Marcus...no problem.
Your readings,
1) Plus 22.8v is correct. Its the AC voltage times 1.414.
Measured again (I'm getting good at this!
2) Should be Minus 22 volts ish, tracking the plus 22 volts. It might be a bit low because of a fault drawing current on that rail. If its minus 20 its OK.
-20.9vdc
3) 11.9 is spot on and within spec.
11.9vdc
4) Again, is that minus 20 volts ? If it is then the regulator U6 is faulty or incorrectly fitted. If it really is plus 20 volts then you have a problem with a failed component somewhere shorting the plus rail to the minus rail.
-20.4vdc
Make sure all the opamps are removed for these tests... and they could well have suffered if they were installed when you had the earlier problem with the diode.
Yes, all opamps have been removed while testing.
If U6 is getting hot and the opamps are removed then that again points to U6 being faulty.
Yes, not much else to go wrong
S1 isolates all the circuitry from the power supply. With it off you should have no other contributing factors. The plus 12 volt rail is correct, the negative rail should mirror it at -12 volts.
More readings if it helps:
Across BT1 terminals: 11.8vdc
Across BT2 terminals: 19.2 vdc
Next step is to get another regulator. I'll let you know if this fixes it. Thanks again for your help.
Still daylight here in the states.
Just a thought, make sure you have the two regulators inserted so the metal tabs are toward the edge of the board. Someone had their U6 in backwards once and had similar results. That sends the input voltage into the ouput pin and things don't go well.Still daylight here in the states.
Thanks. I posted a photo here - looks OK. (Ignore the voltage readings!
)http://www.diyaudio.com/forums/headphone-systems/193977-objective2-o2-headphone-amp-diy-project-75.html#post3796506
Ah, I forgot about your posted picture. Yep, you have it in the right way. Mooly likely has it nailed then, a shorted U6 voltage regulator.
I'll bet I know why it is shorted.
I don't believe in coincidences at all and look for a link to the solder-shorted input pins. RocketScientist / NwAvGuy forgot to include the usual reversed-biased diodes on the two power supply rails that dual +/- power supplies often have. Those diodes clamp the rails to no more than 0.7Vdc, or even better 0.2Vdc with Schottky diodes which I use these days for the purpose.When your input pins to U6 got shorted that effectively put 0Vdc into the negative regulator (input to ground pins shorted), which turned it off. But the positive regulator was still on. That put a fairly large positive voltage on the negative power supply rail via the paths through the various op amps' power pins and resistive dividers like R5 + R9 on the O2 board. Turns out that a fairly small reverse-polarity voltage on the output pin of the regulator chips, like +1.5Vdc or so on the output of the negative voltage regulator, is all it takes to kill them.
The same thing happened a year or so ago in this thread when a fellow accidentally used a DC wall wart instead of AC with his O2. That had the same effect of turning on the positive regulator while leaving the negative regulator off, and it killed U6 too.
Last edited:
OK, thanks for the explanation - always good to know why something failed.
I may order another U5 positive regulator as well to be on the safe side.
I may order another U5 positive regulator as well to be on the safe side.
Ah, I forgot about your posted picture. Yep, you have it in the right way. Mooly likely has it nailed then, a shorted U6 voltage regulator.
I'll bet I know why it is shorted.
When your input pins to U6 got shorted that effectively put 0Vdc into the negative regulator (input to ground pins shorted), which turned it off. But the positive regulator was still on. That put a fairly large positive voltage on the negative power supply rail via the paths through the various op amps' power pins and resistive dividers like R5 + R9 on the O2 board. Turns out that a fairly small reverse-polarity voltage on the output pin of the regulator chips, like +1.5Vdc or so on the output of the negative voltage regulator, is all it takes to kill them.
The same thing happened a year or so ago in this thread when a fellow accidentally used a DC wall wart instead of AC with his O2. That had the same effect of turning on the positive regulator while leaving the negative regulator off, and it killed U6 too.
...I'll bet I know why it is shorted.
No, he knowingly omitted it citing those rev. biased diodes as "optional" components. I had pointed out this design flaw to him in his blog(mid 2011) but can't find the reply/comments now.
Anyways my own O2 has suffered rail failure so one of the battery is not getting charged at all(shows 8.4V) at no load . Without batteries 11.4V & 0.7V at point BT1 & BT2 respectively with O2 OFF. FETs seems OK as they turn off O2 at low voltage. Will post my findings.
Haha, I didn't see that you were from the UK..
EDIT: Forgot to mention that the noise is heard when the RCAs are connected to either P2 or to the pins of J2.
