Tube Specs Sheets and Output Transformer Matching

[TomCarlos]

MEF members….

Ok, I am pushing myself to learn some tube and Output Transformer theory today. And if this has been answered in another thread, my apologies. I did find two threads but not sure that either answers my question.

Output transformer observation/question/thought process

output transformer impedence matching

I was watching a video (Uncle Doug - Impedance Ratio: The Easiest Way to Decipher Output Transformers) on Output transformers and calculating Impedance ratios and matching to tubes. The idea is that you have an OT, not sure of specs, etc, and then try to figure out how to use it in an amp. In the video, the scenario seems to be you have the OT and two output tubes, in Class AB, push pull. In the case of the Uncle Doug video (at around the 25 minutes point), he shows you how to (1) calculate the Impedance ratio of the OT, (2) review the spec sheet of a 6L6 tube, look for Class AB, 360v, and there we see 3,800 ohms for the load resistance, and (3) determine the load that can be applied to the OT secondary leads.

In Doug’s video, we have the following (at the 25 minute point).

OT Impedance Ratio = 1003 (measured or calculated)

Tube Load Resistance = 3,800 (spec sheet, at 360 volts, see below)

OT speaker impedance across windings = (Tube Load Resistance) / (OT Impedance Ratio)

OT speaker impedance across windings = (3800) / (1003) = 3.78 ohms for the Yellow and White leads of the secondary.

Ok, I get all that. BUT… what happens if your amp has FOUR output tubes. Do you still use the 3,800 spec or do you change that in the calculation.

Thanks in advance.




Spec Sheet - 6l6g.pdf

[The Dude]

Roughly speaking, if you double the tube count in the same configuration, you'd halve the transformer impedance. You also need to figure plate voltage for the most accurate calculations.

[TomCarlos]

Thanks Dude....

Intuitively, I know that. It is like yanking two 6l6 tubes from a twin. Instead of a 4 ohm load, you can now run a single 12" speaker.

But going back to the tube spec sheet, it appears that instead of using 3,800 ohms as the load resistance, with four output tubes, you now have 1,900 ohms. It is like a pair of tubes in parallel with a pair of tubes.

Is that how to interpret the scenario?

[The Dude]

Yes. That is basically it. I'm sure you noticed, but the specs you posted are for 2 tubes in a class AB push-pull circuit.

[TomCarlos]

Thanks Dude....

And yes, the example I gave was for two tubes. But I was just wondering, what happens if you have a quad ?? It's rare that I would ever work on something like that, or have an unmarked transformer. But I guess I was in the mood to learn something today.

Much appreciated.

[Helmholtz]

2 tubes wired in parallel can supply doubled load current but available voltage swing stays the same.
To allow for doubled load current the plate load is halved.
As power is given by I˛R (I and R being the values for a single tube), we get (2I)˛R/2 = 2I˛R for the 2 tubes, so output power is doubled.

[TomCarlos]

Thank you Helmholtz... that explains it !!

[Mick Bailey]

That's a neat description.

[J M Fahey]

[B]
Ok, I get all that. BUT… what happens if your amp has FOUR output tubes. Do you still use the 3,800 spec or do you change that in the calculation.

Thanks in advance.

Twice the tubes, twice the available current, half the load impedance,twice the power.

Keeping everything else the same , specially voltage, so you will need twice the PT current capability, and double main filter capacitance to keep same ripple under doubled current load. .

So if you pull 2 tubes from a Twin and readjust impedance, you won´t have exact half impedance but a little more, because power supply will hold a little better since it still has larger-than-needed capability for just 2 tubes.

As of Uncle Doug, I have mixed feelings about his videos: his are better than the average for sure (which is a vrey low reference point) and he clearly has experience, I respect that.

But now and then he states some iffy opinion, some blanket statement which is just some more of "Internet lore".

Oh well.

In this case, I applaud his referring to datasheet, only caveat is they show conditions at 360V +V .... which no red blooded Guitar amp maker/designer ever uses, starting with Leo, everybody and his brother *abuses* them with 430V ... or worse.

Among other things, optimum impedance at such voltages is much higher than datasheet suggested.

Proper procedure is to draw load lines over the plate curves, using *actual* supply voltages.

[SoulFetish]

Tom, one of the things I learned in the process of educating myself on tube output stage design is that there is no single "correct" load impedance for any type of output tube.
Manufacturers provide helpful examples in datasheets for some typical operating conditions, which are chosen to keep the tubes in their safe operating region and get the most out of them.
I think that because of this, and the historical influence of Fender/Marshall/Vox designs, you commonly see threads where someone might ask what the right load impedance is for a particular tube type, and someone might answer with a commonly used impedance as the right one for that tube; ie; 3k3 for a pair of 6L6s in push pull (lets say.)
So, it might be great advice, provided that the person is using the tube in a corresponding manner, but the reality is, there is no one "right" load impedance for a type of tube which covers all conditions.

In fact, Pentodes and Tetrodes, can be designed for wildly varying load impedances, with screen voltage arguably having the greatest influence on power output, while keeping the tube from over-dissipating. (Okay, "wildly varying" might be a little hyperbole, but it might be more that you would think).
In order for me personally to make sense of the relationship between a suitable load impedance and the tube's operating conditions, I rely HEAVILY on the published characteristic curves in the datasheets. I'll import them into Illustrator or redraw them cleaner, and use that info to draw up load lines. That allows me to very quickly get a detailed overview of the tube operation, From there, I can create several different example, make changes, see potential faults like over-dissipation, determine bias.
Drawing load lines is invaluable to me. Maybe it's because I'm a visual learner and respond well to graphical information.
Do you know how to draw an output stage load line, Tom?

[catalin gramada]

Hey.
Still found the precious material of Patrick Turner work, who sadly passed away one year ago. RIP Patrick. The sheets was preserved thankfully to Atrad-Audio NZ. and may be found it here: https://atrad-audio.co.nz/turneraudi....au/index.html. The best educational material regard output tubes and loads ever found. You may take a look at load matching chapter. Exceptional educational material Mr. Patrick Turner done. Thanks to Atrad is still available for us. Many thanks !

https://atrad-audio.co.nz/turneraudi....au/index.html

[TomCarlos]

Thank you gentlemen....

It's been a while since I have studied "theory." And while much of this is still over my head, I am slowly working my way through it. So thanks for your thoughtful replies - JMF, SF, and CG.

I can see that the video and datasheets only take you so far. Sadly, when I print a datasheet for my quick reference binder, I have not included the load line pages. But I will. This is something I want to study further. The other thing the video does not mention is current capacity. So when you come across an unnkown Output Transformer, it also needs to be tested for current delivery capabilities.

The Patrick Turner site has some great info. Sorry to hear about his loss. I hope the site will remain up. I plan to review the PP and SE OPT Calculation pages !!

[Helmholtz]

Or use an online Loadline Calculator (bmamps or vtadiy). They also give useful additional data.
Just don't forget to use full load B+ with power tubes.