The Objective2 (O2) Headphone Amp DIY Project

I've built second O2 with the same parts except op amps to see if it will have similar problems like first one. After some experience with first one, second is running completely fine, but i still have some noise problems on first one.When i plug my headphones jack half way in i hear loud white noise coming from left speaker. After i plug the jack fully in i hear quiet alternating buzz which sometimes disappear when moving with the potentiometer. Also i noticed that second o2 has different R5 R9 R25 values measured in circuit (same polarity of the DMM leads). I would really appreciate if you could help me troubleshoot this. Thanks.
 
Administrator
Joined 2007
Paid Member
You can't measure resistors in circuit, it doesn't work. If they are physically marked as the correct value then 99.99999% certainty they are fine. If in doubt isolate one end by unsoldering and then measure again.

The noise... test this in stages working back from the outout stage. First test.

1) Remove C13 and C14 and test. It should be absolutely silent.
 
Administrator
Joined 2007
Paid Member
That is very strange indeed.

At this point I would possibly suspect something really odd with the power supply and this is where an oscilloscope really comes into its own.

OK... leave those caps out and now short out the two FET's as shown. There is no need to remove them.

And even better... do you have two batteries to test it on battery power only ?
 

Attachments

  • Capture.JPG
    Capture.JPG
    66.1 KB · Views: 156
Administrator
Joined 2007
Paid Member
It doesn't matter. Where ever it is easiest.

Doing this is to try and prove that its not the FET's rapidly switching and "chopping the supply" although its very unlikely and the rail capacitance would even that out even if they were. But... the noise has to be coming from somewhere. And if its not the supply then there isn't much else left.

Have you used the correct opamps ?
 
Administrator
Joined 2007
Paid Member
i can clearly read the mark so think they are genuine but after switching to 5532 i found out that 49720 are root of the problem. You saved me again Mooly.
So i will need new op amps could you recommend me some?

The (NJM, JRC) 4560 is a typical choice for a headphone driver. The 5532 is OK ish but has a high current consumption which is important for battery use.
 
I grounded J2 Pin1 with the thinnest wire I had, AWG 28, and then ran that to the back of the chassis and screwed down the front and the back with the ground around the bottom back screw on the same side as J2.

With the ground in place, the left channel of the headphones drops out almost completely. If I jiggle or move around the 3.5mm jack in the connector, I can hear sound, but as soon as I let go, the left is quiet again.

I checked the soldering on J2 and they all look good with no bridges. I do have a DMM, but I am not sure what I should be measuring.
 
With the ground in place, the left channel of the headphones drops out almost completely. If I jiggle or move around the 3.5mm jack in the connector, I can hear sound, but as soon as I let go, the left is quiet again.

For any left channel issues with the O2 be sure to check for the one layout error that NwAvGuy made. The front left metal leg of the gain switch (looking at the PC board from the front) can contact a metal via hole directly under it. When that happens the left channel gain goes nutz (like 100x voltage gain) and essentially turns the left channel into a comparator, changing the incoming sound into square waves. Just touching the gain switch or anything in the vicinity, like the input jack, can cause the problem to come and go as that metal leg touches the via on and off.

NwAvGuy and I found that problem at the same time early in the thread. :) I was in the middle of helping someone troubleshoot it for the first time and he realized he had made a mistake with that via placement and did a post about it. The solution is to just cut that leg off the switch. If it is already soldered in just take a small screwdriver blade and bend it under the switch (the metal is very soft), or I found my flush cutters can nip it right off.

This may not be your problem, but I post about this every so often in the thread as a reminder to others who may have left channel issues since that original post is long since buried. :)
 
Hey all,. I stuffed the board for the o2 and began doing the testing described in this post http://www.diyaudio.com/forums/head...eadphone-amp-diy-project-189.html#post3806667.

- Andy (Steve Mannix)

Quoting myself here. I started over with a new board and components, as I decided maybe I damaged the board replacing the U5 and U6 I had installed incorrectly. It went together uneventfully. This time, it passes the tests in the post quoted above for D1, D3, D4 and D5.

Fitting in U2 and measuring the DC voltages on the other empty sockets, I get correct values for pin 4 in each case but incorrect values for pin 8. I get -1 volt or so for pin 8. I replaced U2 with a different one, same result. The results are similar for the sockets for U4, U3, and U1.

At this point I thought I may have a bad Q1, so I replaced it but get the same results.

No components are getting hot and everything seems to be oriented correctly. I checked my solder work and no bridges or cold joints I can detect.

Any thoughts on what could be wrong? Hopefully I'm not missing something obvious here.

Thanks in advance for any ideas.

- Andy (Steve Mannix)
 
If it is already soldered in just take a small screwdriver blade and bend it under the switch (the metal is very soft), or I found my flush cutters can nip it right off.
Thanks agdr. I did not have success getting my flush cutters in there. I bent it instead.

Problem was better, but still not perfect. When grounded and pcb inside chassis, left channel would still drop out intermittently, but not if the pcb was out of the chassis and ground to the chassis. I then started looking at the far right pin of J2 and noticed that it was still long enough to touch the chassis when installed. The right most lead of R7 was also still long enough to touch the chassis when installed.

I cut both back and the problem is now solved.

Now that I have pulled the B2-080 from boxenclosures apart so many times, the heads of the screws are a bit rough and stripped. Anyone know the size for replacement screws?
 
Last edited:
I then started looking at the far right pin of J2 and noticed that it was still long enough to touch the chassis when installed. The right most lead of R7 was also still long enough to touch the chassis when installed.

I had forgotten about that. Probably the other O2 layout error! :) He put the parts a little too close the the side rails and some of those leads will definitely touch if not trimmed really short.

Now that I have pulled the B2-080 from boxenclosures apart so many times, the heads of the screws are a bit rough and stripped. Anyone know the size for replacement screws?

PM sent. :)
 
Administrator
Joined 2007
Paid Member
Quoting myself here. I started over with a new board and components, as I decided maybe I damaged the board replacing the U5 and U6 I had installed incorrectly. It went together uneventfully. This time, it passes the tests in the post quoted above for D1, D3, D4 and D5.

Fitting in U2 and measuring the DC voltages on the other empty sockets, I get correct values for pin 4 in each case but incorrect values for pin 8. I get -1 volt or so for pin 8. I replaced U2 with a different one, same result. The results are similar for the sockets for U4, U3, and U1.

At this point I thought I may have a bad Q1, so I replaced it but get the same results.

No components are getting hot and everything seems to be oriented correctly. I checked my solder work and no bridges or cold joints I can detect.

Any thoughts on what could be wrong? Hopefully I'm not missing something obvious here.

Thanks in advance for any ideas.

- Andy (Steve Mannix)

It does sound like a problem with Q1 or the connections to it.

1) Check that both supplies are present on U2 on pin 4 (minus 12 volts) and pin 8 (plus 12 volts).

2) Check that + 12 volts is reaching the "source" of Q1 (that's the left hand connection on the diagram)

3) Confirm that pin 1 of U2 is at around minus 10 volts or so.

4) Confirm that pin 7 of U2 is around plus 10 volts or so.

If all those conditions are met then check that R24 connects to where it should. Also check that C16 is not blobbed with solder shorting it out (or that its faulty... very leaky or short).

Don't forget to check that the "drain" of Q1 actually connects correctly to all the opamp pin 8's. Make sure that the continuity is there.