Cmoy distortion after 3sec

[Thomazz]

Hi guys, I've build 1 cmoy last week but encountering distOrtion issue after power on for few seconds.

NJM4562
100k input res
0.47uf input cap
3.3k res feedback loop
1k res R3
NO R5

Power Supply:
Single 9v batt
Dual 330uF elec cap
0.1uF bypass cap

I din use any pot or switch since it is still on the breadboard.


I try every method I found with Google but still the same....
Chnaged new batt. Added R5. Switch OPamp to 2134. Change input cap to 0.1.

Nothing works!!! help please

Sorry for bad english

[agdr]

Did you make the power supply like in figure 3 here, with the two capacitors in series and the two 4.7k resistors across the capacitors?

HeadWize - Project: A Pocket Headphone Amplifier by Chu Moy

The "+4.5V" from that diagram should be going to pin 8 of the NJM4562 or OPA2134, the "-4.5V" going to pin 4, and the middle "ground" connects to one side of R2 and R3 as in figure 1.

What brand and model number of headphone or IEM are you using?

The NJM4562 probably isn't the best chip to be using since it has a minimum supply voltage of +/-4.0Vdc and you will only get +/-4.5Vdc from your power supply. And that is with a fresh battery - it will drop of course as your battery runs down. Your OPA2134 would be a better choice with its minimum supply voltage of +/-2.5Vdc. That gives you a lot more supply voltage headroom as the battery wears down.

Also, you probably already know, but if you are using pins 2 and 3 for the input then pin 1 is the output, not pin 6 as on the original diagram.

[Thomazz]

None. I skip the 4.7k res. Other 2 of my cmoy works normally with skipping it.

Tried 2134. Same result. Distortion comes after power on for few seconds. The cmoy works and sounds normal only for a few seconds....

Every pins are connected accurately. I am very sure that none of then has been mixed up.

Ya. Since it is dual channel

Originally Posted by agdr

Did you make the power supply like in figure 3 here, with the two capacitors in series and the two 4.7k resistors across the capacitors?

HeadWize - Project: A Pocket Headphone Amplifier by Chu Moy

The "+4.5V" from that diagram should be going to pin 8 of the NJM4562 or OPA2134, the "-4.5V" going to pin 4, and the middle "ground" connects to one side of R2 and R3 as in figure 1.

What brand and model number of headphone or IEM are you using?

The NJM4562 probably isn't the best chip to be using since it has a minimum supply voltage of +/-4.0Vdc and you will only get +/-4.5Vdc from your power supply. And that is with a fresh battery - it will drop of course as your battery runs down. Your OPA2134 would be a better choice with its minimum supply voltage of +/-2.5Vdc. That gives you a lot more supply voltage headroom as the battery wears down.

Also, you probably already know, but if you are using pins 2 and 3 for the input then pin 1 is the output, not pin 6 as on the original diagram.

[agdr]

Originally Posted by Thomazz

None. I skip the 4.7k res. Other 2 of my cmoy works normally with skipping it.

You might try putting those 2 4.7k resistors in on this one and see if it solves the problem. Those resistors prevent the two series capacitors from charging unequally and developing different DC voltages across them. Usually they can't be skipped. The resistors are what is setting your DC point at 1/2 of the 9V, while the two capacitors are just acting as AC bypasses around the resistors for the signal current. The op amp also needs a DC path back to ground for the input bias current via R2 and R3, which capacitors alone for the power supply would block.

What may be happening is after a couple of seconds one of the series capacitors charges more than the other and your "ground" winds up off the target by a few volts in one direction or the other, which will move your output DC bias off center (away from 1/2 supply) causing the output waveform to clip on one rail or the other.

The series capacitors in your other two amps likely were just well matched, by chance. If by luck the two capacitors are very well matched the center point will be around 1/2 of the supply initally, but it won't be nearly as stable with load as with the two series resistors included. In your other two amps the small amount of leakage current through the electrolytic capacitors may be what is supplying the DC input bias current needs for the op amp. Here is another article about it, the second section on "resistor divider":

http://tangentsoft.net/elec/vgrounds.html

[Thomazz]

Tried with 4.7k on. But no luck. Wondering which part cause the prob.

I tested with denon d2000, hd650, apple earpiece, w3 all distort like hell.

[agdr]

I would recommend leaving the 4.7k's in there, even if they are not causing the distortion problem. You will get better results in the end.

