I do have one question though, despite it working beautifully and no hum, I haven't put in any grounds other than the ones that terminate on the pcb ie. the grounds from the RCA inputs, and the two headphone jacks, in and out.
Should there be another ground, like on the chassis?
On the NWAVGUY blog, under chassis, it's mentioned that a wire goes from the blue input to the chassis:
B2-080 Enclosure Ground – To reduce hum and noise it’s important to ground the enclosure to Pin 1 of the input Jack J2. That’s the center pin closest to the front panel. You can either use a short piece of lead wire from one of the small resistors (you want the thinnest wire possible) and make a loop where the screw hole is for the lower right corner of the front panel. Or, if you prefer, use fine gauge wire (like AWG30 wire wrap wire) from J2’s ground pin to the lower screw hole on the back panel behind the batteries. Run the wire under the board keeping it to the far right of the enclosure
Should there be another ground, like on the chassis?
On the NWAVGUY blog, under chassis, it's mentioned that a wire goes from the blue input to the chassis:
B2-080 Enclosure Ground – To reduce hum and noise it’s important to ground the enclosure to Pin 1 of the input Jack J2. That’s the center pin closest to the front panel. You can either use a short piece of lead wire from one of the small resistors (you want the thinnest wire possible) and make a loop where the screw hole is for the lower right corner of the front panel. Or, if you prefer, use fine gauge wire (like AWG30 wire wrap wire) from J2’s ground pin to the lower screw hole on the back panel behind the batteries. Run the wire under the board keeping it to the far right of the enclosure
Excellent 
Grounding the chassis is really to stop the chassis picking up stray noise and allowing it to contaminate the audio. We call it grounding but the O2 isn't mains ground referenced in any way, we simply call the zero volt line 'ground' and refer all points to that. Standing the 02 on top of something with a transformer in it would be one example of where it might help.
So its good practice to tie the chassis to zero volts but how you actually do it shouldn't really be that critical.
Grounding the chassis is really to stop the chassis picking up stray noise and allowing it to contaminate the audio. We call it grounding but the O2 isn't mains ground referenced in any way, we simply call the zero volt line 'ground' and refer all points to that. Standing the 02 on top of something with a transformer in it would be one example of where it might help.
So its good practice to tie the chassis to zero volts but how you actually do it shouldn't really be that critical.
Congratulations on getting there. Have you hooked it up to the ACA for a test yet?
Thanks NealJ and Mooly.
Yes it's lovely with the ACA, and as RafaPolit says, perfectly good output without the need to double up, there's enough power to fill all but the largest of rooms.
The gain at lower level is more than enough, I briefly turned it up to high gain but it was too much.
The only way (perhaps) it could be bettered would be to have less gain and bridge to increase output with a second ACA.
The detail and soundstage on this combo is awesome.
I actually found that it was difficult to get actual hands-on build tips, but using pics that NealJ posted, aewulf, and Mooly's diagrams, not to forget NWAVGUY's own blog helped.
To that end, I did a quick build blog, of sorts, here on headphone systems to try and give back a bit to other novices like me.
A picture is worth a thousand words as they say.
What I'd change if I were doing the pcb's perhaps for those like me that want a dual-purpose unit? make it larger.
Also there's the SIP sockets. I ended up frustrated with them and soldered the resistors straight into them and it ended up looking like a dog's dinner-my fault, but the pcb being small made it difficult to execute, so I'd look to do that another way to make the different gain resistors hot-swappable.
I'm about to post the last picture (before the one I'll take once I've got the burr walnut on it to match the ACA) and you'll see what a mess I made of it, it works at least_ Novice's blog Objective 2 headphone amp
I'd be tempted to find another solution to the SIP sockets idea that would allow for a quick change of resistors.
The selector switch was from Hifi Collective :Elma 2 pole 6 way switch, 04-1264 The pot is an alps RK-27 10k
Last edited:
NealJ, one other thing, because the pin configuration on the switched headphone socket provided with the kit was well documented, it was easy for me to wire it in. However, I would not out of choice used this particular make of headphone socket as they can only be used on a max 2mm panel.
I would find another make that maybe had a nut fixing on the front so a thicker panel could be used.
I only used a 1/4" socket on both inlet and outlet for symmetry.
I would find another make that maybe had a nut fixing on the front so a thicker panel could be used.
I only used a 1/4" socket on both inlet and outlet for symmetry.
I apologize if this has been asked before but I tried to search in this thread and online and could not find an answer that made sense to me. I have read that the 02 is a Class B amplifier, but on the schematic I see two output transistors per channel which are paralleled and being fed the same input signal, and are not operating in opposite voltages of one another. I am a bit confused as to why is this amp considered class B?
The 02 uses opamps rather than transistors, and there are two parallel opamps in the output to increase the current drive ability. Both opamps are fed identical input signals and the identical output voltages are summed by the 1 ohm resistors.
The opamps themselves will be class AB internally and the transistors within the chip will run at a small bias current of a couple of milliamps (class AB).
The opamps themselves will be class AB internally and the transistors within the chip will run at a small bias current of a couple of milliamps (class AB).
