Originally posted by macka
Series 7
X Aleph setup
Input devices * 2SJ74 2x diff pair (cascoded)
R19/29 22K
R16/30 100K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 33mv =6.19 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 449mv =19.23 db from input voltage
From node of R19/29 and R16/29 = 31.45 db
2SJ74 2x diff pair (cascoded)
Open loop = 18.56 volts =51.54 db from input voltage
From node of R19/29 and R16/29 = 54.99 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 32.31 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 23.54db
Series 8
R19/29 44K
R16/30 100K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 37mv =2.43 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 451mv =19.27 db from input voltage
From node of R19/29 and R16/29 = 31.49 db
2SJ74 2x diff pair (cascoded)
Open loop = 22.4 volts =53.19 db from input voltage
From node of R19/29 and R16/29 = 55.63db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 33.92 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 24.14db
Series 9
R19/29 22K
R16/30 442K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 33mv =3.40 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 1.842 volts =31.50 db from input voltage
From node of R19/29 and R16/29 = 43.69 db
2SJ74 2x diff pair (cascoded)
Open loop = 18.56 volts =51.54 db from input voltage
From node of R19/29 and R16/29 = 54.99 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 20.04 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 11.30 db
Series 10
R19/29 44K
R16/30 442K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 37mv =2.43 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 1.875 volts =31.65 db from input voltage
From node of R19/29 and R16/29 = 43.86 db
2SJ74 2x diff pair (cascoded)
Open loop = 22.4 volts =53.19 db from input voltage
From node of R19/29 and R16/29 = 55.63 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 21.54 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 11.77 db
Series 11
R19/29 10K
R16/30 442K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 23mv =6.56 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 1.762 volts =31.11 db from input voltage
From node of R19/29 and R16/29 = 43.28 db
2SJ74 2x diff pair (cascoded)
Open loop = 12.72 volts =48.26 db from input voltage
From node of R19/29 and R16/29 = 54.85 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 17.15 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 11.57 db
Series 7
X Aleph setup
Input devices * 2SJ74 2x diff pair (cascoded)
R19/29 22K
R16/30 100K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 33mv =6.19 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 449mv =19.23 db from input voltage
From node of R19/29 and R16/29 = 31.45 db
2SJ74 2x diff pair (cascoded)
Open loop = 18.56 volts =51.54 db from input voltage
From node of R19/29 and R16/29 = 54.99 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 32.31 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 23.54db
Series 8
R19/29 44K
R16/30 100K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 37mv =2.43 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 451mv =19.27 db from input voltage
From node of R19/29 and R16/29 = 31.49 db
2SJ74 2x diff pair (cascoded)
Open loop = 22.4 volts =53.19 db from input voltage
From node of R19/29 and R16/29 = 55.63db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 33.92 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 24.14db
Series 9
R19/29 22K
R16/30 442K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 33mv =3.40 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 1.842 volts =31.50 db from input voltage
From node of R19/29 and R16/29 = 43.69 db
2SJ74 2x diff pair (cascoded)
Open loop = 18.56 volts =51.54 db from input voltage
From node of R19/29 and R16/29 = 54.99 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 20.04 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 11.30 db
Series 10
R19/29 44K
R16/30 442K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 37mv =2.43 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 1.875 volts =31.65 db from input voltage
From node of R19/29 and R16/29 = 43.86 db
2SJ74 2x diff pair (cascoded)
Open loop = 22.4 volts =53.19 db from input voltage
From node of R19/29 and R16/29 = 55.63 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 21.54 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 11.77 db
Series 11
R19/29 10K
R16/30 442K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 23mv =6.56 db loss
2SJ74 2x diff pair (cascoded)
Closed loop = 1.762 volts =31.11 db from input voltage
From node of R19/29 and R16/29 = 43.28 db
2SJ74 2x diff pair (cascoded)
Open loop = 12.72 volts =48.26 db from input voltage
From node of R19/29 and R16/29 = 54.85 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 17.15 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 11.57 db
CheffDeGaar said:Hi Ian,
Great work !
If it can help you, I once did the maths for the gain of an ugs stage, and the same formula should apply here. It includes the effets of the input to ground resistor. and the feedback network as well.
It's here :
http://psykok.homelinux.org/diy/preamp_ugs/pdf/UGS pour les nuls.pdf
Sorry, it's in french, but the maths should be sufficient per se.
