What Makes the Epiphone Constellation a Bass Amp?

[Kalamazoo 2]

What Makes the Epiphone Constellation a Bass Amp?

More generally, what makes any bass amp a bass amp relative to a guitar amp, maybe besides having to pump more watts to get the same loudness?

More specifically, when I plug my Strat into the EA-17, where do all the highs go? Fender and Marshall bass amps work well with guitars. Why doesn’t my Epiphone?

Before any more talk from me, here’s a fuzzy jpg schematic, a much better pdf is linked at end of this post:



Epiphone produced the EA-71 Constellation V bass amp from 1965 to 1967. Mine is a combo that maps to 1966. It is a tube-driven push-pull with a solid-state rectifier, 6EU7 pre-amp, and a 6C7 cathodyne phase inverter feeding another 6EU7 to drive a pair of fixed-biased 6L6 power tubes to a claimed 40 watts.

The closed-back cabinet with a tuned port in the baffle houses a 15” diameter 16 ohm speaker. The thing is heavy. And loud. And pretty clean until turned way up and pushed hard, which I don’t do much. It does sound cool then, but it’s hard to say whether my old eardrums or that old paper cone will blow first.

A triple-ganged bass pot—quite unfamiliar to me—follows the first gain stage, then feeds a volume pot, the second triode of the 6EU7 and a treble control that also looks unfamiliar.

The inverter and power section look pretty generic to me.

Ampbooks.com’s Cathode Bypass Capacitor Calculator, with a 12AX7 in lieu of the installed, specified 6EU7, and C and R values from the accurate schematic, yields a curve that’s nearly flat from 20Hz (two octaves below low E on a guitar, right?) on up.

CK (μF)  20
RG (kΩ)  150
RL (kΩ)  100 [a guess, but doesn’t seem to move curve]
RK (kΩ)  1.5

gain at 82Hz (low E)  48.58 (33.73 dB)
gain at 10kHz 48.98 (33.80 dB)

I had a 10 and a 5 μF handy, so of course swapped them in and thereby moved the curve.. but it was still flat above 80Hz. That didn’t really add treble, though it did change the sound some—more on that later.

Next on the signal path is the mysterious (to me) bass part of the tone stack, a completely unbypassed gain stage, and the equally enigmatic treble control. I have no idea what is going here. My instinct is to pitch some caps and resistors in there and disconnect the controls—as I have done on the simple Kalamazoo—but I don’t know enough to come up with some values.

I can say that twisting the knobs while playing the Strat is not like my experiences with guitar amps. Running the bass knob at 3/10 minimizes the muddiness found at other settings. Running the treble knob at 6/10 delivers clearest high notes. In other words, the tone stack changes the timbre more than it adds high frequencies on top. The bass can make it thumpier or boomier, but doesn’t make the low notes louder. And in all cases, turning up the volume makes the sound prettier—richer, more complex?—and a whole lot louder.

I’m not likely to use this amp much to make music—unless I get a bass or baritone guitar—but it does seem like an opportunity to learn some things. It’s probably as hard to sell as it is to lift. It was dumped on me a long time ago. I got it working and learned a ton doing so. Now I’m curious.

Y’all have any ideas?

[nuke]

There are a few interesting things here. First the schematic is drawn kinda funky and a litle harder to read.

First stage, V1a. Negative feedback applied through C2.

The triple-ganged bass control is going to some very heavy bass shaping. I'd really have to sit down to think that one through.

Volume control after the bass control complex, feeds V1b grid. Nothing much interesting here, just a triode stage with no cathode bypass. Lower gain.

Treble control stack: Yeah, this one is weird and I think it is going to really limit the bandwidth down a good bit.

V2 is an odd configuration that's doing phase inversion.

V3 amplifies the phases from V2 to drive the 6L6 grids.

Seems like they could have just dropped V2 and made V3 a LTP phase inverter. I dunno what they were thinking.

Output stage, mostly standard, V4 and V5 mostly normal fixed bias push-pull.
But C14 and C15 are kinda weird and would probably further limit treble. And take a lot of abuse.

No negative feedback from the speaker, which is also a bit weird for a bass amp.

The 15" speaker is probably selected for bass, with not much high-bandwidth from the cone.

That's my back of the napkin analysis.

Why is it bass amp?
Mostly the tone control sections and I'm not sure of the total effect C14 and C15.

[martin manning]

The triple-ganged bass control is going to some very heavy bass shaping. I'd really have to sit down to think that one through.

This is an adjustable twin-t band stop filter, mid range is 20 Hz. It sets BW on the low end.

Treble control stack: Yeah, this one is weird and I think it is going to really limit the bandwidth down a good bit.

Notch filter, basically a variable scoop centered on 1 kHz. It sets BW on the high end.

V2 is an odd configuration that's doing phase inversion.

Pretty standard cathodyne inverter.

V3 amplifies the phases from V2 to drive the 6L6 grids.

Diff amp with opposite phases fed into inverting and non-inverting inputs, maybe higher gain than a LTP?

[nuke]

This is an adjustable twin-t band stop filter, mid range is 20 Hz. It sets BW on the low end.

I recognized the t-configuration, but didn't bother to work out any details. I'll take your word for it.

Treble control stack: Yeah, this one is weird and I think it is going to really limit the bandwidth down a good bit.

V2 is an odd configuration that's doing phase inversion.

V3 amplifies the phases from V2 to drive the 6L6 grids.

Pretty standard cathodyne inverter.

Diff amp with opposite phases fed into inverting and non-inverting inputs, maybe higher gain than a LTP?

Yeah, I meant these two sentences to be taken together: It seems kinda weird to follow a cathodyne with a pair of gain stages like that. It seems reasonable just to delete V2 (a 6C4) , add the few components required to turn V3a and V3b (6EU7) into a long tail pair. That's a very proven approach to drive the 6L6 pair.

If they needed more gain than the LTP, just adding a cathode bypass cap at V1b would provide the gain, unless they thought it was clipping in some undesirable way.

C14 and C15 are kinda interesting too. Any thoughts on the why?

But back to the OP's question, what makes this a bass amp? The tone controls mostly, along wtih the speaker choice and ported cabinet.

[martin manning]

C14 and C15 are kinda interesting too. Any thoughts on the why?

Has to be aimed at removing HF content from the plate circuit, doesn't it? Maybe there was some ringing or something going on?

But back to the OP's question, what makes this a bass amp? The tone controls mostly, along wtih the speaker choice and ported cabinet.

Yea, in general extended low-end bandwidth, and right from the input you have 100k grid stoppers cutting highs.

[Phil_S]

It's a crappy design. This amp shares it's topology with other Gibson/Epi amps of the period. I suspect they were chasing clean tone, but the pubic rejected their idea of it. We know this from numbers manufactured and that they never were very sought after. Because the amp sounds like poo, I suggest there is nothing unholy about modifying these amps. Go ahead and swap tone shaping components to make sound that is more pleasing to your ears. While you are at it, swap the factory speaker for something more to your liking. Don't be afraid to rewire the 6EU7 tube sockets for 12AX7's. The two tubes are identical with different pin outs. If you can't find a 6C4, drill the chassis a little bit to take a noval socket and use a 12AU7 in its place. You've got a viable chassis and cab to work with. Why not make it into something you like?

Please don't tell me I'm a heretic who should be burned at the stake. I already know that I don't always have a proper amount of respect for historical preservation. If you don't mod this amp, it will continue to gather dust, and what good does that do?

[nuke]

The 6EU7 is an arguably better preamp tube than the 12AX7A/7025. It is basically a re-pinned AX7, with 6v filament leads only and pins re-arranged to improve noise figures.

There are current production 6EU7 and used/NOS 6EU7 are available for reasonable money. No need to re-wire the sockets unless you just want to.

[Kalamazoo 2]

Wow. This is good stuff, a great feast that has taken me awhile to digest, and I'm still at it. I want to explore some of what you've mentioned, but it will take me a little longer to gather and organize my thoughts. But right now, I want to thank each of you. I appreciate your response very much.

Y'all pointed out some things that inspired me to look at a few schematics with a different eye, comparing gain stages and component values, and I went a little deeper into cathode resistor bypass caps and what they do. So I went back to my Kalamazoo--you remember that thing, right?--and modeled a few values, and dug out a 10uF cap and bypassed the originally unbypassed resistor of the first gain stage for a whopping increase in gain. Boy howdy, that makes a difference! My Strat no longer sounds skinny and weak, and googly moogly, this rig is loud. Not sure I like it.. wait, yes I do. It is for sure a big step down the path to what I'm seeking.

So, thank you.

[nuke]

I once had a Traynor YBA-4, kinda the same idea, 50-watt bass combo with a ported 15-inch speaker.

Basically a YBA-1 and speaker combo.

I did use it a few years as a guitar amp. Stuck a Weber 15 in it with an aluminum dome cap, and tweeked the amp a little and it was pretty good, but man, it was heavy. I wound up converting it back to bass and sold it.

[Kalamazoo 2]

Sorry about the inline schematic. The PDF attachment is a lot better, but I couldn’t seem to upload it as an inline graphic, and the jpg conversion obviously did not go well.

Local NFB via C2. So that’s what that is. Why would such a low-gain stage need NFB?

Early on, I assumed the speaker was junk, but a top-boost pedal yielded quite a bit of highs. Later, I played a rented Traynor YGL-1 through the EA-71 speaker, and it certainly wasn’t bright, but yielded a much more balanced frequency response. Not bad. Not great, but clearly not entirely to blame for the way the Epiphone sounds. A friend just got a pretty new amp with a 16 ohm 12” speaker, so I’ll hook it up and get an even better idea of what the Epi speaker is contributing. Or subtracting.

Gibson vs Everybody Else: From what I’ve read, by 1966 Gibson was pretty out of touch with the market. But Fender was still using cathodyne PI in the Princeton. Again, from what I read, it’s not an inherently bad circuit but the LTP is cheaper and easier. As to Gibson’s continued pursuit of clean.. well, yeah. Of course, they had moved to humbuckers, and those will push a preamp to coloration. Wouldn’t want to piss off ol’ Les now, would we? 100k input resistors?

