Chipset for Astec AA20040 power supply

[ingalopez]

Greetings!. Ive got an Astec AA20040 power supply to repair. Blown fuse, bad mosfet. Yet, after replacing and looking for further damage, nothing happens. No Stby supply!. Unfortunately the chips on this unit have been sanded (!?¡). The stby chip is a DIL16 type. Any one knows?? Thanks

[shakalnokturn]

Hello!

I'm sure you'll get more help if you add a couple of photos (top/bottom), the SMPS gurus will chip-in with some layout based suggestions.

[mariush]

Post some pictures?

Maybe you're confusing the 16 pin chip with the main pwm controller ... a common chip in those days was SG3525 or variations of it ... it's available in dip16 or soic 16

[amyk]

It could also be a TL494. Agree with the above, post pictures (both sides) for more info.

[shakalnokturn]

The 3rd commonly seen PWM IC  in computer PSU's is the KA7500.

[ingalopez]

Thank you there!.
But no, this chip is a self contained PWM pulser. It drives the tranformer (the small, stby one) directly. It even receive the input DC (+300V), very few other external components. A couple of pins have been removed (not drilled on the board). It appears to be in high impedance, on the high voltage sector, but some 400 ohm on the electrolitic capacitor , where the Vcc is expected to appear, but not showing.
Attached are some pics of the board.
The PWM main controller, is mounted on the vertical pcb, near the power elements. The offending chip, is the one on the main pcb.
Both have been erased. What a nasty practice!
There is another vertically mounted pcb, (green) on the edge of the main board,  where the voltage supervision is, I guess.

Hope the potato camera is enough!

[fzabkar]

Could there be a high value resistor between the rectified 300VDC supply and the Vcc pin of the IC? These resistors often go open circuit.

You may need to trace the circuit around the IC if someone is to identify it for you. There may also be markings on the underside of the IC.

[ingalopez]

Nope.
Nothing on the underside. And no resistor either!. I believe this chip  uses internal hivoltage fet to generate its VCC, without the proverbial high value resistor. Like some TOPs and other families do.

[Chris56000]

Hi!

There is one PWM that operates off the HT+ line direct in a 16 pin DIL package – it is the Microchip/Supertex HV9120PG, but unfortunately it's e.o.l. and supplies are getting very scarce, and its very likely Astec used that one!

Unfortunately it's internal startup feed from the HT+ input at pin 1 does have a habit of going open–circuit internally, resulting in the chip sitting there with 300 V on pin 1 and dutifully ignoring it !

I came across these in some large French made battery charger units that were made for railway rolling stock, and it puzzled me for days that there was no obvious means to start these units up internally until I twigged the startup was inside the device itself!

You might be able to find it in 16 pin SOIC and use an SOIC–DIL adapter tho!

The AA20040 is a standard format 145W ATX PC type power supply unit and by the time you've found a '9120 it'll have cost more than TWO of those complete units new!!


Chris Williams

[ingalopez]

Thank you Chris!

That seems promising, since the hiV is actually applied on pin 1, as in your case, and seems to be internally disconnected, as well!
Little hope of finding such rare chip down here, but I´ll check the datasheet to see if it fits.

Alejandro

[Chris56000]

PS!

https://www.ebay.co.uk/itm/402028588296?mkcid=16&mkevt=1&mkrid=711-127632-2357-0&ssspo=M_wP2V8nR8K&sssrc=2349624&ssuid=qNpFa7hUTgy&var=&widget_ver=artemis&media=COPY

Postage might be expensive tho!

Chris Williams

[coromonadalix]

i think this is the same ic datasheet

https://www.ibselectronics.com/ibsstore/datasheet/Vishay/si9120.pdf

attached too

[fzabkar]

I'm wondering whether the failures in this IC could be a consequence of a failure in the bootstrap VCC supply, eg a dried out electrolytic.

Note: When driving large MOSFETs at high frequency without a bootstrap VCC supply, power dissipation in the pre-regulator may exceed the power rating of the IC package. For operation of +VIN > 250 V, a 10-k, 1/4-W resistor should be placed in series with +VIN (Pin 1). For +VIN > 380 V, a 15-k, 1/4-W resistor is recommended.

[Chris56000]

Hi!

Yes, very possibIy!

The Vishay Si9120 is electronically similar to the HV9120 so if you can locate one it should get it going!

Worth saving a search on eBay in case another the same comes up cheap as a "parts mule" (add filter "for parts or not working" so you don't have to plough through thousands of listings of new units) !

Might be worth de–soldering the vertical sub–panel from the main PCB so you can check the components on it more easily!

It's also trying a search for "AA20040 UL Report" – a "UL Report" which stands for the Underwriter's Laboratories", have to receive a report on each new model from the OEM detailing how the proposed design comies with UL safety requirements, so the UL can then test and certify these supply units before they go into quantity production – these reports sometimes have a circuit diagram of the unit in an appendix at the back!

Chris Williams

[shakalnokturn]

Depending on whether the transformer has an auxiliary winding to supply the PWM:

Go for the cheap and nasty 12V zener between Vcc pin7 and GND, then appropriate high value resistor to rectified mains or rectified mains/2 depending on capacitor configuration. Or capacitive divider from AC. That should run in all cases.

If there is an aux. winding and you want to be less energy wasteful add in an extra diode and capacitor from the aux. supply and use that to switch a high voltage depletion MOS on the above setup closely emulating the original internal function.

[ingalopez]

I´ve looking at the 9120 specs, but it doesn´t include the power mosfet to drive the Stby Xfmr, as the offending chip in my board.
As you said, this power supply is kind of a standard ATX power supply.  Perhaps is just a bin case, after all.

[ingalopez]

I may try that, out of desperation.
But I don´t think the rest of the chip is active... measure a few hundred ohms on the VCC line, it may be deceased altogether.

[ingalopez]

Find another such power supply and the chips aren´t blanked!

The chip for the stby power supply is the MC33363P, a Motorola design. It includes the switching device (700V 1 A) and the start up mosfet.

The primary side PWM controller is an Astec AS2214.

Now I will check if those are available .
Than you all!