Its hard to say for sure. Without seeing the amp and hearing this noise for real I think you have to try the battery option just to try and eliminate the PSU as a possible cause.
Interesting bit of O2 history!
Yeah the diodes are optional as long as nothing fails. To NwAvGuy's credit through there is probably physically no space left on the board for two more through hole diodes. He would have had to go surface mount under the board for something like that.Hey on your board Mooly's tests here are a great start: http://www.diyaudio.com/forums/head...headphone-amp-diy-project-75.html#post3795994
In your case, since your O2 has been working for a long time, I'll take a guess it is one of NwAvGuy's other mistakes on the board, failure to heat sink the voltage regulators. I pinged in on that via email at least once. Yes, the thermal specifications for those are all the way up to 200 degrees F (his reply), but that doesn't make for a long-life part.
Last edited:
Question about troubleshooting?
I just finished soldering all my parts onto the board, and I've just plugged the adapter into the board. I immediately noticed that U5 was getting ridiculously hot (everything else was fine). I unplugged everything, took out the batteries, and checked the bottom, and it seems there may have been a short for U5. I'm actually not sure if there was a short, but I tried to clean off some of the sodder, and my pcb got scratched.
So now I have a couple questions. Is my U5 dead, and do I need a replacement now? Is my PCB okay with some of it scratched off?
I just finished soldering all my parts onto the board, and I've just plugged the adapter into the board. I immediately noticed that U5 was getting ridiculously hot (everything else was fine). I unplugged everything, took out the batteries, and checked the bottom, and it seems there may have been a short for U5. I'm actually not sure if there was a short, but I tried to clean off some of the sodder, and my pcb got scratched.
So now I have a couple questions. Is my U5 dead, and do I need a replacement now? Is my PCB okay with some of it scratched off?
Hey guys, I have a problem.
I had thought I had finished my board, but then when I plugged in my wall adapter, I noticed my U5 was getting ridiculously hot, and I took out the adapter and batteries.
I looked under the board, and it seems there might have been a short between the pins for the U5. Then, while trying to remove some of the sodder, I seem to have melted/scratched part of the pcb between the pins.
Is my U5 fried if it just gets ridiculously hot everytime I plug in the adapter? Is there any way to check if my other components are safe? And is my PCB still usable?
I had thought I had finished my board, but then when I plugged in my wall adapter, I noticed my U5 was getting ridiculously hot, and I took out the adapter and batteries.
I looked under the board, and it seems there might have been a short between the pins for the U5. Then, while trying to remove some of the sodder, I seem to have melted/scratched part of the pcb between the pins.
Is my U5 fried if it just gets ridiculously hot everytime I plug in the adapter? Is there any way to check if my other components are safe? And is my PCB still usable?
Your PCB is almost certainly fine. Just make that there is continuity between the relevant connections in the damaged area and link any damaged traces with wire.
U5 should run pretty much cold normally. If it gets hot it is either because its inserted incorrectly, it is not the 7912 version or there is a short on the output. Just because its got hot doesn't automatically mean its been damaged but I would replace it anyway.
With the units switch set to OFF you should have plus and minus 12 volts relative to ground on D1 and D5.
U5 should run pretty much cold normally. If it gets hot it is either because its inserted incorrectly, it is not the 7912 version or there is a short on the output. Just because its got hot doesn't automatically mean its been damaged but I would replace it anyway.
With the units switch set to OFF you should have plus and minus 12 volts relative to ground on D1 and D5.
Hey, new to the o2 scene.
Anyways i've finished assembling my unit, unfortunately i've encountered a few issues.
First issue was that the output would pulse, similar to heartbeats. I resoldered most connections and it seemed to solve the issue.
Next thing i know after 5 minutes of use, the amp just goes "Buuumb" and goes silent...
This is the second time it's happening, the first time i removed R4 and resoldered it.
Also i realised one thing, i tried to measure the voltage between the 2 batteries, i'm getting 48V instead of the 24v stated on NwAvGuy's post. and most values i get on my tests are two times what it is on the post.
Also somehow when i measure the resistance of R1 and R2, it seems to give me different values.
PS: I bought the Head-n-Hifi DIY package.
Anyways i've finished assembling my unit, unfortunately i've encountered a few issues.
First issue was that the output would pulse, similar to heartbeats. I resoldered most connections and it seemed to solve the issue.
Next thing i know after 5 minutes of use, the amp just goes "Buuumb" and goes silent...
This is the second time it's happening, the first time i removed R4 and resoldered it.
