Going to build a SE 2204. Ideas/suggestions please.

[Madison]

I am going to build a JCM800 type amp.
The front panel will be the same layout as the JCM800 but it will run a single EL34.
I will also be including VVR on this. (Somebody wants to "crank it up" in their apartment)
I'd like to run the VVR right off the master vol on the front existing MV pot if possible.
Or, if I must, I will add a pot on the back of the amp but I'd really prefer not to do this method.

Is there a SE 2204 out there as a reference someone has already built?
(It doesn't have to have the VVR, I think I can figure that out later)
Suggestions, links?

Thanks.

[TubeGeek]

I emailed you a copy of a SE 5W Plexi layout...enjoy!

[topbrent]

SE 2204 = AX84 High Octane: 
http://www.ax84.com/hioctane.html


Schematic, Layout, Drill Plan, Bill of materials:
http://www.ax84.com/static/hioctane/AX84_Hi-Octane_101004.pdf

High Octane Kits:  
http://shop.dobermanamps.com/product-p/kit-ho.htm



6550 SE version: 
http://shop.dobermanamps.com/product-p/kit-sel.htm

[Madison]

>>I emailed you a copy of a SE 5W Plexi layout...enjoy!

Might work.
It's very close.

>>SE 2204 = AX84 High Octane: 
This was one of my first thoughts but.... extra gain pot.(2 gain pots plus MV)
I suppose I could ace one?

I have built a couple of the AX84 P1ex from scratch. Liked 'em.

[ncusack]

You could also try one of the SE Plexi builds found on the sewatt forum. Here is a link to the build information and schematic.

http://guitaramplifierpcbs.com/PSE_BuildInfo.aspx

If you check out sewatt there are a number of threads where people have documented different tweaks. There are two versions one with a paralleled V1 and one with a cascaded V1. As far as I can tell it isnt a 2204 but its pretty close.

[Fresh_Start]

All of these designs have the tone stack feeding directly into the power tube and there is no negative feedback.  Plexi and JCM800 Marshalls both have NFB returning to the phase inverter which is between the tone stack and the power tube.  Having experimented with a NFB switch before, I believe that it is an important part of an amp's tone and response.  OTOH the Hi Octane and other designs obviously hit the power tube with a big enough signal.

I've been working on a single-ended Plexi idea for a very long time, but this thread just triggered an idea.  Could you use a concertina splitter in place of the PI and only connect the anode to your single power tube?  Use the cathode for NFB insertion?  Possible circuit attached.

The only question would be "What to do with the unused triode?"  You could have parallel triodes for the first stage and use something lower gain than a 12AX7.  Or have the "extra" triode as a switchable boost in parallel with either the first or second gain stage.

Before hitting "send" I went through my design folder to look at other ideas and came across an "AX84 Plexi SE" design.  Apparently I'm not the first person to think in terms of a concertina splitter with a single output.  There's a boost stage after the tone stack, a switchable LED clipper, and an interesting fixed/cathode bias switch.  I'm not sure why, but the AX84 designers seem to be allergic to NFB. 

The other SCH file is my NOT-built design so far.  If you use ExpressSCH, this file might be helpful as a starting point regardless of whether or not the design actually works.  Note that this circuit shows a different approach to the psuedo-PI, basically dumping signal through two voltage dividers: one before the grid and the second a split plate resistor.

This probably just muddies the water, but at least the SCH file might be useful.

Cheers,

Chip

[tubenit]

Chip,

That is a very intriguing idea!  Thanks for sharing it.

With respect, Tubenit

[ncusack]

I know there has been a lot of discussion about a "fake PI" on the amp garage forum and I know one of the engineers from Tonic Amps has weighed in on that idea for a trainwreck style SE amp he built. I think its called the Absynth and its like 30W with EL34's.

In any case from what i've read they all say that the PI is a huge part of any amp and it never hurts to simulate it. How they implement it is another question I haven't gotten to answering yet but what you have there looks like it would be worth trying Fresh_Start.

[Fresh_Start]

I know there has been a lot of discussion about a "fake PI" on the amp garage forum and I know one of the engineers from Tonic Amps has weighed in on that idea for a trainwreck style SE amp he built. I think its called the Absynth and its like 30W with EL34's.

In any case from what i've read they all say that the PI is a huge part of any amp and it never hurts to simulate it. How they implement it is another question I haven't gotten to answering yet but what you have there looks like it would be worth trying Fresh_Start.

