Naylor Superdrive Presence and OT Tap

[jhaas]

Inspired my cbass and FusionBear's recent Naylor Superdrive builds in this thread:
https://tubeamparchive.com/viewtopic.php?t=24344

I decided to gut an old 2x5881 build that never went anywhere and has been gathering dust for the past three years, an fill it with some superdrive goodness.

This OT for this build has 2/4/8 ohm secondaries. The Naylor Presence circuit is designed for a 16 ohm tap. Since this presence circuit isn't a voltage divider like an old school Presence, it's not clear to me how to modify it when moving from 16 ohm tap to an 8 ohm tap.

Anyone got their head around this circuit enough to suggest a change, or should I wire it up with the same values used on the 16 ohm tap?

[cbass]

It should still work fine on the 8 ohm tap not sure how you would modify it to make the frequncy response the same.
BTW this amp rips I couldn't be happier with the overdrive sound we don't need no stinkin dumbles

The NFB in this circuit is like an anti-presence- its a high-pass shelving filter, with a -6dB/octave slope from ~7500 Hz when the pot is at max resistance, and the pot moves the level -30dB (by sliding the curve up 5 octaves), reducing the NFB.

[cbass]

https://tubeamparchive.com/viewtopic.ph ... ght=naylor

[Mr. Lime]

I have got the same problem and can't get my head around how it works..

Using the 8ohm tap just like the 16ohm tap doesn't give a nice control over the feedback.

Does someone has an advice for me/us?



Thanks in advance!

[pompeiisneaks]

The negative feedback is a calculation done based upon the impedance tap used, and a specific resistance. In the circuit you show, there's a 10k + 470ohm resistor between that feedback and the cathodes. If you change the tap, you need more resistance in line somewhere (before or after that capacitor) to match the same amount. I'm not good with the calculations, someone probably has a simple formula to figure it out, but I've seen them do the adjustments in the past.

Just switching the taps isn't recommended, always adjust the feedback resistance before you switch taps.

~Phil

[sluckey]

I would just temporarily insert a pot in that top most line on your schematic. Probably use a couple gator clip test leads. Turn the pot until you are happy. Carefully disconnect the pot and measure it. Then permanently insert a fixed value resistor in that same line.

EDIT... Disregard. I read this backwards.

[Tony Bones]

Hmmm... The voltage from the 8 ohm tap is only about 70% of what would come from a 16 ohm tap.

So, just shooting from the hip, what would happen if you applied it higher up the 'long tail' so that it would have greater effect. The tail is two 10k resistors in series (plus 470 ohms, but that's small in comparison.) Suppose you replace the bottom 10k resistor with 15k and the top one with 5k, or 4.7k, or something like that. It seems (without proof) that that ought to restore about the right amount of NFB.

Pretty easy to try. Might need to mess with the 0.005 and 0.0015 caps. Maybe parallel each of them with something about half as big.

[martin manning]

Most people will adjust the resistor connected to the OT to compensate for the change in the voltage from one secondary tap to another. That avoids messing up the PI or presence. The adjustment is the square root of the impedance ratio. For example, going from a 16 ohm to an 8 ohm tap, that would be sqrt(8/16) = 0.707, so multiply the existing value by 0.707. This one is different... can you point to the schematic?

[Ten Over]

Replace the 0.0015uF capacitor with a 0.0022uF. Replace the 0.005uF with a 0.0033uF. Leave everything else the same.

[martin manning]

Replace the 0.0015uF capacitor with a 0.0022uF. Replace the 0.005uF with a 0.0033uF. Leave everything else the same.

Ah now I see it’s just above. Can you explain your logic?

[Ten Over]

Ah now I see it’s just above. Can you explain your logic?

It's basically just an impedance divider formed by the 0.0015uf cap. and the 0.005uF cap. He wants more voltage at the junction to compensate for the lower voltage at the 8 ohm tap, so reduce the impedance of the 0.0015uF by making it larger and increase the impedance of the 0.005uF by making it smaller.

[martin manning]

But the 10k tail has to be changed as well to get the same frequency response, no?

[Ten Over]

But the 10k tail has to be changed as well to get the same frequency response, no?

I don't think so, but I would like to see your reasoning. Are you talking about the 10K PI tail resistor or the 10K going to ground?

[martin manning]

The 10k to ground (which I'm calling the tail). That resistor and the presence pot are in parallel with the 5n, and together they determine the corner frequency on the high end. I would leave all that alone and just increase the 1n5 to 2n2, which will restore the feedback voltage to the appropriate level. A 3n3 could be tried too, for a little more feedback.

[Ten Over]

The 10k to ground (which I'm calling the tail). That resistor and the presence pot are in parallel with the 5n, and together they determine the corner frequency on the high end. I would leave all that alone and just increase the 1n5 to 2n2, which will restore the feedback voltage to the appropriate level. A 3n3 could be tried too, for a little more feedback.

Yeah, the parallel impedance is 3.33K until you get up to 700Hz and it's still 2.69K at 7KHz. There is a slight roll off at 3KHz and above with the 2n2/5n NFB as compared to the original, so I thought I would compensate for that by changing the 5n to 3n3. Once I laid a calculator on it, I found that the 3n3 overcompensated, so I agree with you about leaving the 5n in there.

The frequency response is extremely similar with the 2n2/5n version and the NFB voltage is 1.47 times larger. He needed 1.414 times, so that's pretty close.