the best tool for this sort of repair. The filament inside will keep a fairly low resistance while cold and won't reduce the supply voltage too much. Should your circuit overload the supply, it will quickly heat up and raise the resistance to limit the current to a tolerable value.

[Miktor]

The weird thing is not the specific resistor but the question about it!

Except of the fuse bypass procedure ...
What is a "dead diode"?
How was the transistor tested? What kind (BJT/MOSFET) is it?
How many rectifier diodes does this circuit have? It was referred that "some of them remain on the board"!
...

I have no intention of discouraging anyone, but it is strange to ask for the last component ... while "the world is burn"!

Haha, well I was trying to understand the circuit, and I just didn't get that part.

Dead diode in this case means they were shorted out. There were 4 rectifier diodes (not so surprisingly). I of course was planning to remove them before actually connecting it to power again. I just didn't bother while figuring the board out and testing components. However they're all taken off now.

It's a MOSFET of the type FQPF8N60C. I just tested the resistance from lead to lead. All read 0 ohms. Might mean it's dead, open circuit, I'll be sure to test it.

Also the transformer seems to be fine. Reads ~0.6 ohms.

Edit: tested the mosfet on my breadboard, and it seems to be working as it should.

[BMack]

I hate to be "that guy" but to be honest, I'm a little scared to see someone work on a mains-powered circuit who doesn't seem to be aware of the purpose a fuse serves. Not trying to sound like a dick but you could've set your apartment on fire if your apartment breaker had malfunctioned. Please don't do this again. The diode was already shorting the input section before you put the wire in and it would've been easy to check the diodes with a multimeter before trying to power it up again.

Also, the tried-and-true SMPS debugging method of inserting a lamp into one of the supply lines would've saved you here, too. If you don't want to keep on blowing fuses in case more things are broken, you may want to consider that approach.

A good reason to build a "lamp" imo. It's always good to have something to plug in and see if you're drawing excess current before you blow anything back up and have to start over. It's an old trick but it's a damn good trick...the only problem is it's getting harder to find incandescent bulbs.