-Some crazy fellow drew this for me a couple of years ago:
http://www.diyaudio.com/forums/atta...d1279028830-jfet-input-aleph-60-aleph60-j.jpg
http://www.diyaudio.com/forums/atta...d1279028830-jfet-input-aleph-60-aleph60-j.jpg
Since then, playing with this sort of circuit and with different devices and such,
I decided that there is a stability advantage to having some resistance off the
Drains of the input devices when cascoding. This means 100 to 200 ohms
or so between the Drain and the Emitter of the cascode transistor. I have
seen several examples tend toward oscillation, often fixed by a lag resistor
(Drain to Gate) on the outputs, but traced to the low impedance seen by
the input device Drains (the Cascode).
So if you run into this issue, you can try this. Or try it anyway...
Please tell me what I'm measuring here...
Currently I'm building an Aleph 30 with BrianGT boards. This Thread came just in time before soldering the mosfets. Now I am sourcing 2sj313.
But here is another 'problem':
I was wondering about the modification of the resistor in series with the zener diode.
On some other sites I read that zeners below 5v6 (real zener effect) have lower noise than higher 'zeners' (avalanche effect).
I took a 4v3 and a 5v1 zener in series and here are some 'screenshots' of my old hameg scope.
First picture: 9v1 zener. I see some noise.
Second picture: 4v3 and 5v1 in series. At least on my sccpe very low noise.
Could someone please tell me why?
Ulf
Currently I'm building an Aleph 30 with BrianGT boards. This Thread came just in time before soldering the mosfets. Now I am sourcing 2sj313.
But here is another 'problem':
I was wondering about the modification of the resistor in series with the zener diode.
On some other sites I read that zeners below 5v6 (real zener effect) have lower noise than higher 'zeners' (avalanche effect).
I took a 4v3 and a 5v1 zener in series and here are some 'screenshots' of my old hameg scope.
First picture: 9v1 zener. I see some noise.
Second picture: 4v3 and 5v1 in series. At least on my sccpe very low noise.
Could someone please tell me why?
Ulf
Attachments
Pass DIY Addict
Joined 2000
Paid Member
The 9.1V are somewhat noiser. They run quieter with more current, which
is why I lowered the value of the bias resistor - it was close enough to the
edge that occasionally there was excess noise.
That's what I figured with that change...
What is your experience, or any fearless diyer out there, on running 2SJ74s at 170mW (20V, 8.5mA)? Will they last inside a hot Aleph chassis?
... This means 100 to 200 ohms
or so between the Drain and the Emitter of the cascode transistor. I have
seen several examples tend toward oscillation...
Is this what they are trying to describe and figure out as Cascode Local Feedback at the other thread?
The subject of caps across the 9.1 reference zener was discussed many years ago in this thread, in posts #18 & 38 Mr. Pass gives you some details on this cap. The main concern is a thump when you power the amplifier on with larger values.
For my Alephs I tried a 4.7uF electrolytic and had no thump. IIRC I ended up putting a 10uF/50V low esr Panasonic FC across this zener. I measured the momentary DC at the speaker terminals at power on and I do not remember exactly what it was, but it was extremely low, like < 200mv. Upon powering up one of my Aleph mono's has the faintest thump and the other is completely silent. This is using my Vandesteen 2ci speakers with reasonably large woofers, if I use my little bookshelf speakers with a 6.5" woofer there is no audible thump.
you can do various things with these resistors , but you can hardly call that a feedback
however , I have a feeling regarding few things :
- Papa certainly tried these resistors , same as various sorts of real feedback brought to cascode potential node , but
- I'm sure that he could call them (various iterations of said modulation) whatever he want , just for humor sake
the great 2015 front end fet shoot-out
i made up a small front end test bed to compare a few of the candidates for the Aleph. The lineup includes the 2SJ74 JFET ($4), SJ313($2.50), Euvl's suggestion of the SJ76 ($10) and a fet from the Infineon BSP range (0.50)
For all the devices except the 2J74, the load was 390R in the drains of both fets. The devices were run with 10mA bias each. I wanted to test intrinsic linearity so I measured just the input stage (naked, without any feedback) across the load resistors with H2 nulled out.
For the 2SJ74, the load resistors were 1K and bias was 5mA each. I used 10mA Idss fets.
All measurements were made at 0.5V RMS and the fets were driven at high impedance (10K) to get realistic results.
results below ....
i made up a small front end test bed to compare a few of the candidates for the Aleph. The lineup includes the 2SJ74 JFET ($4), SJ313($2.50), Euvl's suggestion of the SJ76 ($10) and a fet from the Infineon BSP range (0.50)
For all the devices except the 2J74, the load was 390R in the drains of both fets. The devices were run with 10mA bias each. I wanted to test intrinsic linearity so I measured just the input stage (naked, without any feedback) across the load resistors with H2 nulled out.
For the 2SJ74, the load resistors were 1K and bias was 5mA each. I used 10mA Idss fets.
All measurements were made at 0.5V RMS and the fets were driven at high impedance (10K) to get realistic results.
results below ....
Attachments
