• 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

There are cable ties, different coloured wires that look to adhere to a convention, and plenty of room to work with a clear layout. If after testing there are no issues, then there is no need to break what ain't broke. I don't see a problem with the wiring, and would hate that someone felt concerned about the reactions of people so had to ask not to have his work displayed.
 
  • Like
Reactions: 1 user
“There are cable ties, different coloured wires that look to adhere to a convention, and plenty of room to work with a clear layout. If after testing there are no issues, then there is no need to break what ain't broke. I don't see a problem with the wiring, and would hate that someone felt concerned about the reactions of people so had to ask not to have his work displayed.”

Tomas was a modest person. He had only been making custom amps commercially for 2 years so I think that he still considered himself to be a newbie in the tube world. Alas, looks like his Ultrasonic Studios page has already come down. He was only 40 years old, such a loss…

Anyway, I’m glad to hear that his construction is at least competent. I suspected it was since the amp sounds so good. Hope others might be inspired by the unusual heater arrangement.
 
I have been searching the whole tread but i could not find any best practice regarding the "phase" issue. What is considered best practice regarding the phase of AC-heater supply, what is the result of doing it wrong, how can you tell which tube needs this or that phase? Does this aplly to certain tubes only, for example directly heated cathodes?
 
Phase of AC, that is an interesting question!
Common knowledge (read: practice) is that the centre of the AC is grounded, such that the two remaining halves alternating get an opposite, halved voltage, signal. And the time constant of the layer is such that there is no alternating temperature. So we talk about residual (magnetic, inductance, of plain voltage) influence on the sugnal.
Also, as I understand: The wires are often wound such that the filament (indirect heated) will cancel AC.
As a system, the effect is mostly very very minimal AC in the signal.

In a DHT it is different, but even then some have heaters that minimise hum per the layout (the original WE 300B is claimed to have that)

That said - what happens if to tubes in series are attached one way and the other (pin 4-5 versus 5-4), would that be heard? Dunno. But an interesting thought.
I once changed the heater in a phono (I heard some hum, ECC88) to an SMPS, and all at once I heard the switching noise! So the benefits of the structure go for 50/60 Hz, not 30 KHz. . .
 
Holy resurrected tech post Batman!

I bought an amp a couple weeks ago. Heard it played through a pair of VERY pricey Tannoy speakers (99dB sensitivity), no noise, just bliss.

TEKTRON TK2 2A3/50I-S(ET). It was - after some emailing to Don Corby of Corbysaudio.com and his query to the factory rep - a dealer DEMO model, probably 20 years old. NO schematic available. I’ve looked at dozens of schematics, found one VERY close to TEKTRON’s (Don did get me a copy of one from the factory dated 2015, which, while CLOSE, has some major differences from the prototype).

This seemed like a bargain at the time, and it IS/WAS. With a NEW set of EML 300b tubes courtesy Scott Bierfeldt of Verdantaudio and financing it sounds even better than it did with the 21 year old Sophia 300b/2.5v tubes it came with but…

there’s STILL a little bit of residual hum despite bias adjustment. Not enough that it’s noticeable with music playing, but JUST enough to annoy me. It’s 120 Hz (definitely, did a sound sweep with an analyzer and then a tone generator, it’s exactly that).

So, I’m contemplating running the 300b filaments as DC but… the cathode bypass cap on the tube(s) now (defunct Tech-Cap) is polarized
1694348135343.jpeg

1694349125341.jpeg

1694349190586.jpeg





As you can see from this schematic - and many more - the bypass caps are usually shown as bipolar. IF, changing ONE thing at a time, I were to add in a DC filtering circuit like this:

1694348422442.jpeg


vs. schema originale /OEM factory schematic:

1694348859954.jpeg


Differences in my amp: no meter, NO negative grid circuit through potentiometer as shown bottom right, no standby circuit. For orientation, smack dab in the middle of the drawing is the 470 uF electrolytic less the 1k 25W panel mount resistor.

QUESTION:

1- Does the polarity of “X, X” matter with respect to the bypass capacitor’s + terminal pointed to the heater bias pot (ground is ground)?

Meaning, for now, I leave the ORIGINAL caps in place, add the DC heater circuit, and see what changes?


BEFORE doing this, I will recheck wiring, twist the AC pairs a bit more, reroute as necessary, but even though the EML’s cut the hum about 3dB (still audible at 3 feet), the possibility of reducing it to zero intrigues me.

Thanks in advance.

Norm

(System: Ortofon Cadenza Black, SME V TA with 33g Litz wire to Quadratic MC-1 SUT, QuirkAudio refurbished APT-Holman preamp, TEKTRON amp, Klipsch Heresy IV speakers)
 
Last edited:
D’OH!

One last photo, this from an extensive websearch for the amp I came across last night.

Looks like this HAS been done to THIS model amplifier, but there’s some loss of detail, and the capacitor between the filament and the grid is curiouser and curiouser.

I wonder how well this worked out.
 

Attachments

  • IMG_2374.jpeg
    IMG_2374.jpeg
    123 KB · Views: 94
Hi NormB, I wonder if you would get a better response if you asked the moderators to move your posts to a new thread 'Solving 120Hz hum in 300B amp'? This thread is more about the general placement of wires for best results, and conventions for good practice, but is not really for fault finding specific issues.
Cheers, Richard
 
  • Like
Reactions: 1 user
Going to go ahead with the project soon, just as soon as I replace a few of the bypass caps, check resistor values. Running 21 year old Sophia tubes the hum was louder, MUCH less with new EMLs, especially since I have about 20 hours on them this past 10 days. But that’s not a bad idea.

I ran my question past an IEEE friend of mine who just looked at me and said “there is no polarity on the filament.”

???

He meant it won’t matter which side I connect + lead to, that will be +, the bias pot will correct any variances. Ground to ground.

Thanks for the suggestion.
 
  • Like
Reactions: 1 user
@Diabolical Artificer The cathode circuit of the 300B is its heater reference, I just wonder why they waste the time making it DC though.

The only relevance in heater referencing, is only with the Cathode in any tube. But I've noticed with a lot of designs people referencing the heaters to the DC ground. Which isn't always the best application.

But I do question other people's transformer windings on their heater secondaries. As I noticed some I bought because I don't want to spend the time to wind the transformer I wished I did, because they didn't power match the 6.3CT winding correctly and noise is rectified on to the cathodes.

So anytime I use a transformer I don't wind, I hook up just the heaters (ac) and apply power and look to see if they wound the transformer winding correct or not. If they wound it correctly, there should be no noise on the scope in the 2mv/div range.
 
Correct.
Either way, it shouldn't reference any other power source or power supply. Nor any ground
There is some drift of electrons between heater and cathode (talking of an indirectly heated cathode that by itself is 'isolated', i.e. not connected). Hence there is potential. And the flow of electrons is intermittent, maybe call it shot-gun style, or like rain, intermittent. Anyway, any differential needs to be braought to 'earth', in DC and AC. To remedy the flow=DC level, it attached so to prevent this on an appropriate DC level; the AC is earthed through e.g. capacitor.
So yes, it inherently IS connected and should be taken into account.
[If not, noise arises and the heater will arise; maybe life time is reduced but I would need to look up that, sorry]
 
Typically the concern regarding the heater relates to hum in the final equipment. A simplistic view from an outsider (eg. the IEEE friend) is that the heater is isolated from remaining circuitry and so it doesn't matter how it is connected. Sadly, audio circuitry can have quite high gain and parasitic effects can play a significant part related to noticeable hum.

This article goes in to that aspect to show how the heater may play a part: https://dalmura.com.au/static/Hum article.pdf
 
There is some drift of electrons between heater and cathode (talking of an indirectly heated cathode that by itself is 'isolated', i.e. not connected). Hence there is potential. And the flow of electrons is intermittent, maybe call it shot-gun style, or like rain, intermittent. Anyway, any differential needs to be braought to 'earth', in DC and AC. To remedy the flow=DC level, it attached so to prevent this on an appropriate DC level; the AC is earthed through e.g. capacitor.
So yes, it inherently IS connected and should be taken into account.
[If not, noise arises and the heater will arise; maybe life time is reduced but I would need to look up that, sorry

I was talking about the 300B because the heater is the cathode and referenced there in the circuit at cathode voltage.

But since you want to discuss heaters in indirect heated tubes, generally, you keep them just referenced to the DC supply and do so within a voltage limit.

I would reference the heaters on the 6em7 to 70 volts on the 300b amp. Because 180V is close to 200V heater to cathode limit.
 
Typically the concern regarding the heater relates to hum in the final equipment. A simplistic view from an outsider (eg. the IEEE friend) is that the heater is isolated from remaining circuitry and so it doesn't matter how it is connected. Sadly, audio circuitry can have quite high gain and parasitic effects can play a significant part related to noticeable hum.

This article goes in to that aspect to show how the heater may play a part: https://dalmura.com.au/static/Hum article.pdf


Beyond all of that, it is when the heater secondaries is not wound correctly is what I'm talking about. The 6.3 CT winding is two 3.15V windings. if they are not exactly the same, when tying them together for the 6.3V winding, there will be a little distortion on the sine wave, that shows up on the cathodes of the tube. And I noticed some cheap power transformers are not wound correctly.

I test to see if its there when I use someone's else's transformer. Otherwise I would just wind it my own correctly. I actually don't have hum problems when I build.
 
"because they didn't power match the 6.3CT winding correctly and noise is rectified on to the cathodes." Never heard of power matching heater windings, could you explain please?

So anytime I use a transformer I don't wind, I hook up just the heaters (ac) and apply power and look to see if they wound the transformer winding correct or not. If they wound it correctly, there should be no noise on the scope in the 2mv/div range.
Could you expand on this please? Also, how would one test for this? Are we talking N transferred to the cathode by the htr wiring, phase noise or something else?

Are we talking about two different things, IE directly heated valves, so filaments and indirectly heated valves?

I'll re-read the last few posts to try & get my head round this phenomenon but have not heard of it before.

