I'm sorry, that's a typo on my part. Its just removal of the opamps and comparator U1 to U4. I'll amend the notes.
If you have the regulators out then you should seen the plus 15 to 28 volts on C2 and C4 and the negative 15 to 28 volts on C3 and C5.
When you switch off to refit the regs I would advise discharging those caps first.
If you have the regulators out then you should seen the plus 15 to 28 volts on C2 and C4 and the negative 15 to 28 volts on C3 and C5.
When you switch off to refit the regs I would advise discharging those caps first.
In your original post you say,
However, when I Turn the O2 on, the Voltage drops to 18V and doesn't stay at 24V. I checked all the Joints and reheated them but the Problem still occurs. Do you guys have any ideas what could be wrong?
Does the amp actually work OK ? The unregulated voltages falling under load is "normal". How much it falls depends on how "beefy" the AC adapter is. Small transformers have poor load regulation (not a problem) and the voltage output falls as current is drawn.
The crucial voltages are the plus 12 volts across C6 and the minus 12 volts across C7.
However, when I Turn the O2 on, the Voltage drops to 18V and doesn't stay at 24V. I checked all the Joints and reheated them but the Problem still occurs. Do you guys have any ideas what could be wrong?
Does the amp actually work OK ? The unregulated voltages falling under load is "normal". How much it falls depends on how "beefy" the AC adapter is. Small transformers have poor load regulation (not a problem) and the voltage output falls as current is drawn.
The crucial voltages are the plus 12 volts across C6 and the minus 12 volts across C7.
OK, that Helps.
But i think i have bigger problems here.
Something is broken now i think. After reattaching U5 and U6 and connecting power, both U5, U6 and C7 get really hot... there was also a "spark situation"
when putting them on the Board again.(Capacitors)
Which parts are now definetly broken? If its only U5 and 6 it wouldnt be so bad, otherwise, not so good.
Thx for helping a noob.
Edit: I Haven't connected any headphones or even completed the Amp completely (no U1,U3,U4) because i wanted to do the Testing first. I don't want to reconnect it now, because the U5,6 and C7 are damn hot now. *sigh
But i think i have bigger problems here.
Something is broken now i think. After reattaching U5 and U6 and connecting power, both U5, U6 and C7 get really hot... there was also a "spark situation"
when putting them on the Board again.(Capacitors)Which parts are now definetly broken? If its only U5 and 6 it wouldnt be so bad, otherwise, not so good.
Thx for helping a noob.
Edit: I Haven't connected any headphones or even completed the Amp completely (no U1,U3,U4) because i wanted to do the Testing first. I don't want to reconnect it now, because the U5,6 and C7 are damn hot now. *sigh
Last edited:
No problem but I'm going to have to leave this for tonight.
Check you have the correct regulator in the correct place (the 7812 for U5 and the 7912 for U6). If things are getting hot (it should all run pretty much cold) then as long as the power switch S1 is OFF and no batteries are fitted then we can fix the power supply easily.
If you are still stuck then check each rail in turn.
If you remove D4 (to isolate the negative rail) and U5 and switch on you should have plus 15 to 28 volts across C2 and C4. If that's no good then it can only be a problem around D3 or the caps fitted incorrectly.
If that's OK then switch off and discharge those caps.
Refit U5 and check the 12 volt output. If that is not correct (and the previous step had the correct 15 to 28 volts present) then U5 is faulty.
Once the plus 12 volts rail is OK we refit D4 and follow the same procedure for the negative rail.
More time tomorrow
but I'm going to have to leave this for tonight.Check you have the correct regulator in the correct place (the 7812 for U5 and the 7912 for U6). If things are getting hot (it should all run pretty much cold) then as long as the power switch S1 is OFF and no batteries are fitted then we can fix the power supply easily.
If you are still stuck then check each rail in turn.
If you remove D4 (to isolate the negative rail) and U5 and switch on you should have plus 15 to 28 volts across C2 and C4. If that's no good then it can only be a problem around D3 or the caps fitted incorrectly.
If that's OK then switch off and discharge those caps.
Refit U5 and check the 12 volt output. If that is not correct (and the previous step had the correct 15 to 28 volts present) then U5 is faulty.
Once the plus 12 volts rail is OK we refit D4 and follow the same procedure for the negative rail.
More time tomorrow
Phew I think it's completely gone now... U5,U6 get really hot, (cannot touch them) And there are 3,4V over C7 and 0V over C6. 
Guess I'll have to reorder it. Trying not to mess this up again...