C2, C14, C15, filter choke on B+, all kinda seem like afterthoughts, or solutions to problems created by lazy design. This amp is grounded all over the place, hums some. Heater wiring not impressive. I’m not sure I’d call it a “crappy design” but it’s not great, and perhaps the implementation was not fully thought out. The build quality itself is not bad. Mostly good soldering and component placement. Mostly.

Tubes. Original tubes. None microphonic. Amazing.

I’d like to understand the tone stack, if only to figure out what kind of component values I should use to replace them with. And no, Phil, nothing heretical about modding this thing. The chassis and cab are sound, but the face is rusty. I replaced the original 4-position power switch as it was corroding, noisy, and looked like an ignition source. It’s a clever thing, but like so much about this amp, just not great. It weighs a ton and has no value. One on Reverb, been there like three years.

Before I start a great experiment, I will plug a bass in just to spin the knobs and listen. I expect it to sound pretty thumpy, like an acoustic bass stuffed with rags only louder, but we’ll find out.

In the fullness of time, I’d like to mess with it some, just to learn stuff. It’s easy to work on—roomy chassis that’s simple to remove. It’s really not far from something like a great honking Princeton. It doesn’t seem far from being a loud clean amp. The trick would be to lose its dullness.

[martin manning]

I’d like to understand the tone stack, if only to figure out what kind of component values I should use to replace them with.

The first thing I would try is paralleling a 33k across the 100k input grid resistor R1 to get more HF content through the first stage. Use Input 1 and put a 1/4" plug in the other input to lift the other 100k's ground so it doesn't divide the guitar signal. To get the full effect you can lift the ground on the treble control.

Then you could build a Fender BF tone stack in a box and alligator clip it in place of the triple gang bass control thing.
You would disconnect C3 from V1a pin 7, and the wire going to the top of the volume pot, and clip the new stack in its place.

[Kalamazoo 2]

The first thing I would try is paralleling a 33k across the 100k input grid resistor R1 to get more HF content through the first stage. Use Input 1 and put a 1/4" plug in the other input to lift the other 100k's ground so it doesn't divide the guitar signal. To get the full effect you can lift the ground on the treble control.

This is just what I hoped to hear, corroboration of reducing resistance on the input, and a method to bypass the treble control. Even better, you've shifted my way of looking at the tone controls. Fantastic. Thank you.

Then you could build a Fender BF tone stack in a box and alligator clip it in place of the triple gang bass control thing.
You would disconnect C3 from V1a pin 7, and the wire going to the top of the volume pot, and clip the new stack in its place.

I thought there might be a way to hang something in there, but hadn't gotten to "build it in a box"--yeah, that makes it easy!

Even easier, can I not just lift the ground of the bass control? Doesn't it work like the treble, just dumping different frequencies to ground? I don't mind really awful tone for awhile as I learn things, but I really want to keep the smoke in.

Oh this is going to be a fun way to hide out in the air-conditioned shop/garage/studio/laboratory while the heat index goes over 103F this afternoon.

[martin manning]

I thought there might be a way to hang something in there, but hadn't gotten to "build it in a box"--yeah, that makes it easy!

Even easier, can I not just lift the ground of the bass control? Doesn't it work like the treble, just dumping different frequencies to ground? I don't mind really awful tone for awhile as I learn things, but I really want to keep the smoke in.

Yes, these are passive filters, and no problem doing that, but it will not have any of the shaping of a TMB stack with mid scoop.

[Kalamazoo 2]

Replaced the input R with a 68k because I had one.

That made a very impressive difference. Almost Fender-like. Almost. Certainly not scooped, but getting right to it. Spinning knobs on amp or guitar now has a much greater effect than before. Beginning to overdrive the preamp. Smells like success. Sometimes the low-mids sound quite piano-like, which is kinda weird. Amp definitely more sensitive to the resonance of this Strat.

Studied the schematic, traced some wires, all set to put the chassis back on the bench and disable the treble control.

[Kalamazoo 2]

Lifted the ground on the treble control. MADE IT VERY VERY LOUD.
Overall, not nearly as tonally interesting as with treble control active. Shoot, overall it was worse. Crummy. It made it hard to play, at any volume or combination of settings. Just wrong.

Very gainy sound. About the time it started yielding some touch, I had to quit because it was so loud.
Guitar volume took effect before 3. With amp at 5, guitar at 5 was painful. Amp at 3, guitar at 5, it all was predictably dull.
Turning up guitar volume helped a bit.
Turning up amp helped more.
Turning up amp bass control made everything a lot louder but did make lows sound better. Sustaining, creamy up to a touch rough at the edges.
Guitar tone controls had little effect on highs, but could bring out the sweetness of the lows. Bass strings really sounded pretty good, but overwhelmed chords.

As I have time, I'll start piling up parts to build the TMB-in-a-box. I think I already have a box .