Fender 5E3 Output Transformer Winding Spec

[FredB]

Anybody know where I can get the original Fender 5E3 output transformer winding spec?

[Chuck H]

I've read that the original 5E3's used "off the shelf" 6.6k Pri Z OT's. If Bruce Collins reads this me may correct this. I think that the contemporary belief favors an 8k primary.

[FredB]

I've read that the original 5E3's used "off the shelf" 6.6k Pri Z OT's. If Bruce Collins reads this me may correct this. I think that the contemporary belief favors an 8k primary.

I was wondering about the number of turna and the interleave of the windings.

The case of 6k6 vs 8K primary impedance is an interesting one. The peeps from the other form I haunt seem to think 8K is the proper impedance for a pair of EL-84 tubes.

My modeling of the the 5E3 power section, which could also apply to a Marshall 18 Watt, seems to suggest that 8K is too much for those tubes. As a result the damping in the power section is higher than 0.5, and so there is a little positive bump in the high end of its frequency response.

I have yet to sort out what primary impedance I want to use, although that is a lesser issue. The information that is difficult to find without having access to Fenders original winding specs the what I mentioned the turns and the interleaving.

[Chuck H]

I've read that the original 5E3 OT's were straight wound. No interleaves. As in wind half the turns, add a center tap, wind the other half of the turns. This has been evidenced many times in threads here as well. They are notorious for having dissimilar ohms from each end to CT. And FWIW I do think they were spec'd such that a clipping 5E3 does stress the OT. Sorry I don't have more specific info.

[soundmasterg]

I do have an original 1956 5E3, but I don't feel like taking it apart to measure anything. THat said, I think Chuck is right in that there was no interleaving.

Greg

[FredB]

I do have an original 1956 5E3, but I don't feel like taking it apart to measure anything. THat said, I think Chuck is right in that there was no interleaving.

Greg

As I understand it without any interleaving the high end response would would be a bit diminished. This makes sense for the 5E3 for a couple of reasons. First they are know for having a generally warm tone, rather than extremely defined highs, and my modeling of the PP EL-84 power sections suggests that such a power amp inheriently has a very good high frequency response to begin with. A transformer designer almost has to go to extra effort to keep it from running flat up to 50-100kHz by using a relatively massive primary inductance.

Well, that at least will be easy enough to try. I would like to make the halves of the primary as balanced as possible though. I doubt ringing would add any pleasantness to the tone.

I bought a Hammond unit before I set myself to the task of find out how to roll my own. I can measure that one and compare with my DIY one. I think I put a terminal block in the amp for quick OT swaps.

[R.G.]

Welcome to the can of worms that transformer design resembles.

As I understand it without any interleaving the high end response would would be a bit diminished.

Yes, in a Class A amp. It gets rather more complicated in a Class AB amp where one half-primary turns off while the other one is on. The issue is the leakage inductances, and what happens when the current in the leakage inductance is turned off in switching from output tube to output tube. In some circumstances, this can make something that looks and sounds like crossover distortion, but isn't.

... in addition to worse high frequency response.

... my modeling of the PP EL-84 power sections suggests that such a power amp inheriently has a very good high frequency response to begin with.

Be very suspicious of simulation/modelling. Treat it as a close, beloved friend that sometimes lies to you for no reason you can see.

A transformer designer almost has to go to extra effort to keep it from running flat up to 50-100kHz by using a relatively massive primary inductance.

Mmmm, not really. All transformers of a size useful for audio power output have significant leakage inductance, and that is what limits high frequency response, almost to the exclusion of other things. Not totally, but nearly. The primary inductance has almost nothing to do with high frequency respones. Primary inductance is the chief governor of low frequency response though. In the Golden Age of tube audio, the ratio of primary inductance to leakage inductance was used as a key indicator of the transformer's audio quality - the bigger the better.

I would like to make the halves of the primary as balanced as possible though. I doubt ringing would add any pleasantness to the tone.

A 3-2 interleave is not tough to do and would be a vast improvement over the generations of MI transformers with no interleave. Leakage inductance goes down as the square of the number of interleave interfaces. If you want the sound to be warm, use a good, low leakage inductance OT and stick a "warming" high-cut capacitor in the preamp somewhere to cut it back. Highs are easy to lose, hard to get in the first place.

[pdf64]

'The case of 6k6 vs 8K primary impedance is an interesting one'
The 6V6 and EL84 data sheets I've seen indicate that 8k is the suggested primary impedance for a p-p pair, with the type of conditions used in a guitar amp, or at least close to.
Don't recall seeing 6k6, at least for regular .
Regards the 5E3, weren't most 50s amps under Western Electric / RCA IP, hence Fender and Gibson using closely similar circuits.
So the transformer spec may be available available from their publications.
Pete.

