Reply With QuoteOriginally Posted by Skepticist
That's a good idea...I'll try that tomorrow.
The Vcc is a bit on the low side but still above the minimum
The old chip was already fairly flat
Maybe try connecting an external 12V supply to that old chip just to see if there's 2V on RFmin
That's a good idea...I'll try that tomorrow.
Wow.....that'll get the ratings up.The Vcc is a bit on the low side but still above the minimum
The old chip was already fairly flat
Maybe try connecting an external 12V supply to that old chip just to see if there's 2V on RFmin
Maybe the Vcc is on the low side, I notice the spec says it has a 2.55v hysteresis.
I measured 9.7v and it looks like it needs 10 to start.
Last edited by loopyloo; 02-08-17 at 01:25 AM.
So I checked out the old chip.
12v supply puts 20mV on pin4. Then I noticed pin4 is 500mV below rail if referenced to Vcc pin.
The new chip has 1.99V on pin4 with it lifted (2volts Yay!), 6.25V if referenced to Vcc pin. The supply measured at the chip is 8.4V ... interesting !
Last edited by loopyloo; 02-08-17 at 11:05 AM.
Very interesting indeed
Makes the internal comparator on pin5 STBY look guilty I'd say if there's absolutely no possibility of external overload
STBY is critical to regulation so I don't think anything good is likely to happen with pin4 RFmin soldered down and pin5 STBY lifted (possibility of runaway high volts on the secondary)
I'd be very tempted to try it though (with a series lamp and TV board disconnected) - seems STBY below 1.24V puts the chip into 'burst mode' which is what you're seeing with it trying to start. If Css C101 can charge up from RFmin 2V it should get into some sort of 'normal' operation - maybe.
EDIT: STBY below 1.24V puts the chip in idle mode - burst only happens when STBY rises 50mV above 1.24V (which is not happening)
Last edited by Skepticist; 02-08-17 at 04:33 PM.
loopyloo (03-08-17)
New chip coming.
Wouldn't mind putting my oscilloscope around IC101 but I don't have an isolating transformer.
I'd be a bit hesitant about probing around in there with CRO too
Did you do a test with STBY lifted and RFmin soldered down?
Almost up to the 8000 views now - should write a book about this adventure
loopyloo (03-08-17)
No I didn't. Do you think it would be safe enough to try it ?I'd be a bit hesitant about probing around in there with CRO too
Did you do a test with STBY lifted and RFmin soldered down?
Almost up to the 8000 views now - should write a book about this adventure
'Looks' like it will be ok.
STBY <1.24V puts the chip in idle mode
STBY >1.3V is normal running
If the output of the STBY comparator is not changing state due to an internal fault then having STBY disconnected won't alter anything but you should see the RFmin 2V not collapsing if the fault is in that comparator.
If RFmin voltage is still low with STBY disconnected then the fault is not inside the chip I'd say.
loopyloo (03-08-17)
OK talked me into it.
At least I haven't given the wife's sewing needle back yet.
Ok well it didn't change the voltage.
ISEN isn't responding either do you think I could have a faulty C113 ?
Did the RFmin voltage hold up around 2V? and Css also charging up to ~2V?
ISEN is for the primary current sensing so should be next to nothing at very low load, 0.8V will start the current limiting
Nah still no good, but I noticed something.
Rfmin sits at 320mV and CSS sits at 120mV for around 30secs then they both drop to 85mV and 20mV respectively.
This thing just refuses to reveal its innermost secrets so it remains indeterminate as to whether the fault is internal, external or both
The RFmin voltage is considerably better, briefly at least, than before which might make it easier to work out where the extra load is
I'll have a really close look under the chip when I change it tomorrow
Give it a good scratching between all the tracks around the chip in case there's some tracking going on and try probing around with an ohmmeter again in that vicinity just to verify the resistor values
And while it's all de-energised, check over R121, R122, R123, D121 (gate of Q102). There was something odd about the pin 11 (LVG) voltage earlier which I put down to the frequency & duty cycle. I don't have specs handy for that mosfet but Vgs would have to peak at 5V or more. It's probably nothing and I can't see any connection between that and the RFmin loading problem.
Last edited by Skepticist; 03-08-17 at 10:30 PM. Reason: added a bit
Yep I'll do that and also give the pads a good clean with circuit cleaner.
Then I'll put my microscope over it before placing the new chip.
So while I'm waiting for the part to arrive :
This is what I made years ago to remove IC's in the easiest way.
Fashioned out of a piece of copper tube 1/4" or 3/8" can't remember but split with a saw, annealed then flattened and cut to shape.
Last thing is to bend it around the soldering iron tip.
It's just a press fit so it's easy to install and remove til needed again and it doesn't slip when using it.
This one is just a nice width for 16 pin DIL's.
Last edited by loopyloo; 04-08-17 at 12:36 PM.
I mentioned earlier that pin 11 was running at 150mV, now it's 55mV. Dunno why these readings keep changing.
And that FET is :
Last edited by loopyloo; 04-08-17 at 04:56 PM.
Posting Permissions
Bookmarks