Tube Preamplifier Heater Methods

[dmp]

I compared heater methods for a tube preamp to see how different options compared for noise performance ( this V76 I built a few years ago )
I had built this with AC heaters with 100 ohm artificial center tap to audio ground. Tightly twisted heater wiring.
The noise spectrum of the preamp was flat without spikes at 60/120 Hz (white noise spectrum) when measured with RMAA

I compared:
- switching from the 100 ohm resistors to audio ground to a humdinger pot
- elevating the artificial center tap to ~50 volts from a voltage divider off the B+ rail
- Switching to DC heaters with the DC ground connected directly to the power audio ground

I didn't see an improvement or any difference in the noise performance with these different setups.
My takeaway is when building a tube preamp with good heater wiring and ground layout, simple AC heaters are just as good as the more complicated options.
But I wonder in what situations would these alternatives show an improvement? Poor AC heater wiring? Or is there a characteristic of certain tubes that allow heater hum to bleed into the audio?

 

[rock soderstrom]

I didn't see an improvement or any difference in the noise performance with these different setups.
My takeaway is when building a tube preamp with good heater wiring and ground layout, simple AC heaters are just as good as the more complicated options.
But I wonder in what situations would these alternatives show an improvement? Poor AC heater wiring? Or is there a characteristic of certain tubes that allow heater hum to bleed into the audio?

That is very interesting, but does not correspond with my experience!
My test setup was simpler and without negative feedback, if that matters.

What definitely matters with AC heating is the type of tube used. The V1 position in an amplifier is dominant in terms of hum.

Some tubes are optimised for AC heater and show much better hum resistance than ordinary tubes like I used (in my test setup). This is also mentioned in some data sheets of these premium tubes.

Example tubes would be EF37a, EF86, EF806, EF804s. I think you have one of these in V1 position, I only had a 6C5 or 6J7 in triode mode.


Addendum: A good example is the evolution from the EF40 to the EF86 (and close SQ relatives). Actually, these tubes are the same, except for the base. The difference is a bifilar winding of the heater in the EF86, which suppresses the hum of the AC heater. AC heating was demanded by manufacturers of consumer devices such as tape recorders, radios and record players for cost reasons. The tube industry responded with appropriate types of tubes.

 

[My3gger]

Poor heater and signal wiring practice can be seen quite often, another reason for DC heaters are pcb's where heater is not wired separately with wires.

 

[rock soderstrom]

Poor heater and signal wiring practice can be seen quite often, another reason for DC heaters are pcb's where heater is not wired separately with wires.

I also thought for a long time that you can't achieve hum-free results with AC heaters on PCBs without separate wiring.

But that's not true, as long as the designer knows what he's doing. You can achieve very good and completely equivalent results in this respect with PCB-based AC heaters. I have seen it several times, even guitar amps with plenty of gain and highly efficient speakers that were completely hum-free, despite PCB.

 

[My3gger]

I recall Ian making pcb with wired heaters due to space constraints.

 

[rock soderstrom]

I recall Ian making pcb with wired heaters due to space constraints.

[silent:arts] D-LA2A version does it too. If a lot of current flows, this also has some advantages over running fat PCB tracks to many tubes.

 

[Lampie519]

The V76 and family all use AC heating. Never heard anyone complain,

 

[rock soderstrom]

The V76 and family all use AC heating. Never heard anyone complain,

Absolutely right. Very well thought out layout, EF804s plus mumetal shielded mains transformer do their magic.

 

[Lampie519]

Very well thought out layout, EF804s plus mumetal shielded mains transformer do their magic

Indeed, and all in very confined space.

 

[rock soderstrom]

Indeed, and all in very confied space.

Ahh, E283CC is not a low budget tube either.
The difference to the ECC83 is the pinout, the 6.3V heater connections are directly next to each other, which reduces the hum influence significantly with AC.

 

[Lampie519]

E283CC is not a low budget tube either.

