@NicMac with nothing plugged into the input jack, your readings for R14 and R20 are normal as you're also reading R7 and R3 in parallel.
@Abraxalito, let's please keep this thread on topic. I was talking about noise not impedance and most are not into winding their own custom tube output transformers for that matter. Input noise and output noise are totally different things. Regardless, it's a bit too far off topic. If you want to start a tube debate, please do it in a new thread. Thanks.
@habarradah, it's fairly easy to see the ground pads on the VR1 footprints. See the board layouts in the PDF footprints if necessary. You can use the signal pads from either of the two sets.
@Abraxalito, let's please keep this thread on topic. I was talking about noise not impedance and most are not into winding their own custom tube output transformers for that matter. Input noise and output noise are totally different things. Regardless, it's a bit too far off topic. If you want to start a tube debate, please do it in a new thread. Thanks.
@habarradah, it's fairly easy to see the ground pads on the VR1 footprints. See the board layouts in the PDF footprints if necessary. You can use the signal pads from either of the two sets.
Last edited:
yep, theres a standard way to figure it out; you look at the datasheet =) as no there really isnt a hardline standard, some will have extra ground pins, some will have L or R duplicated etc. failing access to a datasheet, just put a mini without the barrel into the jack and use your DMM to ohm it out
Last edited:
@HaroldHill, you're welcome! I'm glad it turned out as expected.
@NicMac, seeing as how you apparently have a DMM, if you can't find anything else wrong, you might want to download this Pink Noise MP3 and measure the voltage at each output at P2 with the above file playing at a volume high enough to read around 0.1 - 0.2 AC volts on your DMM (hopefully on the AC millivolt range if it has one--i.e. around 150 mV, or use the lowest AC volts range if it's not autoranging). The reading in both channels may fluctuate some due to the random nature of pink noise, but the idea is to measure how much difference there really is between the channels.
If there's not a significant difference (more than 12% which is about 1 dB), perhaps it's your headphones?
If there is a difference, try to measure the input voltage at P1 for each channel. If it's off by a similar percentage there, the problem is your source. Try a different source.
If the problem isn't the source, carefully measure the pink noise at pin 1 and pin 7 of U1. if it's the same there, yet different at P2 (the headphone output), the problem is after the gain stage (i.e. it could be the volume pot, or anything between U1 and the output). If it's different between pin 1 and 7, but the same at P1, there's something wrong with the U1 gain stage, gain resistors, etc.
All measurements are with respect to ground. You can leave the black DMM lead connected to one of the center battery terminals.
@NicMac, seeing as how you apparently have a DMM, if you can't find anything else wrong, you might want to download this Pink Noise MP3 and measure the voltage at each output at P2 with the above file playing at a volume high enough to read around 0.1 - 0.2 AC volts on your DMM (hopefully on the AC millivolt range if it has one--i.e. around 150 mV, or use the lowest AC volts range if it's not autoranging). The reading in both channels may fluctuate some due to the random nature of pink noise, but the idea is to measure how much difference there really is between the channels.
If there's not a significant difference (more than 12% which is about 1 dB), perhaps it's your headphones?
If there is a difference, try to measure the input voltage at P1 for each channel. If it's off by a similar percentage there, the problem is your source. Try a different source.
If the problem isn't the source, carefully measure the pink noise at pin 1 and pin 7 of U1. if it's the same there, yet different at P2 (the headphone output), the problem is after the gain stage (i.e. it could be the volume pot, or anything between U1 and the output). If it's different between pin 1 and 7, but the same at P1, there's something wrong with the U1 gain stage, gain resistors, etc.
All measurements are with respect to ground. You can leave the black DMM lead connected to one of the center battery terminals.
I've just finished building 2 of these amplifiers it only took a couple of hours once I got the parts sorted. I'm using the AKG 600ohms at the moment and
they sound brilliant, I'm not real flash with the descriptions of the sound, but it'll do me
We owe the designer of this amp a big thank you not just for doing the leg work but for freely releasing it so we can all very cheaply have a great new toy
the compact design, all on one board is a master stroke and makes the construction a doddle
the only thing is why use those sky blue jacks??
cheers
they sound brilliant, I'm not real flash with the descriptions of the sound, but it'll do me
We owe the designer of this amp a big thank you not just for doing the leg work but for freely releasing it so we can all very cheaply have a great new toy
the compact design, all on one board is a master stroke and makes the construction a doddle
the only thing is why use those sky blue jacks??
cheers
RS,
I will do as you suggest. I certainly have DMM's and a scope if necessary
I admit that I have been to lazy (and busy) to fire up a serious trouble shooting thinking that maybe somebody else would have encountered the same problem and located the fault.
I would like to take the opportunity to thank you for the great circuit and great support you have put into the O2 project. I my time at diyAudio I have never seen some many new registrations as over the last months and I'm sure the O2 is to 'blame' for many of these. Certainly some of these new comers will become addicted to DIY audio and move on to other circuits for the best of the community. Certainly, the no-nonsense nature of the O2 is a great way to start out in this hobby.
