Rumble 100 SMPS

[lowell]

Thanks Nick. I did. There is not start up resistor in the datasheet. It's tied to the B+. I'm thinking I'll throw a 100ohm in there and see what happens.

https://www.onsemi.com/pub/Collateral/NCP1271-D.PDF

[nickb]

Thanks Nick. I did. There is not start up resistor in the datasheet. It's tied to the B+. I'm thinking I'll throw a 100ohm in there and see what happens.

https://www.onsemi.com/pub/Collateral/NCP1271-D.PDF

Don't do that. You need to figure out what it does. For starters it's in the range 1k-9.99k. since it ends with a two.

[lowell]

Ah ok so SMD is like small capacitor values...IE 102 for 1nf.

[lowell]

Ok so tried 1k 1/2w. It fried. Have a 10k 2w in there now and its smoking. Measured v drop real quick and its 310v.

Did some calculations and even with a 10k in there the wattage is way too high. So it seems to me the PWM is pulling too much current on its HV pin...pin8.

Thoughts? I am willing to order and replace for the experience of troubleshooting SMPS. However, I don't get ANY reading from pin8 of PWM to "hot" ground. Not even on 2M range.

[nickb]

The maximum current thru that pin should only be 6mA (from the datasheet). Sounds like the chip and maybe more is fried. You want to use a fusible/flameproof thin film resistor in that location. It's purpose is to prevent the chip from exploding and possibly causing a fire, I would guess.

[J M Fahey]

Ok so the switching MOSFET is dead short on all 3 pins. It's an 08N80C3. Other than drain current and D/S breakdown voltage, what specs do I need to meet if I were to sub. I don't have any of these in stock but do have a handful of MOSFETS in TO-220.

edit: nevermind, I don't even have any TO-220 N-channel enhancement types with a 80A rating Will have to order.

It´s *eight* Ampere.

Dug in a litle and am surprised at the puny cheesy supply.

NCP1271 is a laptop brick class SMPS driver/controller, meant for relatively inefficient and low powered single sided driven flyback type supplies.
Datasheet example is a 19V 3A supply, go figure, so we are talking 60W DC output.
A 100W amplifier needs at least 150/200W DC power available.
I didn´t recognize in your pictures the switching MOS, where is it (they)?
Didn´t see its heatsink either.

I expect to see *at least* something equivalent to what an old 200W PC supply uses.

To boot, PCsupply hot parts are straight in the wash of an PC fan ... don´t see that here.
Bewildering.

[lowell]

Makes sense that when the MOSFET shorted it took the PWM with it. I'll try replacing the chip...but after that I think I'll call it on this. First gonna remove the chip and make sure something else isn't pulling current through the startup.

The data sheet says maximum current there is 100ma. Where do you see 6ma? Thanks!

[nickb]

Makes sense that when the MOSFET shorted it took the PWM with it. I'll try replacing the chip...but after that I think I'll call it on this. First gonna remove the chip and make sure something else isn't pulling current through the startup.

The data sheet says maximum current there is 100ma. Where do you see 6ma? Thanks!

You are looking at the maximum "blows up if more than" ratings, not the operating conditions. Look under "Startup Current Source".

100mA is the max current you can pump in there under fault conditions w/o destroying it. To me that would suggest a series resistance of at least 3.7K to ensure that is never exceeded.
.
Say you choose a 6.8K, at 6mA it will dissipate 0.25W. So you pick a 0.5W one to act as a fuse. At 6mA it will drop 41V. I don't see a spec for the minimum HV voltage but the current is specified at HV=50V so we can that as the min. IOW, a 40V drop of your supply of (probably) 320V is fine.

[nickb]

It´s *eight* Ampere.

Dug in a litle and am surprised at the puny cheesy supply.

NCP1271 is a laptop brick class SMPS driver/controller, meant for relatively inefficient and low powered single sided driven flyback type supplies.
Datasheet example is a 19V 3A supply, go figure, so we are talking 60W DC output.
A 100W amplifier needs at least 150/200W DC power available.
I didn´t recognize in your pictures the switching MOS, where is it (they)?
Didn´t see its heatsink either.

I expect to see *at least* something equivalent to what an old 200W PC supply uses.

To boot, PCsupply hot parts are straight in the wash of an PC fan ... don´t see that here.
Bewildering.

I've seen this over and over again Juan. Things are just not designed for continuous full power operation ( unless they are made by you ). The justification that is it's music so the average power is quite low.

[lowell]

Juan there are 4 SMD schottky diode rectifiers, clamped under the 2/8" sliver of metal to the left of the toroidal thing that is towards the middle of the board. See the metal piece just to right of it? They're under there, if that's what you're referring to as MOS. Under them is an aluminum block that's thermocoupled to the chassis.

[Enzo]

MOS as in MOSFET, the main switching transistors on the primary side.

[lowell]

Oh ...I had the one MOSFET removed cause it was shorted.

[lowell]

Side note - any suggestions on a largish sheet of thermal silicone? Just to have a round for cutting pieces for different projects. I searched mouser but couldn't find a whole lot of options. And the ones I found were $$. Looking for the best price.

[J M Fahey]

This is a Newark ad.
SIL-PAD K6 300MMX300MM SHEET - BERGQUIST - SIL-PAD K-6, .006", 12"X12" SHEET | Newark element14
Not expensive at all, U$44 for a 12" by 12" raw sheet, you can make 180 LM3886 Silpads out of it so 25 cents each.

[lowell]

Ok so the power amp chip is a TDA8950J. Replaced the PWM ...no help. Still in shutdown.