Hello guys, my question is pretty simple.
What is the best sounding pentode SE amplifier that you have heard?
Can it be so good when built nice that it can compete with SET amps?
Best regards. Michal
What is the best sounding pentode SE amplifier that you have heard?
Can it be so good when built nice that it can compete with SET amps?
Best regards. Michal
UnSET is a Single Ended Triode design that is UN because it contains no Triodes. It is a two stage amp that runs a pair of pentodes with direct feedback from the plate to the control grid to generate triode curves from a pentode. Tubes that have low screen grid voltage requirements can be used since the screen supply is not required to be at plate potential. I called this topology the Composite Electron Device since it is a tube / mosfet pair with a P channel mosfet follower driving the cathode of the output tube.
Some links about the tech are scattered throughout the Tubelab and Tubes / Valves forums. I'll eventually get all of it into one place.....for now I'm busy building test boards....amps will follow.
UNSET is coming?
https://www.diyaudio.com/forums/tub...sing-mosfet-ultralinear-feed.html#post6267790
Some links about the tech are scattered throughout the Tubelab and Tubes / Valves forums. I'll eventually get all of it into one place.....for now I'm busy building test boards....amps will follow.
UNSET is coming?
https://www.diyaudio.com/forums/tub...sing-mosfet-ultralinear-feed.html#post6267790
Does it have to be a Pentode?
Or, do Beam Power Tubes also qualify?
(Beam Power/Beam Tetrode)
You leave out a lot of good tube types if it has to be a Pentode.
I bet you were not restricting it to Pentodes.
And if it is a real Pentode, my favorites are EL84 and EL34.
6BQ5 "used" to be Beam Power tubes, not Pentodes.
Our club here, mostly uses triodes for SE amplifiers.
I did that too, mostly DHT, but some indirect heated.
In recent years, I mostly use Beam Power tubes for SE, either Triode wired, or UL (and some Pentodes too).
I did not like Pentode unless Triode Wired, or UL in SE amplifiers, but I will try again with a different topology using regulated screen voltage, and some negative feedback.
I am getting ready to try those same tubes in Beam Power mode, but not for SE, instead in push pull amplifiers.
I do not consider my hearing (and my personal bias) to be of general value to qualify to answer your question, but I do appreciate your Quest.
Or, do Beam Power Tubes also qualify?
(Beam Power/Beam Tetrode)
You leave out a lot of good tube types if it has to be a Pentode.
I bet you were not restricting it to Pentodes.
And if it is a real Pentode, my favorites are EL84 and EL34.
6BQ5 "used" to be Beam Power tubes, not Pentodes.
Our club here, mostly uses triodes for SE amplifiers.
I did that too, mostly DHT, but some indirect heated.
In recent years, I mostly use Beam Power tubes for SE, either Triode wired, or UL (and some Pentodes too).
I did not like Pentode unless Triode Wired, or UL in SE amplifiers, but I will try again with a different topology using regulated screen voltage, and some negative feedback.
I am getting ready to try those same tubes in Beam Power mode, but not for SE, instead in push pull amplifiers.
I do not consider my hearing (and my personal bias) to be of general value to qualify to answer your question, but I do appreciate your Quest.
Last edited:
No, the active tube can even be a triode. The plate to grid feedback can be adjusted to make a linear low Mu triode from a higher Mu triode.
Beam power tubes, that's the reason for the UNSET's existence. This started about 10 years ago while looking for a way to get triode like performance from TV sweep (line output) tubes without blasting their screen grids.
Beam power tubes, that's the reason for the UNSET's existence. This started about 10 years ago while looking for a way to get triode like performance from TV sweep (line output) tubes without blasting their screen grids.
I have never heard a normal, high quality, SEP amp compete with a normal SET amp.
Most SEP amps have a hard edge to the music. However, you can definitely minimize it by using small, well implemented NFB loops. For instance some of the best SEP amps I have made use plate to grid feedback vs global feedback. Another type was one that used a special output transformer that had a cathode winding. But neither amplifier was quite as good as enjoyable and natural as an SET amp IMO.
I haven't tried building an Unset yet, but it's very promising.
Discopete,
A Fisher SA-100 is a push pull amplifier operating in Beam Power Mode.
I guess I may have caused your question, since I mentioned push pull amplifiers in my earlier post in this thread.
The Fisher SA-100 is a lot different than either a Pentode single ended amplifier, or than a Beam Power single ended amplifier.
Are you asking for the differences of a typical push pull amp, versus a typical single ended amp?
A Fisher SA-100 is a push pull amplifier operating in Beam Power Mode.
I guess I may have caused your question, since I mentioned push pull amplifiers in my earlier post in this thread.
The Fisher SA-100 is a lot different than either a Pentode single ended amplifier, or than a Beam Power single ended amplifier.
Are you asking for the differences of a typical push pull amp, versus a typical single ended amp?
Last edited:
I had RH84 SEP - rather nice sounding amp, no trace of pentode sound what-so-ever... actually - triode sound with pentode power...
I'm slowly building now parallel version with 2xEL84 per channel, will see how it competes with other amps... should be good for 10watts of power per channel - so, potentially opening the door for almost any speaker on the planet...
I'm slowly building now parallel version with 2xEL84 per channel, will see how it competes with other amps... should be good for 10watts of power per channel - so, potentially opening the door for almost any speaker on the planet...
RH807 modified with a little gnfb and bypassed screen resistor. Best SEP amp I have every built and heard. But I only built and heard one.
I fell in love with CFBed output stages the first time I heard a tube GU50 CFB pentode amp from Thrax audio. CFB seems to heal the micro detail smearing of most PP amplifiers I've always hated, while in SE it fights muddyness at medium to high power levels.
I love using it deep even with high mu TV tubes, from 14% to 20%. Low mu tubes will need more turns. Right now I'm prototyping a 14% CFB PL36 SE.
As far my taste goes, once I got used to CFB SEs I cannot easily go back to non feedback SET amplifiers. Too much smearing. Only DHTs do at some point. Deep CFB SE gives me that hard rock, solid sound. You can listen to metal, overdriven stuff, low bassy electronics and in the meantime not losing that soundstage realism and microdetails SEs have.
If you do it right, it also gives you the pentode anode efficiency.
I love using it deep even with high mu TV tubes, from 14% to 20%. Low mu tubes will need more turns. Right now I'm prototyping a 14% CFB PL36 SE.
As far my taste goes, once I got used to CFB SEs I cannot easily go back to non feedback SET amplifiers. Too much smearing. Only DHTs do at some point. Deep CFB SE gives me that hard rock, solid sound. You can listen to metal, overdriven stuff, low bassy electronics and in the meantime not losing that soundstage realism and microdetails SEs have.
If you do it right, it also gives you the pentode anode efficiency.
Last edited:
I was expecting to hear more like 50% for "deep" CFB, but its effectiveness does depend on the output tube's gm too. The CFB will lower output impedance and remove OT magnetizing current effects (including hysteresis) plus taking out some tube distortion too.
Other local N Fdbk methods should give similar results. The "old" "shunt Schade" method however had a small flaw in it, due to the grid drive voltage signal messing up the current thru the plate feedback resistor.
"New" series Schade, or UnSet, fixes that, plus provides more screen and plate B+ when delivering maximum current. So the tube's capability is boosted some during peaks. Interestingly a little different from UL or CFB with inherent UL.
Other local N Fdbk methods should give similar results. The "old" "shunt Schade" method however had a small flaw in it, due to the grid drive voltage signal messing up the current thru the plate feedback resistor.
"New" series Schade, or UnSet, fixes that, plus provides more screen and plate B+ when delivering maximum current. So the tube's capability is boosted some during peaks. Interestingly a little different from UL or CFB with inherent UL.
Hi smoking-amp, nice seeing you around!
I was hypothetically thinking about CFB helping fighting magnetizing effects in OPT during hysteresis, but can you help me explain how it works? So far my potential idea what might be going on is:
-during small initial permeability where inductance is low, loadline gets steeper due to loading and stage gain decreases.
-gain decreasing means less voltage swing across the primary, hence less injected into the cathode via CFB winding, resulting in less negative feedback
-less NFB means more gain resulting in more injection in OPT primary voltage, for more flux density and fighting the initial permeability phenomena?
Do you find this hypothesis might make sense?
I was hypothetically thinking about CFB helping fighting magnetizing effects in OPT during hysteresis, but can you help me explain how it works? So far my potential idea what might be going on is:
-during small initial permeability where inductance is low, loadline gets steeper due to loading and stage gain decreases.
-gain decreasing means less voltage swing across the primary, hence less injected into the cathode via CFB winding, resulting in less negative feedback
-less NFB means more gain resulting in more injection in OPT primary voltage, for more flux density and fighting the initial permeability phenomena?
Do you find this hypothesis might make sense?
Yes, a typical N Feedback scenario. The magnetizing current will increase with low permeability at low current, so the cathode will not pull up as easily. Causing the grid-cathode V to increase until the extra current is met. The sensed error fix at the cathode side also inserts the necessary extra magnetizing current at the plate to fix that side too.
The primary winding also has some resistance effect, which allows the magnetizing current to insert some small V dist. in there too. A small amount of neg. resistance from the driver stage could fix that too, I think. Just enough to cancel the primary side winding resistanece. (one of the reasons some Amps have a separate feedback winding on the OT without load current going thru it. )
Any local N Fdbk should do the same, but the CFB is about as direct as it gets.
The primary winding also has some resistance effect, which allows the magnetizing current to insert some small V dist. in there too. A small amount of neg. resistance from the driver stage could fix that too, I think. Just enough to cancel the primary side winding resistanece. (one of the reasons some Amps have a separate feedback winding on the OT without load current going thru it. )
Any local N Fdbk should do the same, but the CFB is about as direct as it gets.
Last edited: