Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: JustMike on August 14, 2020, 03:38:42 pm
-
In my journey into tube land I've been slowly learning and as some of you know, I built a design by Tubenit (Ryan's Amp) and he and a few others have really helped me along the way. I think I've been asking reasonable questions and I understand the principles behind most of the answers. So as I've studied the circuit and noted the general similarities to the Fender AB763, I also noticed the other AB763 circuits like the Super and on have higher plate voltages and therefore more gain. But in my experience as a guitar player, a Deluxe on 3 sounds pretty similar to a Super on 3 gain wise. So what do they mean by more gain? Taken to the extreme, I guess the Twin has the most gain, but I wouldn't call a Twin a gainy amp at all!
Rob Robinette's site says all AB763 circuits use the same NFB circuit throughout the series, so as the voltage goes up through the higher powered models, so does the NFB. Is that all it is? More NFB to compensate for higher plate voltage and gain?
-
gain?
the easy answer is Vo/Vin, so 1vpp in 10vpp out = gain of 10
want more surf up something like "understanding gain" you might find better math :icon_biggrin:
-
Gain in a tube* amp can be reckoned either as Voltage Gain, or dB Gain.
Either way, it’s a measure of the amount of increase in signal amplitude
* In a transistor amp its mainly ‘Current Gain‘
-
So as I've studied the circuit and noted the general similarities to the Fender AB763, I also noticed the other AB763 circuits like the Super and on have higher plate voltages and therefore more gain.
Plate voltages in the AB763 series has very little to do with gain.
So what do they mean by more gain?
Who?
Rob Robinette's site says all AB763 circuits use the same NFB circuit throughout the series, so as the voltage goes up through the higher powered models, so does the NFB. Is that all it is? More NFB to compensate for higher plate voltage and gain?
The NFB circuit may look the same (either 820/100 or 820/47), but the amount of NFB applied in each model is not apparent just by looking at this. In fact, there isn't enough information given on the schematics to determine how much NFB is being applied in any of the models.
-
https://en.wikipedia.org/wiki/Gain_(electronics) (https://en.wikipedia.org/wiki/Gain_(electronics))
-
So what do they mean by more gain? Taken to the extreme, I guess the Twin has the most gain, but I wouldn't call a Twin a gainy amp at all!
Gain in a tube* amp can be reckoned either as Voltage Gain, or dB Gain.
Either way, it’s a measure of the amount of increase in signal amplitude
This is the correct definition of gain.
So what do they mean by more gain?
Who?
The same guitarists that say tremolo when they mean vibrato, and vice versa, have used the term gain (or gainy) as a synonym for overdrive/distortion. I guess it's just easier to call an amp "high-gain" instead of "mucho-overdrivey".
Gain is involved in overdrive/distortion because if you have a lot of gain (signal amplitude) preceding a stage that can't handle that level of signal without clipping, then the result will be a clipped, "gainy" sounding (overdriven) stage.
The reason we don't consider the Twin gainy is because the stages don't easily overdrive.
But, technically, there is a whole lot of signal gain in a Twin.
Think of "gainy" as the dumbed down guitarists term for "distortiony" and GAIN as the definition you have been given.
You'll sound smarter than a guitarist that way.
-
... In fact, there isn't enough information given on the schematics to determine how much NFB is being applied in any of the models.
The is enough information on the schematics, but there is a burden on the viewer to do quite a bit of work to figure it out.
-
What is the impedance ratio for the output transformer on a Vibroverb Amp AB763?
-
... In fact, there isn't enough information given on the schematics to determine how much NFB is being applied in any of the models.
The is enough information on the schematics, but there is a burden on the viewer to do quite a bit of work to figure it out.
What is the impedance ratio for the output transformer on a Vibroverb Amp AB763?
Probably 4KΩ:8Ω, as the same output transformer was used in the AA763 Pro (with a single 15) and the AA763 Vibrolux (with a single 12).
However, we don't really need to know that. Kuehnel has a couple-chapters in his 5F6-A Bassman book (https://ampbooks.com/mobile/books/bassman/) going through the math derivation underpinning his Long-Tail Calculator (https://ampbooks.com/mobile/amplifier-calculators/long-tailed-pair/calculator/) and his LTP Feedback Calculator (https://ampbooks.com/mobile/amplifier-calculators/feedback/calculator/).
Cheating? Sure, but it's telling that there's no requirement to input anything about the tube type or output transformer, only the gain of the long-tail (gotten from the first calculator), the output power, transformer feedback tap, and the input of the power amp (which corresponds to the bias voltage of the output tubes).
- That tends to make obvious that the amplification from output tube grid-to-plate and step-down due to the output transformer can be ignored because they're part of a system that only creates a specified speaker-voltage when a specified tube grid-voltage exists.