[FredB]

RG,

Yes, I've just studied up on the equations for determining the low and high frequency corners. Now that you mentioned it, I was presuming there was more of a connection between the primary and leakage inductance, because I had been previously modeling this with mutual inductance and coupling coefficients, which resulted in the high frequency corner moving in direct proportion with changes in the primary inductance.

Pete,

My simplified modeling suggests that an 8K:8 ratio is too high for PP EL84 tubes, and causes a slight rise in the high end frequency response before it rolls off. Whether this happen in an actual 5E3 transformer I have yet to find out. I have a Hammond unit that I could test at a low power level easily enough to see if it's there.

The sight rise might be counter-acted by other paracitics, and in that case would just extend a flat response a bit further. I guess I'm just going to have to test that.

[Chuck H]

Is all this speculation and modeling being done on account of a guitar amp design? I'll guess 'yes' since the 5e3 circuit is the basis for the discussion. So are you doing this as an exercise? Any rise in HF prior to roll off won't always sound bad, or "harsh" or "tinny" or whatever. Remember that there are still many factors like speaker impedance and damping @ frequency and miller capacitance along with others. Most guitar amps were built to "work" without such fastidious attention to these details. As a result there are many happy accidents that resulted in the coveted vintage designs. That rise in HF shown in your model may be just the thing, in combination with other factors, that gives YOUR 5e3 the right amount "clarity", "note separation", whatever, while still sounding "warm". Since guitar amps are used as signal processors, and not strict reference amplifiers, it's often a combination of less than ideal parameters that gives any given amp a particular tonal character. Or, put another way, it really doesn't matter that much. As long as your design operates safely that is. With this in mind you could ignore all this trouble and just clone the part to the best your research allows. Then tweak the resulting amp to sound the way you want it to. That's what everyone else does. Not that you have to be like everyone else, but the alternatives seem to reach a point of diminishing returns. Suppose you idealize this OT to the point where the resulting tone is bland and plain sounding? Diminishing returns.

And, regardless of your model, I know a lot of people would argue in favor of the generally accepted 8K inpedance for a pair of el84's

[J M Fahey]

Do not forget that Fender did not have or define the transformer winding specs, they just ordered samples from a good reliable supplier (think Chicago, Stancor, UTC and such); Fender just said "I like it" (or not).
Same with Marshall and Drake and Dagnall, or even more the original surplus types they bought off the shelf .
I'm sure your Hammond is *very* good.
Maybe even "too good" ... but as RG says, I'd rather use it anyway and if necessary, "warm" the sound in an earlier stage.

[FredB]

Yes, my current project is a 5E3 influenced amp. This evening I tested the Hammond transformer. I got figures for primary leakage 48mH, primary DCR 312, Secondary leakage 53uH, Secondary Inductance 22mH, secondary DCR 0.8 ohms and inter-winding capacitance 223p.

I plugged all that in to RG's transformer model, using a 8000:8 impedance ration to calculate a primary inductance. The primary inductance is greater then my LC meter will measure. I suppose I could to some voltage measurement to verify my estimated primary inductance. Anyway, using those values gave a result without a rise in the frequency response at the high end. I guess that was an artifact of the simple mutual inductance/coupling coefficient transformer model.

I know all this is extraneous to actually building a clone by the numbers. I got to these lengths to understand to my own satisfaction what going on in the transformer. My goal rather than to just clone any particular amp it to thoroughly understand what makes them sound the way the do, and then perhaps use that knowledge to improve on the iconic classics.

[Enzo]

That is a laudable goal, but consider that there are literally thousands of other factors that go into making an amp sound like it does. None of them act in isolation.

[FredB]

That is a laudable goal, but consider that there are literally thousands of other factors that go into making an amp sound like it does. None of them act in isolation.

Well, I wonder about thousands, although there is certainly a multiplicity of interrelated factors involved from the pickups to the speaker. Say, are you familiar with and able to articulate the tonal differences between Cotton, PVC, and Teflon insulated wire? That is one of the finer things I have yet to study.

***

Thinking a little bit more about it, perhaps I erred or simply spoke foolishly when I said "improve" upon the iconic classics. for such a statement can be politically offensive to some, because by achieving iconic status particualar amps become evolutionarily fixed and unimprovable. Maybe it would be better to say something like, become knowledgeable enough to build iconically influenced amps with tone good enough that people will buy them.

[loudthud]

Thinking a little bit more about it, perhaps I erred or simply spoke foolishly when I said "improve" upon the iconic classics. for such a statement can be politically offensive to some, because by achieving iconic status particualar amps become evolutionarily fixed and unimprovable. Maybe it would be better to say something like, become knowledgeable enough to build iconically influenced amps with tone good enough that people will buy them.

But what about the happy accidents? Like Neal Youngs' Deluxe. Is there really someting special about that amp or did someone just stick a 12AX7 in the preamp before they dumped it at a pawn shop? Neal claims it's special.