The Objective2 (O2) Headphone Amp DIY Project

Administrator
Joined 2007
Paid Member
OK, so you are saying with the 2068 fitted that the voltages are now correct... very odd but it seems to prove the IC's are faulty.

The chips should be readily available all that will depend on your locality. You can fit other chips in order to get it all working.

4558's, NE5532's, 4560's even TL072's will all work in order to prove everything else is OK.

Look at the circuit... with the 2068 now fitted in place of a 4556 you could link either pins 5 and 7 or pins 3 and 1 of the original 2068 location (which is now empty) and you should hear music from whichever channel the 2068.
 
Administrator
Joined 2007
Paid Member
Mouser have them although I would suggest you prove it all good first by using something readily available such as the devices I mentioned:
 

Attachments

  • Annotation 2019-11-24 141132.jpg
    Annotation 2019-11-24 141132.jpg
    98.1 KB · Views: 167
I get +11 volts at pin 8, -11 volts at pin 4. All other pins are at 0V.







I could hardly call it music, more like very loud noise/popping and silently in the background some music coming through.



Are 4556's hard to get for replacement?
As Mooly said 4556s are easily available (and cheap), but I do have a few pairs of spare 4556s lying around that I'd be happy to send you should you need new ones. Just PM me, and we can work out the details!
 
Look at the circuit... with the 2068 now fitted in place of a 4556 you could link either pins 5 and 7 or pins 3 and 1 of the original 2068 location (which is now empty) and you should hear music from whichever channel the 2068.

Yeah, it plays. Nice :)

The chips should be readily available all that will depend on your locality. You can fit other chips in order to get it all working.

4558's, NE5532's, 4560's even TL072's will all work in order to prove everything else is OK.

I'll try to find alternatives in local stores, though I will need to ship the 4556's - i couldn't find any around.

So do you think that's what happened? Faulty opamps?
 
Administrator
Joined 2007
Paid Member
Yeah, it plays. Nice :)

I'll try to find alternatives in local stores, though I will need to ship the 4556's - i couldn't find any around.

So do you think that's what happened? Faulty opamps?

Opamps are pretty tough overall, so much so that its difficult to kill one unless you are very determined... so I don't know what has happened here tbh.

Could they be fakes? i.e. not even a dual opamp...

The only thing that bothers me is you mentioning the crackling when you moved things because whatever the problem is, it should be immune to physical inputs like that. I think all you can do is fit known good parts and see how it all plays out.
 
Opamps are pretty tough overall, so much so that its difficult to kill one unless you are very determined... so I don't know what has happened here tbh.

Could they be fakes? i.e. not even a dual opamp...

The only thing that bothers me is you mentioning the crackling when you moved things because whatever the problem is, it should be immune to physical inputs like that. I think all you can do is fit known good parts and see how it all plays out.

Opamps with DAC/AMP came from JDS so hardly believable it is fake. It also could be that there is malfunction somewhere else on the board that causes to break the opamps, idk.
Well, I'll give a try with the chips you mentioned before trying with 4556. Will report back.

Thanks Mooly much for all your help :cheers:
 
Administrator
Joined 2007
Paid Member
You're welcome... and good luck :)

(and as to opamp indestructibility, the only things that will realistically kill an opamp are incorrect supply polarity and/or feeding supply voltage into any of the other pins when the main supply rails are missing. Even then you have to be very very determined to kill it off by ensuring lots of current is available and also that it has somewhere to go through the other pins... and once its supplies are present its all pretty much protected anyway)

:xfingers:
 
Administrator
Joined 2007
Paid Member
Try moving/bending whatever you mentioned at the start and see if you can provoke any weird behavior.

The only scenario I can think of is one of the two rails disappearing and the opamps then placing either a - or + 12 volts across the headphones which would draw a fair bit of current... actual value depending on the DC resistance of the 'phones and how a 4556 behaves under a missing rail condition. Would that be enough to zap them though, and both in the same way... I still have big doubts on that.
 
Try moving/bending whatever you mentioned at the start and see if you can provoke any weird behavior.

The only scenario I can think of is one of the two rails disappearing and the opamps then placing either a - or + 12 volts across the headphones which would draw a fair bit of current... actual value depending on the DC resistance of the 'phones and how a 4556 behaves under a missing rail condition. Would that be enough to zap them though, and both in the same way... I still have big doubts on that.
Isn't the amp protected against one of the rails disappearing, by the comparator circuit? I thought it should shut off the amp should i.e. one of the battery come loose. Shouldn't it do the same trick, even when AC powered, shoud one of the rails fail for whatever reason?
 
I had used LM4562 and OPA1688 in the O2 (one of the very first DIY experiments) and the sound of LM4562 was much nicer and cleaner, but it sounded like it struggled with the 50r cans compared to OPA1688, most noticeably at higher levels (these were very low sensitivity planar HPs). I had always wondered if rebuilding the O2 output stage with twice the LM4562s would be enough to fix that.


With OPA1656 around the corner it would be a better idea to see how they compete with these OPA1688, OPA1656 has better distortion figures comparable to LM4562 though I feel like it had more to do with BJT input than distortion as LM mustve been distorting more into those loads.

Did you used these opamps in voltage-gain stage or as output buffers?

I wonder if anyone here tested OPA1656 in output buffer, paralleled, of course. That should do 150mA, but not sure about it's long term stability, and short-circuit protection.

Thanks!