Grid bias of the AB763 Vibroverb (https://el34world.com/charts/Schematics/files/Fender/Fender_vibroverb_ab763_schem.pdf) is -55v, so max clean output power is happening at ~54v peak to one grid. Kuehnel's Long-Tail Calculator (https://ampbooks.com/mobile/amplifier-calculators/long-tailed-pair/calculator/) says gain is 24.56 to one side, and 26.72 to the other side (when set for 12AT7 and a 22kΩ tail). Using the LTP Feedback Calculator (https://ampbooks.com/mobile/amplifier-calculators/feedback/calculator/), set Rp to 0.047kΩ to match the Vibroverb's 47Ω resistor, 40w RMS output, 8Ω feedback tap, and 54v peak power amp headroom (from above).
It would be convenient to be able to directly enter the feedback resistor value and get the closed-loop gain percentage. However, entering 60 percent spits a result of 0.8kΩ as the resistor value.
√(40w * 8Ω) = 17.89v RMS at the speaker
54v peak needed at the 6L6 grid
54v / 24.56 (gain to one LTP output) = 2.2v peak
2.2v peak /1.414 = 1.55v RMS
17.98v RMS / 1.55v RMS = open-loop gain of 11.6
Closed-loop gain is 60% of open-loop gain (from LTP Feedback calculator), so
Closed-loop gain = 0.6 * 11.6 = 6.96
We might expect 40w at the Vibroverb's speaker when the LTP is fed 2.58v RMS or 3.65v peak.
I have no idea how any of this speaks to JustMike's original question.
-
... I also noticed the other AB763 circuits like the Super and on have higher plate voltages and therefore more gain. But in my experience as a guitar player, a Deluxe on 3 sounds pretty similar to a Super on 3 gain wise. So what do they mean by more gain? ...
"They"? It appears you're the one saying there's "more gain." :icon_biggrin:
Joking aside, if we drew loadlines for a low-voltage 12AX7 gain stage and a high-voltage 12AX7 gain stage, we might be able to show the high-voltage scenario has a little more voltage gain.
More likely though, we will find the high-voltage scenario can output a larger signal. That larger signal can more-quickly eat up the headroom of what follows, causing distortion. In all likelihood, people noticed "more distortion" as a result of giving their preamp a higher supply voltage, conflated the terms "gain" and "distortion," and then said "higher supply voltages give more gain." :think1:
On the other hand, the Deluxe Reverb schematic (https://el34world.com/charts/Schematics/files/Fender/Fender_deluxe_reverb_ab763.pdf) shows a bias voltage of -35v for its 6V6 output tubes. The Super Reverb schematic (https://el34world.com/charts/Schematics/files/Fender/Fender_super_reverb_ab763_schematic.pdf) shows a bias voltage of -52v for its 6L6 output tubes.
It takes a peak drive signal nearly-equal to bias (so 35v peak for the Deluxe, 52v peak for the Super) to push the output tubes to their max clean output power. We start with the same guitar signal at the input jack. Somehow we grow that same guitar signal to 52v peak in the Super Reverb, and only 35v peak in the Deluxe Reverb.
So there's "more gain" in the Super Reverb, from input jack to output tube grid. There's 52/35 = 1.49x more. Is that "a lot"?
The average Fender 12AX7 gain stage provides an amplification of ~61.5x, and a 12AU7 in the same socket gives ~15.8x amplification. So "1.49x" looks like "almost nothing."
It matters when you're picking how much voltage you need in your preamp to drive your power amp, but it doesn't matter in the way I think you meant.
... More NFB to compensate for higher plate voltage and gain?
No.
... Rob Robinette's site says all AB763 circuits use the same NFB circuit throughout the series, so as the voltage goes up through the higher powered models, so does the NFB. Is that all it is? ...
Stop for a sec & look at the schematic of one of the amps above. The feedback loop runs from speaker to phase inverter. Intuitively, you'll realize the feedback loop can't have too much relevance on the preamp, other than to require a bigger drive signal to get the output section going.
And all the blackface amps are similar when it comes to their negative feedback (though not exactly the same).
-
Probably 4KΩ:8Ω
That's what I think and what Mercury Magnetics thinks, also. At the very least, the impedance of the OT feedback tap must be known in order to determine the NFB and that information is not given on the schematic.
I only submitted my original statement in order to give JustMike a heads-up that certain conclusions about NFB are likely to be erroneous due to a lack of information.
-
Thanks guys. There's a lot of good info in your answers (and questions). I try to search for the answers, but sometimes I'll read something and not knowing the fundamental background, my mind asks what must seem like odd questions. But y'know, there are different kinds of students and different kinds of teachers.
Thanks Again
-
You'll sound smarter than a guitarist that way.
:l2: :laugh: :l2: