• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Heater Wiring - the Good the Bad and the Ugly

George (tubelab) has commented on this in the past.

In the early development of the Tubelab SE I was chasing down a vague "foggyness" for lack of a better word, on simple music. It is masked fairly well on complex music, but a single vocalist or acoustic guitar reveals something that didn't sound right. I was using AC heating on a 45 tube with the hum carefully nulled down into the -60 db (referenced to full power) range. Playing a single 1000 Hz tone through the amp revealed IMD products at 940 and 1060 Hz. They were less than -60 db below the 1000 tone and rose sharply as the amp hit clipping. DC heating removed the IMD products, and removed the "fog". Sorry for sounding like an audiophool. Hey, I didn't say that it rolled back the fog, revealing the pure music, just for the few true believers to experience......send $1295 to.......

DC heating on a DHT does have a flaw. One end of the filament will have more voltage on it than the other. This means that the bias voltage between the filament and the grid will vary along the length of the filament. This causes one end of the filament to have higher emission than the other end. Some say that this will wear out one end before the other, while others say that it destroys the sound.

I find on a tube like the 45 which needs 40 to 50 volts of bias and has a 2.5 volt filament, it doesn't make any difference. My Tubelab SE (all DC heating) uses a pair of NX-483 tubes that have a 5 volt filament. The tubes are now 83 years old! They still work great.....I hope I work that good when I'm 83!

Things may be different on tubes like the 811A which has a 6.3 volt filament and needs a very low bias voltage. Here the voltage drop across the filament is such that one of the filament is cutoff and the other end is responsible for most of the plate current.
 
In the early development of the Tubelab SE I was chasing down a vague "foggyness" for lack of a better word, on simple music. It is masked fairly well on complex music, but a single vocalist or acoustic guitar reveals something that didn't sound right. I was using AC heating on a 45 tube with the hum carefully nulled down into the -60 db (referenced to full power) range. Playing a single 1000 Hz tone through the amp revealed IMD products at 940 and 1060 Hz. They were less than -60 db below the 1000 tone and rose sharply as the amp hit clipping. DC heating removed the IMD products, and removed the "fog".

Sorry for sounding like an audiophool. Hey, I didn't say that it rolled back the fog, revealing the pure music, just for the few true believers to experience......send $1295 to.......

That would be money well spent!
This is really useful information concerning the IM distortion.

DC heating on a DHT does have a flaw. One end of the filament will have more voltage on it than the other. This means that the bias voltage between the filament and the grid will vary along the length of the filament. This causes one end of the filament to have higher emission than the other end. Some say that this will wear out one end before the other, while others say that it destroys the sound.

Great intro into this part of the problem.
Your two following examples here also bring it into proper perspective:

I find on a tube like the 45 which needs 40 to 50 volts of bias and has a 2.5 volt filament, it doesn't make any difference. My Tubelab SE (all DC heating) uses a pair of NX-483 tubes that have a 5 volt filament. The tubes are now 83 years old! They still work great.....I hope I work that good when I'm 83!

Things may be different on tubes like the 811A which has a 6.3 volt filament and needs a very low bias voltage. Here the voltage drop across the filament is such that one of the filament is cutoff and the other end is responsible for most of the plate current.

The second case is where the problems may be more severe.

I was going to discuss this next,
and your discussion is the perfect lead in to the next bit.
Great practical info and observations.
Thanks for this
 
I've made a chart to help understand the Direct Heated Triode problem:

The 811a is a good example, because the heater-voltage is significant, and the bias might reasonably be set quite low, even zero volts, or a +ve value for class A2 operation.

Click to Enlarge:

279943d1335842319-heater-wiring-good-bad-ugly-811a-powersupply1.jpg


The very nature of a tube is such that the grid, with very little voltage, can control large current and voltage swings through the plate.
In part the reason why this works is that the grid is much closer to the cathode than the plate, which carries the High Voltage attractive force. The greater distance of the plate in turn lessens its influence.

It follows that small changes in distance between cathode and grid have the same effect as small changes in voltage on the grid. Although distances can't be changed, the effects can be understood by the analogy of distance.

For this purpose, I've redesigned the diagram for the triode,
in order to simulate the effect of the difference in voltage at each end of the cathode,
and its influence on cathode current flow through the screen,
when D.C. voltage is used instead of A.C. .
 