A great thanks also go to all the GB organizers that certainly have their work cut out for them for a very limited return (if any).
Cheers,
Nic
I will do as you suggest. I certainly have DMM's and a scope if necessary
I admit that I have been to lazy (and busy) to fire up a serious trouble shooting thinking that maybe somebody else would have encountered the same problem and located the fault.
I would like to take the opportunity to thank you for the great circuit and great support you have put into the O2 project. I my time at diyAudio I have never seen some many new registrations as over the last months and I'm sure the O2 is to 'blame' for many of these. Certainly some of these new comers will become addicted to DIY audio and move on to other circuits for the best of the community. Certainly, the no-nonsense nature of the O2 is a great way to start out in this hobby.
A great thanks also go to all the GB organizers that certainly have their work cut out for them for a very limited return (if any).
Cheers,
Nic
Thanks everyone for sharing your positive thoughts on the amp. I'm also glad it's brought some "new blood" to diyAudio and perhaps into the hobby itself. There's something really satisfying about listening to a piece of gear you built yourself. The O2 is designed, in more ways than one, to make headphone listeners smile. To me, that's what ultimately matters most.
I've learned a lot too. I've done plenty of commercial designs before, but never a DIY project like this. In hindsight, I realize the O2 isn't so easy to source parts for outside the USA. But thanks to the hard work of several others, the group buys helped a lot, and now there's also the option of buying assembled boards. Many others have helped in other ways with suggestions, finding problems early on, etc.
I'm trying to put what I learned into the upcoming ODA. For one thing, it will hopefully be more "global" and easier to source parts for outside outside North America.
As for the sky blue jacks, they're not my favorite either. There's a standard color code for 3.5mm audio jacks on PCs: Light Blue for Line-In and Light Green for Headphone/Line Out. There were suggestions to use those standard in/out colors for those not wanting to label their panels. But Mouser ran out of the green jacks somewhere along the way so we ended up with 2 blue ones. They do have black ones for $0.20 more each--they're listed in the Notes column of the BOM.
I've learned a lot too. I've done plenty of commercial designs before, but never a DIY project like this. In hindsight, I realize the O2 isn't so easy to source parts for outside the USA. But thanks to the hard work of several others, the group buys helped a lot, and now there's also the option of buying assembled boards. Many others have helped in other ways with suggestions, finding problems early on, etc.
I'm trying to put what I learned into the upcoming ODA. For one thing, it will hopefully be more "global" and easier to source parts for outside outside North America.
As for the sky blue jacks, they're not my favorite either. There's a standard color code for 3.5mm audio jacks on PCs: Light Blue for Line-In and Light Green for Headphone/Line Out. There were suggestions to use those standard in/out colors for those not wanting to label their panels. But Mouser ran out of the green jacks somewhere along the way so we ended up with 2 blue ones. They do have black ones for $0.20 more each--they're listed in the Notes column of the BOM.
RocketScientist, I'd like to ask two question, if you don't mind:
1) do you already have ODA prototype? More specifically, I wonder if you can hear any audible improvement from using that new power supply circuitry;
2) What DAC chip are you planning to base ODAC upon? I'm at the the beginning of designing my own DAC, but I'd rather assemble yours if it fits my taste...
1) do you already have ODA prototype? More specifically, I wonder if you can hear any audible improvement from using that new power supply circuitry;
2) What DAC chip are you planning to base ODAC upon? I'm at the the beginning of designing my own DAC, but I'd rather assemble yours if it fits my taste...
@Alexium, no official ODA proto yet, just some hacked together bits until the ODAC is stable so I can do a proper ODA PCB layout, know which enclosure to design for, etc. PCB spins are expensive and time consuming and I did more than I wanted to with the O2 (four of them). So I'm waiting for things to become a bit more stable first.
The DAC chip hasn't been finalized. I'm fairly pleased with the chip on the prototype I've been testing and tweaking, but there may be even better options. For now, we're not disclosing more details as they're subject to change.
@Jokener, the ODA will very much be an open source design and you're free to offer those in whatever form you like. The ODAC, however, will require contractual agreements with one or more vendors, buying in fairly large quantities, custom firmware with licensing restrictions, etc. The company taking the financial risk to do all that will be the one assembling, programming, testing, selling and shipping the pre-assembled ODAC board. I expect the schematic will be published.
Even if you met the contractual and vendor obligations, I'm not sure there will be enough total ODAC sales volume to split the production between two buyers/assembly contractors/etc. It may fall under the MOQs (Minimum Order Quanties). Instead, it probably makes more sense for both of you to consider combining your requirements and share the production run in some way that helps both of you. But that will be between you and the ODAC supplier. I'm staying out of anything financial or sales related.
I'm sorry for not having more solid details at this point, but I wanted to let everyone at least know it's coming for those who might want to wait versus buying something else only to find out about the ODA/ODAC next month. For those who are patient, I'm doing my best to make it worth the wait.
