# Output transformer question

Discussion in 'Glowing Bottle Tube Amp Forum' started by radiocaster, Nov 5, 2017.

Aug 18, 2015
europe
I was looking at Champ transformers, most had a 7K ohms primary, but I also saw one which was supposed to be vintage correct and had a 17K primary instead. Why such a big difference?

Also, is there a way to calculate what primary an amp would need? I am aware of this and that trannie recommended for whatever set of tubes in the output, but is there some kind of mathematical formula?

2. ### clintjFriend of Leo's

Apr 4, 2015
Idaho
Ballpark formula is expected plate voltage divided by expected idle current.

Plate voltage can be estimated from:

HT winding voltage (i.e. a 330-0-330 would be 330) times either 1.1 (tube rectified) or 1.4 (solid state).

From there, you can use your tube data or bias calculator equations to estimate idle current. 70% for fixed, 90 for cathode bias, 100 for single end class A.

MitchMiami likes this.

Aug 18, 2015
europe
Well yeah, if you do a simple R=E/I calculation using the typical plate voltage and amperage from the manufacturers' charts, you get the most common output transformer primary resistances.

But here's what bugs me, besides that vintage Champ transformer:

If I calculate the value for a 12AU7 as output tube using the typical characteristics, I get about 25K ohms, which is the usual value for trannies for this type of amp.

But if I use the values for a 12AX7, I get over 200K ohms, and the only amp I know that uses a 12AX7 (half of one in single-ended mode) is the Blackheart Killer Ant, but it uses only a 50K primary.

Aug 18, 2015
europe
.

Aug 18, 2015
europe
Also, if say you have a single-ended amp with one tube that requires an 8K primary. If you make it parallel single-ended with the same type of output tube, would it then need a 4K primary?

Mar 24, 2015
Here and there
It would. My gloriously feisty hot rod champ/ax84 P1ex will go either single 6L6/EL34 or 2x6V6 without change thanks to that.

Aug 18, 2015
europe
O.k., but can anyone explain those two stock transformers with weird primaries that I have mentioned?

8. ### WyattTele-Afflicted

Nov 3, 2004
Primary impedance is not an absolute. Tube data sheets give all sorts of examples depending on plate voltage and desired output and THD. Designers after start with an end point in mind and design backwards; since it's audio, they aren't necessarily maxing out sheer performance. And sometimes its a matter of what parts they can source easily.

But I don't see how you came up with 200K for one triode of a 12AX7. Calculating with the maximum plate voltage, I only got 110K.

Assuming 250VDC, SE Class A, 1/2 of a 12AX7 (1.1W max diss.)...
• Using R=E/I, I get ~82K
• Using R=E^2/P, I get ~56K
Both these formulas are only for ballparking only, they ignore a lot of other variables. I can't find a schematic or any OT specs on a Killer Ant anyway.

Last edited: Nov 14, 2017
9. ### Commodore 64Friend of Leo'sAd Free Member

Age:
43
Mar 1, 2010
Kent, OH
Impedance varies with frequency, too.

Tweed Champ (5F1) is 17k

I'm guessing the 17k in 1958 is just what Leo was able to get in quantity at reasonable price. I would think the 5F1 OT would be stable with 6V6 at 2 ohms or 4ohms, but would hesitate to run 8 ohms (34k reflected).

Aug 18, 2015
europe
They're both at 4 ohms, there's one that also has a 5K ohm primary for 8 ohm. Look at the list below the transformer description:
http://www.classictone.net/40-18110.html

Mar 24, 2015
Here and there
17k into 4 ohms sounds almost like a reverb transformer! Match it to an 8 ohm tank giving 34k, sounds perfect for that.

12. ### T PriorPoster Extraordinaire

Mar 17, 2003
Charlotte NC
Correct, as stated above, don't overlook cost, and bulk purchase by Fender. The two OT's although different, at the end of the day either will work just fine and the user will notice no differences in output. 3 watts, 4 watts, 5 watts, it's all the same within relativity. But the bulk costs may not be the same.

I recently acquired an early Blues Deluxe, 1996 I believe, it's marketed as 40 watts. The schematic clearly states 38 ! I'm missing 2 watts. I'm gonna sue.

Last edited: Nov 17, 2017

Aug 18, 2015
europe
Another question: I was looking up Marshall JTM45 transformers and most replacements have an 8K ohm primary, although they run two 5881s or KT66s. The Marshalls that have two EL84 have just a tiny bit bigger primary, 8.3K or whatever.

Now I know the JTM45 doesn't put out a full 50 watts, still I would like someone to tell me how it is possible that such values stray so far from the norm when many people constantly nag about blowing the OT with the wrong value on the primary.

14. ### T PriorPoster Extraordinaire

Mar 17, 2003
Charlotte NC
another thought to throw in the fire, who is to say that the VINTAGE OT was the perfect one for the design, maybe back then Leo used what he could get his hands on, availability. It wasn't like he had Amazon.com to go searching for an OT. Many times we assume that because the VINTAGE amp had it, it was perfect and correct. Leo used what was available within reason. Today, we can search and search to the end of the internet until we find the EXACT specs we are looking for. Leo built amps basically to the RCA Tube Redbook with variations.

Mar 24, 2015
Here and there
Funny this thread came back into view. I think I read something (on the Tone Lizard website?) about using higher than expected OT impedances being common practice to limit current draw for battery operated radio sets so the battery won't run flat so fast. Maybe Fender got a batch of 6V6 single ended OTs, but just that they were originally intended for radios?

Last edited: Feb 20, 2018

Aug 18, 2015
europe
Yeah, but I haven't seen many people go on forums and say "do not use the original OT, you'll fry it if you crank your amp". Which also brings my next question: are you more likely to fry an OT if you crank your amp? And, would you be less likely to fry an amp with a weird primary resistance if it is of super low wattage (a 12A*7 in the output stage)?

17. ### Commodore 64Friend of Leo'sAd Free Member

Age:
43
Mar 1, 2010
Kent, OH
If you look at tube data sheets, you'll see they run at many different primary impedances. It affects how the tubes operate on performance curves. And things like plate and screen voltages change how the primary impedance interacts with the tube operating points. It's a complex system.

You gotta remember that the impedance of a transformer is just a multiple of the turns ratio...so a lot of times we say 4 or 8 or 16 ohms, and we think of mismatches related to that. But that's too simple. 4, 8 0r 16 doesn't matter. At all. It's just a multiplier. And you get scared hearing about FLYBACK voltages about running in the wrong tap. It's the reflected impedance that matters! I've got a 1.2k 100W output transformer. I run a 16 ohm speaker in the 4-ohm tap to get it in a safe operating place for a pair of 6550 tubes. 6550 see 4.8k, and that's just fine and safe. (I wouldn't have spec'd an OT like that if I was buying, but it was free.)

https://www.tubesandmore.com/products/transformer-fender-output-hot-rod-deluxe

Check that, it's a 4k transformer. You can hook an 8-ohm speaker into the 4-ohm tap and run at 8k for a pair of 6L6. It will be perfectly happy. The fact you are in the 4-ohm tap with an 8-ohm speaker doesn't mean a darn thing to the OT. It's the reflected impedance that matters.

Now, how much wattage can an OT handle? Look at the Pro jr or Blues Jr. OT. Tiny. It's made to saturate. Running a small OT is part of the tweed tone, too. On the other side of the spectrum,a 25W Edcor CXPP can push 30-40W no problem. The wattage rating of an OT is murky. It's set by the manufacturer. A dark art. You can't really quantify it unless you are an engineer, and even then, I think most designs are set on precedent (make, test, change until design fits needs). In general, more iron means more wattage. Why? It can dissipate more heat (watts) before the ambient temperature of the OT gets hot enough to melt the enamel of the copper wiring (or plastic bobbin). That's it. An OT is not complicated. It's copper, iron and paper/plastic bobbin. If you push more current through it than it can dissipate, heat will increase until damage occurs (just like a transistor or diode or resistor or computer chip).

I have some transformers at work from an old lathe. I took some of the paper apart to look at the windings of copper. They touch. Or consider a guitar pickup, like fishing line on a reel...wraps on top of each other. Why isn't there a short? --a coating of enamel on the wire. A very fragile coating of non-conductive enamel. When they talk about insulation melting in a transformer, it's not a pvc or teflon...it's enamel. You don't want to melt that.

jsnwhite619 and Nickfl like this.

Aug 18, 2015
europe
Some datasheets do indeed have load resistance at several different voltages and operating classes.

Many datasheets do not have this at all. Plate resistance is something else. More than a few datasheets do not even have a plate current reading @ the nominal 250V or whatever.

Otherwise I suppose you could get that resistance figure by dividing the plate current from the operating voltage, if I'm not mistaken.

I agree with you, but I don't think those transformers I posted about, respectively the one for the Champ and JTM45, are sold to replace them in those amps and use a different tap if you use the "standard" output tubes.

Here's one more dilemma. I found this pic of a Palmer Eins OT:

The amp has an 8 ohm and a 16 ohm tap. I would guess they use the 4 and 8 ohm connections to get 20K.

But that is not as relevant as that they encourage people to experiment with pretty much anything in the output: 12AU7, 12AT7, 12AX7. The 12AU7 and 12AX7 have vastly different resistances, and furthermore this is an amp made to be cranked.

19. ### Commodore 64Friend of Leo'sAd Free Member

Age:
43
Mar 1, 2010
Kent, OH
Perhaps I misunderstood your question initially. My point was that the "norm" is a very broad range, and that perhaps those values you referenced don't actually stray very far...

I was also trying to elucidate some of the factors, because in a lot of cases, the "nagging" you refer to comes from an incomplete understanding of how OTs actually work.