An old simulation I messed with about a year ago and never built.
Keeping 6DJ8 plate above 12V essential, else grid leaks too much.
90V would indeed be better, but maybe get away with a lot less?
Here the triode is abused in plate to cathode 1/(Mu-1) feedback.
First, input to plate is divided 50% to the B+ rail. Not for fake "UL",
nor for doubling Mu gain, but merely to keep that plate well above
serious grid leakage.
The lower IRF530 is setup up as a Triodlington, and the upper 530
is set up as an Aleph/Anti-triode current sink. The BJTs on the far
right maintain bias, and the diodes below are for correcting offset.
JFET cascode on the left is not cause the grid is expected to leak.
The intent was to flip a common GND reference down to the B-rail.
1K vs 1K, set up just like a unity gain concertina splitter, but only
utilizing the inverted output... Not intended to "split" anything...
All sand here are restricted by local feedback to unity voltage gain.
Except the Triodlington, which I guess the 50% UL feedback allows
a gain of 2x(Mu-1).
Keeping 6DJ8 plate above 12V essential, else grid leaks too much.
90V would indeed be better, but maybe get away with a lot less?
Here the triode is abused in plate to cathode 1/(Mu-1) feedback.
First, input to plate is divided 50% to the B+ rail. Not for fake "UL",
nor for doubling Mu gain, but merely to keep that plate well above
serious grid leakage.
The lower IRF530 is setup up as a Triodlington, and the upper 530
is set up as an Aleph/Anti-triode current sink. The BJTs on the far
right maintain bias, and the diodes below are for correcting offset.
JFET cascode on the left is not cause the grid is expected to leak.
The intent was to flip a common GND reference down to the B-rail.
1K vs 1K, set up just like a unity gain concertina splitter, but only
utilizing the inverted output... Not intended to "split" anything...
All sand here are restricted by local feedback to unity voltage gain.
Except the Triodlington, which I guess the 50% UL feedback allows
a gain of 2x(Mu-1).
Attachments
Last edited:
Voltage and current at R1 (cathode resistor) held constant at gate
threshold of MOSFET M1. Thus Mu loadline is at constant current.
Down here, even on the worst part of the triode curve, Mu is still
a linear function, as long as the current is constant.
threshold of MOSFET M1. Thus Mu loadline is at constant current.
Down here, even on the worst part of the triode curve, Mu is still
a linear function, as long as the current is constant.
Last edited:
No longer related to schematic above...
If one were to operate 6DJ8 below 12V plate?
Leakage is assured. Probably only A2 operation
is possible. But if you allow an emitter follower
to drive the grid into A2, and ground cathode,
and apply constant current to plate. Maybe,
just maybe, you could still have a linear Mu.
Which gives linear Mu? Constant plate current,
or constant cathode current? Cause in severe
A2, these currents won't exactly match.
If one were to operate 6DJ8 below 12V plate?
Leakage is assured. Probably only A2 operation
is possible. But if you allow an emitter follower
to drive the grid into A2, and ground cathode,
and apply constant current to plate. Maybe,
just maybe, you could still have a linear Mu.
Which gives linear Mu? Constant plate current,
or constant cathode current? Cause in severe
A2, these currents won't exactly match.
Last edited:
My guess is that constant mu needs constant cathode current, as it is the cathode current which is determined by the net electric field seen at the cathode. The grid current will steal (nonlinearly) some of this. However, there could be a small amount of cancellation if mu increases with current at the same rate as the grid steals it. But why design a line amp which needs an input buffer?
L1 might be bootstrapped gyrator, ccs, or whatever...
The question: Now is cathode current a constant?
Let us assume a grid leak so severe, Q1 never off.
Does Mu work as intended (in fulltime A2) or not?
Yes, thats +3.3V on the grid...
I'm asking, not saying. I don't know if this is plausible
work of science fiction, or just plain nuts???
The question: Now is cathode current a constant?
Let us assume a grid leak so severe, Q1 never off.
Does Mu work as intended (in fulltime A2) or not?
Yes, thats +3.3V on the grid...
I'm asking, not saying. I don't know if this is plausible
work of science fiction, or just plain nuts???
Attachments
Last edited:
First you are telling me to fix my knowledge, then you are asking if low voltage operating is possible? I wasn't aware that my knowledge needed fixing.
Tubes can work at low voltages. The majority of the tubes that were designed to operate with hundreds of volts on the plate will not work very well on 12 volts or so. Some like the 6DJ8 will function at very low currents with a reasonably low distortion so they "work" but sub optimally.
THere were hundreds of tubes specifically designed to work with 12 volts on the plates for car radios in the 1950's. These are known as space charge tubes. Some numbers: 12U7, 12R5, 12J8, 12GA6, 12FX8, 12FR8, 12FK6, 12F8, 12EM6,
There were plenty of tubes designed for aircraft use at 26 or 28 volts. The only one I remember is the 26A7. It will make 1/2 watt of audio power from a 26 volt supply.
Maybe something got lost in translation, or "fix your knowledge" is considered polite in Taiwan?
Fix my knowledge? Runs contrary to my mission statement: Upgrade your idiocy.
Half baked theories can and must be properly misrepresented as absolute truth.
Half baked theories can and must be properly misrepresented as absolute truth.
Member
Joined 2009
Paid Member
I heard a story, when one wise western man come to eastern Master to learn wisdom. Master gave him a bowl full of water and asked to bring some fresh water from the creek, but keep what is in the bowl.
Do you remember the end of the story?
I don't know the "real" answer to that one, but I'd just stick it in the freezer.
Assuming the same bowl had to be simultaneously used for both... You didn't
specify any rules that forbid another container. Drinking the water. Etc etc...
Nor how the Eastern master is going to know the diff between one bowl of
water and a "fresh" bowl from the creek? What he don't know won't hurt.
Just give him back the same bowl, or a defective one from another master,
what most RMA/Warranty depts would do...
LTC1144 with minimal supporting parts (2x shottky diodes, 3x 1uF caps)
seems able to simultaneously upconvert +18V supply to +35.4 and -18.
With pin1 tied high, the frequency should be 100KHz (50KHz ripple out).
Just my interpretation of the spec sheet, could be completely wrong...
But looks like could supply a small signal triode or two.
LT1144 spec sheet does not show the combined doubler and inverter
application, look to lower voltage and frequency (but otherwise same)
ICL7660 and MAX1044 spec sheets and app notes.
This device is a cheap 8pin DIP inductorless capacitive charge pump.
+12.6V would convert to +25 / -12.6 ?
seems able to simultaneously upconvert +18V supply to +35.4 and -18.
With pin1 tied high, the frequency should be 100KHz (50KHz ripple out).
Just my interpretation of the spec sheet, could be completely wrong...
But looks like could supply a small signal triode or two.
LT1144 spec sheet does not show the combined doubler and inverter
application, look to lower voltage and frequency (but otherwise same)
ICL7660 and MAX1044 spec sheets and app notes.
This device is a cheap 8pin DIP inductorless capacitive charge pump.
+12.6V would convert to +25 / -12.6 ?
Last edited:
My knowledge, is it fixed yet???
This is voltage tripler described above, except I had severe PSRR probs
with the negative supply, so I chose to ground the negative rail instead.
Means the heater now floats a few volts above ground, not a bad thing.
The plate load needs explaining: I have an emitter follower bootstrapped
back to provide CCS , hold the tube's loadline flat on Mu , hold cathode
bias diode at constant current too. This is both output buffer and main
power supply filter. Anyways, Q2's emitter following the plate is held at
constant current (and thus constant voltage drop) by action of Q3. I
chose Sziklai rather than Darlington. Specifically so the emitter of Q3
(which is not at constant current) would not have funky voltage drop
in series with the clean emitter controlling the following action.
