GNR TS902W no power , help please

[lcdcaps1]

Re: GNR TS902W no power , help please

ok measured the resistance on J101 and J102 and get 000.9 ohms
i get 323.2v on drain on the transistor mounted on the heat sink

u100 is SG6841

Ok, check J101 and J102 for resistance they should have no resistance (they are fusible resistors). Also, check i you have any voltage (again be careful) at any of the 3 pins of the mosfet (or transistor) mounted on the heat sink.

Could you also tell us what is the markings on U100?

[lcdcaps1]

Re: GNR TS902W no power , help please

ok so after doing some research and lots of reading people are suggesting that the SG6841 chip in my case might be blown, so i tested for shorts across the pins and there are non
I tested for voltage on pin 7 on U100 that is SG6841 and get nothing

Is there any way to know if this is bad as there is no voltage getting to this seems to me it is but would like some clarification please
thanks in advance

[momaka]

Re: GNR TS902W no power , help please

Ok so I measured the voltage on the startup cap as you suggested using the ground/negative from the main cap and i get nothing

Are you really getting 0 (zero) volts or just a very low voltage in the 100's of mV range?
That being said, sometimes PCBs have a very thin invisible layer of flux left over from manufacturing over the solder joints. This can make you get false readings sometimes. To make sure this is not happening, always poke the solder joints with a tiny bit of force and movement of the multimeter probes to break any layer of flux that may be left over (but of course, be careful not to slip and short-circuit something - for this reason I recommend "poking" any joints you want to measure the voltage on prior to plugging the power supply in.)

ok so after doing some research and lots of reading people are suggesting that the SG6841 chip in my case might be blown, so i tested for shorts across the pins and there are non
I tested for voltage on pin 7 on U100 that is SG6841 and get nothing

Is there any way to know if this is bad as there is no voltage getting to this seems to me it is but would like some clarification please
thanks in advance

Well, if there really is no voltage on that startup cap / pin 7 (Vdd) of the SG6841 chip, then it is very much possible that the SG6841 chip is bad. However, before we jump to any conclusions, check the following:
- With the power supply connected to the wall, check if there is a voltage on Pin 3 (Vin) - you should get a voltage there (though I don't know what value).

Also, check ALL of the components surrounding the SG6841 chip. I know that is very general instruction, but if one of these components is not right, it could make the SG6841 chip appear dead.

So in particular, check that NONE of the SMD ceramic capacitors (tan rectangular components) show short-circuit or a reading of less than 30 Ohms across. Any that do, please note them here.

The resistors around the SG6841 controller should all read either the resistance that is printed on their code or LESS. If any of the resistors shows a resistance higher than 5% of its printed value, remove it from the circuit and re-check it.

Next, check all of the SMD diodes and Zener diodes. Namely, check Zener diodes ZD102 and ZD103, as well as diodes D103 and D104 for short-circuit or open-circuit in both directions. I appears that D103 and resistor R108 is what supplies power to the start-up cap, and most likely ZD102 protects pin 7 (Vdd) of the SG6841 PWM controller. If D103 or ZD102 are shorted, that would certainly cause the no voltage you're seeing across the start-up cap. There is also a small SMD ceramic capacitor between ZD102 and D103 - definitely make sure that one is not shorted, as that can cause the same problem. And R108 should read close to 2.7 Ohms.

Finally, check the small transistor, Q103. I think this might be used for pulling/regulating the voltage on either pin 3 (Vin) or pin 7 (Vdd). It's probably an NPN-type transistor rather than PNP, but it's hard to say. When viewed with the two pins on the bottom and the single pin on top, the left pin on the bottom should be Base, top should be Collector, and bottom right should be Emitter.
For NPN-type, you should get a diode reading when the red (positive) multimeter probe is on Base and black (negative) probe is on either Emitter or Collector. For PNP, same results, but with the probes inverted.

If all of these tests pass, you probably need a new SG6841 PWM controller.

I am considering to replace the startup cap as the esr is out of spec

Just curious, what ESR value did you get from it? And capacitance?
What about the new/replacement cap you used - what brand and series was it, and what ESR and capacitance did it have?
Also what device are you using to measure caps?
Thanks.

[lcdcaps1]

Re: GNR TS902W no power , help please

Ok so more of an update, I was going through the diodes again and found a small one,
I tested in diode mode and got 0.001 v both ways using diode mode
I have highlighted the diode in a pic attached but i dont have one this to replace, can i replace with another value one
please can someone advise
otherwise i will have to order that will take some time

thanks in advance

ok so after doing some research and lots of reading people are suggesting that the SG6841 chip in my case might be blown, so i tested for shorts across the pins and there are non
I tested for voltage on pin 7 on U100 that is SG6841 and get nothing

Is there any way to know if this is bad as there is no voltage getting to this seems to me it is but would like some clarification please
thanks in advance

[momaka]

Re: GNR TS902W no power , help please

Ok so more of an update, I was going through the diodes again and found a small one,
I tested in diode mode and got 0.001 v both ways using diode mode
I have highlighted the diode in a pic attached but i dont have one this to replace, can i replace with another value one

That's your snubber diode. Definitely not good if you get a short-circuit across it. Take it out of the circuit and verify your readings, then post what you got.

If that diode shows bad readings out of circuit too, remove the two resistors next to it from the circuit as well, then check and post their values as well. Also check the small SMD ceramic cap, C103, on the solder-side for short-circuit (it's in the area of that snubber diode).

As for replacement of that diode... yes you can use a different part number. It just has to match specs of this one. Most likely, it's a 1 to 1.5 Amp ultra-fast diode, rated for 1000V DC reverse voltage. But post the part number of your original diode so we can find out what specs it has.

And also please check the parts I asked you to check in my last post above.

[lcdcaps1]

Re: GNR TS902W no power , help please

Hi
I'll get onto this later and post but if this diode is reading both ways then is this bad.
Also csn this be replaced with a different value diode please. Thanks

I'll recheck the stsrtup cap voltage but i really poked and was getting some voltage in mv. I'll recheck and post
The esr was 2.8 of the faulty one and new one was 2.3
I don't know the brand of the new one i used but they are new.

That's your snubber diode. Definitely not good if you get a short-circuit across it. Take it out of the circuit and verify your readings, then post what you got.

If that diode shows bad readings out of circuit too, remove the two resistors next to it from the circuit as well, then check and post their values as well. Also check the small SMD ceramic cap, C103, on the solder-side for short-circuit (it's in the area of that snubber diode).

As for replacement of that diode... yes you can use a different part number. It just has to match specs of this one. Most likely, it's a 1 to 1.5 Amp ultra-fast diode, rated for 1000V DC reverse voltage. But post the part number of your original diode so we can find out what specs it has.

And also please check the parts I asked you to check in my last post above.

[lcdcaps1]

Re: GNR TS902W no power , help please

ok so the resistor next to the diode is actually a fusible resistor, 100 ohms one
which measured fine in resistance mode

[lcdcaps1]

Re: GNR TS902W no power , help please

ok so the diode tested ok out of circuit

so this diode was ok

[lcdcaps1]

Re: GNR TS902W no power , help please

My startup capacitor I am getting 001.6 mV and it climbs but still stays in the mV range
This is a new startup cap

Are you really getting 0 (zero) volts or just a very low voltage in the 100's of mV range?
That being said, sometimes PCBs have a very thin invisible layer of flux left over from manufacturing over the solder joints. This can make you get false readings sometimes. To make sure this is not happening, always poke the solder joints with a tiny bit of force and movement of the multimeter probes to break any layer of flux that may be left over (but of course, be careful not to slip and short-circuit something - for this reason I recommend "poking" any joints you want to measure the voltage on prior to plugging the power supply in.)


Well, if there really is no voltage on that startup cap / pin 7 (Vdd) of the SG6841 chip, then it is very much possible that the SG6841 chip is bad. However, before we jump to any conclusions, check the following:
- With the power supply connected to the wall, check if there is a voltage on Pin 3 (Vin) - you should get a voltage there (though I don't know what value).

Also, check ALL of the components surrounding the SG6841 chip. I know that is very general instruction, but if one of these components is not right, it could make the SG6841 chip appear dead.

