How do I mod this 18W to a 6L6 or EL34, or give it headroom in general?

[paintballnsk]

Hello,

I have a lot of questions, so please bear with me.

With your help from a previous post, I've heavily modded this 18W amp. Please see attachment for the mods I made (and if you see anything messed up that is "wrong" please tell me). I'm aiming for a modern rock/metal kind of sound. I typically aim for moderate gain, and hit the amp with a TS808 overdrive. I implemented some design ides from the Splawn schematics I've seen on the Quickrod and they helped with a few things a lot. The original design was very flubby and boomy and an unreasonable amount of bottom end. I basically just experimented with a bunch of components to dial it in to where I like it.

However I'd like to to have a little more pick snap and decay and "bigness". It's a hard term to describe, but it feel to me like the notes don't swell and decay very well. Turning the gain down doesn't seem to help this. They lose distortion, but the mushy feel is still there. Note that the gain knob has a push/pull pot that bypasses some or all of the second gain stage. The compression is still felt on this clean channel as well. It doesn't seem to be an EQ thing at least as far as the final output is concerned. The bottom end is still a little on the warm side, but there's still plenty of upper mids and highs.

I'm thinking of modding it to take 6L6's or EL34's, OR maybe switching it from a solid state rectifier to a tube rectifier. I very much prefer solid state rectification on my bigger amps (6505, Mesa Roadster, Mesa Mark IV) But maybe on EL84's this is just taking the compression too far. I'm not sure.

Also any suggestions for the preamp section to fix this would be appreciated. Am I pushing any of the stages too hard and killing my headroom? I've played a little bit with the bypass caps, but I'm still a little unsure how the resistor values play into what it really does to the sound. From what I've read the larger cathode bypass resistor like a 2.7k put it a little more towards asymmetrical clipping, where the 1.5 will lean towards symmetrical clipping. Also the 1.5 will cut about an octave more bass than the 2.7 assuming the same capacitor value. Should I instead be measuring something to make sure that value is biasing the tube more correctly? I have noticed with the mods I've made that the gain knob will cause a massive high end drop if put above about 3:00 (Miller Effect?), and the Earth knob will motorboat if put to 0.

What would I need to do? Is it as simple as changing the output transformer and tube jacks, or is there a lot more to it than that?

Thanks,
Nick

[EDIT]:
The main issue was the input grid series resistor. It was 56k. I changed it to 66k and it really opened it up. Here's a clip I threw together today to demonstrate the update:
https://soundcloud.com/crossfirerecording/zedmod_1

[Sowndman]

3 gain stages is a lot for el84's. i have an amp with 1 12a and 2 el84 in self inveter. it will distort quite easily.

Can you describe the sound you want with a song or recording from a particular artist ?

[paintballnsk]

3 gain stages is a lot for el84's. i have an amp with 1 12a and 2 el84 in self inveter. it will distort quite easily.

Can you describe the sound you want with a song or recording from a particular artist ?

I'm kind of going for something like a 6505 and Splawn Quickrod. The Quickrod feels a bit like a Marsahall 1987 re-issue but much more gain on tap. And what I mean by that is it's pretty "plucky". That initial pick snap punches right through. The 6505 has a ton of gain, but somehow doesn't sound mushy and compressed, and really has great palm muted chugging type tone. To me it's a hard amp to beat.

I probably couldn't give you an artist. I want this to be my own thing. I could probably find you 5 different artists that use a 6505 and all sound completely different. All I know is that with this 18W, when I palm mute, it's tonally "there", but it just doesn't punch very much.

[HotBluePlates]

... I'm thinking of modding it to take 6L6's or EL34's ...

Takes new tube sockets, output transformer and probably power transformer to do that.

... maybe switching it from a solid state rectifier to a tube rectifier ...

That's going in the wrong direction for what you're saying you want.

... Turning the gain down doesn't seem to help this. They lose distortion, but the mushy feel is still there. ...

Did you ever just bypass the whole Earth control? Clip or solder a wire from the 47kΩ feedback resistor to the top of the 5kΩ Presence ("Sky") control.

And I'm not sure you ever said what speaker setup you're playing through. Naturally, if it's an open-backed cabinet, well...

[Sowndman]

I think you need more watts. 18 watts will work for credance and even zz top but I think you are needing 50 watts to have the headroom you desire. you might rya a kustom defender 50. it will take 6l6 or EL34's factory.

[Paul1453]

... I'm thinking of modding it to take 6L6's or EL34's ...

Takes new tube sockets, output transformer and probably power transformer to do that.

I was thinking the same but waited for a more experience voice to chime in.
Most 6L6 designs I have seen run around +450Vdc for the B+, something this PT is not capable of.
2 6L6s PP can put out about 50W output, current OT probably can't handle that either.
Not a simple plug and play for bigger output tubes.

You might be able to go to 6V6s with what you have now.   

[PRR]

> as simple as changing the...

Funny(??) story. My neighbor got in a 1939 Hudson, straight-Eight engine, been sitting for 50 years. The original rating was 128 HorsePower (pretty spiffy for 1939). (Among many other things) the carb was bad, bummer! It happened he had a new carb for a 1963 Plymouth, semi-hemi engine rated 250 HorsePower, which dropped right on (bless SAE standards).

As we all know, you need a big carb to make big power. Is he expecting 250 HP now? No, because the valves and cam and pistons and crank are all still scaled to 128 HP. He might get 130 HP screaming, but nowhere near 200+ HP until he bores and strokes and ports and puts in an Isky cam, then replaces all the bottom-end crank and rods the engine spits-out under the increased strain.

(In this case, he just wants it to "run". Sunday drive, maybe a parade. Short-term, the 250HP carb works well enough for him to find out what else is wrong (it has a bearing knock). If the client can afford to fix what was wrong when it was parked and what went bad in storage, then he can get the original carb fixed so it runs the way it was meant to.)

OK, tubes-and-OT may be like carb-and-cam, not just a carb. But think weakest-link, or most restrictive-link, of a chain. Every chain link has to take full pull. Every major circuit element has to pass full power.

Any of the usual tubes will "work" with change of socket and bias resistor. They won't make more power without a PT change. At this point you have changed everything (preamp, tubes, all iron). Maybe this is just not the amp for you?

[Paul1453]

I like that car analogy, PRR!

As you said maybe this just isn't the amp he wants.

I know the 6V6s aren't going to give him more power, but maybe the palm muting sound he wants?   

The 6V6s seems to me like the only viable option to be able to still use the PT and OT?   

Probably still won't give him the sound he wants, but a much cheaper try at it.
Before he finally decides to just sell this amp and get what he really wants.   

[paintballnsk]

Well I'm not giving up on the amp. I have an amp tech in town that can do it for me, but this has become quite a tinker toy for me. Buying another amp isn't really a solution to what I'm trying to accomplish. I have other amps that can do the job, but that's not the point haha.

I'd like to learn how to do it myself. I don't mind putting a couple hundred into new parts to make it happen, I just want to learn how to do it, or if you think there's anything in my mods that could be causing the problem, please let me know.

I'm playing through a Splawn v30/k100 loaded 4x12, and I've been testing mostly between Les Paul, PRS Single Cut, and boosting with an Ibanez TS 808.

I will try as you suggested by bypassing the Earth control and see how that goes. I wouldn't mind re-purposing the knob for something else if that happens to work out. Would that bypass that 22uf cap as well? Can you draw me a spot on the diagram to help me out so I'm clear. I don't wanna blow something up :)

From what I'm reading, the feedback loop is to help the power tubes drive into distortion sooner. I would imagine EL84's don't really need any help doing that. Can I just decouple that whole component?

Also, looking at the schematic, what was the original design trying to do after the second gain stage between V1b and V2a? I don't understand why it originally had 1M and 220k resistors before the grid.

[Paul1453]

I don't mind putting a couple hundred into new parts to make it happen, I just want to learn how to do it, or if you think there's anything in my mods that could be causing the problem, please let me know.

From what I'm reading, the feedback loop is to help the power tubes drive into distortion sooner. I would imagine EL84's don't really need any help doing that. Can I just decouple that whole component?

For a couple hundred I think you could pick up a conversion project on Ebay that would more likely give you the results your looking for with a PT and OT designed for a pair of 6L6s.   

Then you'd have two amps not just one and some parts.

I think you have the NFB loop concept backwards.  Disconnect it and I think you will see what I mean.   

[VMS]

I think the first thing to do is figure out is it the preamp or poweramp that is limiting the headroom.

With master volume almost all the way down how high can you turn the gain knob before it starts to distort?

[HotBluePlates]

I will try as you suggested by bypassing the Earth control and see how that goes. ... Would that bypass that 22uf cap as well? ...

Yes, 22uF cap and Earth pot get disconnected (you could just use a wire to bypass, but if you're unsure just disconnect them); see next section & below.

Why? You keep taking about the bass being loose. Negative feedback tightens the outpu stage's control over the speaker, but throws away some signal strength to do that. The typical Presence (Sky) control reduces the amount of NFB at highs to boost highs back to where they'd be without negative feedback.

The Earth control is the same at lows: it reduces the NFB to some very low frequency to boost that range relative to mid-frequencies. But that also loosens the feedback's control of speaker-flap.

... Can you draw me a spot on the diagram to help me out so I'm clear. I don't wanna blow something up :) ...

See below. Just connect the 47kΩ resistor directly to the non-grounded side of the Sky control. You can unsolder the Earth pot & 22uF or not.

From what I'm reading, the feedback loop is to help the power tubes drive into distortion sooner.

Feedback reduces distortion in the output stage. However, you may have read that EL84's would normally have a slow, smooth onset of distortion, but when feedback is applied the distortion occurs at higher signal levels and the transition from clean to distorted is more sudden.

Imagine a gas pedal that moves your car 20mph at 1/4-down, 40mph at 1/2-down, 60mph at 3/4-down, 80mph at full-down. Feedback is added to give you 20mph at 1/4-down, 25mph at 1/2-down, 30mph at 3/4-down, but 70mph at 7/8-down and 80mph at full-down. You stay slow (clean) a lot longer, but when you push far enough you very suddenly transition to high speed (distortion).

What is really happening is the signal input to the output stage is getting bigger & bigger, and eventually overwhelms the feedback's ability to keep the output stage clean, at which point there's a sudden transition into distortion.

Separately, feedback also lowers the apparent source impedance of the output section of the amp, raising the amp's damping factor, which tighten's the amp's control over speaker motion. If you push the output stage with feedback into distortion, that speaker damping is also out the window (you've overwhelmed the feedback already, and so undermine all of its benefits).

[jjasilli]

If I understand you state 2 issues:  too much bass & lack of headroom.


too much bass:  Your preamp mods seem good to tame bass.  But, the tonestack is not set for Fender tone, which you may prefer:  especially the large treble cap & mid pot.  Maybe try Fender values for a cathode-driven tonestack like, for e.g., the '59 Bassman.


lack of headroom:  Your preamp mods also reduce gain.  As a result the voltage divider after the treble pot may now be robbing needed signal strength.  That being said, 18W has only so much headroom if you need to compete with loud bass & drums.

[Paul1453]

Sorry if I'm off target or missing your point, but instead of trying to modify the family station wagon into a street rod.

Maybe starting with an old barn find street rod would be better?

For $125 including shipping this might be a platform to consider?   

http://www.ebay.com/itm/NICE-1962-CONN-95007-Tube-Amplifier-Chassis-6L6-6V6-7199-Guitar-Amp/131642405409?_trksid=p2047675.c100009.m1982&_trkparms=aid%3D777000%26algo%3DABA.MBE%26ao%3D1%26asc%3D34494%26meid%3D3a971a5b8bd64ee2a9dc605b3f136e62%26pid%3D100009%26rk%3D1%26rkt%3D1%26sd%3D141845493893

I'm sure there are others on Ebay that might be even better for your proposed upgrade.

[paintballnsk]

Thanks for the clarification on bypassing the earth control. I will try that when I get he tonight.

And thanks for the links to the amp chassis. That looks like a bit of an undertaking for a noob like me. If it was at least laid out on a turret board I think I could figure it out.

As for playing live, this is intended to be a studio amp first. As it is, I don't really care if it doesn't carry a drum kit, though it has gotten me through a few practices just fine.

[paintballnsk]

Bypassing the earth control made everything monumentally better.

After bypassing the earth, I noticed the Sky/Presence control motorboats below about 90%, and the master volume doesn't do much through the first 3/4 of the turn. It gets really loud after about 3:00 though. I think I modified that cap .1 cap on the bottom side of the LTPI after the Sky control to a .022 and didn't have it in that diagram. I'm pretty sure it's a .022 now. I'll change it back to a .01 and see if that fixes the motorboating.

[Willabe]

I think I modified that cap .1 cap on the bottom side of the LTPI after the Sky control to a .022 and didn't have it in that diagram.

Why did you do that? Not that's there's anything wrong with trying it. But, what were you thinking would happen?

[paintballnsk]

It was a suggested mod from our other thread. It helped with the tone at the time.

[Willabe]

Maybe so, I don't recall that or why it was suggested. 

[paintballnsk]

No worries, it wasn't one of your suggestions, and .. well.. I'm a self proclaimed newbie, so I'm learning :)

While I have your attention, what do those plate resistors do between V1B and V2A? The 1M is bypassed, but I don't understand what the 200 is for. Was the voltage too high for the 3rd gain stage? Why not just lessen the 330k?

[Willabe]

The 1M is not a plate R. The B+ is feed to the tubes plate through the plate R's. In your amp 100K and 220K.

The R's after the plate coupling caps going to the next tubes grid are there to knock down the AC (guitar signal) so it doesn't SLAM the next tubes grid to hard. So those R's throw away AC signal.

At this point I would strongly suggest that you need to do some reading on basic tube amps circuit function. You are still confusing what part is doing what/why it's there and what it's common name is. At the same time your wanting to mod a very high gain amp, which are more complex then normal gain amps. In other words 'you have to learn how to walk before you run'.

There are many free internet sites on line, but this is a very good place to start;

http://el34world.com/schematics.htm       

[paintballnsk]

Right, but then what is the 330k doing? Does that just divert more power into the grid of V2A? That's why I don't understand why there's both there. Why not just lower the 330?

[Willabe]

Right, but then what is the 330k doing? Does that just divert more power into the grid of V2A? That's why I don't understand why there's both there.

No, not right. You still don't understand/see the difference between a plate R or a series R or a grid leak R or a cathode (K) R .

It does NOT divert more power into the grid of V2A, it 'diverts' LESS ACV signal into the the grid of V2A, because that grid leak R changes the bias of that triodes gain.

 This is Exactly what I just wrote you about, you don't understand. You need to do some home work for yourself.

Your not seeing/understanding the very basics of a single triode gain stage. The 330K is a grid leak/grid return R that is there for that triodes bias set up. If/when you learn that, then you'll see/understand what the extra parts (resistors/caps) are there for. If you don't understand that, then how can you understand how to 'tweak' an amp for what you want to hear?
 

Why not just lower the 330?

You could/can do that, but....... there's more to it. Like do you lower that R's value OR do you lower the series R going to that tubes grid? Different things will happen doing either 1 or both.

Your shooting in the dark because you don't understand the very basics.

You can learn a LOT of this on your own, if you put in the time.

Do you really think it's right/fair for the forum members here to teach you every little single thing you want to know about tube amps without doing any basic work on your own? 

The way I see it is, many guys here have spent more time helping you and answering you questions, then you have spent researching the basics for yourself just to be able to understand our answers.

You are certainly welcome here, as you've experienced, but some of this you have to do for yourself.   

When I was a teenager, my car brakes went out, 4 wheel drums, with all those springs, with out auto shop tools designed for pulling them AND putting them back in, although, putting them back in was way easier.   

I went to my father and asked him for the money to get the brakes fixed. He laughed at me, he was right to do that. Said "I'll give you the money for the parts for you to fix them". And he didn't stand over my shoulder the whole time, he just let me work through it and if/when I got stuck, I would go to him and THEN he'd set me straight. He knew what he was do'in.   

[John]

http://www.valvewizard.co.uk/gainstage.html


Highly suggested reading.

[jojokeo]

I'm a little lost in the weeds and can't spend the time to read over every detail in this thread but:
 
Dude - more headroom = don't use 12aX7s!!! Simple and easy. Use 12aY in V1 for sure. Put a 12aT in the phase inverter slot V3. Remove the bypass cap on the cathode follower. Replace the 10K tail resistor on the phase inverter tail (you're driving the power tubes way too hard) to 47k or 56k for starters. Don't touch the other one at this time. These simple items will help greatly without anything else to your original circuit.

[paintballnsk]

http://www.valvewizard.co.uk/gainstage.html


Highly suggested reading.

This is an awesome read. Thanks for the suggestion. Let me give this a thorough read before I ask any more stupid questions and frustrate you. Trust me, this is more frustrating for me than it is for you and I apologize. I asked you to be patient with my stupid questions :) I'm not blindly asking. I'm asking because I tried reading up on them and either couldn't find a specific answer to my question, or didn't understand it. Sometimes I get some supremely intelligent and helpful answers like you've given me, and I guess I have a hard time following them, but then I have to weed through a few other really frustrating and unhelpful replies like "just buy another amp"...

I miss-spoke when I said plate resistor. I just meant the resistors coming out of the plate on V1B. I thought you knew which ones I was talking about, but I should have asked why are the values on the V2A grid stopper R and grid leak R so high? I can't find in any diagram that has 1.22M of grid stopper resistance. I would image that was a huge contribution of why my guitar's tone knob sounded like it was completely rolled off when I first got the amp. So I'm asking because I'm concerned I should be checking for something. Maybe he putting 1M resistors in there to cut highs, or maybe band aid a problem I'm not seeing.

[Willabe]

Maybe he putting 1M resistors in there to cut highs, or maybe band aid a problem I'm not seeing.

From reply #20;

The R's after the plate coupling caps going to the next tubes grid are there to knock down the AC (guitar signal) so it doesn't SLAM the next tubes grid to hard. So those R's throw away AC signal.

[Willabe]

but I should have asked why are the values on the V2A grid stopper R and grid leak R so high?

See quote in the above reply.

There is no grid stopper on V2A. The 1M in series between V1A and V1B and the 1M and 220K in series between V1B and V2A are not grid stoppers.

This is a hi-gain amp and because of that he's putting in those hi value series R's to knock down the AC signal from SLAMMING into the next tubes grid which will cause a very nasty sounding distortion called grid blocking distortion. Each gain stage can only take so much input before driving that gain stage into grid blocking distortion. It is referred to as 'gain scheduling'. There are different ways to achieve this, why he chose to do it the way he is......         

A grip stopper goes directly to the tubes grid after the tubes grid leak R. It is 1 of the methods that can be used to knock down some AC signal. But grid stoppers are also used for other reasons. There's no need to knock down the input signal with a grid stopper because it's such a small AC signal coming in.

The 58K on V1A is a grid stopper and the 2.2K's on the power tubes grid are grid stoppers.

Read this on grid stoppers;

http://www.valvewizard.co.uk/gridstopper.html

[VMS]

There are different ways to achieve this, why he chose to do it the way he is......         

Probably because buzzbomb preamp is pretty close to this one:

http://el34world.com/charts/Schematics/files/soldano/soldano_atomic16.pdf

...but I have always thought that there is an error in that schematic and that the 220k should really be a grid stopper, as we can see on other soldano amps.

[paintballnsk]

There are different ways to achieve this, why he chose to do it the way he is......         

Probably because buzzbomb preamp is pretty close to this one:

http://el34world.com/charts/Schematics/files/soldano/soldano_atomic16.pdf

...but I have always thought that there is an error in that schematic and that the 220k should really be a grid stopper, as we can see on other soldano amps.

Holy cow, they're almost identical! Thanks for the find!

[paintballnsk]

I've dialed in a few things for tone and gain, and they're about where I like them. However still no "snap"..

So I played with the volume knob on my guitar and I was like "Hmm, I'm not crazy about how that sounds, but I like better how it feels."

Turns out the first gain stage, even though it's very clean, was still compressing the signal quite a bit. I'd more prefer fair amount of snap on it. So, I looked up the Marshall schematics and noticed on the High inputs they have a 68k resistor. I added 10k to that 56k resistor and it made a really big improvement. The bottom end is now a lot punchier instead of warm and mushy

Thanks again for the help. I'll keep you posted.

[Willabe]

So, I looked up the Marshall schematics and noticed on the High inputs they have a 68k resistor. I added 10k to that 56k resistor and it made a really big improvement. The bottom end is now a lot punchier instead of warm and mushy.

Good work. 

[paintballnsk]

I updated the OP with some notes and a sound clip I threw together to show the metal tones I got with it.
https://soundcloud.com/crossfirerecording/zedmod_1
Thanks!

[John]

I added 10k to that 56k resistor and it made a really big improvement.

Not sure why that would make such a big difference. Or did you parallel the 10K with the 56K?

[paintballnsk]

Series. Huge difference.