Hi,
I just built 2 LM3886 based amps and am powering it with a powersupply based on carlosfm's design. When i switch it on the speakers just buzz but it goes away when an input is fed. The amps are not in a cabinet and i haven't used a DC blocking cap on the input and the amps are built on strip boards. Is this buzz normal?
Daniel
I just built 2 LM3886 based amps and am powering it with a powersupply based on carlosfm's design. When i switch it on the speakers just buzz but it goes away when an input is fed. The amps are not in a cabinet and i haven't used a DC blocking cap on the input and the amps are built on strip boards. Is this buzz normal?
Daniel
does it buzz with no input connected?
or does it buzz with input connected but no music played?
if it buzz'es with the input shorted to ground then you need to source the problem.
How does your lm3886 circuit look?
How much does it differ from the national semi application circuit?
or does it buzz with input connected but no music played?
if it buzz'es with the input shorted to ground then you need to source the problem.
How does your lm3886 circuit look?
How much does it differ from the national semi application circuit?
I've posted a picture of the circuit i used, i dont use Ci and i dont use R IN or R2 instead i use a 56k resistor between pin 10 and pin 9 before R1.
It buzzes when no music is played when i first start it up and then once i start play'n music it is ok even when i stop the music and turn off the pre.
It buzzes when no music is played when i first start it up and then once i start play'n music it is ok even when i stop the music and turn off the pre.
Attachments
You could try doing it like national's application.
You don't need the Rin 25k pot , but try the R1 and R2 just like that and see.
You don't need the Rin 25k pot , but try the R1 and R2 just like that and see.
I don't think the box will make any difference.
I haven't tried the circuit in the way that you did , so i won't know if it is normal.
But i think if you tried the circuit the way that application note shows it might be better.
just do it to one channel en see what happens.
At this stage it could be anything that brings on the noise , even your preamp.
you need to try and cut out and narrow down where the noise are created.
disconnect the input,
short out the input to ground , does it quiet down.
is the noise louder with the preamp plugged out?
I haven't tried the circuit in the way that you did , so i won't know if it is normal.
But i think if you tried the circuit the way that application note shows it might be better.
just do it to one channel en see what happens.
At this stage it could be anything that brings on the noise , even your preamp.
you need to try and cut out and narrow down where the noise are created.
disconnect the input,
short out the input to ground , does it quiet down.
is the noise louder with the preamp plugged out?
danielnaveen said:I've posted a picture of the circuit i used, i dont use Ci and i dont use R IN or R2 instead i use a 56k resistor between pin 10 and pin 9 before R1.
It buzzes when no music is played when i first start it up and then once i start play'n music it is ok even when i stop the music and turn off the pre.
So you have a 56k reference resistor tying the inverting and non-inverting inputs together and no ground reference? That might be the problem if I am interpreting what you mean properly.
R2 is the ground reference, but more importantly it also equals Rf=22k.
The source resistance seen by both inputs should be the same.
The chipamp amp does tolerate a slightly unbalanced pair of source resistances but you are stretching things a bit far.
Because you have removed the DC blocking cap on the the inverting input, it sees 680//22k=660r.
The non-inverting input sees 1K + [Rsource//56k].
If you disconnect the source then that resistance is 57k. This does not equal 660r. Basic ohms law!
You will find that the output offset changes significantly when you re-insert the source equipment.
I think the 57k is acting like an aerial and picking up magnetic/electrostatic fields from the PSU or the household and sending these to the amplifier.
The source resistance seen by both inputs should be the same.
The chipamp amp does tolerate a slightly unbalanced pair of source resistances but you are stretching things a bit far.
Because you have removed the DC blocking cap on the the inverting input, it sees 680//22k=660r.
The non-inverting input sees 1K + [Rsource//56k].
If you disconnect the source then that resistance is 57k. This does not equal 660r. Basic ohms law!
You will find that the output offset changes significantly when you re-insert the source equipment.
I think the 57k is acting like an aerial and picking up magnetic/electrostatic fields from the PSU or the household and sending these to the amplifier.
Thanks for the info i put a 27k in the place of R2 and put another 27k between pin 10 and 9 before R1 . the buzz ahs reduced but its still there i'll try a 22k resister for R2 and see if it improves. I also have a DC of 50mV on one channel and 100mV on the other.
lm3886 humming
i have also made amp with lm3886 but when i switch on it gives big hum but as i connect the input to ic no hum is produced and works fine.is this ok and what should i do to avoid hum?
