Voltage values on Leo's amp schematics not always correct

[sookwinder]

OK ... time for some train- spotting....

I like Fender BF amps and generally speaking SF amps as well.
I have based most of the amps I have built on BF amps.
Leo used the same preamp design essentially for all the BF designs and subsequently it was carried into the SF era.

This preamp stage always had no matter what the size of the amp:
- a 7025 valve (low noise 12AX7)
- 100k anode resistor connected to the DC voltage from the rectified voltage rail.
- cathode resistor = 1.5k
- cathode bypass capacitor = 25uF



What would change in the design between the various BF amps was the value of the DC voltage at the anode (VA). This in turn would change the voltage at the cathode (VC). So an amp like the BF Princeton reverb has a lower VA and VC, less distortion occurs, it has a cleaner preamp sound than say the Vibro Champ or Twin Reverb. The higher you make VA (assuming you keep the cathode resistor constant) the more chance overdrive will occur quicker.

Now if you do a series of experiments and measure a whole series of VA values and then measure the corresponding VC .... USING THE EXACT SAME preamp set up, the values should all fall onto a single line. This graph line is basically linear for the majority, in the middle section, but curves / flattens outs at the top and the bottom, where it is no longer linear.

The thing is, over the years when I have used the exact component size and voltages that are specified on Leo's BF amp schematics, I can not always get the voltages to match. Sometimes there is quite a large discrepancy. So I have plotted the VA nd VC values from all the BF and SF schematics that I can find and accurately read the stated voltages. This is what I got:



here is the data used:



So what does this tell us ?
Well first off Fender (maybe Leo, assuming he did the majority of the electronic design/tinkering) did not use standard/calibrated components when he developed his amp designs and then wrote/drew up the schematics. Either the brand new RCA 12AX7 (7025) valves varied considerably in their Gm value and/or the real value of the 1.5k resistors he used varied considerably from the nominal 1.5k. Remember there is that statement on the schematics that the component values have a 10% tolerance and that stated voltages are +/- 20%.

But more importantly it seems that they did not go back and check the voltages on subsequent builds of each amp to double check the values and ensure that the stated values were nominally in the centre of the "bell curve" distribution.

So 50 years later when someone is trying to build a BF Deluxe Reverb preamp with exact components and VA as shown in the "AA763" schematic (ie. VA =180V), it is near impossible to achieve the VC of 1.3V as shown in the original design drawings. What you do get is a VC value just over 1.4 - 1.5V. That is a big difference !

So next time you are checking an amp and measuring some voltages or maybe building your own clone, just remember that the values stated on the schematics may not in fact be the value you will measure or what you should be aiming for.

[cbrown]

Good to know and something to think about when trying recreate those classic designs. Thanks.

[saxjag]

Valuable info! Mods, maybe this should be preserved in the Library.

[sookwinder]

I had intended to move this thread into the technical section after a week or so or whenever it falls off the first page of threads







great minds think alike ....

[46346]

cool observation! i'm curious what your AC voltage is where the mains are plugged in. being that you are down under, and build your own amps, i imagine you are converting reliably from 220V-240V. i doubt it would influence the consistency of the DC, but i guess Leo was working with 110 VAC?

[sookwinder]

you have miss read the initial post... I think.
This has nothing to do with the wall voltage (AC voltage)... whether the wall voltage is 110 or 240V
once you go past the rectifier it is all DC.
The transformers for a 240V country or a 230V country, 220 Volt country or even a 110V country, will all do the same thing and produce the particular DC voltage you want, that you have specified at the rail.

It is the DC voltage at the two positions (pin 1 and pin 3) or the first preamp triode that I measured and graphed
What I have labelled as VA is the DC voltage at the anode (A for Anode)
What I have labelled as VC is the DC voltage at the cathode (C for Anode)

[46346]

aye, sook - my question is somewhat ancillary to your observations. perhaps your wall or bench voltage is exactly what the mains of the PT want to see, but i know that in my studio, the wall voltage is higher than Leo's designs called for, and the plate voltages seem to reflect that, relative to the schematic. and when my neighbors are running their AC and amplifiers too, the wall voltage ticks down a bit, and i would expect that to affect the DC in my amp circuits. i've seen different plate voltages on the same circuit at different times of day or season. unless i have the amp on the variac and keep an eye on the meter.

not saying that this would be the cause of the variations you've noted, but unless you're keeping an eye on the incoming AC, might it be a contributing variable? i don't know how widely it affects the DC

[sookwinder]

Ok I understand what you question/statement was ... yes wall voltage can vary quite considerably and I have learnt not to check circuit voltages between 5pm - 8pm and never when it's "damn hot" outside ! I also check wall voltages each as well.

I agree ... variations of 10 - 15 volts in a 240V wall voltage system can also cause the whole system to perform differently enough to have your "favourite setting" sound a little different

But the "issues" I have found when building amps were (again) independent of the wall voltage.
This is because I would/can adjust the dropping resistors along the way in the voltage rail to ensure that the DC voltage at the preamp (V1A) was what Leo had on his schematics. In this situation, if the voltage at the anode is what Leo had, then the cathode should also be OK. But as my graph shows, this may not always be the case.

I just looked at the schematic for the BF Deluxe (not used in the above data) and compared it to the BF Deluxe Reverb.
BF Deluxe preamp V1A Normal channel
Anode 170V
cathode 1.6V

BF Deluxe Reverb preamp V1A Normal Channel
Anode 180V
cathode 1.3V

This "Leo" data just doesn't make sense...

[46346]

word. i figured you might have had a solid reference for the incoming voltage

I just looked at the schematic for the BF Deluxe (not used in the above data) and compared it to the BF Deluxe Reverb.
BF Deluxe preamp V1A Normal channel
Anode 170V
cathode 1.6V

BF Deluxe Reverb preamp V1A Normal Channel
Anode 180V
cathode 1.3V

This "Leo" data just doesn't make sense...

well then... that is just plain wacky

[sookwinder]

yeah certainly "wacky" ... but this is also the classic era of the "TV repair man" style of electronics for expensive consumer products, where someone with a small amount of electronic knowledge would come and replace a few parts (valves, resistor) to a spec sheet and hopefully the TV worked again. Solving the symptom rather than the underlying cause. Maybe that is how guitar techs also worked ?

So having schematic where voltages are +/- 20% from the get go, can account for the variation/errors in stated "correct" values as well.

It sort of shines a light on what appears to be almost an amateurish process of developing electronic equipment/manufacturing that existed back then. One could not really at all on one amp on the production line sounding the same as the next ... yeah they sounded close, but not the same.
This was not a problem as such and would probably only be highlighted when
(a) the amp needed a real service/fix up and a proper electrical tech had to trace through a fault
(b) someone is comparing two amps of the same time and can hear quite a difference in sound even though the two amps are set up the same
(c) more recently when people have started to build clones of these classic designs and are striving to get as close to a sound that they can hear in an original.

The above is also why I have no issue is "adding" what is now regarded essential components to these designs (where Leo had left out or various reasons), when I build them.... grid stoppers and screen resistors for example ...

[46346]

yeah! it's kind of like cooking with someone else's recipe... in the early 1600's when a tablespoon was carved from some kind of wood by someone in your village. who knows what loose tolerances were involved in the daily bench voltages, test equipment, etc. at Fender back then. not to mention between manufacturers. and to a lot of people's ears, the miracle of an amplified guitar in the 50's and early 60's was not so familiar... it could be that even George Martin himself may not have noticed much difference between tweed Deluxes when they were new. but to our ears with decades of rock music engineering behind us...

maybe the voltages at various stages were general ballpark goals that they would try to reach by using whatever was at hand. maybe they were hopeful guesses with out bothering to reference the previous model's schematic. maybe he was trying not to look exactly like Western Electric's designs.

you might remember a few years back when i bought an unadulterated very early 6G6 Bassman. i replaced the filter caps back to spec, and then the amp would choke to silence on any hard attack chords. i checked all the voltages along the circuit, component values... everything. then someone pointed me to the previous, and subsequent, schems - 5F6-A and 6G6-A. they both had a little disc capacitor leading to the input of the PI tube. this was completely absent from the 6G6 schem AND my amp's board. like you said - i have no qualms about adding what apparently Leo forgot. or left out intentionally?? either way, it took several of us to finally ignore the schematic and make the amp work