Weber copper cap Module

[Old Tele man]

...simple, one-word answer: it's the "...voltage..."

...the plate voltages (affecting ALL the tubes) "sag" exponentially with a vacuum tube rectifier (due to Child-Langmuir 3/2's Law equation) while a simple resistor yields a "linear" voltage drop for the same current load.

...one Mullard GZ34 and one Z34 from Ted...so, yes, you might call it a "...test of ones..." if you wish, but a valid test nevertheless--one vs. one.

...to be honest, my 64-year old ears couldn't detect much of a change, but the younger (mid-20s) players claimed the above sonic changes...and, that's why I also did the 'voltage' tests...to quantify the reason for the "changes", and I used a swept sinewave as input signal for consistant & repeatable input and measured the actual DC-voltage "sag" at the 6L6GC plates (on both DMM and oscope)...and, when plotted, the Coppercap™ showed essentially a straight-line "sag" plot, while the GZ34 showed the exponential / cantenary "sag" plot one would expect from the equation: Vp = (Ip/K)^(2/3).

...uh, definetly "hair dryer" warmth! (wink,wink)

[soundmasterg]

When the original chinese GZ34 blew in my '93 AC30, I compared a Weber CopperCap WZ34 and a NOS Amperex GZ34 in it. They were within a couple percentage points of each other as far as the overal effect on the sound. The tube GZ34 made the amp sound just a tad warmer and more touch sensitive and I went with that, but I wouldn't hesitate to use the Copper Cap if needed.

My 2 cents....

Greg

[pdf64]

'Does someone know something on this circular component and the effect that produces?'
I thinks it's a NTC inrush current suppressor type thermistor. Cold the resistance is high and so it limits the start up current. As the resistors conduct and heat up, the thermistor's resistance goes down and is chosen to be effectively a short at working currents.
Check out modern fender schematics, most all have one of these in series with the PT primary to limit the switch on surge. Peter.

[R.G.]

...to be honest, my 64-year old ears couldn't detect much of a change, but the younger (mid-20s) players claimed the above sonic changes...

Also in the interest of honesty, I've done the same test, but quietly, and without the younger players' knowledge. Of the ones that said their amps sounded different, the majority said it was better somehow, in a way they could not put their finger on. But it was not either a big enough sample number to allow good statistical significance testing, to be fair.

Listening test which do not control foreknowledge (and expectations) on the part of the listeners are at best useless, as the result will be whatever the listeners expect if the results are anywhere close. At worst, they're easy to manipulate by the tester - or the testee.

and, that's why I also did the 'voltage' tests...to quantify the reason for the "changes", and I used a swept sinewave as input signal for consistant & repeatable input and measured the actual DC-voltage "sag" at the 6L6GC plates (on both DMM and oscope)...and, when plotted, the Coppercap™ showed essentially a straight-line "sag" plot, while the GZ34 showed the exponential / cantenary "sag" plot one would expect from the equation: Vp = (Ip/K)^(2/3).

Did you happen to pop those results into something like Excell and fit the curve to see if it really was the 3/2 power equation?

Also, what was the amount of variation from straight line to curve? We may be on to something big here if we can relate a specific voltage to a specific sound from an amp...

Just offhand, I can't see why sweeping a sine wave would cause any changes in sag. It's current out of the power supply that causes sag anyway, and within the flat(-ish) passband of the amp, the frequency wouldn't matter much, except as that caused issues with speaker impedance curves. At a guess, you'd have to control that by using a resistor instead of a speaker or only testing at one frequency where the speaker was a relatively tame impedance on its curve to get any useful info.

[TD_Madden]

'Does someone know something on this circular component and the effect that produces?'
I thinks it's a NTC inrush current suppressor type thermistor. Cold the resistance is high and so it limits the start up current. As the resistors conduct and heat up, the thermistor's resistance goes down and is chosen to be effectively a short at working currents.
Check out modern fender schematics, most all have one of these in series with the PT primary to limit the switch on surge. Peter.

Exactly. Ted sold me a couple to use in replacing the "brimistor" in my '66 AC50.

[Pedro Vecino]

'Does someone know something on this circular component and the effect that produces?'
I thinks it's a NTC inrush current suppressor type thermistor. Cold the resistance is high and so it limits the start up current. As the resistors conduct and heat up, the thermistor's resistance goes down and is chosen to be effectively a short at working currents.
Check out modern fender schematics, most all have one of these in series with the PT primary to limit the switch on surge. Peter.

Then, his only purpose in the cooper cap is to slow down the start-up..?
I observed a "strange" phenomenon in the Fargen Miniplexi with the EZ81 substitute: on having activated the standby It was necessary to wait two or three seconds until the sound was appearing.
Then, with the amp sounding, the only "active" component to emulate the functioning of the rectifier tube is the resistor?
Regards

[pdf64]

Resistor and diodes. But the current limiting switch on surge suppressor is a valuable feature, though the warm up is a whole lot quicker than a gz34. Peter.

[TD_Madden]

yes, the "warm-up" is about 3 seconds....

[Old Tele man]

Did you happen to pop those results into something like Excell and fit the curve to see if it really was the 3/2 power equation?

...yes I did, and although the GZ34 value was close to 3/2-power at rated current (slightly above 3/2, ~ 1.58, as I recall), it was way off at idle current (closer to 5/2-power, as I recall)...gotta find my notes for the exact values.

Oh, something else interesting I recall, the Excel-calculated K-constant (representing Perveance) was "different" for the two points and neither was the same as the value calculated when backsolving from GZ34 specs of 17Vp and 225mA, ie: G ~ 0.0032 amp/volt^(2/3).

Also, what was the amount of variation from straight line to curve?

...more "difference" near low current end of the curve than at the high end current...looked sorta like "tuck under" commonly associated with triode curves.

We may be on to something big here if we can relate a specific voltage to a specific sound from an amp...

...I just attribute it all to 'production' variances between different production runs/lots and different companies.