Hi,
I just built 2 LM3886 based amps and am powering it with a powersupply based on carlosfm's design. When i switch it on the speakers just buzz but it goes away when an input is fed. The amps are not in a cabinet and i haven't used a DC blocking cap on the input and the amps are built on strip boards. Is this buzz normal?
Daniel
i have also made amp with lm3886 but when i switch on it gives big hum but as i connect the input to ic no hum is produced and works fine.is this ok and what should i do to avoid hum?
The Input when connected to a Source is loaded with a low impedance. This is usually between 100r and 1000r.
Try testing your amplifier with a range of dummy test plugs fitted to the input.
Try 0r0, 10r, 1k, 100k, 10M.
Measure the output noise of the amplifier.
Now remove the dummy plug. The input is now loaded with the Zin resistor and maybe the shunting resistor that is intended to short capacitor leakage to Signal Ground.
This combination of loads is usually in the range of 20k to 1M0.
Measure the output noise of the amplifier.
Does the open noise voltage compare to any of the dummy test plug results?
Now insert a long interconnect into the amplifier input. It can be a coaxial screened cable. leave the "other end" open. Measure the amplifier output noise.
How does that result compare to all the others?
Learn to understand your equipment.
Try testing your amplifier with a range of dummy test plugs fitted to the input.
Try 0r0, 10r, 1k, 100k, 10M.
Measure the output noise of the amplifier.
Now remove the dummy plug. The input is now loaded with the Zin resistor and maybe the shunting resistor that is intended to short capacitor leakage to Signal Ground.
This combination of loads is usually in the range of 20k to 1M0.
Measure the output noise of the amplifier.
Does the open noise voltage compare to any of the dummy test plug results?
Now insert a long interconnect into the amplifier input. It can be a coaxial screened cable. leave the "other end" open. Measure the amplifier output noise.
How does that result compare to all the others?
Learn to understand your equipment.
Place a 0.1uF between the power pins (1 and 4). I hope you have done the grounding properly. The ground of Cz and the input ground should be separate. The ground of Cz should directly go to the speaker return ground. I follow three star grounding. One main ground attached to the
chassis where the signal star ground and the power ground connected by separate paths. All the grounds of the decoupling caps go directly to the main ground. The speaker return ground wire and the ground of Cz go to the power ground (at the main power supply cap pin)....I know it is more than necessary but sure success...I just completed one and at first powering up the amp was completely silent...
All the best
Roushon.
chassis where the signal star ground and the power ground connected by separate paths. All the grounds of the decoupling caps go directly to the main ground. The speaker return ground wire and the ground of Cz go to the power ground (at the main power supply cap pin)....I know it is more than necessary but sure success...I just completed one and at first powering up the amp was completely silent...
All the best
Roushon.
I have tried so many different grounding arrangements, I can't recall doing this........... The ground of Cz and the input ground should be separate. The ground of Cz should directly go to the speaker return ground. I follow three star grounding. One main ground attached to the
chassis where the signal star ground and the power ground connected by separate paths. All the grounds of the decoupling caps go directly to the main ground. The speaker return ground wire and the ground of Cz go to the power ground (at the main power supply cap pin)....I know it is more than necessary but sure success...I just completed one and at first powering up the amp was completely silent....
It's another I must experiment with.
Why are you putting anything between the input pins at all? Apart from perhaps a 100pF cap. What is the reasoning for it? I'll be very impressed if there is a reason as it's nearly impossible to think through what might happen. And the input stage of the 3886 is hardly straightforward anyway.
Why are you putting anything between the input pins at all? Apart from perhaps a 100pF cap. What is the reasoning for it? I'll be very impressed if there is a reason as it's nearly impossible to think through what might happen. And the input stage of the 3886 is hardly straightforward anyway.
I put 220pf between the input pins.
Roushon
I have tried so many different grounding arrangements, I can't recall doing this.
It's another I must experiment with.
It is messy, but I took the trouble. Also I put 10uF on the power pins to ground in addition to the big caps and a 470pf (parallel to R2) from input to ground.
Roushon
- Status
- Not open for further replies.
- Home
- Amplifiers
- Chip Amps
- LM3886 Hum