I use 5-pin Mini XLR.
So I don't worry about any (self-made) problems with TRS plugs.
Patrick
Guess I will power down before changing phones just to be safe. In hindsight, with a trs jack current limiting would have been a good idea. Iam thinking the mosfets won't like being shorted for very long.
So then it should be no trouble running alot higher current and 18-20V, when mounting them on the side sinks with my own pcb's.
By the way Patrick, the large electrolytic caps, they are only for decoupling right? If so, wouldn't removing them (or replacing with really small value film) give possibility to use capacitance sensitive PSU's like salas shunts?
By the way Patrick, the large electrolytic caps, they are only for decoupling right? If so, wouldn't removing them (or replacing with really small value film) give possibility to use capacitance sensitive PSU's like salas shunts?
I have 16V rails and 150ma bias.
My feeling is it could take a lot more before the chassis temp was an issue. 18-20V should be no problem. 300ma and 20V rails feels possible if you can deal with 45-50c chassis temps. I have not tried though. But do you really want a headphone amp chassis temp that high?
My feeling is it could take a lot more before the chassis temp was an issue. 18-20V should be no problem. 300ma and 20V rails feels possible if you can deal with 45-50c chassis temps. I have not tried though. But do you really want a headphone amp chassis temp that high?
I would choose them mostly because that i have my own made compact PCB's (two regulators per board built with SMD/through hole components), and also components for them at home. And they do sound very good with your SEN I/V, and Pass B1 buffers i already have.
So i see no reason why they wouldn't sound good here also - if they would be stable with removed capacitance of course.
There is no other reason than for decoupling the caps are there i guess? If so, then i would guess there is no harm replacing them with small film caps.
Also if they would work, i will have room for both the SSLVs and the actual F5-HA on side sinks, so heat woudn't either be any trouble.
I'm probably going to aim for 200-250mA F5-HA's with around +-18V supplies.
Also have som FdW shunt pcb's, but i think i will save those for other projects, because they need alot more space.
So i see no reason why they wouldn't sound good here also - if they would be stable with removed capacitance of course.
There is no other reason than for decoupling the caps are there i guess? If so, then i would guess there is no harm replacing them with small film caps.
Also if they would work, i will have room for both the SSLVs and the actual F5-HA on side sinks, so heat woudn't either be any trouble.
I'm probably going to aim for 200-250mA F5-HA's with around +-18V supplies.
Also have som FdW shunt pcb's, but i think i will save those for other projects, because they need alot more space.
Yes, i read about that - but looking at schematics i was thinking that the instability was because of the big caps on the F5-HA boards..
I have read many people that have had trouble with SSLV's oscillating when there is big caps at the load.
That is the reason why i put small (0.25 - 0.5R) series resistors before your SEN i/v's, which made the circuit stable at that place.
So my thought was to try to remove the 1000uF caps, and replace with small 0,1uF film - then feed with SSLV. If there is not any reason they need to be there for F5-HA stability.
Otherwise i'll probably design some cap multiplier boards and order those also.
I have read many people that have had trouble with SSLV's oscillating when there is big caps at the load.
That is the reason why i put small (0.25 - 0.5R) series resistors before your SEN i/v's, which made the circuit stable at that place.
So my thought was to try to remove the 1000uF caps, and replace with small 0,1uF film - then feed with SSLV. If there is not any reason they need to be there for F5-HA stability.
Otherwise i'll probably design some cap multiplier boards and order those also.
Concerning noise, I listening to one channel only (I am still working on the board for the other channel) last night.
Extremely low levels of background noise with just a 317/337 regulator. Transformer is in the same case as amp. And this was with sensitive Grado headphones. Noise was inaudible until the last click of the stepped attenuator with source connected but not playing.
I have not experienced a headphone amp with such low levels of background noise.
I could not judge the SQ with just one channel working. But I could tell it had plenty of snap and bass. But it's difficult to judge vocals, etc. with one channel.
Extremely low levels of background noise with just a 317/337 regulator. Transformer is in the same case as amp. And this was with sensitive Grado headphones. Noise was inaudible until the last click of the stepped attenuator with source connected but not playing.
I have not experienced a headphone amp with such low levels of background noise.
I could not judge the SQ with just one channel working. But I could tell it had plenty of snap and bass. But it's difficult to judge vocals, etc. with one channel.
It's done. Biased up fine, offset is very stable and under 5mv.
Regulators are 317/337, all signal and DC lines are shielded. Pot is 10k stepped smd resistor. Bias is 180ma, rails 16v. Amp stays very cool and has very little noise.
Iam going to listen for few days and then let you guys know how it sounds.
Regulators are 317/337, all signal and DC lines are shielded. Pot is 10k stepped smd resistor. Bias is 180ma, rails 16v. Amp stays very cool and has very little noise.
Iam going to listen for few days and then let you guys know how it sounds.
Attachments
I try not to get too academic.
In his setup, he does not have a X'FD, so the RCA input goes directly to the Pot before the F5-HA.
To measure the noise of the F5 itself (as opposed to the entire audio chain), one needs to short the input to Gnd somewhere, to avoid e.g. RF pickup.
This you can do at the RCA input (I use a special short circuit plug).
And if you then also put the volume to max, the noise measurement will include :
1) the RCA plug-socket contact
2) noise of the volume pot (carbon) resistive track, plus pot wiper contact
3) the F5-HA (plus e.g. relay contact of a protection circuit at the output)
4) the contact of the output jack to the headphone.
Turn the pot to the minimum allows you to take the pot resistive track and the RCA contact out of the equation.
Using an LNA with high Zin to measure the noise at the F5-HA output on the PCB allows you to eliminate 3 & 4 further.
But of course, real life listening includes all of the above.
Patrick
In his setup, he does not have a X'FD, so the RCA input goes directly to the Pot before the F5-HA.
To measure the noise of the F5 itself (as opposed to the entire audio chain), one needs to short the input to Gnd somewhere, to avoid e.g. RF pickup.
This you can do at the RCA input (I use a special short circuit plug).
And if you then also put the volume to max, the noise measurement will include :
1) the RCA plug-socket contact
2) noise of the volume pot (carbon) resistive track, plus pot wiper contact
3) the F5-HA (plus e.g. relay contact of a protection circuit at the output)
4) the contact of the output jack to the headphone.
Turn the pot to the minimum allows you to take the pot resistive track and the RCA contact out of the equation.
Using an LNA with high Zin to measure the noise at the F5-HA output on the PCB allows you to eliminate 3 & 4 further.
But of course, real life listening includes all of the above.
Patrick