Attachments

  • 811a-powersupply1.jpg
    811a-powersupply1.jpg
    24.1 KB · Views: 10,406
Last edited:
Rongon had asked:
"Could one heat the filament with
a well-filtered or regulated +3.15V and -3.15V supply?"
The answer is that using a split-supply would make no real difference,
because the effect here takes place INSIDE the tube.

Secondly, looking at the diagram, one could in theory ground the cathode in three places, at either end of the D.C. supply, or in the middle (of a split supply).
But grounding in the middle would be disastrous, for another reason:

Instead of flowing directly to ground from cathode,
a center-connection would cause the plate current to flow through
the INTERNAL RESISTANCE of the power-supply!


This internal resistance will be unknown, and unpredictable.
For one thing, this resistance is likely to vary with current, heat-effects, and frequency-sensitive factors.

Thus,
(1) There is no advantage in relation to the cathode emission in using a split-supply (i.e., +/- center-tap).

(2) There could be a serious negative effect on the quality of the plate-current through the cathode to ground.​
 
Last edited:
Lets look again:


Click to Enlarge:



While a center-tap connection is not beneficial,
it indeed DOES make a difference which end of the D.C. supply is grounded!

If one grounds the +ve side, one can force the current
to flow all the way along the cathode. This will equalize both current loads and temperature across the cathode.

In the diagram above, the -ve side is grounded, exasperating the imbalance.
 
Could one not simply switch heater connections every few hours of listening by means of a toggle switch?

After all, a situation 1.4 x more severe than this (as far as voltage differential) is concerned arises 120 times every second under AC heater operation?
 
Just wanted to add this info, so it isn't forgotten,
while we're discussing heater circuits:


...
Some nice sound can be had by cutting the line voltage to ~90v from ~120v
Eddie Van Halen - Wikipedia, the free encyclopedia
I've been told repeatedly that this was one of Van Halen's (off the cuff interview) pranks.

That is, he really didn't use a variac,
unless perhaps experimentally in studio once, and its not a part of his 'sound'.

Those running out to try this ought to take serious heed of the obvious warnings, such as:

(1) A variac on most amps would also drop the heater voltage,
and this directly contributes to tube death.


So much so that RCA did a long comprehensive study
and warned designers not to vary the heater voltage
from recommended values more than 4%:
LOWER VOLTAGES especially KILL TUBES QUICKLY.



With 85% of the rated voltage on a tube, it goes from a 5,000-hour+ tube to a 3-hour tube!.
Letting your tech experiment with a variac on your amp for a couple of hours will cost you a whole set of tubes within weeks of installation!

Don't let the heater voltage on your tube drop below 96% for any extended length of time:

12.6 v heater (signaltubes): greater than 12.1 volts or else! (Never use 12 volt regulators (7812) without resistor adjusts)

6.3 v heater (powertubes): greater than 6.05 volts or else! (Never use 6 volt regulators (7806) without resistor adjusts)


Thus rather than a Variac, for most amps, we really want an ANTI-Variac (surge/voltage regulator) that provides correct heater voltage regardless of MAINS or LINE INPUT.
To do the opposite is to destroy tubes, drastically shorten tube life, and is only sensible if you are an idiot millionaire.

"cutting the line voltage to ~90v from ~120v" may give you some 'nice sound',
but it would cut your heater voltages to about 75%,
and for any length of time, cut your tube-life expectancy to only a few hours!

You can get the same results without destroying your tubes several other ways
 

Attachments

  • Heater voltage affecting valve life.gif
    Heater voltage affecting valve life.gif
    6.6 KB · Views: 5,159
  • Like
Reactions: 1 user
strikes me that more than dropping the heater voltage, any amp running fixed bias output tubes would have the negative bias decreased significantly. As the tube multiplies the effect of variations in the fixed bias applied (as opposed to cathode bias where there is effectively a self regulating feedback loop determining anode current) dropping the supply voltage would massively increase the output stage standing current...

(not heater related, but food for thought)

Does this mean that gradually bringing an amplifier up to running by means of a variac on the supply is potentially very damaging rather than a safe way to start up? Seems that yes, it could well be...
 
strikes me that more than dropping the heater voltage, any amp running fixed bias output tubes would have the negative bias decreased significantly. As the tube multiplies the effect of variations in the fixed bias applied (as opposed to cathode bias where there is effectively a self regulating feedback loop determining anode current) dropping the supply voltage would massively increase the output stage standing current...

