The Objective2 (O2) Headphone Amp DIY Project

I’m very late to the O2 party. I bought the kit from JDS Labs a few months ago and assembled it. It ran perfectly until a few days ago. It has developed an audible hiss in the left hand channel when plugged into the AC transformer. It is silent when running on batteries.

I’m following the testing and troubleshooting steps on NwAvGuy’s blog. Resistances are good. I popped out U1, U3, and U4 to test the voltages. So far, everything is within spec.

Any suggestions on what to test next would be greatly appreciated. I have a basic understanding of electronics and I really want to learn more.
 
Caps from power supply?

I got a new multimeter for my birthday that test caps, so I'm doing that now.

Is the level of hiss affected by volume control setting?

Hiss is taken to mean white noise, like an FM radio between stations. Is that the sort of noise you hear?

The hiss is affected by the volume control, implying that the issue is upstream from the pot, no?

Yes, the noise is like FM radio between stations.
 
The hiss is affected by the volume control, implying that the issue is upstream from the pot, no?

Yes, the noise is like FM radio between stations.

Yes, upstream. That is a very strange issue indeed, hiss is strange as a fault in one channel.

Ideally this is oscilloscope territory to see where the noise is coming from but you can check voltages for clues.

Are all the opamps in sockets? If they are then remove only U1 and see if the hiss is still present or not.
 
Yes, upstream. That is a very strange issue indeed, hiss is strange as a fault in one channel.

Also, level of hiss is affected by the gain I select. Higher gain = louder hiss.

Ideally this is oscilloscope territory to see where the noise is coming from but you can check voltages for clues.

Don't tempt me as I really want to get a decent entry level oscilloscope. Right now, all I have is one of those DSO138 kits you can get from Ali Express.

Are all the opamps in sockets? If they are then remove only U1 and see if the hiss is still present or not.

I had U1, U3, and U4 out for testing. All of the voltages are within spec. Popped U3 and U4 back in and there's no hiss.
 
This certainly isn't a typical sort of fault. Hmm...

So with U1 removed there is no hiss.

What I would do now is remove U1 again and try linking pins 1 to 3 and 5 to 7 which by passes the opamp. If you have some low value resistors or bits of resistor leg you can just push those in the socket pins as a quick test to link them. Any noise present before where the opamp was will still pass through but with no gain.

Screenshot 2025-03-16 143633.png


Hiss isn't normally produced by a power supply fault but it is remotely possible the comparator and FET rail switching could be at fault but try the above first.

Also make absolutely sure it isn't an issue with anything connected to the input and make sure there is no damage to any of the grounding around the inputs.

You can link the FET's out like this which should prove if the noise is caused by the comparator switching. Hard to see how it could be really given all the decoupling caps. Try all the above first before suspecting something power supply related.

If you suspected an opamp you can swap the 2068 with the 4556's
 
  • Like
Reactions: beffie
What I would do now is remove U1 again and try linking pins 1 to 3 and 5 to 7 which by passes the opamp [...]

Bypassing 2068 gives me zero noise.

I happen to have an extra 4556 from another project I've yet to mess around with, so I tried that and the amp is dead quiet again. Put the 2068 back in temporarily and the left channel sounds like my old Corolla's radio.

So, the 2068 is out of spec?

Edit: Also, thank you for the quick responses!
 
Apparently so, and how very very unusual. More than just out of spec, its suffered some bizarre failure. I can count on the fingers of one hand opamps I have had genuinely fail and that is over many many years when it was the day job.

One possibility for failure could be something plugged into the input that gave a voltage spike, there are possibilities here for that to happen such as connecting to something mains powered with an SMPS supply and there being high leakage currents. Hot swapping inputs in such situations can be problematic.

You can use the 4556 or choose something different, There are zillions to choose from and if you use battery then low power devices can be beneficial. Don't discount something like a TL072 or TL082 or there are the OPA devices like the OP2134
 
  • Like
Reactions: beffie
Could a power surge or static discharge cause this? Granted, I have the AC transformer plugged into a surge protector. Also, I have pin 1 on the input jack grounded to the case as per assembly instructions.

I hot swapped between my two iPods, but that's about it, and I usually run the Pods off battery.

You can use the 4556 or choose something different, There are zillions to choose from and if you use battery then low power devices can be beneficial. Don't discount something like a TL072 or TL082 or there are the OPA devices like the OP2134

Now you're just giving me excuses to order more stuff.

Again, thank you for the help!
 
  • Like
Reactions: Mooly
We'll never know I guess. It may just be some bizarre random one off failure (although incredibly rare its not impossible) or maybe it got impaired or zapped during construction with a soldering iron with high leakage that led to a fault occurring many hours down the line.

I can't see your power supply doing this as there is just so much decoupling on the rails. A voltage spike could not have a fast rise time with all those caps. There would not be enough energy to transmit to the rails.

A spike on the inputs could do damage in theory.... maybe you were carrying the iPod across a nylon carpet in dry air and you and the iPod were able to give a static discharge into the inputs as you plugged it in. All guesswork and conjecture, we just do not know.

Also the 2068 opamp is just a normal bjt type and so one of the least sensitive to static.
 
  • Like
Reactions: beffie