Thanks ncusack! [/sarcasm on/off]  I just spent an hour or so searching for "Absinthe" and then "fake pi" on Amp Garage.  AX84 search didn't turn up much additional info.  There's a circuit called "Derailment" and another called the "Dirty Little Monster" (gotta love that name, eh?).  The consensus among the Trainwreck folks is that the PI is an essential part of that sound.  Makes sense to me.  The question is how to get there. 

Dartanion (Tonic Amps -> Absinthe) appears to use a full LTPI and simply uses only one of the two outputs.  But it's a commercial circuit so no schematic available.

Others use 1 triode, hence "Fake PI", but in a wide variety of ways.  That AX84 schematic I posted is even weirder than I thought before - negative voltage on the cathode of that last stage...  There are a lot of regular common-cathode gain stages with really cold bias like 100K plate resistor and 22K cathode resistor.  That shouldn't have much gain, but I don't know what it sounds like.  There's also a version similar to the PTP SE-34 circuit I posted earlier, except there's a big resistor under the cathode/grid return resistor junction.  I'm not sure what that's all about.

There does seem to be some consensus that parallel triodes for the first gain stage to reduce noise is a good way to utilize an "extra" triode.

Madison, if any of these rambling thoughts interest you, I can keep digging.  Otherwise I'll wait until I pick up my project in a month or two.

Cheers,

Chip

[LooseChange]

My design just eliminates the cathode follower. I drive the tone stack from the plate of V2a and run the volume pot into the grid of V2b. Set V2b like a typical cathode biased triode sending the plate to the grid of the power tube. At times I've added a master on the plate of V2b.
Now you can send the NFB to V2b.
Simply stated: Wire up V2 to be a Champ and change the cathode resistors and caps to match the 2204... Or tweak to taste.

[Madison]

Madison, if any of these rambling thoughts interest you, I can keep digging.

>>No no, it's all very intriguing and educational.
Ramble away.
I am getting closer to knowing.

BTW.
I PMed you.

[tubenit]

Only had about 5 min to draw this ............... but here is an idea for you to consider.

I took my Carolina SongBird and rearranged some of the topology for you to make it more 2204-ish. So I'm confidant the PT & OT should work for you.

This would give you a volume, tone stack into a "trimmer", and then a master volume (& MV tone stack).  Should give alot of tonal choices.

Plus it has a mid-boost, PAB, and a volume bright switch.

With respect, Tubenit

[Madison]

Thanks Tubenit.

Some great ideas in there.
I am going to have to stick with the standard JCM800 front panel on this project so I am limited to no addional switching, etc.

I'll have a closer look at what you did.

[tubenit]

Well, what I posted isn't really 2204 topology.  The first two gain stages are somewhat Marshall-ish but with an additional gain stage following.

But I drew it that way because I have more confidance that it would work and sound good.  I tried a "hot switch" in a Marshall topology once prior to the 2nd gain stage (which would be closer to 2204 topology) & I never could get it to sound right to me.

My guess is you're probably looking for something different than what I posted to consider?  But if not, you could modify the layout for the Carolina SongBird to match the "Madison's idea" schematic.

IF you used a single (mini-toggle) DPDT with center off ............. you could wire it for the mid-boost one direction and the PAB on the other direction.  Both are great additions to tone, IMO.  You could hardwire the volume bright cap.

With respect, Tubenit

[LooseChange]

Madison, Take a closer look at what I explained above... It works well and is 2204 topology.

[Madison]

>> Take a closer look at what I explained above... It works well and is 2204 topology.

Yes yes!
Very close to what I am looking to do.
This seems like the right way.
I am thinking the TMB plus MV after V2b.
NFB (adjustable. presence?) as you stated.
Thanks!

[jojokeo]

I'm sure this will work but the CF with the tone stack immediately following it is largely responsible for giving the JCM800 it's "Marshally" signature sound. I've messed with placing the tone stack in other places using the same circuit and it does indeed change the sound and the way the controls interract. With it placed earlier in the circuit ala Fender, it yields a more bluesier type of disortion/overdrive - not the same crunchiness. Just something to consider or be aware of.

