FritzS
Been here a while!
Sound of Blue Danube
Sound of Blue Danube
Posts: 1,364
|
Post by FritzS on Jul 18, 2006 10:37:50 GMT
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 18, 2006 17:27:37 GMT
Anything that will provide current gain is worth a shot in a high current follower type circuit maybe the strength of the power fet over the mosfet less gate capacitance to overcome and possibly it is more linear in operation. Mosfets are tricky if you want to avoid a rolled top end and even though some seem to like that sonic signature (and why they like the mosfet) is not accurate.Is in fact as tone control used to tame a hot system or hot transducers. The way to avoid this tendency to rolled highs is found in the Nelson Pass design papers-Pure Class-A Operation (the most inefficient mode of operation known to humans for you EU members ) combined with my own findings that the source impedance driving the gate needs to be extrtemely low or you are in effect setting up a low pass filter in the audible range that is a combination of the impedance (the "R" componant) and the Gate (the "C" componant of an R/C filter).To get this up to the inaudible range the source Z,the "R" of the single pole -6dB Low Pass R/C filter needs to be very low and that means a buffered gate (an additional stage in thew case of the Szekeres) AND a tight layout So the power FET looks like it could be less of a headache used in the same type circuits for the above reasons though I have zero time with these parts so is only conjecture. BTW-you don't need to use those huge current pumping devices as used in the Pass FET amp if you are looking to drive headphones where you only need to be in the 1-2 watt area
|
|
FritzS
Been here a while!
Sound of Blue Danube
Sound of Blue Danube
Posts: 1,364
|
Post by FritzS on Jul 19, 2006 8:20:02 GMT
Anything that will provide current gain is worth a shot in a high current follower type circuit maybe the strength of the power fet over the mosfet less gate capacitance to overcome and possibly it is more linear in operation. Mosfets are tricky if you want to avoid a rolled top end and even though some seem to like that sonic signature (and why they like the mosfet) is not accurate.Is in fact as tone control used to tame a hot system or hot transducers. The way to avoid this tendency to rolled highs is found in the Nelson Pass design papers-Pure Class-A Operation (the most inefficient mode of operation known to humans for you EU members combined with my own findings that the source impedance driving the gate needs to be extrtemely low or you are in effect setting up a low pass filter in the audible range that is a combination of the impedance (the "R" componant) and the Gate (the "C" componant of an R/C filter).To get this up to the inaudible range the source Z,the "R" of the single pole -6dB Low Pass R/C filter needs to be very low and that means a buffered gate (an additional stage in thew case of the Szekeres) AND a tight layout Is the Szekeres with the additional JFET an good joice? Szekeres and his substitutes are good for low impedance headphones (Grado, ...) but for Sennheiser, Bayer, etc. (250 .... 300 Ohm) we need more voltage - HD600/650 ~ 6V till 10V peak (not RMS). Some decades ago Yamaha and a second Japanese firm has build pre- and power amps only with JFET's. I subscribed audioXpress too and there Pass presents some of his ZEN ideas - they are for winter operating but not for summer - in winter you can turn off your house-heating :-) I know about the smaller need of power for class A headphone amps. I am unsure which next headphone amp I would build - a WNA MKIII substitut, a tube amp or one of the "simple" solid state amps ..... a Szekeres/Murdey design with a pre gain-stage? HeadWize headwize.com/projects/index.htm have a lot of ideas .....
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 19, 2006 8:55:50 GMT
Yes.Not as a gain stage (though could be) but as a voltage follower to buffer the gate of the mosfet.What many forget is the input impedance "seen" by the mosfet gate is a combination of the driving source,the interconnect AND the varying resistance of the input pot which together is very high in value because it is a ZERO GAIN device and Grados need NO VOLTAGE GAIN.To use with headphones that do need voltage gain it is a simple matter to add a front end to provide it or use a different topology for the mosfet to provide both voltage and current gain though THEN it will not be suited for Grados so a switchable front end the only way to true dual use Vetical FETs,and as far as I know were custom devices no longer made have one and it runs in parallel (x2 Zen Amps) so you can imagine the heat AND electrical bill the murdey design has "issues" though I understand he has a V-2 out.If you want to experiment with the idea of an SE Pure Class-A Follower with gain you could do the Szekeres "honest",straight from the original schemo then experiment with various front ends to see what you personally prefer with YOUR headphones I know.there was a time when I was the most active participant of Headwize.Check the archives
|
|
FritzS
Been here a while!
Sound of Blue Danube
Sound of Blue Danube
Posts: 1,364
|
Post by FritzS on Jul 25, 2006 8:35:31 GMT
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 25, 2006 14:16:53 GMT
I have one as a headphone monitor amp siding my mixer.Superb amplifier that will drive everything including small loudspeakers on the desk if they are efficient this is the original: www.quadesl.com/pdf/hpart.pdfpcb available @ $45 per stereo amp though easy enough to do on a perf board (looks less professional but works as well ) www.quadesl.com/schematics.htmlhighly recommended *
|
|
FritzS
Been here a while!
Sound of Blue Danube
Sound of Blue Danube
Posts: 1,364
|
Post by FritzS on Jul 26, 2006 10:24:01 GMT
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 26, 2006 15:30:57 GMT
it should,they copied it then added a third channel for the ground so they could call it their own. The pcb shown is still the original I beleive so if you decide to purchase a board you need to specify to Sheldon Stokes which biasing version you want.The LED string IS an elegant solution and maybe best for most users. I just like control over everything in my life (control freak ) so suggesting the adj. bias version. As for the MOSFETs.Mine has the original parts listed being OLD as dirt,built when those were still readily available,uses an OP275 as the front end,has some minor ps changes (shottkey bridge,smaller PS caps.filtered IEC inlet),otherwise is dead nuts from the Headwize project modify the layout man ! It is DIY afterall ! Option #2 and one I use all the time for the AD825 or OPA627/637 dual to single opamp mod is an adapter like the Browndog Nice enough amp but nothing like the SDS.The Black Gate is a traditional opamp gain stage/current pump output stage using mosfets in place of the usual bipolars.The SDS actually biases the Op-amp output into class-A (read the article then look at the schematic ).It is all about the mosfet/opamp biasing scheme that makes the SDS amp what it is which is very smart even though called "Dated" or "flawed" by the imitators. Is the wheel flawed or dated ? Should we make a sqwuare one just because we can ? ;D
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 26, 2006 15:37:24 GMT
BTW : Mine is well travelled having been built then sold to finance another project and as time passed the "customer" decided to ove into an all triode system so contracted me to build him a new amp and I was able to weasle the SDS back for cheap (cost of my services reduction ). Damn glad I did too.A nice steady amp that will drive anything (including 600 ohm AKG cans at the flip of a switch !) sitting alongside my mixer in combination with a cue/monitor control
|
|