(not heater related, but food for thought)

Does this mean that gradually bringing an amplifier up to running by means of a variac on the supply is potentially very damaging rather than a safe way to start up? Seems that yes, it could well be...

I'm not quite sure of some of this.

IF you have a soft-start for the heaters, that is good.
You should have a SECOND system which both softstarts the HV, and also delays applying it for at least 30 seconds.
A third independent system should allow the BIAS supply (if it exists) to fire up immediately and which (if it fails to start or stops) also blocks the HV via a relay.

Although both the heaters AND the HV should 'slowstart' to minimize tube wear and protect diodes/caps from surges,
They should operate at different rates. The heaters only need a current limiting thermistor or some such, while the HV needs a (later bypassed) resistor during 'ON' and charge.

Variacs are definitely damaging to tubes when applied to heater circuits, and also to BIAS supplies too!.
But its not the 'slow start' part, but the fact that a variac can be left for extended periods at the wrong voltage.

Slow-starting heaters is definitely good for tube wear because it allows slow expansion of heaters as they rise in temperature, while also simultaneously limiting "on" surges.

The reason lower heater voltages damage cathodes is because cathode surfaces get stripped of their coatings
while HV is applied to cold tubes.
This can't happen if you have the proper start-up sequence in the first place.
You can warm up the heaters slowly, check that the bias is on the grids, then turn on the HV safely.
In fact that is the only way to properly operate.
Part of the effectiveness of a Bias supply is its fixed nature and stability.
Without this, runaway conditions can easily arise.
This means that blindly applying Variacs to Power Transformers that also feed both HV and Grid/Scrn bias
is dangerously foolish, as you have noted.



If you have all the power supplies coming on randomly or willy-nilly, then certainly the instability of the entire system can cause just about every kind of damage, from stripped cathodes to runaway currents and screen meltdowns.
 
Last edited:
what about tubes that are operated whit a valve rectifier ?

the lower voltage on the preamp tubes might not be such an issue. it is true that the cathodes will not reach their optimum temperature but the rectifier may delay the HV upto a point where the tube's cathodes will not exhibit cathode stripping.

cathode stripping might also vary on the state of the tube. as stronger tubes can deliver more current at lower temperatures. (note this be course its a good tube testing procedure)

and your quite correct . variacs shouldn't be used on systems . only to reform the electrolytic s whit a SS rectifier plug in and output preamp tubes removed.

v4lve.
 
Really good article and hints....

Hi there to ALL the contributors on this article/thread.... VERY helpful and well explained... Thanks to ALL.... What a great forum and I appreciate all the selfless inputs... Being a bit of a 'newby' at valve/tube equipment, the knowledge of others is always so welcome.... Best wishes to diyAudio and its readers... Cheers.... Keith.... New Zealand
 
--
(1) A variac on most amps would also drop the heater voltage,
and this directly contributes to tube death.


So much so that RCA did a long comprehensive study
and warned designers not to vary the heater voltage
from recommended values more than 4%:
LOWER VOLTAGES especially KILL TUBES QUICKLY.



With 85% of the rated voltage on a tube, it goes from a 5,000-hour+ tube to a 3-hour tube!.

Would this go for all tubes or specifically for power tubes? I remember that undersea tubes were deliberately starved in heater current to increase the life time (no maintenance possible ...).
 
I'm also sceptical about this starved filament stuff. I'm thinking DHTs because that's all I use. It is just not believable that DHTs last 3 hours at 85% voltage. I have them on all day for weeks and months. Thinking 26 in particular here. Plenty of evidence of lower distortion at starved filaments also.

Is RCA talking about indirectly heated tubes here?

Andy
 
Would this go for all tubes or specifically for power tubes? I remember that undersea tubes were deliberately starved in heater current to increase the life time (no maintenance possible ...).

- its not "power tubes" vs. "others";

The study included currently manufactured power tubes, signal tubes, sweep tubes, TV tubes etc.
It presumably included all tubes with modern cathodes and filaments, and those with coated cathodes and anodes.

The large directly heated triodes of the previous era were probably not even under consideration. The reason would have been that RCA did not make them. I think for a while that Western, and Taylor continued to make the older tubes, thoriated tungsten, carbon anodes etc., and probably RCA continued to make its 813s and 814s for a while under army contract for transmitters and radar.

