I can see some spikes possibly from the signal wiring around transfos and by having the IEC near enough also but at -120dB vicinity level they present no practical problem.
They contribute to the full audio band merged noise number worsening but you wouldn't hear them individually.
BTW for 33 Ohms the manual recommends 150mA bias optimally but since this is a compact build, and not to heat it up even more, prefer high sensitivity 33 Ohm HPs if you must. It produces distortion of only the 2nd harmonic highly dominant at such loads anyway.
Hi Salas, the 33R is just a test load I made to test HP amps, not an actual headphone I have. Indeed I'll stick to the 115ma bias I have so far.
Hi xrk971. The dual 50VA was recommended by Salas, and yes they are way overkill, but Salas had some very convincing arguments, core temperature, etc...
Part of the heat in the regs is because our line voltage is 120V, not 115V as these transfo are specs. That caused an higher voltage before the regs, 28Vdc, or a 11V differential, causing increased heating of the pass transistors.
A 16V would had been better, but it is not available in these kind of encapsulated frame. Also a lower VA rating would had been possible, but I had these 50VA in stock, and they were recommended by Salas, so...
Hi xrk971. The dual 50VA was recommended by Salas, and yes they are way overkill, but Salas had some very convincing arguments, core temperature, etc...
Part of the heat in the regs is because our line voltage is 120V, not 115V as these transfo are specs. That caused an higher voltage before the regs, 28Vdc, or a 11V differential, causing increased heating of the pass transistors.
A 16V would had been better, but it is not available in these kind of encapsulated frame. Also a lower VA rating would had been possible, but I had these 50VA in stock, and they were recommended by Salas, so...
Your Swiss cheese frame idea worked very well for assisting the output stage direct to box sinking in such a small non ventilated space though. The PSU sinks are calculated for 7V differential and some ventilation so 11V and no ventilation pushes them more but not terribly yet. Your circular mesh windows to add in their vicinity will surely help. Although a big enough rectangular mesh window over the sinks and bargraph Leds area would be possibly nicer for their warm deep orange glow to come out at night.
Cans only, my long time favorite AKG 702, I can't stand enclosed models. I have also A Sennheiser HD600, very nice a well, but I find the AKG more confortable on the long run. I listen to my music almost exclusively cans with when I'm working on the bench. Source is Mark Levinson ML390S
A shot in the dark like this
A shot in the dark like this
Attachments
Last edited:
Cans only, my long time favorite AKG 702, I can't stand enclosed models. I have also A Sennheiser HD600, very nice a well, but I find the AKG more confortable on the long run. I listen to my music almost exclusively cans with when I'm working on the bench. Source is Mark Levinson ML390S
A shot in the dark like this
I know what you mean, I have used a friend's K701 during the prototype's final evaluation period along my HD600 and other models while on the bench. Maybe seven different cans across 33 to 600 Ohm. Fostex planar even. For some reason the K701 benefitted most noticeably from the DCG3 preamp. It really fleshed out in tone and also breathed more vs directly off my Aune S16 DSD DAC which brags a special quality 3W at 33 Ohm HP amp output. My friend who owns this 701 for many years claimed that it was the first time he really heard what is it all about and that he finally liked it. He was always finding it dry before. He had MF X-Can tube SS hybrid, now he uses a Benchmark DAC1 as HP amp. The 701 and 702 are acoustically the same model I believe. You know what is the funny thing though? I have designed and tested the DCG3 as a line preamp mainly but I added the HP output as a convenience later just because it could.
Attachments
And one of the other cans amp I like is my Borbely all FET (Audial clone), also originally designed to be a preamp Same ADN, discrete dual mono supply, MOSFET final stage, JFET input, DC servo, Class-A bias. Even the ventilation is pretty similar. I guess I like this approach...
Same ADN, discrete dual mono supply, MOSFET final stage, JFET input, DC servo, Class-A bias. Even the ventilation is pretty similar. I guess I like this approach...Attachments
Last edited:
Try a wire from your ALPS ground pins to make a connection nearby to chassis. Direct or via few Ω resistor for loop break. It usually helps against various fields when bringing the box to the same electrical level as the preamp's signal return. Especially the pot's metal body parts must. This was the open plan prototype's noise floor with its "chassis" both mains earthed and signal grounded. The black wire from the pot's return pins goes to the aluminum frame and the test lead at the other end goes from the frame to the bench PSU's mains earthed chassis. It was just a handy nearby binding post at mains earth potential.
*Cursor at 50Hz mains frequency, inputs shorted via the pot set at minimum, much of the 1/f rise is the audio card's own.
*Cursor at 50Hz mains frequency, inputs shorted via the pot set at minimum, much of the 1/f rise is the audio card's own.
Attachments
I have to report, after I spend like month with AK4490_DCG3_GradoPS500, that this combo sounds really really really nice
I can listen them for hours no matter how Grado are uncomftroble
Thanks once again to Salas, Tea and Lyben
(one pic from office, I don't mind the heat...haven't drill holes like Algar did)
I can listen them for hours no matter how Grado are uncomftroble
Thanks once again to Salas, Tea and Lyben
(one pic from office, I don't mind the heat...haven't drill holes like Algar did
)