Nice tube indeed. Not seen often though. Maybe similar to the ECC803s

 

[rock soderstrom]

Nice tube indeed. Not seen often though. Maybe similar to the ECC803s

Yes, similar with small differences. The ECC808 is perhaps even closer in terms of some design goals. Not exactly cheap today either, unfortunately! All the tubes mentioned are "brummarm", i.e. hum-reduced with AC heating.

Nice collection of tubes, by the way!

 

[dmp]

What definitely matters with AC heating is the type of tube used.

Yes, good points - interesting to hear that some other tubes did 'bleed' hum from the AC into the audio. This test was on the V76 which used a EF804S. I also have built RCA designs with 1620 and 12AY7 tubes that also showed a white noise spectrum, no heater hum, (AC heaters with 100 ohm CT). I have in guitar amps over the years found 12ax7s that had hum while swapping. So an unhealty tube may show more issues.

Poor heater and signal wiring practice can be seen quite often, another reason for DC heaters

Yes, DC heaters could 'fix' poor wiring or ground layout. Connecting the heater CT to the ground in a way that it affects the first tube could pollute the audio.
Elevating the heaters might fix this because it would force the builder to move the CT ground.

This does make me continue to like AC heaters - done with proper wiring and ground layout. Simpler is good.

 

[rock soderstrom]

I also have built RCA designs with 1620 and 12AY7 tubes that also showed a white noise spectrum, no heater hum,

1620 and 12AY7 are also special quality tubes. I think they make the difference with AC heating. With DC heating, the differences are smaller or no longer relevant.

This does make me continue to like AC heaters - done with proper wiring and ground layout. Simpler is good.

I agree, especially with the right tubes. Next step is an AC heated microphone.

 

[Lampie519]

Try AC @ 250KHz no hum !
Here a direct heated tube with center tapped transformer:

 

[rock soderstrom]

Try AC @ 250KHz no hum !
Here a direct heated tube with center tapped transformer:

I don't want to hijack the thread, but what is this exactly? Can you explain it further?

 

[Lampie519]

We can continue here not to hijack this tread :
https://groupdiy.com/threads/tube-otl-and-zotl-amps.83620/

 

[My3gger]

1620 and 12AY7 are also special quality tubes. I think they make the difference with AC heating. With DC heating, the differences are smaller or no longer relevant.
I agree, especially with the right tubes. Next step is an AC heated microphone.

When saying DC, do you mean regulated/stabilized or filtered?

 

[Tubetec]

Hum depends on the level at the grid of the tube ,
A good high step up ratio transformer raises the signal level to where tube noise doesnt matter all that much ,
A tube mic your potentially dealing with a much smaller signal at the grid , so ac induced hum is much more a factor , likewise to some degree a tube pickup input stage , if you want to drive the following stage hard, DC heating can have an impact on noise levels in that case .
AC heating can be good enough in certain cases , others times it wont .

Ive modded a few first stage preamp tubes in guitar amps to run off rectified and smoothed DC from the 6.3v ac heater line , you might end up with a little less than 6.3v dc but thats usually not an issue with a first preamp tube in a guitar amp , it certainly can make a difference to hum levels with humbucking pickups ,single coils less so .

 

[rock soderstrom]

When saying DC, do you mean regulated/stabilized or filtered?

In this context, I am talking about a regulated and well-filtered DC voltage. Specifically, I had an LM317 power supply with ample RC filtering compared to the AC variants mentioned by dmp.

Edit:

I would summarize my position this way. You can get very good results with AC heating with the right tubes and careful wiring and layout in terms of residual hum.

This was also a stated design goal of the manufacturers at the time, to optimize certain tubes for this.

DC heating relaxes the situation somewhat, one achieves very good results faster and more reproducibly, even with ordinary tubes in demanding applications.

The disadvantages of regulated DC heating are the increased component costs, the thermal load on individual components and the somewhat reduced reliability in the long term.