So in particular, check that NONE of the SMD ceramic capacitors (tan rectangular components) show short-circuit or a reading of less than 30 Ohms across. Any that do, please note them here.

The resistors around the SG6841 controller should all read either the resistance that is printed on their code or LESS. If any of the resistors shows a resistance higher than 5% of its printed value, remove it from the circuit and re-check it.

Next, check all of the SMD diodes and Zener diodes. Namely, check Zener diodes ZD102 and ZD103, as well as diodes D103 and D104 for short-circuit or open-circuit in both directions. I appears that D103 and resistor R108 is what supplies power to the start-up cap, and most likely ZD102 protects pin 7 (Vdd) of the SG6841 PWM controller. If D103 or ZD102 are shorted, that would certainly cause the no voltage you're seeing across the start-up cap. There is also a small SMD ceramic capacitor between ZD102 and D103 - definitely make sure that one is not shorted, as that can cause the same problem. And R108 should read close to 2.7 Ohms.

Finally, check the small transistor, Q103. I think this might be used for pulling/regulating the voltage on either pin 3 (Vin) or pin 7 (Vdd). It's probably an NPN-type transistor rather than PNP, but it's hard to say. When viewed with the two pins on the bottom and the single pin on top, the left pin on the bottom should be Base, top should be Collector, and bottom right should be Emitter.
For NPN-type, you should get a diode reading when the red (positive) multimeter probe is on Base and black (negative) probe is on either Emitter or Collector. For PNP, same results, but with the probes inverted.

If all of these tests pass, you probably need a new SG6841 PWM controller.


Just curious, what ESR value did you get from it? And capacitance?
What about the new/replacement cap you used - what brand and series was it, and what ESR and capacitance did it have?
Also what device are you using to measure caps?
Thanks.

[lcdcaps1]

Re: GNR TS902W no power , help please

Hi
I have done all the checks you mentioned

22.68V pin 3
0001.7 mV and climbing

tested SMD Caps surrounding SG6841 chip
all read above 30 ohms

all resistors read ok
one resistor, r130 i dont know what the reading should be
its a h95

there is one resistor, R108. high lighted in attached pic
rating is 2.7 ohm and i get 3.7 ohm, tested in and out of circuit but the same value, is this needed to get this working ?

q103
diode voltage reading
0.632v

All diodes are ok, non are bad
tested all in diode mode

I am using a esr meter, one similar to bob parker design

Are you really getting 0 (zero) volts or just a very low voltage in the 100's of mV range?
That being said, sometimes PCBs have a very thin invisible layer of flux left over from manufacturing over the solder joints. This can make you get false readings sometimes. To make sure this is not happening, always poke the solder joints with a tiny bit of force and movement of the multimeter probes to break any layer of flux that may be left over (but of course, be careful not to slip and short-circuit something - for this reason I recommend "poking" any joints you want to measure the voltage on prior to plugging the power supply in.)


Well, if there really is no voltage on that startup cap / pin 7 (Vdd) of the SG6841 chip, then it is very much possible that the SG6841 chip is bad. However, before we jump to any conclusions, check the following:
- With the power supply connected to the wall, check if there is a voltage on Pin 3 (Vin) - you should get a voltage there (though I don't know what value).

Also, check ALL of the components surrounding the SG6841 chip. I know that is very general instruction, but if one of these components is not right, it could make the SG6841 chip appear dead.

So in particular, check that NONE of the SMD ceramic capacitors (tan rectangular components) show short-circuit or a reading of less than 30 Ohms across. Any that do, please note them here.

The resistors around the SG6841 controller should all read either the resistance that is printed on their code or LESS. If any of the resistors shows a resistance higher than 5% of its printed value, remove it from the circuit and re-check it.

Next, check all of the SMD diodes and Zener diodes. Namely, check Zener diodes ZD102 and ZD103, as well as diodes D103 and D104 for short-circuit or open-circuit in both directions. I appears that D103 and resistor R108 is what supplies power to the start-up cap, and most likely ZD102 protects pin 7 (Vdd) of the SG6841 PWM controller. If D103 or ZD102 are shorted, that would certainly cause the no voltage you're seeing across the start-up cap. There is also a small SMD ceramic capacitor between ZD102 and D103 - definitely make sure that one is not shorted, as that can cause the same problem. And R108 should read close to 2.7 Ohms.

Finally, check the small transistor, Q103. I think this might be used for pulling/regulating the voltage on either pin 3 (Vin) or pin 7 (Vdd). It's probably an NPN-type transistor rather than PNP, but it's hard to say. When viewed with the two pins on the bottom and the single pin on top, the left pin on the bottom should be Base, top should be Collector, and bottom right should be Emitter.
For NPN-type, you should get a diode reading when the red (positive) multimeter probe is on Base and black (negative) probe is on either Emitter or Collector. For PNP, same results, but with the probes inverted.

If all of these tests pass, you probably need a new SG6841 PWM controller.


Just curious, what ESR value did you get from it? And capacitance?
What about the new/replacement cap you used - what brand and series was it, and what ESR and capacitance did it have?
Also what device are you using to measure caps?
Thanks.

[momaka]

Re: GNR TS902W no power , help please

tested SMD Caps surrounding SG6841 chip
all read above 30 ohms

Okay, good. Anything less than 100 Ohms, though?

one resistor, r130 i dont know what the reading should be
its a h95

Heh, that's cause you're looking at it upside down. Turn it around, and it should look more like "564", no? It's the same value as R128 and R129 - i.e. 560 kOhms.

In any case, since you're getting about 22 Volts on pin 3 of the PWM chip, I don't think that R130 resistor is open... but it doesn't hurt to check it again.

there is one resistor, R108. high lighted in attached pic
rating is 2.7 ohm and i get 3.7 ohm, tested in and out of circuit but the same value, is this needed to get this working ?

Sound to me like your multimeter has offset at low resistances. This is typical with cheaper/budget multimeters. When you touch the probes of your multimeter together on the lowest resistance scale, what resistance do you get? If it's 0.5 to 1.0 Ohms (some cheaper meters even read as high as 2.0 Ohms), that would be the reason why your 2.7 Ohm resistor reads as 3.7 Ohms. If that is the case, you shouldn't worry.

Also, 3.7 Ohms for that resistor still shouldn't prevent the circuit from working.

q103
diode voltage reading
0.632v

Alright, that looks like it's okay too.

All diodes are ok, non are bad
tested all in diode mode

Welp... at this point, I think the problem points to a faulty PWM chip then. So time for a new SG6841. Then we can continue from there if the problem still persists. I don't think there's anything else we can test until that's replaced.

On that note, maybe buy two of these PWM chips, in case one turns out a dud or somehow gets blow up while repairing the PSU.

[lcdcaps1]

Re: GNR TS902W no power , help please

Thanks for the reply, really appreciate every ones help here
I dont know how i did not see that h95 resistor

I measured it and got a reading of 564.5 K ohms
is this ok

please let me know anyone, someone
thanks

Okay, good. Anything less than 100 Ohms, though?

Heh, that's cause you're looking at it upside down. Turn it around, and it should look more like "564", no? It's the same value as R128 and R129 - i.e. 560 kOhms.

In any case, since you're getting about 22 Volts on pin 3 of the PWM chip, I don't think that R130 resistor is open... but it doesn't hurt to check it again.


Sound to me like your multimeter has offset at low resistances. This is typical with cheaper/budget multimeters. When you touch the probes of your multimeter together on the lowest resistance scale, what resistance do you get? If it's 0.5 to 1.0 Ohms (some cheaper meters even read as high as 2.0 Ohms), that would be the reason why your 2.7 Ohm resistor reads as 3.7 Ohms. If that is the case, you shouldn't worry.

Also, 3.7 Ohms for that resistor still shouldn't prevent the circuit from working.


Alright, that looks like it's okay too.


Welp... at this point, I think the problem points to a faulty PWM chip then. So time for a new SG6841. Then we can continue from there if the problem still persists. I don't think there's anything else we can test until that's replaced.

On that note, maybe buy two of these PWM chips, in case one turns out a dud or somehow gets blow up while repairing the PSU.

