I found this 'stereo' 50s 'Hi-Fi' valve amp in a skip down my street and thought it'd make a great low watt guitar amp conversion.
It has ECL83 "TV Tubes" which are a triode preamp and a 2.5 watt power pentode in the same envelope, configured as a 5W push-pull pair with the triodes as a paraphase inverter.
My dilemma is, should I stick with the paraphase or convert it to a Long-tail-pair (differential)?
My understanding is that paraphase inverters are considered older 50s technology and the Long-tail-pair is a 60s "improvement". Bearing in mind that, at the time "improvement" meant cleaner, more Hi-Fi, less tube character and it's this tube character that is exactly what we want in a guitar amp.
So, has anybody had any experience of the sonic differences between these two phase inverter configurations?
Also, the more usual twin-triode (same bottle) long-tail-pair is likely to be reasonably inherently more balanced, where as my separate triodes are not. So, has anybody also had any experience of the sonic effect of an un-balanced phase inverter? Might it be better for guitar? More asymmetrical distortion (even ordered harmonics)?
While I'm at it, anybody know what C6 - R9 is for?
Would greatly appreciate some guidance on this.
It has ECL83 "TV Tubes" which are a triode preamp and a 2.5 watt power pentode in the same envelope, configured as a 5W push-pull pair with the triodes as a paraphase inverter.
My dilemma is, should I stick with the paraphase or convert it to a Long-tail-pair (differential)?
My understanding is that paraphase inverters are considered older 50s technology and the Long-tail-pair is a 60s "improvement". Bearing in mind that, at the time "improvement" meant cleaner, more Hi-Fi, less tube character and it's this tube character that is exactly what we want in a guitar amp.
So, has anybody had any experience of the sonic differences between these two phase inverter configurations?
Also, the more usual twin-triode (same bottle) long-tail-pair is likely to be reasonably inherently more balanced, where as my separate triodes are not. So, has anybody also had any experience of the sonic effect of an un-balanced phase inverter? Might it be better for guitar? More asymmetrical distortion (even ordered harmonics)?
While I'm at it, anybody know what C6 - R9 is for?
Would greatly appreciate some guidance on this.
Attachments
...or a concertina splitter. Easier implementation, and certainty of balance.
Assuming all else is equal, there should be no "sonic" difference between the various phase splitter options once they are implemented - they are just different ways of getting to the same goal after all.
Assuming all else is equal, there should be no "sonic" difference between the various phase splitter options once they are implemented - they are just different ways of getting to the same goal after all.
The 'push' and the 'pull' sides must be as similar as possible, but the paraphase circuit
adds an entire extra stage (unity gain inverter) in one side. Differential is much better.
The RC series branch to ground adjusts the open loop characteristics so that closed loop
operation will be stable, with no overshoot.
adds an entire extra stage (unity gain inverter) in one side. Differential is much better.
The RC series branch to ground adjusts the open loop characteristics so that closed loop
operation will be stable, with no overshoot.
Last edited:
Is it working, or are you able to hear it as it is? I think it is a good philosophy to make incremental change, so you learn what works, and it is easier to back out what doesn’t.
If I've understood right, both the paraphase and long-tail-pair splitters have gain that pushes the power pentodes into overdrive (when turned up) which gives a softer, rounder distortion sound. While the Concertina has no gain, so less power stage overdrive and more preamp distortion, which gives a more shrill and harsh sound, not really what I want.
Also, remember this is a found amp. I already have all the parts for free, so I don't gain or save anything by using fewer stages.
I'd imagined that the Differential would probable be technically better as most designers at the time seem to have adopted it in favour of the paraphase, but "better" how? "Better" why? Bearing in mind I'm looking for 'tube tone/character' rather than technical fidelity.
Not sure I understand what "closed loop operational stability" or "overshoot" is/are. (I'm more of a musician than a technician). It's just that it's attached to only one phase leg which seems odd to me. Wouldn't it upset the balance?
Yes it's working… well, I had to replace the smoothing caps and one of the output trafose is dead (I think corroded rather than browned out). But it's stereo so I have 2 channels and the other on is fine, and sound good, though with a bit of a mid range hump.
You understand wrong. The gain across a ltp versus a va/concertina is the same. Both provide half the gain of a va on its own. The parts count is about the same for both as well.
If you want to include a fistful of harmonics, there are better ways than by selecting a particular phase splitter.
If you want to redesign it as a guitar amp you should visit How to design valve guitar amplifiers. Lots of simple clear explanations of technique and design that perfectly match the amp you have and your purpose.
