Spitfire-ish

[bean]

I had an 2x12AX7 and 2xEL84 amp that I never bonded with. I ripped out the preamp and inserted a spitfire preamp leaving the EL84 power section and power supply unchanged.

I used the following schematic for the Spitfire preamp:
http://www.diyguitarist.com/Schematics/ ... ematic.bmp

The amp actually sounds pretty good.... it's got some of the spitfire character, but it is higher gain and not as much clean headroom. Looking at some numbers, I noticed my PT wasn't up to the task of providing Spitfire voltages. I used a 220-0-220 instead of a 285-0-285

My B+ readings are as follows:

My Build Spitfire
B+ 1 285 352
B+ 2 276 349
B+ 3 274 320
B+ 4 273 329

Tube voltages are as follows (V1 is parallel):

My Build Spitfire
V1 Plate 109.8 150
V1 Cathode 1.1 1.26
V2 Plates 211 & 216 250
V2 Cathodes 56.2 62
EL84 Plates 278 348
EL84 Screens 274 348
EL84 Cathodes 8.77 11.2

Why do I have a mismatch on the plates of V2?
Should any changes be made since my voltages will be lower? Should I try to get more voltage to the preamp tubes? How would that be accomplished?

The power supply I used is 47uf-22uf-22uf-22uf with 1K resistors. SS rectification and no choke.

The existing power amp section I left in place has the following differences from the Spitfire:

I used 10K grid stoppers instead of 1.5K
I used 470 ohm resistors on the screens instead of 1K
I used a 130 ohm cathode resistor instead of 120 ohm
I also have the addition of 120 ohm resistors between the plates and OT

I'm just trying to wrap my head around what the differences in the power amp contribute to the sound and if I should be changing any bias resistors in the preamp section or trying to get the preamp voltages up since my voltages are lower.

Thanks!

[Phil_S]

You are never going to have the high voltages on that Spitfire schematic from 220-0-220. I think you know that.

You might convert to a solid state rectifier and that might buy you 10v +/- more at the power tubes, but what you are seeing now isn't going to rise significantly.

You can, however, raise the preamp voltages closer to Spitfire levels by changing the dropping resistors in the B+ ladder. Tell us what your power supply ladder looks like and I'm sure you'll get some recommendations. Resistors in the dropping string can be as low as 470r or 1K to allow voltages to increase.

[bean]

Thanks Phil!

What does the higher voltage do? Is that where the headroom is coming from?

You are never going to have the high voltages on that Spitfire schematic from 220-0-220. I think you know that.

Yeah... I knew my voltages wouldn't be spitfire high on the power tubes. Just wondering if I could get the preamp tubes closer.

You might convert to a solid state rectifier and that might buy you 10v +/- more at the power tubes, but what you are seeing now isn't going to rise significantly.

Already using SS with no choke.

You can, however, raise the preamp voltages closer to Spitfire levels by changing the dropping resistors in the B+ ladder. Tell us what your power supply ladder looks like and I'm sure you'll get some recommendations. Resistors in the dropping string can be as low as 470r or 1K to allow voltages to increase.

My power supply ladder is:
47uf - 1K - 22uf - 1K - 22uf - 1K - 22uf

I see the on the spitfire schematic that the dropping resistors for both B+3 and B+4 come off of the 2nd filter capacitor. Would that help any?

[morcey2]

Switch to a choke instead of a 1k resistor between B+1 and B+2. That will get the voltages on the screen, PI, and Preamp up. I use the standard fender deluxe style choke on lots of my builds when I need one. It's cheap and it works.

Get rid of the resistors between the power tube plates and the OT. That will give you more range from the power tubes. At the moment they're just dissipating power that would be better served being dissipated in the output tubes.

Don't worry about the difference between the plate voltages on the PI. That's well within range for the resistor tolerances.

With a post-pi master volume, you're going to be getting lots of distortion from the PI itself. If you change the bias resistor on the PI from 1.2k to 680 or even 470 ohm, it will warm up the sound quite a bit. I'd also change the "coupling cap" on the far side of the PI (pin 2 on V2 to ground) to be 0.1µ from 0.01µ.

Matt

[surfsup]

El84 are sensitive so you can clip a 470k to each 100k PI plate resister to make them 82k. This would reduce gain but also raise the PI's plate voltage. Quick and easy test.

I am not familiar w the spitfire pre but the first stage voltage is fine. You might want to reduce signal into the second stage with a divider on a volume pot if there is one after the first stage or add a divider.

those things will help raise headroom some...

[Darkbluemurder]

Switch to a choke instead of a 1k resistor between B+1 and B+2. That will get the voltages on the screen, PI, and Preamp up. I use the standard fender deluxe style choke on lots of my builds when I need one. It's cheap and it works.

Get rid of the resistors between the power tube plates and the OT. That will give you more range from the power tubes. At the moment they're just dissipating power that would be better served being dissipated in the output tubes.

Don't worry about the difference between the plate voltages on the PI. That's well within range for the resistor tolerances.

Fully agree to all of that.

With a post-pi master volume, you're going to be getting lots of distortion from the PI itself. If you change the bias resistor on the PI from 1.2k to 680 or even 470 ohm, it will warm up the sound quite a bit. I'd also change the "coupling cap" on the far side of the PI (pin 2 on V2 to ground) to be 0.1µ from 0.01µ.

Normally, that is the effect of a Post-PI MV but in the Spitfire, most of the overdrive comes from the power tubes. There is only one gain stage ahead of the PI so the PI will never clip hard. I once rewired one of my amps to the Spitfire schem, and turning down the PPIMV drastically changed the amp's tone.

Agree on the bias resistor change.

I doubt that increasing the PI input cap yields a noticeable change. This is due to the fact that the PI input is "bootstrapped", i.e. the impedance is much higher than the 2 meg resistors. As a result the 3dB downpoint is already very low with the 0.01uf. A better place to gain bass is to increase the coupling cap after the first gain stage from 2n2 to 4n7.

Cheers Stephan

[Darkbluemurder]

El84 are sensitive so you can clip a 470k to each 100k PI plate resister to make them 82k. This would reduce gain but also raise the PI's plate voltage. Quick and easy test.

Good suggestion, and easily reversed if you don't like it.

I am not familiar w the spitfire pre but the first stage voltage is fine. You might want to reduce signal into the second stage with a divider on a volume pot if there is one after the first stage or add a divider.

those things will help raise headroom some...

The spitfire has only one gain stage (paralleled tube at the input). Before adding a voltage divider at the gain/volume pot I would reduce that common 220k plate resistor to say 150k or even 120k. Since the tube sections are paralleled, a 220k is like 440k on a single tube stage. A lower plate resistor would give you more headroom - at the expense of some gain. It will also be beneficial if you use your guitar volume a lot since the treble roll-off will be lessened.

Cheers Stephan

[surfsup]

I see now. I was on my phone's small screen last night.

bean, do you have the line out wired up? I'd be interested to hear your comments about how well it works to send a signal into a delay pedal then into a 2nd amp set clean to add delay to a distorted signal? Or anyone reading that's done that, does it work well?

[bean]

Thanks guys! Now to convince the wife that I need to get back in the shop to do some very important tweaking to my amp... Ha!

I will definitely try some of these suggestions!

I did not wire in the line-out. So can't report on that.

I also wired the MV as a cut control with the addition of a cap (can't remember value off hand)

Thanks again! Will report back.... eventually!

[morcey2]

Normally, that is the effect of a Post-PI MV but in the Spitfire, most of the overdrive comes from the power tubes. There is only one gain stage ahead of the PI so the PI will never clip hard. I once rewired one of my amps to the Spitfire schem, and turning down the PPIMV drastically changed the amp's tone.

Agree on the bias resistor change.

I doubt that increasing the PI input cap yields a noticeable change. This is due to the fact that the PI input is "bootstrapped", i.e. the impedance is much higher than the 2 meg resistors. As a result the 3dB downpoint is already very low with the 0.01uf. A better place to gain bass is to increase the coupling cap after the first gain stage from 2n2 to 4n7.

Cheers Stephan

I'm talking about the other side of the PI, not the normal coupling cap coming from the previous gain stage. It's basically a bypass cap on the grid of the second side of the PI. Since that side is cathode coupled to the input side, it serves as an AC ground for the grid. Making it about 10X larger than the input coupling cap will even out the frequency response of the PI without making it muddy the way increasing the other coupling caps in the signal chain can. That's assuming that the other input isn't being used for something else, like mixing in another channel.

In experimenting with different values on an 18watt LiteIIb, (quite similar to the spitfire) if I made that cap double or less than the input cap, the amp was too bright. It's one of those things that can be clipped in to see the effect without any permanent changes to the amp.

Matt

[Darkbluemurder]

Matt

OK, you were referring to the "grounded grid cap". I agree, the bigger value is better here since the long tailed pair PI is a differential amp, i.e. the more low frequencies you bleed to ground in the other side, the more they get amplified.

I think the notion of making the input PI caps the same size comes from the Vox AC30 where the vibrato channel was fed into the other side of the PI - which is perfectly fine if the channels are out-of-phase with each other. Matchless did the same thing in the DC-30 which also has two channels. But they then kept on doing the same thing in all the other models, including the single channel amps and also in the Spitfire. This does not make as much sense since no signal input is fed into that side of the PI.

Cheers Stephan