I've been thinking of getting a pair of D2000s or D5000s! Nice phones. Unfortunately the 25 ohm impedance on those may not work that well with a CMOY. Chu Moy gets into that down in that "appendix 1" section of the main article. His fixes for that issue, the "load resistor", is really just limiting the available output volume and potentially affects the frequency response a bit. Low impedance headphones will cause the output DC bias point to shift when that simple virtual ground power supply gets unbalanced, which in turn will put DC across your headphones.

For nice headphones like your D2000s you might want to just give up on the CMOY and look at something like an O2 headphone amp in one of the threads here

The Objective2 (O2) Headphone Amp DIY Project

NwAvGuy: O2 Details

That one does have the necessary current output to drive your 25R headphones. The fellow who designed it even has published measured results into low impedance headphones. There is a fellow in Germany who is selling kits of parts worldwide in the group buy thread here

O2 Amplifier Kit & PCB Groupbuy, Worldwide

And a fellow in England who ran the last group buy is selling a completed amplfier, along with a fellow in Canada I believe. They have posts in that forum too.

Anyway, back on your distortion problem with this CMOY. At this point if it were me I would test each and every part (resistor and capacitor) with a multimeter to make sure the value is what I'm expecting, or alternatively substitute in another similar part. Could you post a photograph of your build? Maybe I or someone else here can spot something.

Some voltage readings would also help, if you have a meter. Measure from ground to each of pins 1, 2, 3, 4, and 8 on the op amp.

[Thomazz]

Planned to Build an O2 and MiniMax soon. My currect project is with the Cmoy with EQ. So i started on a breadboard. But this distortion pops up before the equilizer circuit, which is too bad.

The breadboard is damn messy. I guess i am the only 1 who can read it. LOL.

Get u update for the Voltage from each pin once i reach home.

Originally Posted by agdr

I would recommend leaving the 4.7k's in there, even if they are not causing the distortion problem. You will get better results in the end.

I've been thinking of getting a pair of D2000s or D5000s! Nice phones. Unfortunately the 25 ohm impedance on those may not work that well with a CMOY. Chu Moy gets into that down in that "appendix 1" section of the main article. His fixes for that issue, the "load resistor", is really just limiting the available output volume and potentially affects the frequency response a bit. Low impedance headphones will cause the output DC bias point to shift when that simple virtual ground power supply gets unbalanced, which in turn will put DC across your headphones.

For nice headphones like your D2000s you might want to just give up on the CMOY and look at something like an O2 headphone amp in one of the threads here

The Objective2 (O2) Headphone Amp DIY Project

NwAvGuy: O2 Details

That one does have the necessary current output to drive your 25R headphones. The fellow who designed it even has published measured results into low impedance headphones. There is a fellow in Germany who is selling kits of parts worldwide in the group buy thread here

O2 Amplifier Kit & PCB Groupbuy, Worldwide

And a fellow in England who ran the last group buy is selling a completed amplfier, along with a fellow in Canada I believe. They have posts in that forum too.

Anyway, back on your distortion problem with this CMOY. At this point if it were me I would test each and every part (resistor and capacitor) with a multimeter to make sure the value is what I'm expecting, or alternatively substitute in another similar part. Could you post a photograph of your build? Maybe I or someone else here can spot something.

Some voltage readings would also help, if you have a meter. Measure from ground to each of pins 1, 2, 3, 4, and 8 on the op amp.

[agdr]

Here is where those voltages should sit. All taken with the amp input connected to ground, the headphone plugged in, and into the 3+ seconds after the distortion starts.

ground to pin 1 = 0Vdc
ground to pin 2 = 0Vdc
ground to pin 3 = 0Vdc
ground to pin 4 = -4.5Vdc
ground to pin 8 = +4.5Vdc

The first sim plot below is with 0V signal in, the second with 1Vpeak in.

[Thomazz]

[QUOTE=agdr;2808794]Here is where those voltages should sit. All taken with the amp input connected to ground, the headphone plugged in, and into the 3+ seconds after the distortion starts.

ground to pin 1 = 0Vdc
ground to pin 2 = 0Vdc
ground to pin 3 = 0Vdc
ground to pin 4 = -4.5Vdc
ground to pin 8 = +4.5Vdc

The first sim plot below is with 0V signal in, the second with 1Vpeak in.[/QUOTE

without headphone
8.19V batt

1=191.6mV
2=17.5mV
3=17.3mV
4=3.94V
5=17.4mV
6=17.3mV
7=190.8mV
8=3.75V


with Denon D2000
8.03V batt

1=15.3
2=1.4
3=11.4
4=6.96
5=11.4
6=1.4
7=15.3
8=0.79v

[Thomazz]

Added R5 for 100ohm and 330ohm.

Both gimme same reading from the meter. Distortion....