BH with EL84 is 10%?
PI is referred to -150V, but what worries me is the low-to-no current part of the cycle.
Has anyone applied current feedback from the secondary of the OT to the cathodes of the power amplifier tubes, to shorten the path?
It's something I?ve never seen in amps, but would shorten the CFB path:
one of the sides of the OT secondary is con nected to the speaker, while the other side (usually grounded) is connected to a 0,1 Ohm resistor, then grounded, then to another 0,1 Ohm resistor. then to the other side of the speaker. This will give two current feedback signals that can be connected to the power amp tubes' cathode bias sensing resistors, giving current feedback on both sides of the PP.
If not clear, I'll post a sketch of it.
Has enyone experienced something like this, together with the BH shunt feedback?
Thanks
It's something I?ve never seen in amps, but would shorten the CFB path:
one of the sides of the OT secondary is con nected to the speaker, while the other side (usually grounded) is connected to a 0,1 Ohm resistor, then grounded, then to another 0,1 Ohm resistor. then to the other side of the speaker. This will give two current feedback signals that can be connected to the power amp tubes' cathode bias sensing resistors, giving current feedback on both sides of the PP.
If not clear, I'll post a sketch of it.
Has enyone experienced something like this, together with the BH shunt feedback?
Thanks
Or maybe something like this after the PI, scaled to the scope, to accomplish gain and swing?
Positive Current Feedback simple Zen amp
I draw this loadline for the EF86, 430 V B+, 150 V g2, 47 kOhm plate resistor, 4,9 mA. So... I should design the working conditions and the DC load line to have that AC load line.
With an input swinging between -0,5 and -3 V, the output swings between 75 V at 7,5 mA and 327 V at 2,2 mA. It's a gain of 101.
PS
I've found this amp has a EF86 as phase inverter.
https://www.keith-snook.info/amplifier-hifi-schematics/QUAD-II-Schematic.pdf
~ QUAD II Valve Power Amplifier triode cascode phase splitter driver ~
With an input swinging between -0,5 and -3 V, the output swings between 75 V at 7,5 mA and 327 V at 2,2 mA. It's a gain of 101.
PS
I've found this amp has a EF86 as phase inverter.
https://www.keith-snook.info/amplifier-hifi-schematics/QUAD-II-Schematic.pdf
~ QUAD II Valve Power Amplifier triode cascode phase splitter driver ~
Attachments
Last edited:
Hi Esa, the loadline is for one pair only, so the load for the hextet is 2 kOhm Raa.
Hi Kay,
thanks for the head-up. I used: 280 V on g2, 15V on g3, 870 V on plates, 28 mA at idle (60% of the dissipation) and a load Raa of 6 kOhm. I'm aware I'm over the 40 W dissipation curve, but being off almost 50% of the time, using force ventilation and being bass a non constant source of sound, shouldn't it be acceptable?
As for the needed swing, I've considered to go AB2. I'm searching infos on how positive I can go into AB2 before causing troubles to the tubes. I've seen some amps going around +20 V on g1, and in that scenario if I'm right I would need around 250 Vpp (considering feedback).
thanks for the head-up. I used: 280 V on g2, 15V on g3, 870 V on plates, 28 mA at idle (60% of the dissipation) and a load Raa of 6 kOhm. I'm aware I'm over the 40 W dissipation curve, but being off almost 50% of the time, using force ventilation and being bass a non constant source of sound, shouldn't it be acceptable?
As for the needed swing, I've considered to go AB2. I'm searching infos on how positive I can go into AB2 before causing troubles to the tubes. I've seen some amps going around +20 V on g1, and in that scenario if I'm right I would need around 250 Vpp (considering feedback).
