"Black Flag" JTM100 - the build has begun!

[Roe]

.. nominally they were both 32+32, but the F&T measured about 36-37uF, and the Erie around 46-47! So yes, it was way off spec, but that didn't make the amp feel any stiffer at all, as one would expect. It did sound slightly different (warmer, woodier), but not stiffer per se.

...nominally 1M, but actually measures 1.4M, w... Alpha 1M pots from my stash - they varied from almost exactly 1M to 1.15M. And dig this: the one in my Black Flag high treble channel measures as low as 0.9M! ...

Roe: I forget...with the bass pot, what effect does it give to use a smaller value pot? If I weed through the 1M pots, I figured I'd use the low value ones for the bass pot, and the higher ones for the volumes, but perhaps not if it leads to muddier bass...

lower value on the bass pot gives less bass. try the duncan tonestack calculator. I use it quite a lot. my 1m pots measure from 750k to 1.3m. for a jtm I suggest a treble pot that reads high (260-70k) and a bass pot that reads high. also a mid pot that reads low to medium (22k). a vol pot that reads high wont hurt but it will also affect how the bright cap works. however, avoid 1+m vol pots in gainy amps since it leads to hiss.

I've also seen filter caps wary. I think 16uf on the screens, and 32uf on the mains, is as low as you can go, and that higher reading caps can help.

[parkhead]

Potentiometers are often overlooked as a source of variability because they're regarded as variable carbon comp resistors and that's where the discussion ends. The truth is different brands of audio taper pots from different eras vary considerably in rates of taper and resistance, both of which affect the overall picture.

One thing I've learned is when dealing with guitar volume and tone pots, I want to be assured that the pot delivers at least the rated resistance. I've changed plenty of guitar volume and tone pots for 'reworked' pots that deliver the full rated resistance (or more), and the differences are plainly audible. For example, many '500k' volume pots give maybe 440-450k. These are often installed unchecked. When the pot is reworked to deliver 500k+, suddenly the pickup gives greater output and clarity.

There is no reason why this observation doesn't apply to every pot within an amplifier, which is a worthwhile note to DIY amp builders.

When dealing with a pot that gives a substandard reading, one can modify it to increase the total resistance. Here's a video that illustrates how to disassemble Bourns and CTS pots, and the physical difference between log and linear taper:

http://www.youtube.com/watch?v=rUkrpqEmXb8

As for modifying the pot to increase the resistance, I believe it's a matter of scraping the carbon conductor with an Xacto knife. I've purchased reworked CTS guitar pots inexpensively from one Ebay seller (strat100), who I imagine would be willing to modify one's amplifier pots for a nominal charge.

excellent stuff I totally agree ith TED

I originally stumbled on this when repairing an amp with a bad pot and discovered that the "sound" had changed completely
like most of you I have spent countless hours documenting and measuring the "good old stuff" and found that placing a resistor
between the hot terminals of a volume pot immediately resulted in a sound that was closer to the original amps...

there have been many discussions, arguments and threads on various forums, here are some unscientific points
ie why I think it works and sounds better

a. the taper of the volume control is faster & so there is slightly more signal to work with at lower volume settings
b. older pots were offered in more ranges of tapers than we can get now
c. there is a fixed resistor on the hot side in paralell with a mechanical connection ie better signal transfer
d. I suspect there is a capacitance between the wiper and the carbon strip.. modern pots using thinner lighter mechanical components might have a higher capacitance between the wiper and the carbon

Like Ted I do not build or use anything, guitars, amps & effects without inspecting the pots and making sure they are mechanically solid and meet official specs I will also add a resistor as high as 10 meg in parallel with the wiper/carbon to
reduce capacitance & mechanical noise

in terms of time spent at the workbench this area of investigation has proven to be among the most profitable

I hang around with a lot of CAR people and a good engine builder will know every spec of every part he works with and be able to build a "stock" engine with higher horsepower just by cutting down on frictional mechanical losses & knowing where a tiny beneficial tweak in tolerances will improve flow, friction or compression

The vintage factories were not building with this attention to detail, however the circuits were tuned to sound good with the components they used... production then ran the same designs for 20 years with hundreds of minor supplier changes

lets not forget that if the final test bench saw the correct wave on the scope and output numbers met specs the amp was good to go ... any listening tests would be for noise rather than "fatness of tone"

p

[parkhead]

.. nominally they were both 32+32, but the F&T measured about 36-37uF, and the Erie around 46-47! So yes, it was way off spec, but that didn't make the amp feel any stiffer at all, as one would expect. It did sound slightly different (warmer, woodier), but not stiffer per se.

...nominally 1M, but actually measures 1.4M, w... Alpha 1M pots from my stash - they varied from almost exactly 1M to 1.15M. And dig this: the one in my Black Flag high treble channel measures as low as 0.9M! ...

Roe: I forget...with the bass pot, what effect does it give to use a smaller value pot? If I weed through the 1M pots, I figured I'd use the low value ones for the bass pot, and the higher ones for the volumes, but perhaps not if it leads to muddier bass...

lower value on the bass pot gives less bass. try the duncan tonestack calculator. I use it quite a lot. my 1m pots measure from 750k to 1.3m. for a jtm I suggest a treble pot that reads high (260-70k) and a bass pot that reads high. also a mid pot that reads low to medium (22k). a vol pot that reads high wont hurt but it will also affect how the bright cap works. however, avoid 1+m vol pots in gainy amps since it leads to hiss.

