My "audiophile" LM3886 approach

35v offset

Hi everybody
I have built stereo rev A and all the voltages are correct and looking nice excerpt that I have the full dc across the speaker terminals. i used a dummy load which is good. Help, are the lm3886 dead, seems unlikely as there has been no pops and sparks, I have seen a 3875 go pop before, do they die without issueing a death rattle. I have checked everything but my family are fed up of me fiddling with this thing
thanks in advance
 
Re: 35v offset

flodge said:
Hi everybody
I have built stereo rev A and all the voltages are correct and looking nice excerpt that I have the full dc across the speaker terminals. i used a dummy load which is good. Help, are the lm3886 dead, seems unlikely as there has been no pops and sparks, I have seen a 3875 go pop before, do they die without issueing a death rattle. I have checked everything but my family are fed up of me fiddling with this thing
thanks in advance


Check the metal tab of the LM3886 to see the voltage on it. It is well insulated from ground?


George
 
Hi
With regard to the caps that I messed up, it was c9 and c24 and their interconnection with r8,r9,r10,and r31,r32,r33. There seem to be some fine tracks here that mean you have to concentrate. It really teaches you to put aside a day where you are doing that alone, with good light and in one sitting and not snatching moments when you can. We live and learn, but I will mess it up next time as well.
Regards
Steve
 
Hi Flodge,

One of the great things about DIY is making your own PCBs 🙂 Very cool, but one of the harder things to do is get an effective solder mask like you get with a commercially made PCB.

You may already know this but one way to get a pretty decent solder mask is to lightly spray the copper tracks with a coat of clear acrylic spray paint.

The reason this works is the the heat from the iron and the flux found in most solders will "clean" the acrylic away from the copper where you are applying the iron/solder, but the copper will not fow out much beyond the area that is diectly heated. This is a technique I have used with very good success it helps prevent solder bridges and such.

I am very glad it is sorted out.

Cheers!
Russ
 
LOUD Buzz and nothing.

One channel of my Ref C died while playing Christmas music this morning. All voltages look good, except there is 32.4 volts on the output 0.47 resistor. The rail is 33.4.
Something may have fell across a pin somewhere. Guess my Christmas eve will get spent troubleshooting.
That output protection circuit seems to have done it job. The wire in my Lowthers voice coil is about 40 ga. It will not take a lot of dc.


George
 
Re: LOUD Buzz and nothing.

Panelhead said:
One channel of my Ref C died while playing Christmas music this morning. All voltages look good, except there is 32.4 volts on the output 0.47 resistor. The rail is 33.4.
Something may have fell across a pin somewhere. Guess my Christmas eve will get spent troubleshooting.
That output protection circuit seems to have done it job. The wire in my Lowthers voice coil is about 40 ga. It will not take a lot of dc.


George


Sorry to learn of your problem. One thing to check after things are sorted out are the three transistors in the speaker protection circuit.
 
Hi...

I finally got 'round' to powering up my RevC boards last night... and sadly, it seems I can't even 'paint by numbers'. Left channel looks fine, but the right channel has full pos rail voltage going to pin 7 of the lm318n. I can't see any physical differences between the two boards, so I'm guessing its a bad zener. I haven't had a chance to pull it off the board yet to test it (Christmas and all...), but if anyone has any other thoughts I sure would appreciate the input...

thanks
steve
 
Confusing find in My Ref C

Mine, with maybe 200 hours use acted up Saturday morning. Checked and found all voltages okay, except +32 volts on one channels output. The protection relay was holding closed.
Dug into it today, and really found nothing. Cleaned up some solder flux around the LM3886. Found a small blob of solder under one of the jumper wires going to the relay.
Since everything seemed okay, semi reassembled it. On initial power up, it had about 100 mv dc on the channel that had 32 volts before. It slowly dropped. The relay kicked off a couple times.
After five minutes of being on it stabilized. The offset is back to 1 mv, the hum down around 2-3 ma with the inputs unconnected.
Longer powered up, it does the same thing. But now with no dc on the output resistor.
Does this sound like an issue with the LM3886? Replacing the LM3886 looks like a PITA. Hate to go to all that and still have the same issue.
Relay circuit seems fine. The relay will kick back on about 10 seconds after power down momentarily. But it is a lot easier to replace those three transistors.

George
 
thanks

Thanks for the pcb advice, very useful indeed, you guessed right that it is the copper getting grubby on the small tracks.
Now I have got the amp working i have a question as it seems taht a lot of my problems was also due to the opamp. I have one dud one but the diagnosis is rather difficult as the result of putting it into any of the working channels is to produce full voltage across the output. ne5532 produces 16v across the output. The protection circuit is obviously there for a reason but it does make one think twice about connecting it up to my monitor golds.
regards
steve
 
Hmmm I don't think the NE5532 is indicated for the circuit. Try LM318 as used in the schematic and see how that does.

I would not connect the amp to anything until you have < 50mv DC on the output. Really with this amp you will like get < 10mv. When you are ready to give it a go, do yourself a favor and find some cheap speakers to use as test subjects. 🙂

Cheers!
Russ