Edit: I also have big trouble with unsoldering the parts. This Pump thing does not work really well. And the sparks I saw cannot be good.
Guess I'll have to reorder it. Trying not to mess this up again...
Edit: I also have big trouble with unsoldering the parts. This Pump thing does not work really well. And the sparks I saw cannot be good.
Last edited:
Here is a recent write-up on the Disappeared One:
NwAvGuy: The Audio Genius Who Vanished - IEEE Spectrum
That NwAvGuy dot com domain they talk about did get renewed a few days ago. But the renewal happened at exactly midnight on the expiration day so it looks automatic.
Here is my DIY version / interpretation of one, FWIW, which I whipped up when it looked like NwAvGuy had disappeared for the long haul last year:
http://www.diyaudio.com/forums/headphone-systems/229934-version-o2-desktop-amp-oda.html
https://drive.google.com/folderview?id=0B67cJELZW-i8VmhVNk5PODNtZnc&usp=sharing
Same output chips in parallel, just more of them and using the SIP package (upright for better heat dissipation) package version. Uses better (lower distortion, lower noise, 600R drive capability vs. just 2K for the NJM2068) gain stage LME49990 chips. Uses the same pot-in-the-middle design but I use a 1K pot instead of 10K and a 4.99K input bias ground return resistors instead of 40.2K to lower Johnson noise. Those changes run 10x the current through the pot vs. the O2 amp and takes full advantage of that higher output current capability of the gain chips (and requires 10x the coupling capacitance). 5 position gain switch, RCA connectors, input select switch, pre-amp out, much lower noise two stage power supply, heat-sinked voltage regulators, power connector in back, CRC power filter, output DC offset zero adjust in the one I'm working on right now, output headphone relay, holes for a Zobel network and external damping factor select switch, and all sorts of other good stuff.
The whole thing fits in a B4-080 aluminum case (or two in a B4-160 case using 2 board with power supply in back) with CAD designs for front and back panels done. With the B4-080 case an ODAC can be mounted on a top slot board, while with the B4-160 it can be mounted on the rear board.
Last edited:
Phew I think it's completely gone now... U5,U6 get really hot, (cannot touch them) And there are 3,4V over C7 and 0V over C6.
Guess I'll have to reorder it. Trying not to mess this up again...
Edit: I also have big trouble with unsoldering the parts. This Pump thing does not work really well. And the sparks I saw cannot be good.
Try using solder braid for desoldering (and make sure your iron is up to the job).
Solder braid,
http://www.diyaudio.com/forums/parts/127924-working-smd-how-do-without-specialised-tools.html
The 7812 and 7912 should be available from many outlets. When you build it up, do as I outlined earlier. Concentrate on just one rail at a time and make sure you discharge the caps each time you work on it (at least until you have the regs fitted and working)
It will work. What you will notice is that the "feel" of the control will be different because it works into the 40.2k resistor (R12 and 13). In other words the level vs rotation of the pot will be different. Also 100k will roll off the hf earlier than a 10k due to the input capacitance of the opamps. In practice its probably not a problem at all.
It's a 50hz high pitch noise.
I solve this problem by connecting input somewhere in the middle of the pot so that it becomes a ~12k pot. Adjustment is pretty bad now...1st step is my normal listening level I guess I could customize the steps..I can use resistors of the remaining ~11 steps. How would you go about solving this, Mooly?
I solve this problem by connecting input somewhere in the middle of the pot so that it becomes a ~12k pot. Adjustment is pretty bad now...1st step is my normal listening level
I guess I could customize the steps..I can use resistors of the remaining ~11 steps. How would you go about solving this, Mooly? Attachments
It sounds like the pot/and/or wiring is picking up stray interference. Also, if you have built the O2 on anything other than the official PCB then you need to ensure your wiring and grounding scheme is OK.
What would I do I'd use a good quality conventional pot. I don't like stepped attenuators for these very reasons, prone to noise pickup and too course in resolution.
What would I do
I'd use a good quality conventional pot. I don't like stepped attenuators for these very reasons, prone to noise pickup and too course in resolution.@Mooly: OK, I'll go get some of this solder braid. And a better soldering iron.
What would be the best way to discharge the Capacitors? Just use a resistor (which size) or just turn S1 off and wait for a few minutes?
I am starting a new hobby here, so I guess I have to spent a little money for equipment... but I like it.
What would be the best way to discharge the Capacitors? Just use a resistor (which size) or just turn S1 off and wait for a few minutes?
I am starting a new hobby here, so I guess I have to spent a little money for equipment... but I like it.