Considering both the tripler and the bootstrap, I see 36V rail, and the
strap can actually swing a little above that, up to about 40V in total.
Linearity of the tube don't seem bothered by the low voltage, as long
as I am forcing a constant current loadline, Mu is still in full control.
You also need Duncan Amps "dmtriodep.inc" for NH12AT7 triode model.
NH meaning "No heater".
My currents are biased to the limit of 1/2W this chip can dissipate.
Taking them any further might let the magic blue smoke out.
Let us assume a 12.6VDC isolated wall wart that won't be bothered
to float slightly (another 12.6V) above GND for the heater supply.
This is voltage tripler described above, except I had severe PSRR probs
with the negative supply, so I chose to ground the negative rail instead.
Means the heater now floats a few volts above ground, not a bad thing.
The plate load needs explaining: I have an emitter follower bootstrapped
back to provide CCS , hold the tube's loadline flat on Mu , hold cathode
bias diode at constant current too. This is both output buffer and main
power supply filter. Anyways, Q2's emitter following the plate is held at
constant current (and thus constant voltage drop) by action of Q3. I
chose Sziklai rather than Darlington. Specifically so the emitter of Q3
(which is not at constant current) would not have funky voltage drop
in series with the clean emitter controlling the following action.
Considering both the tripler and the bootstrap, I see 36V rail, and the
strap can actually swing a little above that, up to about 40V in total.
Linearity of the tube don't seem bothered by the low voltage, as long
as I am forcing a constant current loadline, Mu is still in full control.
You also need Duncan Amps "dmtriodep.inc" for NH12AT7 triode model.
NH meaning "No heater".
My currents are biased to the limit of 1/2W this chip can dissipate.
Taking them any further might let the magic blue smoke out.
Let us assume a 12.6VDC isolated wall wart that won't be bothered
to float slightly (another 12.6V) above GND for the heater supply.
Attachments
Last edited:
Kenpeter,
My knowledge, is it fixed yet???
I don't know.LOL.
You know what they say if it isn't broken dont fix it!😀
However a few D.I.Y. mods would be O.K.
I used a DC to DC voltage converter with 6.5KV isolation for the heaters of SRPP and powered the HT from an inverter @ 400V with a current limited reg 50mA at 350V. Then connected the DC converter output to the HT to lift the heaters.
Just for fun.
Regards
M.Gregg
My knowledge, is it fixed yet???
I don't know.LOL.
You know what they say if it isn't broken dont fix it!😀
However a few D.I.Y. mods would be O.K.
I used a DC to DC voltage converter with 6.5KV isolation for the heaters of SRPP and powered the HT from an inverter @ 400V with a current limited reg 50mA at 350V. Then connected the DC converter output to the HT to lift the heaters.
Just for fun.
Regards
M.Gregg
You are right, but Eastern Master expected from Western wise man to fix his knowledge before learning something new. Sometimes fixing means amputation. 😉
But in this particular case I don't know what to fix, since on low voltage tubes don't have totally miraculous properties: they are still the same. JFETs, for example, on such voltages perform better.
So.. Has the wise master, fixed our H.V. industry, and offered up schematic's and magic to make myself a 100w 12v powered amp? No you say? Come back later?
Aww shucks... I thought he had cured the worlds power factor...
Ill stick to 200+v for best results 😉
Aww shucks... I thought he had cured the worlds power factor...
Ill stick to 200+v for best results 😉
You need some fuzzy quantum electrons that can be both 200 V and 12 V at the same time. When I blow the fuse in my DVM I seem to get them quite often. Something about measuring and dead cats.... Uh-Oh, my cat heard that.
You know that ancient Egyptians in order to please loved ones preserved their bodies so returning back in next reincarnation they could find them and see how they were loved and respected, so even disposed vehicles of their souls are still respected. All organs they considered significant were well preserved. What was "fixed", was brain...
- Status
- Not open for further replies.