Also i realised one thing, i tried to measure the voltage between the 2 batteries, i'm getting 48V instead of the 24v stated on NwAvGuy's post. and most values i get on my tests are two times what it is on the post.
Also somehow when i measure the resistance of R1 and R2, it seems to give me different values.
PS: I bought the Head-n-Hifi DIY package.
There should be enough information in the last couple of months worth of postings on here to fix pretty much any fault on this amp.
Check the basics. Is the -/+ 12 volts correct on the two regulator outputs. I can't see how you can measure 48 volts DC anywhere (unless the AC input is incorrect and if that were so then things would be going pop)
Check the basics. Is the -/+ 12 volts correct on the two regulator outputs. I can't see how you can measure 48 volts DC anywhere (unless the AC input is incorrect and if that were so then things would be going pop)
As faults and fault finding issues seem to crop up a lot on this amp I have put together some notes.
Fixing the O2 Headphone amp.
This guide attempts to set out a definite fault finding procedure and should cover most eventualities of non functional units. So just work through the steps measuring and checking as you go.
Checking the supplies...
The unit must first be put into a state that allows all critical voltages to be measured.
Preparation.
1) Remove ALL opamps and the comparator. Thats U1 through to U4
2) Make sure S1 is OFF.
3) Remove any batteries.
Measurement.
1) With your meter set to DC volts and with the BLACK meter lead firmly connected to ground (that is the common junction of C2, C3, C4 and C5 etc) measure the voltage on the striped end of D3. It should be higher than PLUS 15 volts and lower than PLUS 28 volts.
2) Keeping the BLACK lead on ground measure the DC voltage on the NON striped end of D4. It should be higher than MINUS 15 volts and lower than MINUS 28 volts.
If those conditions are OK now check the output of the two regulators. If the above readings were incorrect then you have a problem with either the AC input, a problem with the two rectifier diodes or a problem with incorrect or faulty regulators causing a short or partial short. Make sure that a 7812 is fitted for U5 and a 7912 for U6. Also make sure any electroylitics are fitted correctly with respect to polarity.
Checking the output of the regulators.
The BLACK meter lead is kept on ground unless otherwise stated.
1) Measure the DC voltage on the striped end of D1. It should be 11.5 to 12 volts approximately.
2) Measure the DC voltage on the NON striped end of D5. It should be MINUS 11.5to 12 volts DC approximately.
If you have incorrect readings for the plus and minus 12 volts yet have correct input voltages to the regulators, then the problem can only be either the regulators themselves or incorrectly fitted D1 and D5.
Checking the voltage comparator and switched output rails.
1) Disconnect the AC supply and turn ON S1. Wait for 1 minute to allow the caps to discharge and then fit just U2 in its socket making sure it is the right way around.
2) Apply AC power and ensuring S1 is still ON measure the DC voltages at pins 8 and pins 4 of the remaining empty IC sockets. There should be PLUS 11.5 to 12 volts DC on pin 8 and MINUS 11.5 to 12 volts on pin 4.
If the voltages are low or missing then there is a problem around U2 and the FET rail switches. If they are correct then you should now check the comparator operation (next step below)
Checking the comparator operation and the FET switches. For this test FIT U1 in its socket. This allows a small current to be drawn and allows accurate checking of the rail switching circuitry.
The comparator acts as a low battery cut-off (to protect the batteries from over discharge) and should be checked as follows.
1) Rough test of comparator... Apply a SHORT across R9. This will force the comparators to change state and the DC rails on pins 8 and pins 4 of the opamps (U1 only fitted) should fall to zero.
2) Accurate test of comparator. The comparators change state when the voltage on pin 2 of U2 falls BELOW the voltage across the LED. In other words if there is 2.00 volts on the LED, the rails will cut off when pin 2 falls below 2.00 volts. Replacing R9 with a 47k multiturn preset allows different LED's to used and the trip point to be adjusted accurately by monitoring the combined battery voltage and setting the preset to cut off when the combined battery voltage falls to say to 14 volts (two 8.2 volt batteries containing 7 cells each and discharging to 1 volt per cell).
Checking the audio circuitry.
1) As U1 is now fitted a basic check is to measure the voltage on pins 1 and 7, the opamp outputs. This should be near zero to within a few tens of millivolts.
2) Fit U3 and U4 and again measure the DC voltage on the output pins (1 and 7). Again this should be near zero.
That hopefully covers all the basic testing of your O2. If it passes these tests then its basically OK. Any audio only faults, distortion, hum, noise etc would need testing with an oscilloscope to track down. Built on the proper PCB there should be no such worries.
