This one has. One of the bottom pins, relative to the board, does not hug the opamp properly. The resistance from opamp pin to socket pin must change unless I apply pressure on the top of the opamp or move the opamp around in the socket. However, the source of this problem could very likely be from the fact that, during assembly, some of the socket insert pins were raised and I had to push them down. I may have damaged them in the process. I am buying a new socket tomorrow at my campus EE shop and will report back the results.
Thanks! Never mind--I was wrong but very close to the problem. It was either R22 or C20 that had a cold solder joint. I fixed it! Another lesson learned. Now, to enjoy mind blowing clarity.
My O2 kit from JDS Labs came this week. I've read NwAv Guys notes and 6L6's build guide so am forewarned about the corner of S2 and positioning of R10. Any other obvious pitfalls, words of wisdom or advice? I just started to read this thread (349 pages?) and it seem that a lot can go wrong. I'm kind of spooked. I've built a few small project but am mostly a "paint-by-numbers" person. Maybe I'll read a bit more before I jump in.
Last edited:
Don't be scared, most folks have built these with absolutely no problems. The via near S2 is the only real "gotcha", and you just need to clip off or bend that leg. Like any project, take your time and pay attention to correct part placement/orientation, and everything should just work. I built two, and they both worked just fine with no hiccups.
Some of the more unusual O2 build problems
Here are some of the more unusual O2 build problems that have cropped up in those 350 pages.
And some personal recommendations..
Here are some of the more unusual O2 build problems that have cropped up in those 350 pages.
- The power adaptor has to be AC, not DC. Double check that the label says "AC" for the output and not "DC". With DC only one power supply rail is active and that has killed the NJM4556A chips before.
- The power adaptor can be bad or not powered. If you LED does not light at all with the O2 switch on, put your DMM on AC volts and verify that you are getting 14 - 20VAC out from the tip. A plugstrip the adaptor is plugged into can be off, wall AC switch off, etc.
- You only have one shot at getting the polarity (orientation) of the power rectifiers and filter capacitors correct: D3, D4, C2 - C5. If you get D3 or D4 reversed it will change the polarity on the rail and cause some of C2 - C5 to quickly overheat, and possibly even shoot the metal can off like a little missle. If the polarity of C2 - C5 get reversed the same overheating will happen.
- Be careful not to set the board down on wire lead clippings then power it up. Surprisingly easy to happen during a build. The wire clippings short out random things on the back of the PC board.
- You MUST use a red LED, at least without making circuit changes in the values of some resistors. The other LED colors won't work (including white) since their forward voltage is different. The O2 uses that forward voltage as a reference for the comparator.
- There are two different voltages of "9V" NiMH cells. The common one that the O2 is designed for is the 7 cell version, which charges up to about 7 * 1.45Vdc = 10.15Vdc, with a nominal voltage of 7 * 1.2V = 8.4. Those are the blue Tenergy batteries that RS has in his photos and most other "9V" NiMH on the market. Harder to find are the 8 cell version that runs at 8 * 1.2Vdc = 9.6vdc nominal. The 8 cell version won't work in the O2 without a modification.
- There is a version of the Alps 9mm 10K pot that has a switch on the back and just happens to fit perfectly in the hole pattern on the O2 board. When the switch is turned on it will short out some of the traces. If you are using the exact part numbers that RS has listed in his BOM you will be OK.
- The problem likely isn't the MOSFETS. When the power doesn't come on properly most folks seem to immediately assume they've blow the mosfets with static electricity and change them, only to find that isn't the probems. The more ususal culprits are the LED being bad or the wrong color (it must be on or the power management circuit won't work), the U2 chip bad or in the wrong way, a resistor in the wrong spot in the power management circuit, a bad solder connection in the PM section, mosfets in backwards or swapped in the wrong positions, or the power supply section simply not producing +/-12Vdc (or +/-9Vdc or so out of the batteries) in the first place.
- The batteries have to be charged equally. If you are inserting a pair of batteries for the first time out of the package or out of the parts bin, let them charge up overnight first in the O2 before trying to use the O2 on battery power to equally charge the batteries. Out of the package one may be charged and the other flat.
- Make sure U1 and U2 are not swapped if you get distorted sound. The op amp U1 may actually work for awhile in place of the comparator U2, but not vise versa.
- Don't forget to connect the ground wire between the center pin of the input jack and the metal case screw.
- Double check (reheat) your solder connections on R12 and R13 for 1-2 seconds. If those don't make contact C13 and/or C14 will charge up and put a large DC voltage on your headphone output.
- Use a magnifying glass to verify that you don't have a solder bridge between the voltage regulator pins or between the mosfet pins. Very easy to happen and rather hard to see without a magnifier.
And some personal recommendations..
- Spend the extra $$ and get the gold plated IC sockets that RS lists in the BOM. The contacts won' t surface oxidize or corrode over time.
- Spend the extra $$ and use the sockets that RS lists on the BOM for the gain resistors. Makes it easy later to change the gain.
- Spend the extra few cents and get the low noise DD version of the op amps
- Use an adaptor that is at least 14Vac and 400mA or greater. Don't use a 12Vac adaptor. If you have headphones that will pull more than 50mA per channel use an adaptor rated at 1amp or greater. The half wave supplies cause a voltage spike in the transformer about 3x the output current.
- Don't use Fairchild voltage regulators. Use the On Semiconductor regs RS has in the BOM. I've run into a curiously large number of bad-out-of-the-box Fairchild discrete semi's over the years for some reason. No choice though on the mosfets.
Last edited:
I think NwAvGuy blogged these, but here's a couple more things to watch out for:
Make sure the regulators are seated down far enough so the tabs will not contact the case.
