oke.. there is one really simple way to scale the SOZ power amp circuit.. multiply every resistor by four and use a +&- 15 Volt supply .. .. that is almost too simple .. I know .. but it works really well.. please understand that I like to keep my final version (both a Briged and a SE design) and specs to myself for a while.. I would realy like to publish an article in Headwize or audioexpress...
Mister Pass: would this not be a realy nice new project for you excelent webside ... ?
I use irf610 MOSFET's with a small headsink and a Grado SR60, but remember, you gonna have to modify your headphone cable to separate the return connections of your headphone transducers..
I think a headphone amp is a real nice simple first project for a DIY'er.... very good results are easyly obtained and probbably amaze how good a headphone can sound...
goodluck
Mister Pass: would this not be a realy nice new project for you excelent webside ... ?
I use irf610 MOSFET's with a small headsink and a Grado SR60, but remember, you gonna have to modify your headphone cable to separate the return connections of your headphone transducers..
I think a headphone amp is a real nice simple first project for a DIY'er.... very good results are easyly obtained and probbably amaze how good a headphone can sound...
goodluck
That's in line with my first thought, except I'm driving the phones single-ended so I multiplied resistors by 8 (SE 32 ohms => Diff 64 ohms). I calculate about 200 mA in each leg of the circuit; I think I can get away with 5 watt resistors.
Okay, here's a precise question: What about raising the supply levels to +/- 25 Volts, adding resistance to the top and bottom resistors to keep the current the same (thereby wasting even more power)? Is this better, because a larger value resistor is a better "current source", or worse, because the damping factor is increased? Or does it matter?
I happen to have a +/- 25 volt supply.
Okay, here's a precise question: What about raising the supply levels to +/- 25 Volts, adding resistance to the top and bottom resistors to keep the current the same (thereby wasting even more power)? Is this better, because a larger value resistor is a better "current source", or worse, because the damping factor is increased? Or does it matter?
I happen to have a +/- 25 volt supply.
Good question... I asked myself the same thing: what is more importend, high resistance or high Iq (both increasing dissipation). It seems the circuit benefits more from a higher Iq then a higher resistance with higher voltage supply. I think Pass mentioned something like that in the Zen articles. So I use lower supply voltage (even 12Volt will do , cheap ) at 250mA max/channel. Higher Iq gives better lineairety but becomes a little out of proportion.
greetings,
thijs
) at 250mA max/channel. Higher Iq gives better lineairety but becomes a little out of proportion.greetings,
thijs
Refreshing this thread...
Okay, I bought myself some Sennheiser 580s and it looks like I can easily change the plug to a 4-wire one so as to be able to drive them differentially.
Thus I'm looking at this again.
When we last left this, tschrama had scaled resistor values but was keeping the details to himself (herself?).
Tschrama, if you're listening, any progress?
Okay, I bought myself some Sennheiser 580s and it looks like I can easily change the plug to a 4-wire one so as to be able to drive them differentially.
Thus I'm looking at this again.
When we last left this, tschrama had scaled resistor values but was keeping the details to himself (herself?).
Tschrama, if you're listening, any progress?
Another way you could go about it is to run a P channel common drain into a n channel common drain. No voltage gain, but who cares with headphones? It would be DC coupled with reasonable offset, especially with a little matching. Not a Zen but one could claim it was Inspired By Zen (tm).
It would also avoid the common ground issue.
It would also avoid the common ground issue.
Hi... Paulb ... progress on the headphone department is slow.. I have choosesn not to modify my headphone wires, but to optimize the single ended headphone amp ... Now I use a MOSFET folower, loaded by a MOSFET/LED constant current source, the output is through a 1000uF electrolyt... the whole thing is powered by a NEGATIVE voltage sypply (16000uF and some MKT bypass cap's) at the CCS... This way the input does not need a bias voltage so no input cap is nessecery .. I think that's nice.... This only works if the voltage demand of the amp is less then 1 volt or so, ... so a low impedance headphone is needed for loudness...
I can publish my balanced designs on my webside or email them... but my computer at home is broken (never buy a Maxtor HardDisk!!) so it might take a few days to find them...
Last week I changed the two opamp in my Philips CD723 CD player ... from JRC4580 to Burr-B opa2312's one fpr the analog out and one for the headphone out.... made a real difference in the high tone like drum Hi-Hat ... but not more pleasing then the were... :-( .....
succes,
Thijs
How do you like your new headphones ??
I can publish my balanced designs on my webside or email them... but my computer at home is broken (never buy a Maxtor HardDisk!!) so it might take a few days to find them...
Last week I changed the two opamp in my Philips CD723 CD player ... from JRC4580 to Burr-B opa2312's one fpr the analog out and one for the headphone out.... made a real difference in the high tone like drum Hi-Hat ... but not more pleasing then the were... :-( .....
succes,
Thijs
How do you like your new headphones ??
Yes, Nelson, I believe what you described is known in Headwize.com land as a Szekeres after the fellow that published the project. I've actually started building one but have revised my thinking. I want to get rid of the output cap.
I thought a SOZ would be more linear than a single FET, and I can afford the power so am trying to decide whether to cut into the cable of my new $350(CDN) headphones.
I thought a SOZ would be more linear than a single FET, and I can afford the power so am trying to decide whether to cut into the cable of my new $350(CDN) headphones.
Hi Ebrewste (?)...
your idea reminds me of a webside that features some 'zero offset J-FET buffers schematics' ... I think you need a symmetrical voltage supply for that... which I just happen to dismis .. it makes things even simpler and cheaper .. .. futhermore .. a balanced design can reduce second order harmanic generation by a factor of 10 or even more ..
I also think that DC offset in a Headphone is deadly for lineairity of the transducers... I have no measurements on this subject.. but speakers can rely on the compliance ratio of their BOX design and (open) headphones can not, headphones are typically 1000x more sensitive than speakers..I do not want 1 mV offset or more!
Still a SOZ for headphones appeals to me .. maybe a DC-offset monitor would be nice , so I could manualy adjust offset by a trimmer.....
greetings,
Thijs
your idea reminds me of a webside that features some 'zero offset J-FET buffers schematics' ... I think you need a symmetrical voltage supply for that... which I just happen to dismis ..
it makes things even simpler and cheaper .. .. futhermore .. a balanced design can reduce second order harmanic generation by a factor of 10 or even more ..I also think that DC offset in a Headphone is deadly for lineairity of the transducers... I have no measurements on this subject.. but speakers can rely on the compliance ratio of their BOX design and (open) headphones can not, headphones are typically 1000x more sensitive than speakers..I do not want 1 mV offset or more!
Still a SOZ for headphones appeals to me .. maybe a DC-offset monitor would be nice , so I could manualy adjust offset by a trimmer.....
greetings,
Thijs
I just went on simulating again ... . this time with great results..
I desided to use a POWER MOSFET.. because of the reply of N. Pass ..
I tried IRFP140's mosfet folower, biased by a 47 Ohm resistor at 130mA.
this seems to be more lineair than a IRF510 or IRF610 at 200mA.. but they also have a higher Cap. than the IRF510's
hmmmmm.... a SOZ design might only be of benifit when it's driven by a BOSOZ or something like that...
..
Wouldn't this be a great addition to the Pass projecs?? ;-)
greetings,
Thijs
I desided to use a POWER MOSFET.. because of the reply of N. Pass ..
I tried IRFP140's mosfet folower, biased by a 47 Ohm resistor at 130mA.
this seems to be more lineair than a IRF510 or IRF610 at 200mA.. but they also have a higher Cap. than the IRF510's
hmmmmm.... a SOZ design might only be of benifit when it's driven by a BOSOZ or something like that...
..
Wouldn't this be a great addition to the Pass projecs?? ;-)
greetings,
Thijs
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.