I hope that covers most aspects of working and testing this great little amp.
Note you can not generally test components like resistors in circuit. Everyone will come up with totally different readings that are meter, meter polarity dependent. Any resisdual voltage from the caps will further influence the readings. If the resistors are correctly fitted according to the colour code then its highly unlikely that any will be a problem.
Fixing the O2 Headphone amp.
This guide attempts to set out a definite fault finding procedure and should cover most eventualities of non functional units. So just work through the steps measuring and checking as you go.
Checking the supplies...
The unit must first be put into a state that allows all critical voltages to be measured.
Preparation.
1) Remove ALL opamps and the comparator. Thats U1 through to U4
2) Make sure S1 is OFF.
3) Remove any batteries.
Measurement.
1) With your meter set to DC volts and with the BLACK meter lead firmly connected to ground (that is the common junction of C2, C3, C4 and C5 etc) measure the voltage on the striped end of D3. It should be higher than PLUS 15 volts and lower than PLUS 28 volts.
2) Keeping the BLACK lead on ground measure the DC voltage on the NON striped end of D4. It should be higher than MINUS 15 volts and lower than MINUS 28 volts.
If those conditions are OK now check the output of the two regulators. If the above readings were incorrect then you have a problem with either the AC input, a problem with the two rectifier diodes or a problem with incorrect or faulty regulators causing a short or partial short. Make sure that a 7812 is fitted for U5 and a 7912 for U6. Also make sure any electroylitics are fitted correctly with respect to polarity.
Checking the output of the regulators.
The BLACK meter lead is kept on ground unless otherwise stated.
1) Measure the DC voltage on the striped end of D1. It should be 11.5 to 12 volts approximately.
2) Measure the DC voltage on the NON striped end of D5. It should be MINUS 11.5to 12 volts DC approximately.
If you have incorrect readings for the plus and minus 12 volts yet have correct input voltages to the regulators, then the problem can only be either the regulators themselves or incorrectly fitted D1 and D5.
Checking the voltage comparator and switched output rails.
1) Disconnect the AC supply and turn ON S1. Wait for 1 minute to allow the caps to discharge and then fit just U2 in its socket making sure it is the right way around.
2) Apply AC power and ensuring S1 is still ON measure the DC voltages at pins 8 and pins 4 of the remaining empty IC sockets. There should be PLUS 11.5 to 12 volts DC on pin 8 and MINUS 11.5 to 12 volts on pin 4.
If the voltages are low or missing then there is a problem around U2 and the FET rail switches. If they are correct then you should now check the comparator operation (next step below)
Checking the comparator operation and the FET switches. For this test FIT U1 in its socket. This allows a small current to be drawn and allows accurate checking of the rail switching circuitry.
The comparator acts as a low battery cut-off (to protect the batteries from over discharge) and should be checked as follows.
1) Rough test of comparator... Apply a SHORT across R9. This will force the comparators to change state and the DC rails on pins 8 and pins 4 of the opamps (U1 only fitted) should fall to zero.
2) Accurate test of comparator. The comparators change state when the voltage on pin 2 of U2 falls BELOW the voltage across the LED. In other words if there is 2.00 volts on the LED, the rails will cut off when pin 2 falls below 2.00 volts. Replacing R9 with a 47k multiturn preset allows different LED's to used and the trip point to be adjusted accurately by monitoring the combined battery voltage and setting the preset to cut off when the combined battery voltage falls to say to 14 volts (two 8.2 volt batteries containing 7 cells each and discharging to 1 volt per cell).
Checking the audio circuitry.
1) As U1 is now fitted a basic check is to measure the voltage on pins 1 and 7, the opamp outputs. This should be near zero to within a few tens of millivolts.
2) Fit U3 and U4 and again measure the DC voltage on the output pins (1 and 7). Again this should be near zero.
That hopefully covers all the basic testing of your O2. If it passes these tests then its basically OK. Any audio only faults, distortion, hum, noise etc would need testing with an oscilloscope to track down. Built on the proper PCB there should be no such worries.
I hope that covers most aspects of working and testing this great little amp.
Note you can not generally test components like resistors in circuit. Everyone will come up with totally different readings that are meter, meter polarity dependent. Any resisdual voltage from the caps will further influence the readings. If the resistors are correctly fitted according to the colour code then its highly unlikely that any will be a problem.
Hi,
I'm from Penang and have build more than a dozen of O2.You can reach me and work it out for this issue.
Seems to me it's related to regulator failure.