Hi!
I'm building an LM3886 amplifier, and now I'm about to test it.
I got an psu with around 35V after the bridge (24V secondary).
But what do I do with the input on the amp while measuring the dc offset? In Peter Daniels guide he says "(input shunted to ground)". What does that mean?
I'm building an LM3886 amplifier, and now I'm about to test it.
I got an psu with around 35V after the bridge (24V secondary).
But what do I do with the input on the amp while measuring the dc offset? In Peter Daniels guide he says "(input shunted to ground)". What does that mean?
Connected. Input connected to ground.
It's a fancy way of saying 'with a parallel bypass to ground', since the input appears as a resistance between 'the input' and ground.
Old school.
w
Got to keep these youngsters in line.
It's a fancy way of saying 'with a parallel bypass to ground', since the input appears as a resistance between 'the input' and ground.
Old school.
w
Got to keep these youngsters in line.
Yes put a wire between input positive and input negative. Power the amp and measure the DC-offset on the output. It should be between 0mV and up to about 100mV. I think 100mV is high but acceptable.
Test your power supply first but I read on Faktiskt.se that you already have done that, so just connect the amp and go 🙂 Some people recommend that you use a "Light bulb" in series with the power. Search in this forum if you want to use it.
//CJ900RR aka Bullit på faktiskt.se
Test your power supply first but I read on Faktiskt.se that you already have done that, so just connect the amp and go 🙂 Some people recommend that you use a "Light bulb" in series with the power. Search in this forum if you want to use it.
//CJ900RR aka Bullit på faktiskt.se
Thanks! Now I understand!
I'll try the amps today, so I will come back with results both here and on Faktiskt!
I'll try the amps today, so I will come back with results both here and on Faktiskt!
Well.. Something is wrong. One of the amps show 0V offset and the other show something like 35V..
The one showing 0V, have you connected it to a source and if so, is there music?
The other one showing 35V must be something really wrong. Any pictures, close-up and good resolution would help. Also schematic.
The other one showing 35V must be something really wrong. Any pictures, close-up and good resolution would help. Also schematic.
CJ900RR said:
Yes, there is music (and good music, Dropkick Murphys)!! 😀 😀 😀
What does it mean that the offset is 0V? (What is the offset anyway?)
Schedule
An externally hosted image should be here but it was not working when we last tested it.
Pics of the board
Can you see something thats wrong (except from my bad soldering..)?
Iv'e checked with a multimeter, and as far as I can see there is nowhere the signal is cut. The IC or the capacitators can be the problem, if there is nothing else wrong.
Offset Clarification
The DC voltage with respect to ground, at the output of the amplifier, should ideally be zero volts.
Owing to the input coupling capacitor, whether or not you ground the input or leave it open, it will make no difference to the DC offset.
Question: Do the amps work? Do they play music and sound ok?
If they do, perhaps the issue is misinterpretation of the method of doing an offset voltage test. If you follow CJ900RR's method verbatim, "Yes put a wire between input positive and input negative. "
that is, a connection from pin 9 to 10, then you short out all the negative feedback that keeps the output voltage correct. Don't do it!
Just leave the signal source disconnected or turned down, and measure the DC voltage across the speaker.
Hope this helps
The DC voltage with respect to ground, at the output of the amplifier, should ideally be zero volts.
Owing to the input coupling capacitor, whether or not you ground the input or leave it open, it will make no difference to the DC offset.
Question: Do the amps work? Do they play music and sound ok?
If they do, perhaps the issue is misinterpretation of the method of doing an offset voltage test. If you follow CJ900RR's method verbatim, "Yes put a wire between input positive and input negative. "
that is, a connection from pin 9 to 10, then you short out all the negative feedback that keeps the output voltage correct. Don't do it!
Just leave the signal source disconnected or turned down, and measure the DC voltage across the speaker.
Hope this helps
I haven't tried both amps with music, but one of them works.
The amp I haven't tried is the one that measures 35V DC-offset. I will try with the input loose tomorrow!
The amp I haven't tried is the one that measures 35V DC-offset. I will try with the input loose tomorrow!
be careful...
the amp with 35 volts of offset...please check the offset voltage with the input open BEFORE you hook up the speaker...my last post might have left that ambiguous.
the amp with 35 volts of offset...please check the offset voltage with the input open BEFORE you hook up the speaker...my last post might have left that ambiguous.
C2 will limit the DC gain to 1, so you would need 35 V DC offset at the input to get the same voltage at the output. That is very unlikely, so forget about DC offset until you have the 35 V fault sorted out. You can desolder R4 to activate the muting to be absolutely sure about that.
Finish to solder all those spliced open wires and isolate them, so that they don't make contact to anything else, when you move the IC or cramp it into a small housing. If one of the rail wires makes contact to the output trace or pin, you will get rail voltage there.
Don't forget to use a heatsink, even for short-time testing. Isolate the heatsink electrically from the IC's heatsinking flag. If you have not used a heatsink yet, you may already have grilled the IC, which could also explain the 35 V at the output. That is rail voltage, not offset and can reach the output, if one of the output transistors has shorted after overheating. Desolder the IC and measure continuity between the output and the supply pins. If you have continuity, replace the IC.
Finish to solder all those spliced open wires and isolate them, so that they don't make contact to anything else, when you move the IC or cramp it into a small housing. If one of the rail wires makes contact to the output trace or pin, you will get rail voltage there.
Don't forget to use a heatsink, even for short-time testing. Isolate the heatsink electrically from the IC's heatsinking flag. If you have not used a heatsink yet, you may already have grilled the IC, which could also explain the 35 V at the output. That is rail voltage, not offset and can reach the output, if one of the output transistors has shorted after overheating. Desolder the IC and measure continuity between the output and the supply pins. If you have continuity, replace the IC.
pacificblue said:
I've used a heatsink and I hope it has done it's job!
pacificblue said:
Ok, I'll do that!
I'll be back tomorrow with results!
Re: Offset Clarification
Opps.
As djoffe has stated, my instruction can be misunderstood. What I meant was; connect input of amp to ground! You should not connect the positive input of the LM3886 to the negative input of the LM3886.
Sorry for that, I blame it on my poor English skills 🙂
djoffe said:
Opps.
As djoffe has stated, my instruction can be misunderstood. What I meant was; connect input of amp to ground! You should not connect the positive input of the LM3886 to the negative input of the LM3886.
Sorry for that, I blame it on my poor English skills 🙂
pacificblue said:
I had continuity between pin 1 and pin 3, and pin 5 and pin 3. It seems that the +V has connection to the output...
I have to order a new IC and then we are back in business!
Re: Re: Offset Clarification
The inverting input (-IN) is connected to signal ground (return).
The non-inverting (+IN) is connected to signal hot (flow).
If you connect +IN to -IN it has the same effect as connecting signal hot to signal ground.
not OOPs.CJ900RR said:
The inverting input (-IN) is connected to signal ground (return).
The non-inverting (+IN) is connected to signal hot (flow).
If you connect +IN to -IN it has the same effect as connecting signal hot to signal ground.
pacificblue said:
I've bought a new IC, and i measured it to be sure, but there is continuity between pin 1 and 3?! 😕 That is the +V and output, right?
Is it supposed to be like that?
In one direction, yes. You should get continuity, if the mulitmeter's com is connected to 3. With the com connected to either 1 or 5 there should be no continuity.
- Status
- Not open for further replies.