Hi there, just an all-around newbie here.
I've been soldering a O2 kit recently, and noticed I've got a pair of mismatched C13 C14 (coupling?) capacitors, in which one is a 2.2uf one and the other is a 3.3uf cap.
BOM states it could be ">=2.2uF >=50 V", but I'm wondering if installing a mismatched pair would deteriorate the signal, or is it just irrelevant/insignificant?
I know you're thinking "why ask the internet and not just go out and buy a new pair of caps", but I'm just thinking if it's worth 1 hour+ of time (no electronics shop around where I live unfortunately) to change a cap that may or may not cause any difference.
I've been soldering a O2 kit recently, and noticed I've got a pair of mismatched C13 C14 (coupling?) capacitors, in which one is a 2.2uf one and the other is a 3.3uf cap.
BOM states it could be ">=2.2uF >=50 V", but I'm wondering if installing a mismatched pair would deteriorate the signal, or is it just irrelevant/insignificant?
I know you're thinking "why ask the internet and not just go out and buy a new pair of caps", but I'm just thinking if it's worth 1 hour+ of time (no electronics shop around where I live unfortunately) to change a cap that may or may not cause any difference.
Hi, finally finished my O2 and am very happy with it. Been using batteries this last week whilst I was trying to build a desktop enclosure, that all went a bit wrong today so bought a large Hammond case from my local Maplin(UK).
Using it with an old Pioneer DV-757Ai multi-format player & Sennheiser HD-650's. I've settled on low/high gains of 2.25 & 3.5... had no issues during the build and I really can't fault the O2.
Thanks to RocketScientist for designing such an amazing cheap headphone amp!
Using it with an old Pioneer DV-757Ai multi-format player & Sennheiser HD-650's. I've settled on low/high gains of 2.25 & 3.5... had no issues during the build and I really can't fault the O2.
Thanks to RocketScientist for designing such an amazing cheap headphone amp!
Check around the gain switch and U1 for solder bridges to the PCB ground plane. The metal shell of the gain switch is connected to ground via the two pins at the corners and the two middle pins of the 6 gain switch pins are soldered to ground. You might have a bridge from one of the remaining 4 pins of the switch to one of those. A magnifying glass is highly recommended for this one! Some bridges can be very hard to see just by eye.
If the gain switch were just bad/open (no connection) you would get a gain of 1 in that position. Would still sound OK in that channel, just not as loud. But if the gain switch gets shorted to ground in one channel the inverting input of U1 gets grounded and effectively turns the op amp into a comparator.Your incoming music waveform would get converted into square waves at the same frequency which would sound very distorted. Below is a plot of what happens there.
You can also test for a solder bridge by unplugging the amp from the AC adaptor and pulling out the batteries to disconnect all power. Then measuring with an ohmmeter from pin 6 of U1 to ground (metal shell of the gain switch) and flip the gain switch in both positions. Then do the same with pin 2 of U1. If you measure a dead short to ground at either pin in one of the gain switch positions, there is your solder bridged trace.
The other thing to check is that the values of R17 match R21, and R19 matches R23, of course. With the amp powered down and disconnected as above you can just measure across those resistors with the ohmmeter. In one position of the gain switch you should get the right reading.
Good luck!
Thanks for the reply. Everything checked out okay checking the gain switch to the two pins on U1. I double checked both resistors and they matched up. I may try using the scopes at my campus lab and see if the output is coming out square like you described. I would be checking the output of U1 so pins 7 and 1 of it correct?
I did notice if I toy with the button on the switch, move it around a bit the sound appears to be normal, equal channels but then goes wrong if I let go or mess with it again.
That sounds like a bad switch (open connection). In that case your gain would just drop to 1 in that channel. It would not sound distorted, just not as loud as the other channel. The switch might have been defective from the start or may have been damaged by too much soldering heat.
A scope would tell you for certain! Yep, pins 1 and 7 of U1 for the output vs. pins 3 and 5 for the inputs. It would be a good idea to take a look at the power supply rails too with the scope.
A scope would tell you for certain! Yep, pins 1 and 7 of U1 for the output vs. pins 3 and 5 for the inputs. It would be a good idea to take a look at the power supply rails too with the scope.
Last edited:
Completed my first O2 build on Friday with only a few small self-induced issues along the way. Amazingly, it worked perfectly despite my best attempts at screwing it up. It sounds great! Drives my HD-595s to any volume I can tolerate without any problems. Listening to Radio Paradise now via my iPhone and their 128K AAC link. Sounds fabulous!
Thanks again to NWAVGuy, Ollie and everyone who made this happen. What a great way to start my DIY Audio "career!"
Thanks again to NWAVGuy, Ollie and everyone who made this happen. What a great way to start my DIY Audio "career!"
I imported the bill of the materials into mouser and did a double check of everything just before ordering and I noticed that most of the gain resistors and a few other resistor's watt ratings are 1/2, 1/3, and 1/4 instead of the specified 1/8. I cross-checked the part and manufacture numbers and they're correct, but why are the power ratings getting changed? If order them will they not fit? This problem showed up in both versions of the BOM I've checked and I'm getting a little frustrated because I keep having to change my order and it's getting confusing to check whether or not I have the correct parts selected.
Last edited:
Send me a PM with your email and I'll share my Mouser project with you so you can compare yours with it..
I compared the two and everything seems good now. I didn't know that higher wattage rated resistors can get as small as 3.9 mm or less
I built an O2 on Saturday and it turned out well. Took me about 7 hours total with setting up and prep, soldering and finally testing.
Top of the O2
O2 amp top | Flickr - Photo Sharing!
Bottom of the O2, to show off my not so good soldering work.
O2 amp bottom | Flickr - Photo Sharing!
Top of the O2
O2 amp top | Flickr - Photo Sharing!
Bottom of the O2, to show off my not so good soldering work.
O2 amp bottom | Flickr - Photo Sharing!
O2 Answers
Finally, there have been several posts about AC wall transformers. Someone already quoted the relevant paragraph from my blog, but the best bet is any transformer rated at 14 - 18 VAC and at least 200 mA. The 12 volt WAU12-200 from Mouser works because on typical line voltage it's around 13.5 VAC or higher. But if you have low line voltage, or plan to really stress your O2 with power hungry low impedance cans, you're better off with a 14+ VAC transformer. 20 VAC is the upper limit.
It's worth noting the Power Supply Rejection Ratio (PSRR) of the O2 is so high (at least 80+ dB) that even if the supply voltage is too low and the regulators fall out of regulation, the resulting ripple on the supply rails has very little effect on the audio performance of the amp. The performance is still way under the threshold of audibility even with lots of ripple.
Sorry I've been missing lately but I'm still around. I want to thank AGDR, MrSlim, and all the others who have been helping answer questions. I'm also glad to read reports from O2 builders who are happy with their amps! I've received lots of emails from happy O2 users from around the world and surprisingly few from people having problems.
That should work as long as significant EMI/RFI from the DC-DC converter doesn't find its way into the audio circuitry by other means (inductive, capacitive and ground system coupling). The best would be to have the DC-DC circuitry in a different enclosure. The pre-packaged DC-DC modules tend to have much lower EMI than discrete/DIY DC-DC designs.
This has been answered, but just to be clear with no load on the amp (nothing plugged into the headphone jack) you'll see several hundred millivolts of DC offset with the amp off that will take a while to slowly drop towards zero. I've revised the testing procedure in the O2 Details blog article to account for that. Most direct coupled amps do the same thing and it quickly drops close to zero with any sort of normal load.
The O2 doesn't need any burn-in. Amps with electrolytic caps in the signal path, or devices with certain film caps used in filter circuits (like phono preamp RIAA EQ) can benefit from being powered up for a while. But once the DC operating points of the op amps are close to stable, the O2 is ready to rock and perform its best. And that takes only seconds.
I'm still hoping for a 24 bit USB DAC option for the desktop version of the O2 and should have more to report in the next few weeks. I'll be starting a new thread on diyAudio for the new amp once I have enough to report.
Finally, there have been several posts about AC wall transformers. Someone already quoted the relevant paragraph from my blog, but the best bet is any transformer rated at 14 - 18 VAC and at least 200 mA. The 12 volt WAU12-200 from Mouser works because on typical line voltage it's around 13.5 VAC or higher. But if you have low line voltage, or plan to really stress your O2 with power hungry low impedance cans, you're better off with a 14+ VAC transformer. 20 VAC is the upper limit.
It's worth noting the Power Supply Rejection Ratio (PSRR) of the O2 is so high (at least 80+ dB) that even if the supply voltage is too low and the regulators fall out of regulation, the resulting ripple on the supply rails has very little effect on the audio performance of the amp. The performance is still way under the threshold of audibility even with lots of ripple.
Here is my Mouser project for the O2 BOM. There are a couple of substitutions, and two of the optional gain resistors are back ordered at them moment, but they are for 7x and 10x, which are unlikely to be required in most cases (and 7x is almost the same as 6.5x anyway)
O2 Mouser project parts list
Since it is on the Canadian site, you should click the "Save as my own project button" to make a copy of it in your own profile, if you are not in Canada.
O2 Mouser project parts list
Since it is on the Canadian site, you should click the "Save as my own project button" to make a copy of it in your own profile, if you are not in Canada.
Last edited:
You should build it with the sockets for the gain resistors. That way you can try a few combinations..
Try canada.newark.com, they want $11 for shipping, which still seems like a lot. You would think they could mail it for a couple of bucks, but that is they way of things now, courier only and expensive. They will ship UPS, don't take any guff from the UPS guys though, they Neark are supposed to include all taxes and fees in what they charge you, so don't let UPS charge you any BS "brokerage fees"!
FOUND SWITCH PROBLEM!!!!!!!!
The via below the S on "S2" is to close to the switch case and causes a short. Its not a faulty switch you have Schmidtc89, its the via causing a short. to solve this either cut off that small support leg before installing the switch, or if youve already installed it then bend it inwards slightly so it misses the via.
Hope that makes sense.
The via below the S on "S2" is to close to the switch case and causes a short. Its not a faulty switch you have Schmidtc89, its the via causing a short. to solve this either cut off that small support leg before installing the switch, or if youve already installed it then bend it inwards slightly so it misses the via.
Hope that makes sense.
The above should not be true for the final PCB revision (boards with a September date on them). The via was moved up next to R21. Where did you get your board forsakenrider?
As for gain, see:
O2 Gain Settings