Look at page 13 & 14, and eq 32, where Rfb is the feedback resistor, Rin the input resistor, Rg the ground resistor and Ad the differential open loop gain.
On the last pages, you can see how the variarion of a resistor value impacts the gain ("entree" means input)
Hope this helps, and keep up this good work
Cheers
Thanks Cheff,
I need to take a break...this is painful...no Red Wine during the write up....
I will look at some way of tabulating the results so it is easy to read when time permits
iMac
X Aleph setup
Input devices IRF9610 diff pair
* 2SJ74 2x diff pair (cascoded)
* ZVP2110A diff pair
Series 12
R19/29 22K
R16/30 221K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 33mv =3.4 db loss
Differential Voltage at amplifier output
IRF9610 diff pair
Closed loop = 947mv =25.72 db from input voltage
From node of R19/29 and R16/29 = 38db
2SJ74 2x diff pair (cascoded)
Closed loop = 964mv =25.88 db from input voltage
From node of R19/29 and R16/29 = 38.09db
ZVP2110A diff pair
Closed loop = 937mv =25.62 db from input voltage
From node of R19/29 and R16/29 = 37.84db
IRF9610 diff pair
Open loop = 13.34 volts =48.69 db from input voltage
From node of R19/29 and R16/29 = 52db
2SJ74 2x diff pair (cascoded)
Open loop = 18.25 volts =51.42 db from input voltage
From node of R19/29 and R16/29 = 54.85db
ZVP2110A diff pair
Open loop = 11.24volts =47.00 db from input voltage
From node of R19/29 and R16/29 = 50.62db
IRF9610 diff pair
Feedback taken from input less closed loop gain = 22.97 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 25.54 db
ZVP2110A diff pair
Feedback taken from input less closed loop gain = 21.38 db
IRF9610 diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 14.2 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 16.76 db
ZVP2110A diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 12.78 db
Series 13
R19/29 44K
R16/30 221K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 39mv =1.979 db loss
IRF9610 diff pair
Closed loop = 956mv =25.80 db from input voltage
From node of R19/29 and R16/29 = 38.02 db
2SJ74 2x diff pair (cascoded)
Closed loop = 974mv =25.96 db from input voltage
From node of R19/29 and R16/29 = 38.18 db
ZVP2110A diff pair
Closed loop = 950mv =25.74 db from input voltage
From node of R19/29 and R16/29 = 37.97 db
IRF9610 diff pair
Open loop = 16.28 volts =50.42 db from input voltage
From node of R19/29 and R16/29 = 52.4db
2SJ74 2x diff pair (cascoded)
Open loop = 22.3 volts =53.16 db from input voltage
From node of R19/29 and R16/29 = 55.13db
ZVP2110A diff pair
Open loop = 13.50 volts =48.78 db from input voltage
From node of R19/29 and R16/29 = 50.78db
IRF9610 diff pair
Feedback taken from input less closed loop gain = 24.62 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 27.22 db
ZVP2110A diff pair
Feedback taken from input less closed loop gain = 23.04 db
IRF9610 diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 14.38db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 16.95db
ZVP2110A diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 12.81db
Series 14
R19/29 100K
R16/30 221K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 43mv =1.130 db loss
IRF9610 diff pair
Closed loop = 965mv =25.888 db from input voltage
From node of R19/29 and R16/29 = 38.10 db
2SJ74 2x diff pair (cascoded)
Closed loop = 978mv =25.99 db from input voltage
From node of R19/29 and R16/29 = 38.22 db
ZVP2110A diff pair
Closed loop = 959mv =25.83 db from input voltage
From node of R19/29 and R16/29 = 38.00 db
IRF9610 diff pair
Open loop = 18.14 volts =51.36 db from input voltage
From node of R19/29 and R16/29 = 52.48db
2SJ74 2x diff pair (cascoded)
Open loop = 25.2 volts =54.21 db from input voltage
From node of R19/29 and R16/29 = 55.35 db
ZVP2110A diff pair
Open loop = 15.3 volts =49.88 db from input voltage
From node of R19/29 and R16/29 = 51.00 db
IRF9610 diff pair
Feedback taken from input less closed loop gain = 25.48 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 28.22 db
ZVP2110A diff pair
Feedback taken from input less closed loop gain = 24.05 db
IRF9610 diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 14.38db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 17.13db
ZVP2110A diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 13.00 db
It would seem that the ZVP2110A has about 4-5 db less open loop gain than the 2SJ74 configuration and a bit less open loop gain than the IRF9610.
I have no conclusions at this stage. There are numerous possibilities here and the results on paper are quite interesting.
iMac
Input devices IRF9610 diff pair
* 2SJ74 2x diff pair (cascoded)
* ZVP2110A diff pair
Series 12
R19/29 22K
R16/30 221K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 33mv =3.4 db loss
Differential Voltage at amplifier output
IRF9610 diff pair
Closed loop = 947mv =25.72 db from input voltage
From node of R19/29 and R16/29 = 38db
2SJ74 2x diff pair (cascoded)
Closed loop = 964mv =25.88 db from input voltage
From node of R19/29 and R16/29 = 38.09db
ZVP2110A diff pair
Closed loop = 937mv =25.62 db from input voltage
From node of R19/29 and R16/29 = 37.84db
IRF9610 diff pair
Open loop = 13.34 volts =48.69 db from input voltage
From node of R19/29 and R16/29 = 52db
2SJ74 2x diff pair (cascoded)
Open loop = 18.25 volts =51.42 db from input voltage
From node of R19/29 and R16/29 = 54.85db
ZVP2110A diff pair
Open loop = 11.24volts =47.00 db from input voltage
From node of R19/29 and R16/29 = 50.62db
IRF9610 diff pair
Feedback taken from input less closed loop gain = 22.97 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 25.54 db
ZVP2110A diff pair
Feedback taken from input less closed loop gain = 21.38 db
IRF9610 diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 14.2 db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 16.76 db
ZVP2110A diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 12.78 db
Series 13
R19/29 44K
R16/30 221K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 39mv =1.979 db loss
IRF9610 diff pair
Closed loop = 956mv =25.80 db from input voltage
From node of R19/29 and R16/29 = 38.02 db
2SJ74 2x diff pair (cascoded)
Closed loop = 974mv =25.96 db from input voltage
From node of R19/29 and R16/29 = 38.18 db
ZVP2110A diff pair
Closed loop = 950mv =25.74 db from input voltage
From node of R19/29 and R16/29 = 37.97 db
IRF9610 diff pair
Open loop = 16.28 volts =50.42 db from input voltage
From node of R19/29 and R16/29 = 52.4db
2SJ74 2x diff pair (cascoded)
Open loop = 22.3 volts =53.16 db from input voltage
From node of R19/29 and R16/29 = 55.13db
ZVP2110A diff pair
Open loop = 13.50 volts =48.78 db from input voltage
From node of R19/29 and R16/29 = 50.78db
IRF9610 diff pair
Feedback taken from input less closed loop gain = 24.62 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 27.22 db
ZVP2110A diff pair
Feedback taken from input less closed loop gain = 23.04 db
IRF9610 diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 14.38db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 16.95db
ZVP2110A diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 12.81db
Series 14
R19/29 100K
R16/30 221K
Differential Voltage input voltage at input of R18/28 49mv
Differential Voltage closed loop gain at node of 19/29 and R16/29 = 12mv =12.2 db loss
Differential Voltage open loop gain at node of 19/29 and R16/29 = 43mv =1.130 db loss
IRF9610 diff pair
Closed loop = 965mv =25.888 db from input voltage
From node of R19/29 and R16/29 = 38.10 db
2SJ74 2x diff pair (cascoded)
Closed loop = 978mv =25.99 db from input voltage
From node of R19/29 and R16/29 = 38.22 db
ZVP2110A diff pair
Closed loop = 959mv =25.83 db from input voltage
From node of R19/29 and R16/29 = 38.00 db
IRF9610 diff pair
Open loop = 18.14 volts =51.36 db from input voltage
From node of R19/29 and R16/29 = 52.48db
2SJ74 2x diff pair (cascoded)
Open loop = 25.2 volts =54.21 db from input voltage
From node of R19/29 and R16/29 = 55.35 db
ZVP2110A diff pair
Open loop = 15.3 volts =49.88 db from input voltage
From node of R19/29 and R16/29 = 51.00 db
IRF9610 diff pair
Feedback taken from input less closed loop gain = 25.48 db
2SJ74 2x diff pair (cascoded)
Feedback taken from input less closed loop gain = 28.22 db
ZVP2110A diff pair
Feedback taken from input less closed loop gain = 24.05 db
IRF9610 diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 14.38db
2SJ74 2x diff pair (cascoded)
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 17.13db
ZVP2110A diff pair
Feedback taken from node of 19/29 and R16/29 loop less closed loop gain = 13.00 db
It would seem that the ZVP2110A has about 4-5 db less open loop gain than the 2SJ74 configuration and a bit less open loop gain than the IRF9610.
I have no conclusions at this stage. There are numerous possibilities here and the results on paper are quite interesting.
iMac
I have reviewed the results and with a degree of investigation on the basis of known information and the previous analysis I have narrowed down the options.
In the case of the ZVP2110 R1 is 11K and R2 220K. This gives a differential input impediance of 22K
In theory this should give a closed loop gain of 26.44 db.
In practice here I get 25.38 db.
I can trim this out with adjustment of the feeback resister but I suspect I am not getting enough open loop gain from the ZVP2110 and this is resulting in a gain error as discussed earlier.
I measured the open loop gain at 44.79 db referenced to the differential input voltage. That is 19.41 db feeback.
My program this week will be to install from Silmic capacitors in the current share loop and adjust the feeback and perhaps increase the bias of the ZVP2110 to see what effect his has subjectively and on the open loop gain
I will report in the subjective aspects.
In respect to the 2SJ74 I do not think it requires any further adjustment. I may try the 3rd parrellel set of Jfets .
iMac
In the case of the ZVP2110 R1 is 11K and R2 220K. This gives a differential input impediance of 22K
In theory this should give a closed loop gain of 26.44 db.
In practice here I get 25.38 db.
I can trim this out with adjustment of the feeback resister but I suspect I am not getting enough open loop gain from the ZVP2110 and this is resulting in a gain error as discussed earlier.
I measured the open loop gain at 44.79 db referenced to the differential input voltage. That is 19.41 db feeback.
My program this week will be to install from Silmic capacitors in the current share loop and adjust the feeback and perhaps increase the bias of the ZVP2110 to see what effect his has subjectively and on the open loop gain
I will report in the subjective aspects.
In respect to the 2SJ74 I do not think it requires any further adjustment. I may try the 3rd parrellel set of Jfets .
iMac
I have had some time to listen to the above configuration using the ZVP2110.
It works quite nicely and is certainly an improvement over the IRF9610.
Small adjustments to the feedback resisters R16 and R30 make subtle changes according to taste.
I will try adjusting the bias to 30 ma over the next couple of days.
iMac
It works quite nicely and is certainly an improvement over the IRF9610.
Small adjustments to the feedback resisters R16 and R30 make subtle changes according to taste.
I will try adjusting the bias to 30 ma over the next couple of days.
iMac
An update on the subjective evaluation of the ZVP2110A.
Set up above with 10K to ground and 220K feeback, input 11K.
I initally tried to tim out the gain to 26db with 11K input resister and about 275K feeback resisters.
I sounded good. However on direct comparison with the Pass labs X250.5 ( my unfair advantage) it lacked what I would call a flesh and bone quality.
By process of elimination I determined the issue arose with an additional series 1K (economy) input resister in the test jig, it made the sound brittle. Bypassing it impoved things.
But the balance was slightly titled towards the treble with the bass sounded a tad recessed. So I went back to 220K feeback resister. Its now within a hairs whisker of the reference amp.
All things being equal I really like the ZVP 2110A.
I have not tried increasing the input bias yet.
iMac
Set up above with 10K to ground and 220K feeback, input 11K.
I initally tried to tim out the gain to 26db with 11K input resister and about 275K feeback resisters.
I sounded good. However on direct comparison with the Pass labs X250.5 ( my unfair advantage) it lacked what I would call a flesh and bone quality.
By process of elimination I determined the issue arose with an additional series 1K (economy) input resister in the test jig, it made the sound brittle. Bypassing it impoved things.
But the balance was slightly titled towards the treble with the bass sounded a tad recessed. So I went back to 220K feeback resister. Its now within a hairs whisker of the reference amp.
All things being equal I really like the ZVP 2110A.
I have not tried increasing the input bias yet.
iMac
Re biased the front end to 35ma (total) on the ZVP2110A.
It was not that difficult, just parralleled 392R on the current source 575R resister and 820R in parrallel with the 392R drain resisters.
I calculated the dissipation on each device at about 320 mw. They are hot to touch and I dare not go higher without cascoding.
Subjectively its a hard call as I have not spent a lot of time listening to this set-up but I think it is sweeter and smoother with a bit more resolving power.
I should point out I have not compared the ZVP2110A to the Jfets yet but subjectively it significantly better than the IRF9610.
I will measure the closed and open loop gain tomorrow.
iMac
It was not that difficult, just parralleled 392R on the current source 575R resister and 820R in parrallel with the 392R drain resisters.
I calculated the dissipation on each device at about 320 mw. They are hot to touch and I dare not go higher without cascoding.
Subjectively its a hard call as I have not spent a lot of time listening to this set-up but I think it is sweeter and smoother with a bit more resolving power.
I should point out I have not compared the ZVP2110A to the Jfets yet but subjectively it significantly better than the IRF9610.
I will measure the closed and open loop gain tomorrow.
iMac
Hi Ian,
Thank you again for all of this wonderful data. And thank you too for remaining "dry" for the duration (well at least for the writing portion). A great sacrifice for a good cause.
I was surprised that the open loop gain with the JFET's was higher than for the 9610's. At this point I suspect that 3 pair would only improve the noise figure and even that might be inaudible. But don't let me discourage you if you're keen to do it.
The more data you uncover, the happier I get with what I have. I love this thread!
Cheers,
Graeme
Thank you again for all of this wonderful data. And thank you too for remaining "dry" for the duration (well at least for the writing portion). A great sacrifice for a good cause.
I was surprised that the open loop gain with the JFET's was higher than for the 9610's. At this point I suspect that 3 pair would only improve the noise figure and even that might be inaudible. But don't let me discourage you if you're keen to do it.
The more data you uncover, the happier I get with what I have. I love this thread!
Cheers,
Graeme
Anyway,
I hope this has been useful for those teetering on the edge and ponding if an X Aleph is too difficult to make and perhaps a renaissance for a very fine design.
There is just something about SE amplifiers that sounds right. If you dont need lots of current then balanced SE is possibly all you will ever need.( biamp if you do)
Its a surprisingly flexible and robust design.
While I dont think there are any hard and fast rules about gain or feeback it pays to do some measurements to help guide the way you set it up as what is good for one device may not be good for another dependiing on the open loop gain and other factors like capacitance on the Gate. The ZVP2110A seems to get more bandwidth with 10k to ground in the feedback divider.
I think other factors also come into play like supply rails and the number of parrallel Fet followers and the choice of output device.
These are my observations and are perhaps generalisations but too much feedback and I think its tends to bounce around inside the circuit with the X Aleph and the sound starts to bite and become aggressive. I suspect a lot of early attempts ended up like this.
Modest feedback (20-25db) seems to work well and you get a nice sweet result.
Less feedback, like below 20 db or less and the balance tilts towards the treble. You loose a bit of the warmth and it takes on a live atmosphere.
I think this is real X Series territory and you see large banks parrallel followers to help linearise the output stage. (The SE bias and Jets in the front end of the X.5 series seem to bring them closer to the Aleph.)
Most importantly trust your ears.
iMac
I hope this has been useful for those teetering on the edge and ponding if an X Aleph is too difficult to make and perhaps a renaissance for a very fine design.
There is just something about SE amplifiers that sounds right. If you dont need lots of current then balanced SE is possibly all you will ever need.( biamp if you do)
Its a surprisingly flexible and robust design.
While I dont think there are any hard and fast rules about gain or feeback it pays to do some measurements to help guide the way you set it up as what is good for one device may not be good for another dependiing on the open loop gain and other factors like capacitance on the Gate. The ZVP2110A seems to get more bandwidth with 10k to ground in the feedback divider.
I think other factors also come into play like supply rails and the number of parrallel Fet followers and the choice of output device.
These are my observations and are perhaps generalisations but too much feedback and I think its tends to bounce around inside the circuit with the X Aleph and the sound starts to bite and become aggressive. I suspect a lot of early attempts ended up like this.
Modest feedback (20-25db) seems to work well and you get a nice sweet result.
Less feedback, like below 20 db or less and the balance tilts towards the treble. You loose a bit of the warmth and it takes on a live atmosphere.
I think this is real X Series territory and you see large banks parrallel followers to help linearise the output stage. (The SE bias and Jets in the front end of the X.5 series seem to bring them closer to the Aleph.)
Most importantly trust your ears.
iMac
macka said:
I measured the open loop gain and it actually drop with the above input bias to 42 db.
Thats about 3 db less open loop gain annd my little fets are frying their head's off
.I am not sure why but there you are. Perhaps a slightly lower bias of 9-10 ma each fet would be interesting.
Anyway , I have moved the bias back to the stock setting which is close to 11.5 ma .
iMac
Graeme,
I tried the ZVP2110A with stock bias using 11K to ground and 110K feedback just for grins. The gain is close to 20 db and feeback is around 25 db.
It bought back memories of my first X Aleph.
A bit cold and clinical, lacking soul. Depending on the system this may appeal but not really what its all about.
Aside from better matching (which I will do shortly) for lower offset and that last drop of susy the ZVP works well as a drop in replacement for the IFR 9610. Using the stock bias it should work for all versions up 30 volt rails.
My recommendation is use 10K input resisters 10-11K to ground and 220-240K feedback. The 10-11 K to ground seems to improve bandwidth. Small increase in the feedback resister can be used to fine tune the tonality. I found 220K gave a warm and sweet presentation, 240K has a bit more air and ambience.
Some may wish to attempt cascoding the ZVP2110A to enhance its performance further. I may look at this later on.
Next on the program is to compare the ZVP2110A to the cascoded 2SJ74 2 x dual differential pair.
If anyone is interested I have a stash of ZVP2110A.
iMac
I tried the ZVP2110A with stock bias using 11K to ground and 110K feedback just for grins. The gain is close to 20 db and feeback is around 25 db.
It bought back memories of my first X Aleph.
A bit cold and clinical, lacking soul. Depending on the system this may appeal but not really what its all about.
Aside from better matching (which I will do shortly) for lower offset and that last drop of susy the ZVP works well as a drop in replacement for the IFR 9610. Using the stock bias it should work for all versions up 30 volt rails.
My recommendation is use 10K input resisters 10-11K to ground and 220-240K feedback. The 10-11 K to ground seems to improve bandwidth. Small increase in the feedback resister can be used to fine tune the tonality. I found 220K gave a warm and sweet presentation, 240K has a bit more air and ambience.
Some may wish to attempt cascoding the ZVP2110A to enhance its performance further. I may look at this later on.
Next on the program is to compare the ZVP2110A to the cascoded 2SJ74 2 x dual differential pair.
If anyone is interested I have a stash of ZVP2110A.
iMac
Ian,
The data you are generating is invaluable. At the very least Graeme and I are both paying very close attention. Once I finish building the new deck in my backyard I will be doing some modding to my Aleph X's looking closely at the ZVP2110A's.
Also, since a certain generous benefactor and source of inspiration in this forum bequeathed to me a very large trafo some time ago, and since this trafo has very recently been joined by a set of very large heatsinks, I will soon be building something similar to an Aleph 4 (the trafo has high voltage secondaries and not enough current capacity to support a balanced X design). I think I will change my approach a bit and look at using cascoded ZVP2110A's for the input differential there too.
Your work is greatly appreciated!
Terry
The data you are generating is invaluable. At the very least Graeme and I are both paying very close attention. Once I finish building the new deck in my backyard I will be doing some modding to my Aleph X's looking closely at the ZVP2110A's.
Also, since a certain generous benefactor and source of inspiration in this forum bequeathed to me a very large trafo some time ago, and since this trafo has very recently been joined by a set of very large heatsinks, I will soon be building something similar to an Aleph 4 (the trafo has high voltage secondaries and not enough current capacity to support a balanced X design). I think I will change my approach a bit and look at using cascoded ZVP2110A's for the input differential there too.
Your work is greatly appreciated!
Terry
As final phase in the Jfet / small fet version of the X Aleph I will attempt a cascoded ZVP2110A differential pair and then compare them both to the reference amp.
I also want to tweak the feeback a bit more on the 2SJ74 jfet option because it has higher open loop gain
It will then be a case of deciding which one I will install permanently
iMac
I also want to tweak the feeback a bit more on the 2SJ74 jfet option because it has higher open loop gain
It will then be a case of deciding which one I will install permanently
iMac
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.