But the study refers to common tubes made and promoted in the 1960s and beyond.

Not 'Power vs Signal', but rather 'modern vs. pre-war'.
 
Good thread. Two points though. On DC heating of, say a 12AX7 tube tube where you have a center point in the heater, would feeding +6.3V from either side and exit from the middle not be the best route to go?

Also on cathode stripping. What is the mechanism that strips the coatings? I have seen the migration of elements in metals (used to work in a metallurgical lab in the aerospace industry, also took care of our vac furnaces) but I find it hard to believe a voltage between two points will cause atoms in a vacuum to fly off one surface and presumably deposit themselves on the other. And if the surface coatings have such a tenuous grip on their reality when cold why would they not be more inclined to leave the surface when they are hot? Almost in the realm of thermal spraying done on turbine engine parts, mind you all those require the melting of the source material.
 
Also on cathode stripping. What is the mechanism that strips the coatings? I have seen the migration of elements in metals (used to work in a metallurgical lab in the aerospace industry, also took care of our vac furnaces) but I find it hard to believe a voltage between two points will cause atoms in a vacuum to fly off one surface and presumably deposit themselves on the other. And if the surface coatings have such a tenuous grip on their reality when cold why would they not be more inclined to leave the surface when they are hot? Almost in the realm of thermal spraying done on turbine engine parts, mind you all those require the melting of the source material.

I was thinking this may have more to do with the primitive state of chemical coatings in the 50s, 60s. Everything must have been experimental, but mostly unguided.

I think RCA did the study in the first place because the army forced them to.
 
Filaments run on AC or DC

Hi

While all you are saying is basic and true, you do not know enough about tube amps to really write a tech article covering a lot of time span where various tecniques were used for circuit design.

A. The amp in the picture is a 1935 Seebburg jukebox amp, has a nice sweet tone, very good sonics, good frequency response and dead quiet.

There is a hum balancing pot on the front between the #80 rectifier and one of the #45 tubes. I found this pot had a spot where the hum just vanished, and I mean audiophile quiet!

This amp uses one #45 backwards to provide negative bias for a fixed bias, class-A output, and a very tricky circuit.

This amp is similar to a SE amp, driver transformer with one #45 driver, two #45s out, one #45 bias rectifier

Early amps and Radio transmitters were designed by radio men that knew tubes, and early gear had no hum, a complete myth, as they used good wiring, grounding and tube design to make it work.

This amp, with it's looney wiring, you would think would hum and be unstable, but it is quiet as a mouse, and rock stable.

This amp, is, compared to some of the early amps I have re-stored, actually pretty nice inside, Silvertone small guitar head amps a total mess inside, and others, but all the high-gain guitar amps I have worked on have almost zero hum, AC filaments.

Part of this is the tubes themselves. The mike preamp with the four tubes is one of a group of early style preamps I built, all out in studios, all used every week, and no re repairs, blown tubes or any problems in seven years!

These us a 39/44 very early pentode a #37 buffer, and two #76 triodes in parallel to drive the out transformer so they can run up to +12DB out/600 ohms.

These five-pin tubes were the first group with cathodes, and cathodes were designed to combat the AC hum problems and mostly disconnect the audio or RF signal from the filament.



These preamps use DC for the filaments, and all my gear uses a transformer, bridge, three stage filtering, then a TIP31C transistor as a "choke" filter, for noisless DC to the tubes

I would always use DC, but I am saying that if you had the adventures of being old enough like me to have been inside so many old radios, amps, transmitters, you would know that tubes have a sneaky way of not doing what all your CAD design programs say!
 

Attachments

  • Big_seeb_a_lr.jpg
    Big_seeb_a_lr.jpg
    192.8 KB · Views: 3,884
  • Big_seeb_b_lr.jpg
    Big_seeb_b_lr.jpg
    161.6 KB · Views: 3,848
  • Big_seeb_c_lr.jpg
    Big_seeb_c_lr.jpg
    220.4 KB · Views: 3,788
  • Vpre_4tube_insde_b.jpg
    Vpre_4tube_insde_b.jpg
    167.2 KB · Views: 3,800
  • Vpre_frnt_b.jpg
    Vpre_frnt_b.jpg
    127.2 KB · Views: 2,397
  • Like
Reactions: 1 user