Super Reverb '65 RI - Changing Output Transformer to Match Speaker Load

[Tom Phillips]

I’m in agreement with Chuck & The Dude.

First let’s all remember that there are a lot of design rules that are bent or outright broken by tube type guitar amplifiers. There are also people who tweak their amps purposely to “out of spec”, “over spec” or "mismatched" conditions in order to achieve their desired tone.

…For the past year now, I've been running on 2 speakers only therefore, at a 4 ohm load. The amp is not as loud yet breaks up earlier and - overall; much sweeter … Will I be losing the pleasant audio gains I've achieved by the mismatch and essentially be putting it back to 'stock' - meaning losing my early sweetspot?

Given those statements it appears that you are very happy with the sound of the amp and it is not eating output tubes or otherwise showing signs of stress. Given the situation, if you change the OT, then I think the probability is higher that you will be less satisfied with the tone than the probability that you will unlock some hidden tone.

…Coming from a lifetime of hating to mismatch anything, though no apparent harm has been done to the amp or speakers….

If the goal was to drive a purely resistive load with maximum power then exact matching would be very important. However, the impedance of a real speaker load varies quite a bit across the frequency band generated by a guitar. Therefore, a 4 Ohm rated speaker load used in place of a 2 Ohm rated speaker load is no big deal for a Super Reverb type amp. Since you goal is to play a guitar amp that gives you the tone you like and you already have that I say to ditch the idea of changing the OT and enjoy the amp as is.

Cheers,
Tom

[pdf64]

Yes, if it was a JTM45 (60s or RI) it would be running 6L6 or similar beam tetrode (5881 / KT66) with an 8k load (maybe 6k6 with an RS OT).
So a similar scenario to a SR with 4 ohm load.
Though I think that the JTM45 RI runs with a lower HT of ~400V, which reduces the stress on stuff.

[g1]

Further to the points above, most vintage Fender's with an "ext. spkr" jack are running mismatched as soon as you add an ext. cabinet.
Fender didn't see any problem with it, neither should you.
(I realize the mismatch is in the opposite direction, but don't consider it significant)

[Richard]

This guy found and old SR that was hooked up to a 16 ohm speaker.

'66 Fender Super Reverb Restoration

[loudthud]

In the 70's I did sound for a bar band with two guitar players. They each used a blackface Bassman into a 16 ohm 4x12 Marshall cabinet. The volume worked great in the bars they were playing. Nice overdrive tone.

If you match the impedance into the two Super speakers, they will likely blow due to the increased power.

[Chuck H]

In the 70's I did sound for a bar band with two guitar players. They each used a blackface Bassman into a 16 ohm 4x12 Marshall cabinet. The volume worked great in the bars they were playing. Nice overdrive tone.

If you match the impedance into the two Super speakers, they will likely blow due to the increased power.

That's a good observation! Maybe not the actual watts, since they'd be right on the edge of safe. But IIRC the damping will be lower with the proper match and peaks will be more pronounced. The tremendous dynamics that tubes are capable of in guitar amp circuits would be punching on those speakers like Mike Tyson.

[Richard]

How much of the improvement over stock is due to the impedance mismatch and how much is due to working the speakers harder?

[jbefumo]

Sorry for jumping into this old thread, but I have a very similar 'problem' -- amp sounds great as it is, so I just have to mess with it ... maybe ...

This started life as a beat up old "blackface" super reverb that I could plainly see was Silverface, but was so cheap on ebay that I bought it unheard, figuring on just using it for parts, chassis, etc. When I got it I discovered that (a) it was a big, nasty, ultralinear model in the oversized Twin-reverb-width chassis, and (b) not only had it been 'blackfaced', but the job had clearly been done by someone who really knew their stuff: It had been properly retro-converted, including a complete complement of Mercury Magnetics transformers (power, output, choke, and reverb), a GZ34 rectifier, properly modified bias adjustment, and some of the neatest, boutique-approved wiring I'd ever seen. My initial inclination was to move it all into a narrower chassis I had on hand, and work up a blackface face plate, but after seeing that wiring, I decided not to mess with it at all. I did, however, stumble upon a brand new Mojotone Twin Reverb cabinet at an attractive price, so I grabbed that, and the Super Reverb chassis did indeed fit it perfectly.

I replaced the 6L6s with a pair of Gold Lion (reissue) KT66, which I'd done on some Fenders in the distant past with good results (after contacting Mercury Magnetics to ensure their transformer was up to the increased filament current)

My cab now contains a pair of vintage Altec Lansing alnicos (418-8C), with a claimed frequency response of 60-8000 Hz. I've used these, and the 15" version, in many amps in the past, and think they're the best speakers ever made -- they have a crystalline sparkle without the tendency toward harshness that characterize JBLs of the same era. There's just nothing quite like them available any more (other than, perhaps, the ones being reproduced by Great Plains Audio, which I have not heard.)

The end result is, for my tastes, the best sounding clean amp tone I've ever heard, so I'm reluctant to mess with it further, other than to ensure that it's perfectly adjusted.

I have tried it with a pair of 4-ohm Celestion G12T-100s, and while the load is then correct, I think those speakers, in that amp, sound like crap.

I have also run it, briefly, through an Altec Lansing 15" (418) [45-8000Hz], and THAT is tonal nirvana, but then I'm pushing 8-ohms, which definitely makes me uncomfortable.
All of the amp gurus to whom I've spoken unequivocally recommend changing the transformer. If I do that I'll probably go with Mercury just because that's what's in there now, and I'd like to maintain the entire complement. I'm also pretty sure I that 15" would fit in there with a new baffle board, making it, essentially, a Vibroverb, I suppose.

Anyway, yes, there is a question here:

Someone above observed that the JTM-45 ran the KT66s into higher impedance, but at a lower plate voltage, would swapping the GZ34 for a 5U4GB rectifier tube ameliorate the situation?

[bob p]

Wow. Necrothreading.

I'm a bit surprised that nobody brought up the idea of choosing an OT that allows you to maximize the use of your secondary windings.

I'm sort of going through this same dilemma now, as I have a new (to me) vintage SF 4x10 Fender amp that's about to go through some changes. I'm in agreement with Chuck's idea (seems to be popular among the membership here) that the best thing to do is not to change the OT if you're already getting the tone that you're after with a mismatch. I was about to make the Bassman head / Marshall 4x12 comment, but LT beat me to it. Yes, Fenders tolerate a mismatch well in either direction, but IME they tend to sound best when they're running into too high of a mismatched load rather than in the opposite direction. I can't say that I've ever liked running Fenders mismatched into a low Z load. To me the high Z mismatch always sounds better.

I think that the OP has probably already taken action on his Super, so this advice may be coming in late, but if I had a SR that sounded good with a high-mismatched 4R load, then changing out the transformer is the last thing that I'd consider. (If it ain't broke then don't fix it!) I might consider putting a switch in the speaker cabinet, so that I could choose the load mismatch at the speaker level without having to change anything on the electronics side. This accomplishes several things --

First, it allows you to transition from the matched to the mismatched condition, getting all of the tonal options out of the amp without having t change the electronics. All that you need are robust speakers. From a practical standpoint, only 2 of them need to have higher power handling capabilities.

Second, this kind of switching allows you to control breakup point and the speaker surface area at the same time. We all know that acoustic efficiency is the greatest point of power loss in a guitar amp, and this kind of switching allows you to control both the amp's break up point and acoustic coupling efficiency at the same time. It's a great way to run one particular amp to make it suitable for both small AND large venues, rather than building a collection of special purpose amps. I thin the idea of a SR running mismatched with either 2 or 4 speakers is a particularly good solution to this problem.

Third, switching at the speaker level rather than at the transformer taps allows you to get your matched/mismatched selection while still using the complete transformer winding. Switching between 8/4/2 doesn't do that, and IME the sound suffers if you don't use all of the secondary windings that are available, especially if you're driving the amp hard.


In my case, the amp has 4x32 ohm speakers that are paralleled to yield an 8R load into an 8R OT. I have to decide whether to keep the load the same and doing an intentional high-side mismatch at the speaker level (the stock speakers may not handle it), or changing the OT to go with a 16/8/4 or an 8/4/2 setup. If I do decide to change the OT, which way I go will largely be determined by the speaker compliment I end up with, and my preference to utilize as much of the secondary windings as possible when jumping back and forth between matched and high-mismatch.

[Chuck H]

The 5u4 will plug right in, but it'll draw another amp of current (30% more than the gz34) from the 5V winding. Probably ok. Not sure though. You should expect about another 30V drop.

[J M Fahey]

As also noted above, for *best output* 6L6 optimum impedance is around 5000 ohms, I personally use 5500 on my handwound transformers for the very practical reason (specially here where tubes are relatively scarce and expensive and only a few brands to pick from) that any 6L6, any brand or quality, or even good but somewhat worn ones are very happy with that, and deliver solid 40W RMS.
Tubes are "current limited" devices, so going high is easier for them than going down in impedance.

2) going high lets them cope better with speaker impedance changes, and sound smoother, better mids; going low makes them lose mids (where speaker impedance falls) so they sound grittier and scooped.

Which is best?
Absolutely personal decison, based on hearing.

[robrob]

If you want to be a bit more exacting, a change of value of the feedback resistor would be in order.

The OP can tap the NFB off the 2 ohm tap and keep the NFB resistor and voltage level unmolested.

[pdf64]

As also noted above, for *best output* 6L6 optimum impedance is around 5000 ohms...going high lets them cope better with speaker impedance changes...

It's tricky to be certain, but I get the impression that up till the late 50s, Fender tended to use a ~6k OT for their 6L6 based amps.
Any thoughts on why Fender moved to a seemingly rather low ~4k2 from late 50s to late 70s? Especially as they facilitated an extremely low load impedance by providing a parallel ext speaker output.

My hunch is that it may have been a conscious decision based on speaker impedance being above its nominal value for much of the applicable frequency range; Hence an apparently low primary impedance actually worked out to be a reasonable compromise across the range.

Alternatively, perhaps they thought they may be able to push the beefier plates that were then available to get more output power by reducing the load impedance.

Or perhaps just because (as suggested) they preferred the sound!

[J M Fahey]

Low plate load impedance means current limited tubes can not fully drive a real world speaker, so sound changes, it loses mids , think from 250Hz to 400 Hz, while it can easily cope at resonance (so it has a bump at around 100Hz) and a "built in Presence/trebleboost" from, say, 1500to 6000Hz.
It actually keeps rising above that, but speaker response falls like a brick.

The effect is enhanced if you add an extra cabinet in parallel, although I guess few, if any at all, used extension speakers way back then.

So yes, I guess Leo started designing "by the book" but then experimented, I bet Freddy Tavares or some other "factory worker/Musician" liked it better and said: "boss, that one is best, here, listen to this" .

FWIW the Ultimate Attenuator, basically a load resistor driving an SS back end, uses a single 30 ohm power resistor for any amp , no change for 4, 8 or 16 ohm ones.
This improved version has 2 x 15 ohms resistors in series ... and a PCB; the earlier ones just a single 30 ohm one and wired PTP.

The actual circuit is VERY simple, the raw speaker signal directly drives (through a wire wound pot) a couple TO247 devices, most probably complementary MosFets or even Bipolars (including Darlingtons), no driver transistors needed because the tube amp can provide any current needed, if needed at all.

Most users just use it , but a few tinkerers have opened them and worried about that, designer´s answer is always: "just use it as is, you won´t burn your amp".
So far they are holding fine.

[bob p]

Any thoughts on why Fender moved to a seemingly rather low ~4k2 from late 50s to late 70s?

I think the simple explanation is that Leo was after low distortion.