I've also seen filter caps wary. I think 16uf on the screens, and 32uf on the mains, is as low as you can go, and that higher reading caps can help.

also the .022 cap that drives the bass pot is part of this equation with .025 or .03 having a cleaner sounding bass tone
than say .02 or .018

while you can see this slightly with the tone stack calculator you can hear dramatic differences in an amp
if you play with the .022 middle and bass caps trying say a pair of .018's

you may not hear it at low volumes but as you bring the amp up the it is more a change in the distortion tone rather than an eq change

a while ago someone at one of these forums could not get .022 mustards & built their amp with .018's and posted some sound clips the results were very interesting

p

[Xplorer]

Potentiometers are often overlooked as a source of variability because they're regarded as variable carbon comp resistors and that's where the discussion ends. The truth is different brands of audio taper pots from different eras vary considerably in rates of taper and resistance, both of which affect the overall picture.

One thing I've learned is when dealing with guitar volume and tone pots, I want to be assured that the pot delivers at least the rated resistance. I've changed plenty of guitar volume and tone pots for 'reworked' pots that deliver the full rated resistance (or more), and the differences are plainly audible. For example, many '500k' volume pots give maybe 440-450k. These are often installed unchecked. When the pot is reworked to deliver 500k+, suddenly the pickup gives greater output and clarity.

There is no reason why this observation doesn't apply to every pot within an amplifier, which is a worthwhile note to DIY amp builders.

When dealing with a pot that gives a substandard reading, one can modify it to increase the total resistance. Here's a video that illustrates how to disassemble Bourns and CTS pots, and the physical difference between log and linear taper:

http://www.youtube.com/watch?v=rUkrpqEmXb8

As for modifying the pot to increase the resistance, I believe it's a matter of scraping the carbon conductor with an Xacto knife. I've purchased reworked CTS guitar pots inexpensively from one Ebay seller (strat100), who I imagine would be willing to modify one's amplifier pots for a nominal charge.

excellent stuff I totally agree ith TED

I originally stumbled on this when repairing an amp with a bad pot and discovered that the "sound" had changed completely
like most of you I have spent countless hours documenting and measuring the "good old stuff" and found that placing a resistor
between the hot terminals of a volume pot immediately resulted in a sound that was closer to the original amps...

there have been many discussions, arguments and threads on various forums, here are some unscientific points
ie why I think it works and sounds better

a. the taper of the volume control is faster & so there is slightly more signal to work with at lower volume settings
b. older pots were offered in more ranges of tapers than we can get now
c. there is a fixed resistor on the hot side in paralell with a mechanical connection ie better signal transfer
d. I suspect there is a capacitance between the wiper and the carbon strip.. modern pots using thinner lighter mechanical components might have a higher capacitance between the wiper and the carbon

Like Ted I do not build or use anything, guitars, amps & effects without inspecting the pots and making sure they are mechanically solid and meet official specs I will also add a resistor as high as 10 meg in parallel with the wiper/carbon to
reduce capacitance & mechanical noise

in terms of time spent at the workbench this area of investigation has proven to be among the most profitable

I hang around with a lot of CAR people and a good engine builder will know every spec of every part he works with and be able to build a "stock" engine with higher horsepower just by cutting down on frictional mechanical losses & knowing where a tiny beneficial tweak in tolerances will improve flow, friction or compression

The vintage factories were not building with this attention to detail, however the circuits were tuned to sound good with the components they used... production then ran the same designs for 20 years with hundreds of minor supplier changes

lets not forget that if the final test bench saw the correct wave on the scope and output numbers met specs the amp was good to go ... any listening tests would be for noise rather than "fatness of tone"

p

Very interesting.
This is an old RS 500k pot, inside, measured about 660 k.

[Roe]

yes it turns out that the old pots typically read high, often around 30% higher than nominal value

[Vegard]

What type of chassis was this amp built on? I would think the 45/100, but this looks like steel??

[shakti]

Yes, steel chassis. Same as 12 series.

[ivan H]

Hi Shakti, regarding ur method of connecting the 1 ohm bias sensing resistors between pins 1 & 8 then grounding pin 1, I think u have used the more correct method of connecting them, at least for pentodes. Kevin O'connor advised me on this a while back on another forum. The gist of it is,,, electrons travelling from the cathode strike the anode with such velocity that some bounce off (secondry emission). The supressor grid, positioned between the screen & anode, & being connected to ground (0 volts), supresses this (somewhat). When the supressor (pin 1) is tied to the cathode (pin & both r grounded via the 1 ohm resistor, it puts a small positive voltage on the supressor, meaning it then draws a small amount of current, resulting in slightly inaccurate biasing & also slightly reduced tube efficiency. Actually, to achieve really accurate bias via 1 ohm resistors, the screen current needs to be taken into consideration. This can be calculated by measuring the voltage drop across the screen resistor & then divide this voltage drop figure by the resistance value of the screen resistor to obtain the screen current. This should be subtracted from the current shown across the 1 ohm resistor, tho u can just guesstimate it at around 3 or 4 mA. Back to the supressors, an even more efficient method of connecting the supressors is to tie pin 1 to the negative bias supply voltage, like at the apex of the 220k bias splitters, tho this can change the tone a very slightly. Another nice build u have happening there. Cheers

[beaulieu]

Awesome build after seeing your work makes me want to rip my nos builds apart and redo them.