If I want to use it as desktop only I can then just place a "random" Led and forego the trimmer right?
Yes, you can do that. Just make sure you do fit something in its place though because the circuit may not work correctly with either leaving the LED out or shorting it out. It certainly doesn't in simulation anyway.
And if you are never using batteries then the easiest way is just to link the two FET's out and not fit any of the comparator circuitry components at all.
And if you are never using batteries then the easiest way is just to link the two FET's out and not fit any of the comparator circuitry components at all.
That depends on what you consider "uncomplicated". Do you mind warming up the soldering iron again or do you insist on something prebuilt?
XLR outputs are not all created alike, and you will need different kinds of adapter cables depending on what you have. (This applies to any device with unbalanced input, not just the O2.)
There are two possible cable pinouts:
a) XLR pin 2 = signal, pin 1 = ground
b) XLR pin 2 = signal, pin 3 = ground, possibly 1 and 3 tied together as well
Ground-referenced output stages require type (a) and will not work with (b).
Floating or servo-balanced output stages require (b), they will not work with (a).
Merely impedance-balanced output stages (= single-ended output stages souped up with some passives to get balanced output impedance) will work with both types. These are usually found on gear towards the budget side of things (compact mixers, the Behringer 2496 series, budget audio interfaces, ...).
I would consider going from female XLR to e.g. RCA with Actually Decent Coax(R), then use/make a short 2x RCA to stereo 3.5 mm adapter with some thinner shielded audio cabling. I have even used abominations like 1-piece 2x RCA to 1/4" and 1/4" to 3.5 mm adapters plugged together, but it really is not ideal in terms of leverage and potentially pots a lot of stress on the poor input jack.
I imagine designing a balanced input board wouldn't be that hard either, assuming it can still be squeezed into the case complete with 2x 1/4 TRS inputs (people have fitted RCAs, and those jacks aren't too different in size; XLR might be possible but I wouldn't be getting my hopes up too far right now). It kind of depends on what sort of performance standards you want to apply. A 1-opamp circuit with e.g. a gain of -6 dB would probably work perfectly fine in practice, it's just that the rest of the O2 performs to rather high standards (well, arguably the input stage is needlessly low noise even). I might look into it if there is any kind of demand.
XLR outputs are not all created alike, and you will need different kinds of adapter cables depending on what you have. (This applies to any device with unbalanced input, not just the O2.)
There are two possible cable pinouts:
a) XLR pin 2 = signal, pin 1 = ground
b) XLR pin 2 = signal, pin 3 = ground, possibly 1 and 3 tied together as well
Ground-referenced output stages require type (a) and will not work with (b).
Floating or servo-balanced output stages require (b), they will not work with (a).
Merely impedance-balanced output stages (= single-ended output stages souped up with some passives to get balanced output impedance) will work with both types. These are usually found on gear towards the budget side of things (compact mixers, the Behringer 2496 series, budget audio interfaces, ...).
I would consider going from female XLR to e.g. RCA with Actually Decent Coax(R), then use/make a short 2x RCA to stereo 3.5 mm adapter with some thinner shielded audio cabling. I have even used abominations like 1-piece 2x RCA to 1/4" and 1/4" to 3.5 mm adapters plugged together, but it really is not ideal in terms of leverage and potentially pots a lot of stress on the poor input jack.
I imagine designing a balanced input board wouldn't be that hard either, assuming it can still be squeezed into the case complete with 2x 1/4 TRS inputs (people have fitted RCAs, and those jacks aren't too different in size; XLR might be possible but I wouldn't be getting my hopes up too far right now). It kind of depends on what sort of performance standards you want to apply. A 1-opamp circuit with e.g. a gain of -6 dB would probably work perfectly fine in practice, it's just that the rest of the O2 performs to rather high standards (well, arguably the input stage is needlessly low noise even). I might look into it if there is any kind of demand.
Last edited:
Thanks for your reply
i am connecting my Yamaha CL3 Digital mixer outputs (balanced xlr) to the input of O2
i built six of these O2 for my recording studio setup in order to give each musician their own seperate listening. I just want to make shure i use the proper connection not to loose quality.
im thinking that since the O2 has unbalanced input there is nothing else to do but connecting with two cables xlr to rca with the xlr pin 1 ground, pin 2 hot, and pin 3 cold gong to rca shield - 1 and cold, and rca pin to hot. thereby loosing six db signal but how else could it be done?
BR...T
Oh, that's a Big One. The outputs are called OMNI OUT, with nominal output of +4 dBu und maximum of +24 dBu. The block diagram shows two amplifier triangles with + und - symbols on pins 2 and 3... I am guessing this is a ground-referenced output.
I'd also give the (a) cable a shot first. XLR pin 3 should remain unconnected (unless you can't seem to get much of any signal out at all).
Losing some signal amplitude would be the least of my worries here, you'd still have up to +18 dBu to work with either way. I hope you built your O2s with a minimum gain of 1X, 'cause you're going to need that setting. I think the gain stage should clip at a hair over +18 dBu then. Depending on headphones used, running into volume pot imbalance at low settings might be more of a problem...