Last edited:
The value of R12 and R14 are switched in the schematic.Since the sound is good, I suppose it's only the schematic that's wrong.
Mona
Zebb,
My three questions are:
1. Do you want to use the guitar amp in a small Jazz Club, and want the sound tending towards smooth soothing music?
2. Do you want to use the guitar amp in a small stage Rock Music Club, and want the sound tending towards fuzzy and hard distorted music?
3. Do you want to use the guitar amp in an apartment as a practice amp, and want the sound tending towards fuzzy and hard distorted music, but at super low volume, so that the Landlord does not kick you out of the apartment?
These are 3 completely different guitar amplifiers.
My three questions are:
1. Do you want to use the guitar amp in a small Jazz Club, and want the sound tending towards smooth soothing music?
2. Do you want to use the guitar amp in a small stage Rock Music Club, and want the sound tending towards fuzzy and hard distorted music?
3. Do you want to use the guitar amp in an apartment as a practice amp, and want the sound tending towards fuzzy and hard distorted music, but at super low volume, so that the Landlord does not kick you out of the apartment?
These are 3 completely different guitar amplifiers.
Last edited:
That creates an inverse polarity signal of the same amplitude, to drive the other side of the push-pull circuit.
It's one way of many to design the circuit, but is probably the worst way.
The best balance is with a circuit that is symmetrical, like a differential amplifier. Driving both sides of a
differential amplifier with balanced signals is even better, since this removes the common mode signal.
Last edited:
Zebb,
The second phase leg gets its grid signal from the first phase leg.
Therefore, the dominant pole of C6 and R9 affects the high frequency rolloff and phase of Both phase legs.
Stability is paramount, unless you want to have the possibility of a full power output at one frequency, without even having a guitar connected (self oscillation).
The second phase leg gets its grid signal from the first phase leg.
Therefore, the dominant pole of C6 and R9 affects the high frequency rolloff and phase of Both phase legs.
Stability is paramount, unless you want to have the possibility of a full power output at one frequency, without even having a guitar connected (self oscillation).
"A long-tailed pair usually has a little more output signal swing than a cathodyne using the same valves, and it has the advantage of providing some gain" - "providing lots of output swing for overdriving the power stage" - "The gain of this differential pair is exactly the same as that of a normal stage" [The Valve Wizard]
Paraphase and LTPs use 2 triodes. Concertinas use 1, how can the parts count be the same?
I don't
Of course the splitter is not the only change to think about, it's just that there's an argument for leaving it as it is, if it sounds okay, and an argument for adopting the same topography as the classic guitar amps I want to emulate the sound of.
I am familiar with, and read much of Merlin's (extensive) contributions, but he doesn't talk much about the sonic effects of the different methods which is what I'm asking about. Also, much of the technicalities are over my head, which is why I'm looking at adapting an existing amp rather than starting from scratch.
Actually no, that's the way round they are. I reverse engineered my schematic from the actual circuit, and it corresponds with the published circuit. I did notice that they were the opposite way round to the equivalents in the Fender Deluxe 5D3 circuit (which uses the same PI). Is this definitely wrong? Should I switch them?
Attachments
Quite right, good question, and it's true I didn't make that clear at the beginning.
In answer, I'm looking to create a low watt, home, practice amp. The 5 watts output of this amp should suite that well.
I'm not looking for a heavy rock, fuzz distortion sound. Rather a Fender 'clean' sound with some soft round bluesy overdrive. Sort of a cross between a Fender Blackface Twin and late Tweed Deluxe sound.
So what would be a better way?
Is symmetrical balance necessarily better?
"a slightly unbalanced phase inverter is often quite benificial to guitar tone, due to the additional 2nd harmonic it introduces" [The Valve Wizard]
Oh yer… forgot about the second half being fed by the first.
So in-stability means prone to 'self oscillation' and this RC combo shunts high frequencies to ground, preventing oscillation, right?
Well, that's what you get by inverting the values of R12 and R14.
If you like it leave it as is.
LTP and concertina are both nice inverters.
LTP produce more odd harmonics and concertina more even ones.
The even harmonics are cancled out by a pp-stage, the odd not.
That makes to prefer the concertina.
But the output swing is larger with an LTP so to drive tubes who demand much drive voltage (or with cathode feedback) it's best choice.
Or a concertina followed by (two) voltage ampifiers.
The phase inverter in your amp is simply a x-1 gain stage.A concertina has aproximately the same gain if you include the pre-stage (both two triodes with the gain of one of them).
The LTP also use 2 triodes, each on produce half the gain of one tube.
Mona