Thanks Kay,
I've downloaded the following datasheets of the different names of the same tube:
http://wavebourn.com/pyramid-8/GU-50.jpg
http://www.frank.mif.pg.gda.pl/sheets/018/g/GU50.pdf
?????? ??-50. ??????????? ??????????????
https://frank.pocnet.net/sheets/128/e/EL152.pdf
to check (when possible) what shown here:
Loadline calculator for power stages with reactive load - Vacuum Tube Amplifiers - DIY
based on Normal Karen curves and equations:
http://www.normankoren.com/Audio/Tubemodspice_article.html
From there the g1 voltage at idle is -45.69 V, so without any feedback it would be enough to have 65,7 Vp or 131,4 Vpp.
Considering the feedback (as before), this swing has to be multiplied 1,9 times (if my calculations are correct) and so 248,9 Vpp.
PS
I'm attaching this page ( ?????? ??-50. ??????????? ?????????????? ) in pdf format and translated from Russian to English).
I've downloaded the following datasheets of the different names of the same tube:
http://wavebourn.com/pyramid-8/GU-50.jpg
http://www.frank.mif.pg.gda.pl/sheets/018/g/GU50.pdf
?????? ??-50. ??????????? ??????????????
https://frank.pocnet.net/sheets/128/e/EL152.pdf
to check (when possible) what shown here:
Loadline calculator for power stages with reactive load - Vacuum Tube Amplifiers - DIY
based on Normal Karen curves and equations:
http://www.normankoren.com/Audio/Tubemodspice_article.html
From there the g1 voltage at idle is -45.69 V, so without any feedback it would be enough to have 65,7 Vp or 131,4 Vpp.
Considering the feedback (as before), this swing has to be multiplied 1,9 times (if my calculations are correct) and so 248,9 Vpp.
PS
I'm attaching this page ( ?????? ??-50. ??????????? ?????????????? ) in pdf format and translated from Russian to English).
Hi kokoriantz,
i guess you are calculating the power for a SE.
This is a PP, so the formula it's not the swing divided by sqrt(2) then squared and then divided by the load.
So the rms power of a push pull is:
{[(2 x Voltage swing) / sqrt(2)]^2} / Raa =
= {[sqrt(2) x voltage swing]^2} / Raa =
= (2 x voltage swing^2) / Raa = 2 x 640000 / 2000 = 640 Wrms
But it will sag for sure with 1,5 A of current (that's 3 x 500 mA circa), so that power will never be reached.
This datasheet https://tubedata.tubes.se/sheets/118/e/EL152.pdf contains even more info, al least for those who read German 
.
Years ago I've seen LS 50 plate curves somewhere in the WWW with the screen @ 300 Vdc which might come closer to what you're planning. But no curves with a positive suppressor at all.
Best regards!
.Years ago I've seen LS 50 plate curves somewhere in the WWW with the screen @ 300 Vdc which might come closer to what you're planning. But no curves with a positive suppressor at all.
Best regards!
The kokoriantz math also can be applied, but has to be modified as the swing applies to only one primary half. So the load is 500 ohms instead of 2 kOhms. The result will be the same, though
.With my ruler in the plate curve set I've calculated even more.
Anyway, for sure you won't have a supply voltage of 870 Vdc @ full drive, unless you're planning a (regulated) SMPS for that job.
Best regards!
Edit: Btw, what a pity that Hans-Georg (es345) appears to have disappeared. He as an EE has some considerable bunch of experience with this tube and we've had extensive e-mail and phone conversation on his 6 x GU50 bass amplifier in 2012. He also gave valuable hints to another toroid winding shop in Poland, this one even with a German Department located at Munich (Badel).
Last edited:
Thanks again Kay,
do you think it's a reasonable choice to go with a custom SMPS for this amp?
For sure it will, weight-wise
but are there custom designers with reasonable prices for all those voltages? Or maybe just B+ and screens, then go with a thoroid for other small power windings.
It is indeed. I've written him on various forums, but I've got no replies.
Yes, design in Germany and production in Poland: Ringkerntransformator, Transformatoren, Ringkerntransformatoren | Ringkern, Ringkern-Transformator, 100V-Übertrager - BADEL Elektronik GmbH
PS
in attachment GU50 curves with g2 at 250 and 300 V.