Andy.
 
Last edited:

Diabolical Artificer said:​

Could you expand on this please? Also, how would one test for this? Are we talking N transferred to the cathode by the htr wiring, phase noise or something else?

Are we talking about two different things, IE directly heated valves, so filaments and indirectly heated valves?

I'll re-read the last few posts to try & get my head round this phenomenon but have not heard of it before.

Andy.
I'm talking about when I make a power transformer.
Without really going in depth about it, when i make a 6.3Vct winding, I'm actually making two 3.15V coils and connecting them together to form a 6.3Vct winding.
After I get my first 3.15V winding spun, I load test it with 90% loading to see if I need to spin more on the bobbin if my voltage dropped more than 10%. I make note of the number of turns, spin the other 3.15V winding. Then I tie them together, and run a 5V p-p 60hz sine wave on the primary and look across the 6.3V CT winding that I just wound to look for distortions in the waveform from the center tap not being electrically in center crossing of the AC signal in the secondary. If there is an issue of number of turns, its shows up as a flat spot on the secondary winding sine wave on the scope.Then after fixing that flat spot (by adding or removing turns) I load it again 90% at full voltage on the primary and check the sine wave on the secondary to see if I have any distortions.

For any distortion induced is injected into the signal at the cathodes. You can always check to see if the heater winding is good by removing B+ (plate voltage) on the unit and look at the cathode in a scope (across the referenced ground or itself) and see if there is distortion from the heaters rectifying into the signal, they should be little pulses on the cathode where the heater winding is distorting. If the heater secondary winding is correct, there should be no distortions translated onto the cathode.

When I use some one else's transformer, or have someone like Jensen or Edcor make me one, I test the transformer to make sure its correct before building with it. This is my process that I check heater windings. I also have checks for B+/bias and signal and output transformers.

I find that after 1993, I have to check this when I buy a transformer. Because so many don't do above and arbitrarily spin a transformer with an X number of turns. A lot of the ones that still do it the old manual way is very expensive to buy from. Like $200-300 more for the transformer. I understand the price hike because you spend much more time making a more perfect transformer for the application.

@Diabolical Artificer I don't know why the quote button doesn't show on the post until someone comments afterwards. but hopefully mentioning you will trigger the notification that I responded to your questions.
 
Thanks for that explanation. If I've understood you correctly your referring to the matching of the two windings of a center tapped heater winding so that they have an equal voltage with reference to the CT. However I've found in practice when using non center tapped windings with a humdinger/humbucker pot that the lowest noise is often achieved with the pot offset. How much of this is down to the pots non linearity I don't know.

I've attached an old article that might touch on the subject, I'll admit to not fully understanding all the nuances of this subject.

Lastly there's another aspect, when winding htr windings on a transformer the turns per volt of a core doesn't always allow you to dial in the heater voltage precisely, especially on toroid cores that have a higher T/V.

Regards, Andy.
 

Attachments

  • hum_bucker.pdf
    257.9 KB · Views: 44
Thanks for that explanation. If I've understood you correctly your referring to the matching of the two windings of a center tapped heater winding so that they have an equal voltage with reference to the CT. However I've found in practice when using non center tapped windings with a humdinger/humbucker pot that the lowest noise is often achieved with the pot offset. How much of this is down to the pots non linearity I don't know.

I've attached an old article that might touch on the subject, I'll admit to not fully understanding all the nuances of this subject.

Lastly there's another aspect, when winding htr windings on a transformer the turns per volt of a core doesn't always allow you to dial in the heater voltage precisely, especially on toroid cores that have a higher T/V.

Regards, Andy.
Turns per volt will get you in the ball park, but I put the winding under 90% load and add more turns to compensate for the voltage loss under load.
When I need to make a ct winding, I have to divide these extra turns equally across two the 3.15V coils I join to make a 6.3V winding.
 
Thanks for that explanation. If I've understood you correctly your referring to the matching of the two windings of a center tapped heater winding so that they have an equal voltage with reference to the CT. However I've found in practice when using non center tapped windings with a humdinger/humbucker pot that the lowest noise is often achieved with the pot offset. How much of this is down to the pots non linearity I don't know.

I've attached an old article that might touch on the subject, I'll admit to not fully understanding all the nuances of this subject.

Lastly there's another aspect, when winding htr windings on a transformer the turns per volt of a core doesn't always allow you to dial in the heater voltage precisely, especially on toroid cores that have a higher T/V.

Regards, Andy.
The humdinger is just references the heaters so it will stay reversed biased with the cathode. Most of the time, I reference to B+ instead of gnd to cancel the ripple out of the audio.

But all of this doesn't work if there is something wrong with the transformer. Center tapped or not.

There is another tube diy site I don't belong to but linger and watch and laugh as they make a dc heater circuit with a zillion capacitors when the real problem is their transformers they are using are bad.

There is no real reason to use DC heaters over AC heaters for anything.

Another thing I noticed in the DIY world is arbitrarily grounding the speaker winding. Why? its not going to do anything positive or be an advantage of anything in this 300b circuit.
 
Last edited: