Splitting SRPP over 2 tubes and raising heater

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Curlewcat, piezo are not very flat across the 20 - 20khz range.
20hz is low but you will need 100hz to properly tone the low notes.
electret mics inside the body may be possible. or electric guitar pickup with small magnetized sleeves on the strings at the body before the rest.

otherwise, install a good quality pickup and adjust your preamp / headamp until the response is perfect
My MM Phono cart is now MC quality by adjusting the RIAA eq of the preamp to match it.
 
Last edited:
just do it...if your build has buzzing, then that is the time to consider dc biasing the filaments

humming and buzzing are two different things but both can be helped by dc lifting filaments..

say your SRPP has a B+ of 300 volts so that there is 150 volts across each plates. so the upper cathode sits at 150 volts too...

you can use a common filament supply for both lower or upper, i have done this many times... whether single bottle or separate it does not matter...

now for the heater to cathode voltage ratings....
12at7 spec says +- 90 volts, so since one cathode sits at 150 volts,
you can bias the filaments with a dc lifter of say 75 volts,
this way the upper half and the lower half cathodes are both biased correctly...and within specs...

An externally hosted image should be here but it was not working when we last tested it.


with a B+ of 330 volts, R1 can be 560k 2 watt and R2 180k1 watt C can be 33ufd 100 volts dc..
 
In regards to using a cascode stage or an SRPP stage as a Driver:
+150V cathode - (+75V) filament = 75V. 90V - 75V = 15V.
You are allowed to swing the cathode up to +/- 15V (150 + 15V = 165V), 165V - 75V = 90V.
If the Next stage has 20V grid bias, then it will not ever be able to approach 0V grid to cathode.
 
Am I wrong? I thought grid to cathode voltage was independent of Vhk? It's simply a breakdown voltage guideline, no? The next stage will work as intended but exceed Vhk Max? Maybe I'm a bad tube builder, but I've never elevated a heater, and never had a failure... Fingers crossed, of course...
 
Last edited:
In regards to using a cascode stage or an SRPP stage as a Driver:
+150V cathode - (+75V) filament = 75V. 90V - 75V = 15V.
You are allowed to swing the cathode up to +/- 15V (150 + 15V = 165V), 165V - 75V = 90V.
If the Next stage has 20V grid bias, then it will not ever be able to approach 0V grid to cathode.


Ah, by the way, see here

http://www.diyaudio.com/forums/tubes-valves/49567-unison-research-smart-845-schematics-3.html#post687152

This renowned commercial product used a +B of about 660VDC for the ECC82, and its heaters are elevated at about 160VDC, top triode cathode is at about 350VDC, clearly Vkf~190V is out of specs by about 10V, not to speak that an 845 needs a lot drive voltage, anyway I did not see complains against squealing...
 
The grid to cathode voltage I was talking about is the Signal that drives a next stage, like a KT88, 300B, or pick your favorite tube.
Take an amp with a driver stage that is RC coupled to the output stage. If the output tube grid bias is -30V, you want the stage that drives that grid to swing positive 30V and -30V, or 60Vpeak to peak.

The maximum cathode to filament voltage is rated for each tube type.
Some ratings are for steady state.
Some ratings are for peak.
Some tubes have ratings for both steady state and for peak.
Some ratings are for positive differential values, some ratings are for negative differential values (peak, steady state, or both).
Check all these ratings for the tubes and the circuit voltages in your amp design, including maximum signal swing.

An amplifier has the following:
The filament is set to a fixed positive voltage.
The cathode is set to a positive voltage (no signal).
The cathode moves up and down when signal is applied.
You either are within the maximum cathode to filament voltage ratings during signal conditions or you are not.

Some amplifier designers choose to exceed the ratings (either intentionally, or because they do not know any better).
Run the tubes anywhere you want. Just don't tell me about the failures (performance wise because of leakage resistance in parallel with the circuit; or early tube failures; or longer term tube failures).

By the way, squealing/squeaking is not the only performance or failure mode when cathode filament leaks. There is loss of gain, distortion, etc.
 
Last edited:
Am I wrong? I thought grid to cathode voltage was independent of Vhk? It's simply a breakdown voltage guideline, no? The next stage will work as intended but exceed Vhk Max? Maybe I'm a bad tube builder, but I've never elevated a heater, and never had a failure... Fingers crossed, of course...

you are fine, the vhk is a spec so that in cases where this spec can be exceeded, then the filaments are biased with dc to keep to within the actual vhk specs...

filaments can also be floated, not referenced to ground...
so those specs assumed that filaments were referenced to ground, am i correct PRR?

the dc lifter makes the filament at a certain Vhk referenced to ground...
 