[lcdcaps1]

Re: GNR TS902W no power , help please

OK so replaced SG6841 and no joy
i get 318mv on pin 3
and on pin 7 3v

also my start up capacitor has voltage 025.8mV
no output power

[momaka]

Re: GNR TS902W no power , help please

Thanks for the reply, really appreciate every ones help here
I dont know how i did not see that h95 resistor

I measured it and got a reading of 564.5 K ohms
is this ok

All electronic components have a tolerance for their values. Typically it can be anywhere from 1% to 20%.
Assume most resistors to be 5%, unless they specifically have a lower tolerance specified (i.e. 1%, 2%, etc.). So for that 560 KOhm resistor, anything in the range of 532-588 KOhms is in spec.

For electrolytic capacitors and most inductors, tolerance is usually specified as 20%.

For film and ceramic capacitors, you'll normally see 5-10% parts used.

OK so replaced SG6841 and no joy
i get 318mv on pin 3
and on pin 7 3v

also my start up capacitor has voltage 025.8mV
no output power

Something doesn't make sense here. How can you have only 25.8 mV on the start up capacitor, but 3V on pin 7 (Vdd) of the chip? Re-check your measurements and also re-check all SMD components mentioned previously. This time, please post what resistance/diode readings you get in your post.
For example, post something like this:
D108: forward direction diode test xxx mV, reverse direction diode test yyy mV
R103: zzz Ohms resistance
...
Do this for all components I mentioned above.

That said, also try to familiarize yourself with the SG6841 datasheet to get an idea of some of the values to expect. For example, the minimum *startup* voltage on pin 7 (Vdd) is 16V, and minimum operating voltage for that pin is 10V (page 5, "Under-Voltage Lockout Section".)

[lcdcaps1]

Re: GNR TS902W no power , help please

ok i think i might have found a shorted diode that i missed but im sure i tested before
I checked again but seems that its gone
its d104
0,055v, diode mode both ways

what can i replace this with, can i used one from a old power supply
attached is a pic shown
please help again
thanks

[R_J]

Re: GNR TS902W no power , help please

It looks like the 47Ω resistor is in parallel with the diode. if it is, thats throwing off the reading

[lcdcaps1]

Re: GNR TS902W no power , help please

does it mean i need to replace
i dont think it should show a reading in both directions, right ?

what can i replace with because i cant find any datasheet for it
please

It looks like the 47Ω resistor is in parallel with the diode. if it is, thats throwing off the reading

[lcdcaps1]

Re: GNR TS902W no power , help please

Can i run this power supply board without this diode
i dont have one like this to hand
If i cant run this without it then can i use something else as a substitute please

Please help some one, anyone
Thanks

does it mean i need to replace
i dont think it should show a reading in both directions, right ?

what can i replace with because i cant find any datasheet for it
please

[rddube]

Re: GNR TS902W no power , help please

It looks like the 47Ω resistor is in parallel with the diode. if it is, thats throwing off the reading

What R_J is telling you is that the 47 ohm resistor that is in parallel with the diode is going to give you wrong readings on the diode. To test it fully, you would have to remove it from the circuit.

Also, I am a little curious on how fast you got that SG6841...did you buy it new somewhere or did you take it off a scrap board?

Finallly, I think you need to do what Momaka suggests in post #54 and give us the results because we seem to be hitting a dead end.

[lcdcaps1]

Re: GNR TS902W no power , help please

Hi

Thanks again for the reply
The SG6841 chip i had purchased brand new so had them on hand
I tested all the nearby components as mentioned here and all tested ok.
Tested resistors in resistance mode and diodes in diode mode.

As it is confusing all of you it is to me as well

I will go through and test all again and update thread

What R_J is telling you is that the 47 ohm resistor that is in parallel with the diode is going to give you wrong readings on the diode. To test it fully, you would have to remove it from the circuit.

Also, I am a little curious on how fast you got that SG6841...did you buy it new somewhere or did you take it off a scrap board?

Finallly, I think you need to do what Momaka suggests in post #54 and give us the results because we seem to be hitting a dead end.