The Objective2 (O2) Headphone Amp DIY Project

[Mooly]

And again 1,5 MOHM resistors has turned into 400k resistor! I am soldering very cerefully and I am not overheating components, how this can happen?

Are you holding the resistors in your fingers while measuring ? If so you are adding your body resistance in parallel to the wanted one.

 

[Bichrome]

If you are measuring out of circuit, first thing to do is put a new battery in your DMM. The battery is used for resistance readings and when they start to do dead you will get readings all over the place. Even if it is a brand new DMM out of the package - I've run into a least one over the years that sat on the shelf so long the included battery was flakey. Then make sure your meter test leads are fitting tight in the DMM. I've seen some loose ones over the years that will mess up readings. Highly unlikely the problem is actually your resistors, especially if they are on a cut tape like that (all from the same batch). 236k for a 274k would be 14%, way out of the ballpark.

If you are taking measurements in-circuit, well that doesn't really work reliably as Mooly has noted. In-circuit voltage readings to ground are a much more reliable way to go. At the very least for in-circuit resistance readings you need to make sure the O2 is completely unplugged from the AC, batteries out, and take a small value resistor (like 1 ohm or just a straight wire) and short out every single capacitor on the board for a a few seconds to discharge them all. Charged caps will give you bad in-circuit readings. Even then once you put your ohmeter on something it may start charging up a cap nearby, resulting in a "moving" resistance reading. And resistors in-circuit often have other parts across them, like other resistors and transistor junctions, so that the resistor is unlikely to read what its markings say. If you really want to measure a resistor that is in-circuit already the best thing to do is unsolder and "lift" one leg out of the PC board so you can make an good measurement.

EDIT: I just took a look at the O2 BOM to verify. R4, 5, 8, and 24 are 270K. You probably got that crossed with the line above in the BOM that is 274R. But 236K would still be 12.5% off and 274 is 1.5%, which is unlikely to happen for 1% resistors.

Thank you, it helped me a lot. Turns out that R25 reading was "corupted" by circuit. I soldered out one leg and measured the resistence - guess what... it was 1,5 MOhm same thing for R4, R5, R8 and R24. Than I measured resistor values with 3 different DMM's and they showed same values.

Are you holding the resistors in your fingers while measuring ? If so you are adding your body resistance in parallel to the wanted one.

And again, i should have known this too, next time I will be more careful. Thanks.

By the way I tested the amp for 2h and it is working great. Some DC voltages differs(+/- 0,2 VDC) from NwAVguy's testing tutorial, but values are same and stable(just opposite polarity) in both rails, so I guess that it's ok.

 

[Mooly]

Thanks for the update, great to know you got it sussed

 

[UnknownAX]

I want to build the O2 from a head'n'hifi kit and add a ffew features.

1. An integrated Hifimediy Sabre Tiny DAC

2. A 6.35mm output (propably a Neutrik slim connector)

3. Line out from the dac

4. AC jack on the back of the amp

I still want to use it with batteries, so I can't mount the Dac there but I think there should be enough space for it somewhere on or even under the O2 board.

Would it be possible drill a bigger hole and put the 6.35mm jack to where the AC jack usually goes? Is there enough space for it inside the case? Except for D4 and R6 there shouldn't be anything there and I guess I could even mount those on the other side of the board. The jack is about 15x16x25mm. Panel mounting another AC jack on the back shouldn't be a problem.

I also need line-out from the dac, so how should I connect it? Would I get line out on the input jack if I connect the dac to P1 or does the connected (but turned off) O2 do something the the signal. That would also mean that using the input on the O2 would send the signal to the input of the O2 + the output of the dac (=bad?)
I could also connect the dac to the pins of the input jack that are usually shorted to ground while nothing is connected.(This is how NwAvGuy connects the odac to the O2 in his blog). Then I could use seperate rca's for the line out of the Dac, but I whould need to have atleast a dummy plug connected to the O2's input in order to disconnect the dac from the O2's input. I hope this doesn't sound too confusing....

Thanks for any suggestions!

 

[sgtstan]

Success!

I have just build the O2 again and it runs very nice now! Thanks to your help and enough patience while building it.

But there is one little problem still. I can only use the 3,5mm input as a source. When I connect the RCA connectors, the O2 is silent!? I used the P1 connectors to wire the connectors and checked connections thouroughly. But still nothing.

I connected the headphones to a 6,5mm connector to P2 and removed the "original" connector. May this be the reason? Do P1, P2 only work when the 3,5mm jacks are removed?


P.S.: I just noticed, that U5 and U6 are still getting a little warm when the amp is running. I can still touch them, (not like the last time...) but do I have to worry about that?

 

[Mooly]

Good timing

If the output voltages from the regs are correct (plus 12 volts on D1 and minus 12 on D5) then they are themselves OK. They will be warm... I wouldn't have said they should be hot though. How hot depends on how much current is flowing. Is it charging batteries too ? Quick guess, for the O2 less batteries it should pull under ??? 30 to 40 milliamps per rail depending on the opamps. If the input to the regs is around plus and minus 20 volts DC then each reg dissipates around 0.3 watt so they should not be hot.

If you have doubts then set your meter to read DC current and with the amp ON connect the meter across D1 and note the current. Repeat by measuring across D5 for the other rail. That will show the current drawn per rail without you having to unsolder anything. Be careful... don't short anything out. You are just measuring across those diodes on a current range. The meter effectively shorts the diode out and shows the current flow.

The sockets as drawn are of a type that has continuity when the plug is removed and so that has to happen to allow the signal to pass through (when using the off board input)

 

[sgtstan]

I measured across D1 and D5. They both say 21mA? The Regulators may have 30° to 35°C according to my "fingertip-thermometer". I just want to make sure its OK.

But the off-board input still doesnt work... That's weird?? 3,5mm works fine and sounds amazing



Edit: When I use the RCA-plugs to connect the source (ODAC btw) AND plug a connector (not connected to anything) to the 3,5mm jack, it works!? Why is that? I don't get it.

 

[UnknownAX]

But the off-board input still doesnt work... That's weird?? 3,5mm works fine and sounds amazing



Edit: When I use the RCA-plugs to connect the source (ODAC btw) AND plug a connector (not connected to anything) to the 3,5mm jack, it works!? Why is that? I don't get it.

You forgot to cut the tiny traces which short the inputs to ground when nothing is connected to the 3.5mm input.

Have a look at the second pic of the "tech section":
http://nwavguy.blogspot.com/2012/05/odac-may-update.html?m=1

 

[Mooly]

I measured across D1 and D5. They both say 21mA? The Regulators may have 30° to 35°C according to my "fingertip-thermometer". I just want to make sure its OK.

But the off-board input still doesnt work... That's weird?? 3,5mm works fine and sounds amazing



Edit: When I use the RCA-plugs to connect the source (ODAC btw) AND plug a connector (not connected to anything) to the 3,5mm jack, it works!? Why is that? I don't get it.

21 ma is fine... all is well there. So the amp is all OK Well done.

As for the RCA input, it seems unknownAX knows something so I would have a look at his suggestions.

Remember I have never built an O2 and have never seen the boards and so on... I just work to the circuit as printed.

OK, looking at the circuit follow the audio from pin 1 of P1 to the 3.5mm socket where it goes into pin 5. From pin 5 it links internally in the socket to pin 4 which is ground. Is this where these links are on the board ?

When you plug something into the 3.5mm socket the connection between 5 and 4 opens and the short to ground is removed. In that state either input will work, and obviously you don't want both to mix together.

So check out the article unknownAX mentions

 

[UnknownAX]

You could cut the little traces and connect the RCA's to the pins that are usually shorted to ground (pin 3 and pin 4). This way the RCA's are only used when theres nothing plugged into the 3.5mm jack, ie. you could leave your odac connected to the RCAs even if you use the 3.5mm input.

 

[Steve Mannix]

Hey all,. I stuffed the board for the o2 and began doing the testing described in this post http://www.diyaudio.com/forums/head...eadphone-amp-diy-project-189.html#post3806667.

I realized I had U5 and U6 installed backwards when I went to do the first step of the testing. After fixing that with a new one of each in the correct positions, the voltages for D3 and D4 are correct.

When testing the outputs of the regulators, the voltage for D1 is 0. Is it possible I fried D1 and D5 when I applied power with the regulators backwards? Any other ideas?

Thanks in advance and apologies if this was covered somewhere in the thread.

- Andy (Steve Mannix)

 

[Mooly]

Anything is possible. Zero volts on D5 but with the correct voltages on D3 and D4.

Make sure the voltage on D3 (stripy end and it must be 15v or higher) is getting to the input of the regulator. If it is and there is zero volts on D1 (non stripy end) then either the reg is faulty, the print is broken somewhere around the reg or the reg has a short from output to ground.

 

[availlyrics]

Would it be possible drill a bigger hole and put the 6.35mm jack to where the AC jack usually goes? Is there enough space for it inside the case? Except for D4 and R6 there shouldn't be anything there and I guess I could even mount those on the other side of the board. The jack is about 15x16x25mm. Panel mounting another AC jack on the back shouldn't be a problem.

6.35mm h/p jack near AC/rectifier section might inject noise, if possible locate the 6.35mm somewhere else.

 

[yetanothersteve]

After I finished my O2, I tried it with cheap 24 Ohm IEMs after verifying it would turn on and off without incident. I initially had a loud scratchy noise on turn off with the IEMs.
I am waiting on my DMM to arrive so I visually inspected to see if there were any obvious issues. A couple of solders were suspect and so I reheated those.
Found some solder/lead flakes on the board and gave it a good brushing with an old toothbrush and hit it with compressed air. Should have done the preceding first, but I was too eager to try it.
Now, with the 24 Ohm IEMs, the turn off is simple thump, 32 Ohm earbuds and 120 Ohm AKGs are both turning off with a barely noticeable thump.

This thread was very helpful. Thanks.

 

[Mooly]

Pleased you found it all helpful Steve

 

[TheGimp]

Use Isopropol alcohol with the tooth brush to remove flux that remains on the board. Flux can cause a leakage path in high impedance circuits.

 

[Fsatsil]

Anyone that's having an issue with their O2 build make sure to double and triple check all your solder joints. When I built my first O2 I was having a turn on/off thump problem that for the life of me I couldn't figure out. I finally decided to reflow ever single solder joint on the board and it's been working perfectly ever since. A bad joint isn't always obvious to the naked eye so use a magnifying glass and good lighting and reflow any joints that look suspect.

 

[Wao]

One quick question what would be affected if i buy Q2 with Rds 0.72 .

 

[yetanothersteve]

DMM arrived

Since the O2 is built, I measured according to this diagram.

Using a Fluke 115
P2: .003 and .003
D3: 16.41
D4: -16.38 (D3 and D4 are within 0.1 of each other but on opposite polarities)
U4: 11.66 and -11.75
D5: -12.02
D1: 11.93

Nothing seems to be horribly off, but I am not an expert. It sounds great, so there is that subjective analysis.

If I am done soldering, I will clean with isopropyl alcohol as TheGimp suggested. Found 99% at Fred Meyer's, the local Kroger subsidiary.

 

[Mooly]

One quick question what would be affected if i buy Q2 with Rds 0.72 .

Nothing of note. The higher Rds will drop more voltage across the FET but considering the current consumption of the O2 this amounts to around 20 millivolts... so no problem.