You can check the results of floating the filaments. The Dyna Stereo 70 did that. There were two 6.3V filament windings, one for each channel. There was a center tap on each winding. There were no resistors from the center taps to ground. There were only ceramic capacitors from the center taps to ground. A pair of EL34 for one channel was connected to one of the filament windings, and the other channel EL34s likewise to the other filament winding. The cathodes of each pair was connected to a 15.6 Ohm resistor to ground, the cathodes were at 1.56V. Perhaps the EL34 tubes have quite a bit of cathode to filament leakage, that may have contributed to the following problem:

There was a 7199 pentode / triode in each channel. The pentode plate was direct coupled to the grid of the triode, which was operated as a concertina phase splitter. The concertina cathode was elevated to about 120 or 130VDC. After many hours of operation, the 7199 triode cathode to filament became leaky (some as low as 100k Ohms cold). That unbalanced the concertina signals, with more signal on the plate than the cathode. The result was increased 2nd harmonic distortion from the amp (negative feedback did not eliminate all of it). Some will argue it sounded better that way, and some will argue it sounded worse that way.

See if you can find a copy of issue # 10 of "Sound Practices" magazine, look for the article on pages 35 and 36, "Save those 7199s!" by Matt Kamna. It will tell you all about the problem, and how to fix it.

And that brings another reason why I do not like SRPP and Cascode circuits in Hi Fi amps. How often do you need to check your amplifier performance to see if it has been affected by leaky cathode to filament interfaces.

At least where it could become a problem in long tailed pair cathode coupled phase splitters, the parallel cathode impedance is much lower than the cathode to filament leakage resistance.
 
Last edited:
Some amplifier designers choose to exceed the ratings (either intentionally, or because they do not know any better).

I think that Italian designers have done a magnificent work on the Smart 845.

By the way, squealing/squeaking is not the only performance or failure mode when cathode filament leaks. There is loss of gain, distortion, etc.

Squealing is the most usual complaint against SRPP, at least here on the forum.
 
Squealing is the most usual complaint against SRPP, at least here on the forum.

i never encountered this, that is because i never used this SRPP, instead i used a broskie type circuit, the top triode used a CCS....i got very good results doing this on my 5894 SET amp....
 

Attachments

  • IMG_5817 (2).jpg
    IMG_5817 (2).jpg
    455.3 KB · Views: 124
You can check the results of floating the filaments. <snip>

in many of the amps i have seen, the filaments of the output tubes are not lifted with dc, a center tapped 6.3 volts supply is often enough connected to ground, and the reason for this is that hum and noise are not generated much at the output stage more than in the input and intermediate stages, wherein hum and noise are amplified by the succeeding stage(s), in the output tubes, the succeeding stage is the output transformer...

btw, in case of the output tubes being cathode resistor biased, the voltage developed there are also used to lift the filaments of early stages, in some case, even used to power the filaments, in this case the early tube filaments doubled up as cathode resistor biasing, i saw this on a Fisher receiver with 6bq5 output tubes...powering 12ax7 filaments with dc voltage..
 
You should never do this, regardless of the topology used, a floating filament in the presence of an electric field can reach a voltage which violates the manufacturer specs by far.

I think that this kind of errors is responsible of the bad reputation of SRPP.


how many tube amps have you built?

i have done this on many occasions, no detrimental results, but you are not me, so do not do this yourself....

you only need to know when to do this or not....

for the record i have never done an SRPP and no plans of doing it ever,
Broskie is adequate for me thank you.....

so i am not responsible for the bad reputation of SRPP that you claim....
i had nothing to do with it...my tube amp builds do not hum, do not hiss, do not buzz and is enjoyable to listen to...
 
I've only built one SRPP and I think it was in spec. (300V B+, 6SN7) but I didn't like it compared to grounded cathode. I'd rather use the whole 300V on each stage with 6SN7 :) I almost always use DC for heaters (SMPS) and never an issue so far (Soviet tubes). I routinely exceed Vhk by 50V without issue. Not that there won't be one in the future, but so far, so good. I'm also used to sudden catastrophic failures (like arc over in cheap tubes), not gradual failures creeping in over time.
 
Last edited:
I've only built one SRPP and I think it was in spec. (300V B+, 6SN7) but I didn't like it compared to grounded cathode. I'd rather use the whole 300V on each stage with 6SN7 :) I almost always use DC for heaters (SMPS) and never an issue so far (Soviet tubes). I routinely exceed Vhk by 50V without issue. Not that there won't be one in the future, but so far, so good. I'm also used to sudden catastrophic failures (like arc over in cheap tubes), not gradual failures creeping in over time.

good learning....nothing beats actual experience.....if i am right, my understanding is that the srpp by its nature removed much of the 2nd harmonics...unlike the common cathode amp...

i have done broskie ccda and the aikido....and guess what?
folks who listened to both swears that the ccda sounded better....
they say the aikido was "too honest" for their taste...
very much like a very good solid state amp....
so now i only do ccda's...
my latest, the 6BC4 common cathode line amp is yet another thing....
very enjoyable and very tube like sounding....
and very easy to build and low cost, what is not to like?
 
good learning....nothing beats actual experience.....if i am right, my understanding is that the srpp by its nature removed much of the 2nd harmonics...unlike the common cathode amp...

That's right for a theoretical, optimized SRPP, for the real world optimized SRPP things are a bit different, see measurements here

http://www.diyaudio.com/forums/tubes-valves/74903-survey-aikido-distortion-17.html#post4719609

http://www.diyaudio.com/forums/tubes-valves/74903-survey-aikido-distortion-18.html#post4721214
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.