Should I worry...

[Andrea]

For some reason, there's what seems a tiny, nearly inaudible oscillation in one channel only of my modded Zero amp (see sig), which with the HD650 I hadn't yet noticed.

Yesterday I had this impression of hearing a subtle wisper on the right channel. Switching to the PX100, it became much more clear. Sort of a "sad wisper" if you will.

Now, should I worry for my HD650 ? Is there a remote possibility that such a sort of oscillation could have damaged my headphones? Please be totally frank

 

[vixr]

A nearly inaudible ocsillation, and a totally invisible attachment.

 

[rickcr42]

Oscillation can not only destroy your headphones by hitting them with energy in the ultrasonic region continously thus heating up the voice coils unilt they go "ping" but may actually burn up the amp at some point because it has no way to deal with the increased heat buildup.

Oscillation may be above the range of human hearing but in electronic terms is just another signal and as such will be treated as one even if that signal is harmful

 

[rickcr42]

BTW-what makes you think it "suddenly" went into oscillation ?

BTW 2-simple fix is to check your byapssing/grounding scheme ,clean it up then add loop compensation (bandpass filtration in the feedback loop)

 

[Andrea]

Thanks, but couldn't you be more practical/clear... i.e., might this almost inaudible oscillation have put my HD650 in trouble at all ?

Btw, today I experimented a bit and found that it were the BUF634's that I had mounted on a bouble DIP BrownDog with 150 ohm BW resistors (to play the part of the output follower opamp of my Zero) to oscillate. First I tried cutting the 150 ohm resistors and suddently, in low BW mode, the oscillation went away. Then I removed the buffers and put an AD8620 in, and still not the least (audible at least) oscillation. So I'll definitely keep the AD8620 in, and just add 47 ohm resistors at the output, inside the feedback loop.


Also, I had first taken care to see if one of the buffers was running hotter than the other: no way, both were equally cool (as much as they're supposed to be with 150 ohm BW resistors). I even failed to guess from chip temperature which buffer was the channel showing the oscillation... It proved to be the other one. Is this reassuring about the smallness of the oscillation and thus the safety of my 'phones?


Thanks, and sorry for teasing

 

[Andrea]

Quote:

Originally Posted by rickcr42 BTW-what makes you think it "suddenly" went into oscillation ?

No it wasn't sudden, it's just that I'd been using the HD650 exclusively with the Zero since my buffer mod (a couple of days before), and thus I hadn't even noticed this very faint whisper in one channel only, until I switched to the more efficient PX100.

 

[Andrea]

Quote:

Originally Posted by vixr A nearly inaudible ocsillation, and a totally invisible attachment.

Sorry, but it sounds almost criptical to me non native english speaker... Do I take it as a reassurance?

edit- I think I've figured out what you meant

 

[Garbz]

Do the headphones sound alright? That would be your answer. It's quite alright to cook the voice coils a bit. A mate of mine frequently overdrives his speakers. Pulling the drivers out a while ago the coils were a bit charred but they sounded every bit as good as they used to. So yes damaged but no reason to worry untill they start to distort or blow up.

 

[Andrea]

Quote:

Originally Posted by Garbz Do the headphones sound alright? That would be your answer. It's quite alright to cook the voice coils a bit. A mate of mine frequently overdrives his speakers. Pulling the drivers out a while ago the coils were a bit charred but they sounded every bit as good as they used to. So yes damaged but no reason to worry untill they start to distort or blow up.

If this oscillation was as dangerous as you paint it, there would be at least a bit of overheating of the BUF634 corresponding to the channel that oscillates...

Which is not my case, as I said above.

And that was with a 32 ohm headphone connected (so, potentially, more excess current delivered to the voice coil because of the oscillation), not the 350 ohm HD650.

 

[rickcr42]

Quote:

A mate of mine frequently overdrives his speakers. Pulling the drivers out a while ago the coils were a bit charred but they sounded every bit as good as they used to. So yes damaged but no reason to worry untill they start to distort or blow up. __________________

that may work with "in band" signals,signals the speaker is designed to reproduce but I highly doubt any tweeter will handle real power in the ultrasonic regions for any length of time without totally frying the voice coils.

If you look at raw speakers made for the pro arena (large format studio monitors,P.A. systems for bands) you will find there is often a HUGE heat sink area plus thermally conductive liquid in the voice coil gap to pass this heat to the actual heat radiating device AND tweeter protection circuits !

Why ?

Long term high volume power handling which means HEAT

Take that to a square wave generator whic has infinite additive overtones (multiples of the original go on and on and on and....) in the ultrasonic region.
Power and heat but being inaudible no clue as to why a driver self destructed because other than a slight "brittle" quality or even what some call "detail" may in fact just be the lower edge,the audible part,of oscillation.

Just because humans can not hear a thing does not mean it is not there then you toss into the equation a small driver that likely CAN respond to an ultrasonic pulse (being small) and if that pulse is continous and has high energy can in fact blow a seriously underprotected driver such as is found in headphones.

 

[Andrea]

Funnily, the amp sounded much smoother (mid-treble range) with the high bandwidth mode for the buffers, oscillation and all, than with the low bandwidth and no trace of oscillations anymore. All with the PX100.

 

[rickcr42]

if oscillation IS the problem that could also happen since the buffer sounds better across the board in high bandwidth mode.

the only real test is to put the amp on a high bandwidth scope so you can actually SEE what is going on.If there is signal above the actual signal input frequency on the scope then it is oscillating and you need to find out why.couldbe a simple bypassing scheme screwup