[Fresh_Start]

I sent Madison a PM with this topology but thought I'd share it after jojokeo's observations:

Gain 1 - parallel triodes
Voltage Divider
Gain 2
Volume
Gain 3
Cathode Follower
Tone Stack
Master Volume
Gain 4 - NFB & Presence
Power Tube

Maybe use a 5751 or even lower mu tube for V1.  Set up the rest of the preamp through the Master Volume just like a Marshall 2204, then add Gain stage #4 set up like a BF Champ with an insertion point for negative feedback.

Just a thought.  Sorry I don't have a good way to do a block diagram.

Chip

[phsyconoodler]

ok,I added my 2 cents in another thread,but here goes.Why single ended in the first place?Build a low-voltage 2204,about 250v on the plates and bias it class A.I bet it would kick ass and make the SE sound like a dinky toy.
  Easier,lighter,cheaper and killer tones.

[Madison]

ok,I added my 2 cents in another thread,but here goes.Why single ended in the first place?Build a low-voltage 2204,about 250v on the plates and bias it class A.I bet it would kick ass and make the SE sound like a dinky toy.
  Easier,lighter,cheaper and killer tones.

Great idea for sure.
I have already committed to a single EL34 with 2204 front end.
I have decided I am probably going to do something like what Loose Change described.
The guy wants cranked JCM800 @ below 1W.
A real challenge.

[Matty_V]

Below 1W? I guess not every electric guitarist likes to blow the doors off their hinges huh?  I don't really see the point in that.

[Madison]

>>I don't really see the point in that.
I do; the customer is always right.

[topbrent]

This one is a fun little monster, but it is a plate fed tonestack which gives a bit of a more aggressive attitude. 

Basically the standard list 80's hot rodded JCM800 mods in a single ended package.

[Madison]

Yup.
That's the one.
I will do a few tweaks.

[Fresh_Start]

I didn't feel comfortable posting that Allen schematic, but since someone else did I've got a couple of questions:

What exactly is the 100K/470K connection between the plate & the cathode of the third gain stage doing?  Looks like a form of local feedback, plus a voltage divider to reduce the signal swing.  However, the NFB ratio looks really low (4.7K / [4.7K + 100K + 470K]).  Assuming the purpose is to reduce the signal, why is this better than other approaches?  I've got complete faith in David Allen - just want to understand.

The other thing that puzzles me is that the tone stack is inside the negative feedback loop.  So if you use the Treble control to cut high end frequencies, there will be relatively less high end in the NFB signal and the high end that's left will be corrected less by the NFB, right?  Heck, I don't know enough about NFB to understand this.  Anybody want to help out?

Cheers,

Chip

[tubenit]

Just another option to consider.

With respect, Tubenit

[DummyLoad]

The output tube grid connected to a pot wiper could be bad news if the wiper were to become intermittent or oufright fail.

--DL

[topbrent]

there is a second version of the class act with fewer knobs, also:

[Madison]

Tubenit, thanks for the SCHs!They will come in handy.
I don't suppose anyone has a layout for the ClassAct?

[topbrent]

Here is the "fewer knobs version..." of the class act aka the class act 2.

[Madison]

One more thing and yet more advice needed.

The front panel will be a standard plexi JCM800.
V/MV/T/M/B/P

I will be putting VVR in this amp.
Should I:

option 1- get rid of the normal MV and just have that be the VVR control (preferred just for cleanliness)
Option 2-have an additional VVR pot on the rear of the amp as a sort of a Grand Master Volume?

What are the advantages/disadvantages?

I am not sure I need the regular MV if there is VVR?

[jojokeo]

I will be putting VVR in this amp.
Should I:

option 1- get rid of the normal MV and just have that be the VVR control (preferred just for cleanliness)
Option 2-have an additional VVR pot on the rear of the amp as a sort of a Grand Master Volume?

What are the advantages/disadvantages?

I am not sure I need the regular MV if there is VVR?

option 1 = YES - for SE big tube in my amp having both VVR and MV is redundant. The VVR is a better "MV"
option 2 = Not necessary - one VVR is all you need.
Again all you need is a single VVR for a single ended amp like this. It works great as you master volume and overall output power "limiter". The lower you dial your voltage the easier it it for the preamp to distort/overdrive the power tube.
You do not need a regular MV as a result. For higher power PP amps then yes, have a MV and VVR but not w/ the SE el34/6v6 IMHO.

[LooseChange]

I think the MV is needed with the VVR. When the VVR is turned way down and the preamp is cranked you will overload the grids of the power tube and you won't like the tone.