EDIT: Qusp is mostly correct. It's very much a mutual design. As is almost always the case with such things, testing with the dScope has revealed many areas where we've found ways to substantially improve the performance. I've "finessed" some aspects of the design by trial and error using the dScope. Theory and what might seem right only go so far with the "black arts" of EMI, RFI, jitter, ground currents, PCB layouts, etc.
With a 16 bit DAC it's not that hard to get around 15.5 bit ENOB performance. But with a 24 bit DAC it's much more challenging to get much greater than 16 bit ENOB. The ODAC is currently rocking 18 - 19 bit ENOB performance (about -111 dB SNR or under 6 uV of noise) and that's taken some serious dScope time and effort.
The DAC chip hasn't been finalized. I'm fairly pleased with the chip on the prototype I've been testing and tweaking, but there may be even better options. For now, we're not disclosing more details as they're subject to change.
@Jokener, the ODA will very much be an open source design and you're free to offer those in whatever form you like. The ODAC, however, will require contractual agreements with one or more vendors, buying in fairly large quantities, custom firmware with licensing restrictions, etc. The company taking the financial risk to do all that will be the one assembling, programming, testing, selling and shipping the pre-assembled ODAC board. I expect the schematic will be published.
Even if you met the contractual and vendor obligations, I'm not sure there will be enough total ODAC sales volume to split the production between two buyers/assembly contractors/etc. It may fall under the MOQs (Minimum Order Quanties). Instead, it probably makes more sense for both of you to consider combining your requirements and share the production run in some way that helps both of you. But that will be between you and the ODAC supplier. I'm staying out of anything financial or sales related.
I'm sorry for not having more solid details at this point, but I wanted to let everyone at least know it's coming for those who might want to wait versus buying something else only to find out about the ODA/ODAC next month. For those who are patient, I'm doing my best to make it worth the wait.
EDIT: Qusp is mostly correct. It's very much a mutual design. As is almost always the case with such things, testing with the dScope has revealed many areas where we've found ways to substantially improve the performance. I've "finessed" some aspects of the design by trial and error using the dScope. Theory and what might seem right only go so far with the "black arts" of EMI, RFI, jitter, ground currents, PCB layouts, etc.
With a 16 bit DAC it's not that hard to get around 15.5 bit ENOB performance. But with a 24 bit DAC it's much more challenging to get much greater than 16 bit ENOB. The ODAC is currently rocking 18 - 19 bit ENOB performance (about -111 dB SNR or under 6 uV of noise) and that's taken some serious dScope time and effort.
Last edited:
I would suggest either a daughterboard or unfilled section of the PCB for the ODAC where one can get both an unbalanced line-out (RCA/BNC) as well as a balanced line-out (XLR). If the DAC is as good as you claim it will be (I have no doubt) it doesn't really make sense to have a desktop DAC/HPA unit for ~$130 and not be able to get a top-class line-out for $10-$20 more.
If you already have this functionality planned, accept my apologies - but it seems you only have line-in and not line-out, currently.
If you already have this functionality planned, accept my apologies - but it seems you only have line-in and not line-out, currently.
@454Casull, like it says in the ODA/ODAC article, there will be no balanced outputs for the reasons mentioned there. Balanced stages offer worse performance in most instances. The ODA is all about the best performance for the lowest cost. If you have something with only balanced inputs, you can always use the appropriate cable to go from RCA to XLR.
If there's room, I am considering a footprint for a 4 pin XLR headphone connector using the same pinout as Violectric and other commercial headphone amp manufactures. That would allow those with headphones already cabled that way to use them on the ODA without an adapter. But two of the pins will be ground. It won't be a balanced output as that would degrade the performance and add to the cost. A 4 pin XLR will, however, have better crosstalk into low impedance headphones than a standard 1/4" TRS jack with a shared ground.
@Turbon, the ODAC can be used standalone from USB power in its own (smaller) enclosure. It could also be powered by a dedicated power supply if you wanted but the current prototype has some carefully tested and optimized USB power filtering that seems quite effective at keeping the signal to noise ratio very close to what's ideally possible from the DAC chip itself.
If there's room, I am considering a footprint for a 4 pin XLR headphone connector using the same pinout as Violectric and other commercial headphone amp manufactures. That would allow those with headphones already cabled that way to use them on the ODA without an adapter. But two of the pins will be ground. It won't be a balanced output as that would degrade the performance and add to the cost. A 4 pin XLR will, however, have better crosstalk into low impedance headphones than a standard 1/4" TRS jack with a shared ground.
@Turbon, the ODAC can be used standalone from USB power in its own (smaller) enclosure. It could also be powered by a dedicated power supply if you wanted but the current prototype has some carefully tested and optimized USB power filtering that seems quite effective at keeping the signal to noise ratio very close to what's ideally possible from the DAC chip itself.
Last edited: