[Avro_Arrow]

I even wrote to Nelson and asked if he was interested in doing a "headphone sized" F1.
He said we had thought about it but that was as far as he got.
He said I was welcome to try myself and wished me luck...
 
Quote:

Replace R1, R2 with CCSs.
 
Why not use a more "Supersymmetric" topology if you want an active return path and are taking the very pass approach? It only requires a few more parts and should work a decent chunk better. You would have to reterminate your headphones balanced, but this is a small effort. I have been daydreaming of a headphone-sized F1 for a while (I call it the Head-F1 as a play on the Head-F1 HF2) but can never get off of my lazy bumm to get it built.

 

[Avro_Arrow]

Bump

 

[Avro_Arrow]

I wasn't going to do this but what the heck...
I added and op amp to the design to raise the
gain. The op amp also lowers the output
impedance. I used a single rail supply and an
output cap. You can tailor the value of the output
cap to suit you headphones. The value of 4700uF
is for 16 ohm. Of course you don't have to use
the op amp in the schematic, roll some different
ones and see how they sound...
You can add a ground channel and get rid of the
output caps if you like. The op amp also works
as a servo to keep the DC constant.
 

Edit: Sorry, the feedback was messed up...fixed

 

[Avro_Arrow]

bump

 

[TzeYang]

if you're going to add an opamp anyway, why not simply build PRR's MOSFET + OPAMP combo with two rails?

There's an elegance of using output caps when your circuit is that simple but if you have something that controls the output, having output caps is counter-intuitive.

 

[Avro_Arrow]

bump again...

 

[mugdecoffee]

 
Quote:

I wasn't going to do this but what the heck...
I added and op amp to the design to raise the
gain. The op amp also lowers the output
impedance. I used a single rail supply and an
output cap. You can tailor the value of the output
cap to suit you headphones. The value of 4700uF
is for 16 ohm. Of course you don't have to use
the op amp in the schematic, roll some different
ones and see how they sound...
You can add a ground channel and get rid of the
output caps if you like. The op amp also works
as a servo to keep the DC constant.
 

Edit: Sorry, the feedback was messed up...fixed

EDIT:  mis-read the schematic.  Not sure why you have the pot in there though.  The DC levels don't have to be precise since there's an output cap.

 

[nikongod]

P1 is not the volume control, the volume control is not shown. P1 is a trimpot to adjust the output offset voltage.
 
As drawn, this circuit is limited to a potentiometer at less than ~6Kohms if you want to maintain the rule of thumb that the pot=1/10 the input impedance of the amp. If you have a pot less than 6Kohms (5Kohm pots are abundant) you dont need the buffer to overcome the input capacitance of the mosfet do you now?.... If you want the input buffer to isolate the mosfet and give truly high input impedance R5 and R6 should both be in the meg ohm range.
 
Im not sure that using the three terminal oscillator is such a great idea. LM317 are not terribly expensive and very easy to plug in, but if you dont shown them the respect they demand they turn you around and bend you over. A mosfet CCS would not require many more parts (actually quite a few less if you think about whats in an LM317), and may actually cost less than the LM317.

 

[Avro_Arrow]

These designs are not set in stone...there just here to make you think.
I did present a discrete current sink earlier in the thread.
You don't really need a trimmer in the above design if you are building
a two channel amp with the output caps. I you build the three channel
version with no output caps then Nikongod is right, the trimmer set
the offset to zero.
 
I was kind of hoping other people might chime in and present
their ideas and designs rather than just rating if they think my
designs are worth while or not.

 

[TzeYang]

just so you know, these ideas are all discussed and done thoroughly before here and on headwize. Eventually, you'll end up with a PRR like MOSFET amplifier with two rails.You could add complexity to the output stage by adding a CFP, add some FETs to the rails, cap multiplier, small shunt regs to isolate the output stage and the input stage, but that's just about as complex as it gets before you turn it into a discrete amplifier.

Most of us chose to review and criticize for obvious reasons, it's been done before and we all know where it ends up.




 
 
 

 

 

[Avro_Arrow]

Thanks, thats good to know!
Now I know I don't have to waste my time reading or posting here.
I'm glad you already know everything...
 
Quote:

just so you know, these ideas are all discussed and done thoroughly before here and on headwize. Eventually, you'll end up with a PRR like MOSFET amplifier with two rails.You could add complexity to the output stage by adding a CFP, add some FETs to the rails, cap multiplier, small shunt regs to isolate the output stage and the input stage, but that's just about as complex as it gets before you turn it into a discrete amplifier.

Most of us chose to review and criticize for obvious reasons, it's been done before and we all know where it ends up.




 
 
 

 

 

[jcx]

some diy project collections - now getting dated but useful links:
 
http://gilmore2.chem.northwestern.edu/ubb/showpage.php?fnum=3&tid=6339&&action=seekpost&pid=51081
 
and
 
http://www.head-fi.org/forum/thread/136005/diy-projects-kits#post_1600278
 
 
not a comment on the value of this thread but if I put these links in a couple of relevant threads maybe they won't get lost so easily
 
 
 
I have shown a varition of a Pass design with impedance scaling for a heasphone load
 
http://www.head-fi.org/forum/thread/457265/a-few-caps-questions-urgent-please#post_6189600
 
 
but my opinion is that if you want accuracy then composite op amp circuits make objectively better amps for most dynamic headphones despite the audiophile meme of "discrete is better"

 

[regal]

Quote:

 
but my opinion is that if you want accuracy then composite op amp circuits make objectively better amps for most dynamic headphones despite the audiophile meme of "discrete is better"

Thanks for the thread crap.   This has nothing to do with "discrete"  you are missing the whole point of these project(s).
 
 
Hopefully I'll try to bring this back after that attempt at derailment,   what we are looking for here is a means to eliminate the output cap while retaining a complete bandwidth.  The active ground isn't a good means.  Didn't Melos have a servo taking care of the DC offset?

 

[nikongod]

 
Quote:

The active ground isn't a good means.  Didn't Melos have a servo taking care of the DC offset?

Whats wrong with active ground?
 
Yes, Melos had a DC servo controlling the output stage in the headphone amp.

 

[jcx]

a little quick with the "thread crap" card there regal
 
if you had clicked on the link to my post you would have seen a SE Class A Pass design - seems like another Pass amp and the idea of scaling them for headphones was introduced by the op - is he dragging the thread off topic now because the later one Avro_Arrow posted was actually push-pull instead of single ended?
 
and I have posted circuit ideas showing the principles behind using op amps with their outputs biased in Class A for the whole output range - I could easily show these with SE instead of push-pull - again it seems Avro_Arrow has included chips in input and output in his posts so far