BGW750 G power amp has PS warning LED on.

[CalavoBob]

I've got a BGW750G power amp in my shop which has a power supply warning LED that goes on as soon as the amp is powered up. I have a schematic for the B and C models but not the G. Both power rails (+ and -) are a solid textbook 85VDC with about 110 mVpp of sawtooth squeaking through on top of the massive 21,000 uF caps that follow the rectifier bridge when the output is connected to two dummy 8 ohm resistors with no signal applied. There's 780 uV of noise on the left and 900 uV on the right, mostly random hiss, but there's a little 60 Hz in that noise floor. I've removed both amp boards to check them;, but have not found problems with the banks of power or driver transistors, and the few electrolytic on the boards measure fine with no leakage as do the handful of tantalums.
This got me suspecting the board with the warning LEDs but nothing obviously wrong there either. The electrolytics on this board tested ok. There is an LM339 that is used to make that PS warning LED work.They're an obsolete comparator chip and I've seen these go bad in other gear; but usually the LEDs they drive just go dark. I've got an RS339 in my parts bins that I'm considering replacing it with but I'm not positive that it's the same thing as a LM339. I thought I ask the forum if anyone has seem this problem before.

 

[Brian Roth]

This doesn't directly answer your question, but Jameco has LM339 chips in stock. I've had good luck with Jameco for obsolete chips....most recently for 74xx TTL parts.

https://www.jameco.com/shop/Keyword...=10001&search_type_c=jamecoall&freeText=lm339

Bri

 

[JohnRoberts]

does the amp work?

JR

 

[CalavoBob]

It actually seems to work well with no audible distortion or noise and I'm getting what appears to be the rated power on my test bench. It's just that the pesky power supply warning LED stays on. I think I'll try to replace that LM339 with the RS339 chip that I have here.

 

[ccaudle]

LM339 that is used to make that PS warning LED work

Comparators are not that complicated, measure the voltages at the two input pins and the output and see if the behavior indicates a working or non-working comparator. Maybe you are just glossing over what you have checked on the warning circuit, but it almost sounds like you are going to replace the comparator without even checking the circuit to see if the output of the circuit makes sense based on the input voltages.

 

[JohnRoberts]

339 comparator circuits are (or should be ) pretty simple to troubleshoot.

the 339 uses what's called an open collector output. So when the + input is higher than the - input commanding a "high" output state, the output transistor is turned off resulting in a high impedance output.

When the + input is lower than the - input the output is in an active "low" state, which means the output transistor is turned on pulling the output down to the negative supply.

To illuminate a LED the cathode could be connected to the active low output and the anode through a resistor to the + rail. Of course there are other ways to use that open collector output to light LEDs.

This is a good circuit to learn troubleshooting on.

Of course do no harm, don't blow up your working power amp trying to fix a PS light.

JR

 

[Matt Syson]

The LM339 has a few variants and equivalents and should be relatively easy to find. Some of the variants have different tolerances so sometimes those with hi spec variants are suitable in SOME circuits but to simply put an LED on in an amp like this should not need a 'rare' (close tolerance) part. The open collector design means that you can use 2 (or more) comparator stages to then light ONE lamp (LED) so you may have one comparator 'looking ' for a positive DC offset, and a second 'looking' for a negative DC offset.

 

[AudioFurga]

I've got a BGW750G power amp in my shop which has a power supply warning LED that goes on as soon as the amp is powered up. I have a schematic for the B and C models but not the G. Both power rails (+ and -) are a solid textbook 85VDC with about 110 mVpp of sawtooth squeaking through on top of the massive 21,000 uF caps that follow the rectifier bridge when the output is connected to two dummy 8 ohm resistors with no signal applied. There's 780 uV of noise on the left and 900 uV on the right, mostly random hiss, but there's a little 60 Hz in that noise floor. I've removed both amp boards to check them;, but have not found problems with the banks of power or driver transistors, and the few electrolytic on the boards measure fine with no leakage as do the handful of tantalums.
This got me suspecting the board with the warning LEDs but nothing obviously wrong there either. The electrolytics on this board tested ok. There is an LM339 that is used to make that PS warning LED work.They're an obsolete comparator chip and I've seen these go bad in other gear; but usually the LEDs they drive just go dark. I've got an RS339 in my parts bins that I'm considering replacing it with but I'm not positive that it's the same thing as a LM339. I thought I ask the forum if anyone has seem this problem before.

It doesn't look like the LM339 is obsolete.
https://eu.mouser.com/ProductDetail/Texas-Instruments/LM339N?qs=rshUhwi3fbY%2BaR%2BOl3edNw==

 

[Khron]

It doesn't look like the LM339 is obsolete.
https://eu.mouser.com/ProductDetail/Texas-Instruments/LM339N?qs=rshUhwi3fbY%2BaR%2BOl3edNw==

I was thinking / wondering the same - such jellybean parts rarely (if ever) get discontinued "just like that". Granted, possibly the through-hole packages may be getting more scarce nowadays, but still plenty of choice left (and worst case, soic-to-dip adapter boards are widely available, if need be).

 

[Brian Roth]

It doesn't look like the LM339 is obsolete.
https://eu.mouser.com/ProductDetail/Texas-Instruments/LM339N?qs=rshUhwi3fbY%2BaR%2BOl3edNw==

I didn't even check, because I ASSumed the OP was correct about the LM339 being discontinued. Mouser USA shows 10s of thousands in PDIP.

https://www.mouser.com/c/semiconductors/amplifier-ics/analog-comparators/?q=lm339&mounting style=Through Hole&package / case=PDIP-14&instock=y

I did find a RS339 at Digikey, but it was in a SOIC package AND limited to 5V power rails.

Bri

 

[SoundBlaster]