The coaxial AC jack soldering tab can make contact with the case right at the slot where the PCB slides in. Make sure that tab is soldered towards the far side of the hole and/or trim it slightly with cutters.
Make sure the regulators are seated down far enough so the tabs will not contact the case.
The coaxial AC jack soldering tab can make contact with the case right at the slot where the PCB slides in. Make sure that tab is soldered towards the far side of the hole and/or trim it slightly with cutters.
Wish i'd seen that list when i built my first one. Unexpected issues i ran into,
-AC jack needs its legs trimmed to fit into the PCB, apparently it has changed since the O2 was created.
-If your sourcing the parts yourself from the BOM there was a change to the caps for the two voltage regulators, BOM lists .22uf film caps, vs the later change to 1uf ceramics. I believe this was just a precaution as the 7912 datasheet recomends 1uf.
If you are sourcing parts direct go for the optional gold sockets, gain resistors sockets, and DD amps as per above. On the second one i built i also opted for the better Panasonic/Wima caps and mil spec resistors from the BOM as it only adds a few $'s to the total.
Also if you want it to look really spiffy the PCB's from Head'n'hifi are a sexy red/gold colour. ...i know its sad, but i know its there. That amp always sounded better too, in my head
-AC jack needs its legs trimmed to fit into the PCB, apparently it has changed since the O2 was created.
-If your sourcing the parts yourself from the BOM there was a change to the caps for the two voltage regulators, BOM lists .22uf film caps, vs the later change to 1uf ceramics. I believe this was just a precaution as the 7912 datasheet recomends 1uf.
If you are sourcing parts direct go for the optional gold sockets, gain resistors sockets, and DD amps as per above. On the second one i built i also opted for the better Panasonic/Wima caps and mil spec resistors from the BOM as it only adds a few $'s to the total.
Also if you want it to look really spiffy the PCB's from Head'n'hifi are a sexy red/gold colour. ...i know its sad, but i know its there. That amp always sounded better too, in my head
A correction and a clarification on the unusual O2 problems list -
With a rated adapter secondary voltage of 14Vdc rms to 20Vdc rms the open circuit (just measuring the tip of the plug) voltage would be higher of course, more like 16Vdc - 24Vdc. Just as long as it isn't zero. I've helped at least one person with a dead O2 who had forgotten to switch on the plugstrip their adapter was plugged into.
My adapter sizing recomendation at the bottom of the list also takes into account the 25mA or so of chip quiescent current and 25mA max battery charging current in addition to the 2x 50mA ouput current. Easy to forget about the battery charging current draw through the power supply circuit.
With a rated adapter secondary voltage of 14Vdc rms to 20Vdc rms the open circuit (just measuring the tip of the plug) voltage would be higher of course, more like 16Vdc - 24Vdc. Just as long as it isn't zero.
I've helped at least one person with a dead O2 who had forgotten to switch on the plugstrip their adapter was plugged into.My adapter sizing recomendation at the bottom of the list also takes into account the 25mA or so of chip quiescent current and 25mA max battery charging current in addition to the 2x 50mA ouput current. Easy to forget about the battery charging current draw through the power supply circuit.
Curiously enough it might be my DMM. I've used it for so long it might be the batteries slowly dying. Either way I attached some images. The wires going out to the left lead to my power connector and the wires on the right lead to my DAC.
Front: http://i.imgur.com/s0mNrGn.jpg
Back: http://i.imgur.com/CqZH5c7.jpg
Front: http://i.imgur.com/s0mNrGn.jpg
Back: http://i.imgur.com/CqZH5c7.jpg
Last edited:
I actully don't know... I attached an image and as far as I can tell its just another barrel connector.
http://i.imgur.com/vK1mbM6.jpg
http://i.imgur.com/vK1mbM6.jpg
There's only 2 conductors there in the circuit, so if you use 3 wires you have to make extra sure the connections are right.
Every DMM I've seen has a low battery indicator. My suggestion now is to make the same voltage and visual checks described earlier. In the meantime I'll see what I can find in the photos.
Every DMM I've seen has a low battery indicator. My suggestion now is to make the same voltage and visual checks described earlier. In the meantime I'll see what I can find in the photos.
That pin on the side of the barrel connector dosent go anywhere on the PCB so you can ignore that, i believe its purpose is to detect if a plug is inserted. It either gets tied to ground with nothing connected, or tied to ground when a plug is connected, cant remember which, dosent matter anyway.
You could do with a bit more solder in general, its cheap so dont hold back. One of the four caps jumps out at me as it looks like it has almost none on one of the pins. Basically redo them all, and the voltage regs. I like to add enough so it just starts to flow through to the other side of the board, many of the components arent showing any solder top side.
Also tidy up the input wiring from the ODAC on the bottom of the board, it looks like the ends of the cables are shorting to each other.
And for sanity sake put your meter into AC and measure the AC adapter, then in DC mode one of your 9v batterys just to confirm the meter is a ok.
You could do with a bit more solder in general, its cheap so dont hold back. One of the four caps jumps out at me as it looks like it has almost none on one of the pins. Basically redo them all, and the voltage regs. I like to add enough so it just starts to flow through to the other side of the board, many of the components arent showing any solder top side.
Also tidy up the input wiring from the ODAC on the bottom of the board, it looks like the ends of the cables are shorting to each other.
And for sanity sake put your meter into AC and measure the AC adapter, then in DC mode one of your 9v batterys just to confirm the meter is a ok.
Sorry for getting back to you guys so late but I was curious so I went out and bought a new battery for my multimeter and after a bit of testing I concluded that it was my multimeter that had decided to act up on me. I did all the tests again and all the numbers came out fine so I installed the OpAmps and... it totally worked!
Everything is working fine and I owe it all to you guys, Thanks so much!
Everything is working fine and I owe it all to you guys, Thanks so much!