Thanks, again to you and Papa of course.
Maybe this article shed some light into the black hole, which is at the place of my brain.
(That light will be swallowed up without a trace, I'm afraid.)
Meanwhile, reborn J2 plays some of my favourites from Swiss minimal music experts, throgh my Quad ESLs.
I like it, and this is what matters.
Maybe this article shed some light into the black hole, which is at the place of my brain.
(That light will be swallowed up without a trace, I'm afraid.)
Meanwhile, reborn J2 plays some of my favourites from Swiss minimal music experts, throgh my Quad ESLs.
I like it, and this is what matters.
Questions about C1 and C2:
C1: Can we omit?
C2: Why is it 10 uF? There was 220 uF in your article.
What is the purpose of this change in value?
Thanks, as usual.
C1: Can we omit?
C2: Why is it 10 uF? There was 220 uF in your article.
What is the purpose of this change in value?
Thanks, as usual.
C1 is the phase lead compensation. It's required -- the amp will be unstable without it.
C2 smooths the output of the bias spreader so it doesn't try and follow the music signal. Anything between 1u and 1000u is probably fine.
C2 smooths the output of the bias spreader so it doesn't try and follow the music signal. Anything between 1u and 1000u is probably fine.
Wait a minute, I meant 10 uF input capacitor on balanced -line, not that C3 10pF in feedback loop.
Or I misunderstand something?
Or I misunderstand something?
Last edited:
My previous build has needed input coupling capacitor even on +line input, because of presenting output DC offset.
Perhaps, it was because of the imperfection of my previous preamp.
Perhaps, it was because of the imperfection of my previous preamp.
Yes, the 10uF on the input can be omitted -- but you'd want to be very sure that all your inputs were DC-free.
O.K..
As time goes by, this case seems to be of no effect, as input elco properties evolved. Sound is become better and better.
I usually don't believe snake oils and similar things.
But it is an experiential fact by me, electrolit condensators' sound is improving, as they began works.
Perhaps, I will implement a removable jumper underside the PCB at this point.
As time goes by, this case seems to be of no effect, as input elco properties evolved. Sound is become better and better.
I usually don't believe snake oils and similar things.
But it is an experiential fact by me, electrolit condensators' sound is improving, as they began works.
Perhaps, I will implement a removable jumper underside the PCB at this point.
I just hooked up my KEF Q75s to the J2, because I need more SPL at the moment, and I don't want to kill ESLs.
Then, it turned out, there is need for more power than J2 can provide, so I switched to my monster Aleph.
Then, it turned out, there is need for more power than J2 can provide, so I switched to my monster Aleph.
A little bit off topic:
Well, I don't want to demolish the house.
But the car is also in the garage where I listen to music.
If there was any rust on it, was all fallen down now.
What is Hi-Fi good for?
Well, I don't want to demolish the house.
But the car is also in the garage where I listen to music.
If there was any rust on it, was all fallen down now.
What is Hi-Fi good for?
@Jeff
A couple of things... I have the 20v trans here, told ya I wanted that volt. P
What is the max we want to run on the rails and also, one of the Semis is running a bit hotter than the rest. That one is up to 69-71C and the rest are 56-60C. I know the max temp is 150C, but I find it strange that the one is so much above the rest. When I put the new tranny, in I will double check the resistors, but I checked twice before I fired it.
A couple of things... I have the 20v trans here, told ya I wanted that volt. P
What is the max we want to run on the rails and also, one of the Semis is running a bit hotter than the rest. That one is up to 69-71C and the rest are 56-60C. I know the max temp is 150C, but I find it strange that the one is so much above the rest. When I put the new tranny, in I will double check the resistors, but I checked twice before I fired it.
The input JFETs are spec'ed at 25V (30V if you used SJ109s); the CCS JFETs at 40V. Each of them sees about 1/2 the rail-to-rail voltage. So that would say don't go any higher than +/-25V (or +/-30V).
However, Papa sez the Toshiba JFETs don't actually break down until around 40V, so there's at least some leeway available, depending on how much you want to trust Toshiba quality.
(The optocoupler only sees a few volts and is good for 70V anyway. The SemiSouths are good for 1200V, so no issues there.)
The higher you go the more issues you're going to have with heat, though. 70º on a SemiSouth wouldn't bother me, but I wouldn't want it much higher (remember there will still be a delta between the case temp and the junction temp).
Check the temps of the heatsinks right next to the JFETs. If that's also hotter on the hot one, then it's an electrical issue. If it's the same on all 4, then the hot one is a mechanical issue (thermal paste, screw torque, or something).
However, Papa sez the Toshiba JFETs don't actually break down until around 40V, so there's at least some leeway available, depending on how much you want to trust Toshiba quality.
(The optocoupler only sees a few volts and is good for 70V anyway. The SemiSouths are good for 1200V, so no issues there.)
The higher you go the more issues you're going to have with heat, though. 70º on a SemiSouth wouldn't bother me, but I wouldn't want it much higher (remember there will still be a delta between the case temp and the junction temp).
Check the temps of the heatsinks right next to the JFETs. If that's also hotter on the hot one, then it's an electrical issue. If it's the same on all 4, then the hot one is a mechanical issue (thermal paste, screw torque, or something).
The sink is hotter as well. I don't know if I got a FET acting up, or if something else is going on.
Jeff,
would you please recommend values for 1.7A bias preserving original single ended flavour
would you please recommend values for 1.7A bias preserving original single ended flavour
Lowering R4 will increase the bias contribution of the current source, increasing the push-pull flavour a bit and decreasing the singled-ended flavour.
You can also adjust the flavours by modifying the ratio of the source resistors. If the total resistance stays the same then the bias will be the same.
Conversely, if you adjust the total resistance but leave the ratio the same the bias will change but the flavour will be the same.
Or at least that's my understanding of it.
Sweet!
Target values would be 0R28 and 0R42.
Caddock MP930s in 0R3 and 0R4 would be reasonably close.
Or you could use a pair of 3W Panasonic ERXs in parallel for each (2 x 0R56 and 2 x 0R82). That would get you much closer at the expense of not being quite as neat.
Cheers,
Jeff.
@ Jeff
I pulled the FET and cleaned it, I had one keratherm left from a previous project, I swapped out the other brand and the kero and tightened everything back up. It's now the coolest FET in the box. Me thinks, I will go back to keratherm on all my projects from here out. 🙂
I pulled the FET and cleaned it, I had one keratherm left from a previous project, I swapped out the other brand and the kero and tightened everything back up. It's now the coolest FET in the box. Me thinks, I will go back to keratherm on all my projects from here out. 🙂
So, having got the heat under control on the one FET, I did the tranny swap. I now have the 300VA 20V from the 300VA 18V and where I was getting just under 22v rails, I now have 24.5V rails. I got my volt Jeff. 🙂 It was like those magic $1000 cables it sounds sooooo much better now. lmao. Can I hear any difference? It's to late here to push it up, but this was never meant to be a window shaker anyway, I just wanted the damned volt.
Last edited: