Marsha Valve (inspired by Fetzer Valve)

[vanhansen]

Thank's to ROG's Fetzer Valve, I became inspired to make a little circuit similar to it but using the front end of the Marshall JCM800, the Marsha Valve.



This schematic is the first prototype.  I have since made some revisions to it but haven't tested them out yet.

EDIT: Schematic above is now version 1.1.

With this one, I tested using a NTE458 (watch the pinout, it's DGS) but a J201, MPF102 or similar should work perfectly.  The drain is biased to 4.5V on the dot.  Source was something like .53V, Gate at 0V.  

The 10k source resistor actually comes off the second half of first the 12AX7 to ground in the JCM800 amp schematic but I wanted to see what this would do.  Had to use two 47n caps in series to get as close to 22n as I could.  I also omitted the 470k/470p network and the 1n treble bleed cap on the Volume pot at first but that will be added in during the tweaking stages.

My initial thoughts during design and breadboarding were if it was just going to filter out the lows like my Fetzer Valve build does.  To my surprise, it actually gives a nice full range boost (at least while listening with headphones).  It didn't start to break up until the volume was nearly maxed which I liked.  It really added a nice punch.  On the dirty channel of my little practice amp, I had the gain set to around 6 and when I kicked this circuit on, it really pushed it hard and it sounded great.

Yeah, there's a million boosters out there and stuff like this, but after hearing how well my Fetzer build turned out, I just had to try it myself.  Once I get the other designs tested and tweaked, I'll share them too.

Thanks, ROG for being an inspiration.  Great designs and a great resource for learning.

[ragtime8922]

There it is. Cool. In what ways do you feel it differs from the Fetzer? More gain, less gain? Brighter/Darker?

[vanhansen]

There it is. Cool. In what ways do you feel it differs from the Fetzer? More gain, less gain? Brighter/Darker?

It's got a little more gain using the NTE458 so the JFET used will definitely change the amount of gain.  It sounded a little darker than the Fetzer to me.  It didn't sound like a treble boost or a bass boost, it just boosted, and fairly clean and warm too.

[petemoore]

Notes from 458 "Fanboy', nice sounding transistor, higher gain than the J201, in my little 1 or 2 transistor builds they sound great ! Expensive though.

[vanhansen]

Hey pete.  Yeah, the 458 sounds real nice.  I snagged the last two that were at Fry's last time I was there "just to get trimpots".  Yeah, right.  A few packs of resistors and the JFETs too.    They were definitely more expensive but quite a bit compared to J201s but I didn't care.  Had to try them out.

[vanhansen]

Tweaking Update:

Forget about me saying it boosted clean.  Headphones can be deceiving.  Heck, I first listened to it at midnight so I had no choice.    Anyway, the prototype has a good amount of dirt.  Pushes the amp input real nice.  The only thing clean about it is it's not muddy...that's good.

Couldn't get it to bias with a 2.2k/.47mf combo from source to ground.  It was the closest I had to 2.7k/.68mf that the JCM800 has.  Threw the 10k resister back in and it biased right away.  No problem.

EDIT:  Threw the 2.2k resister back in and got it to bias to 4.5v.  I wasn't turning the trimpot enough. :oops: At first it was .23V at the drain and source.

The 470k/470p network after the 22n cap from drain isn't going in.  It brightens it up way way too much.  It sounds a lot better without it.

More later.

[vanhansen]

Tweak Update #2:

I stayed up way too late playing with this thing.  While messing with it I decided to add another stage to it since I had another NTE458 sitting there.  So, I added it after the gain pot.   100k trim to drain, 22n cap from drain to output, and 10k source resister to ground.  All this is is the second half of the first preamp tube.  Take a look at the JCM800 Emulator and you'll see its really Q1 and Q2 minus the 470p/470k networks.

Ran in to bias troubles on Q2 but was finally able get it right using a 200k trimpot.  It was weird.  With a 100k, if I had the first stage connected to the gate, the voltage at the drain would jump to over 8 volts.  If it was disconnected, it would drop down.  I swapped Q1 with Q2, same thing.  Changing out the trimpot finally got it.

My wife said it was Hendrixy when I played it on my amp's clean channel.  I'll take that as a good sign.  

Here's the single JFET final schematic.  Sounds great.  Very reactive to your picking.



Again, watch the pinout of the NTE458, its DGS rather than DSG like J201's, MPF102's.  There is a NTE458 with DSG pinout but Fry's didn't have it when I picked them up.

[vanhansen]

Hmm, almost 300 views and 2 comments (my updates excluded).  Guess there's no interest.    :cry:

[Arn C.]

Hey Erik!  I am interested, just damn busy.   I will try to perf this up today and maybe if I get a chance, I will have it ready to fire up tonight when I get home.   Thanks for another cool idea to try!

Peace!
Arn C.

[vanhansen]

Hey Erik!  I am interested, just damn busy.   I will try to perf this up today and maybe if I get a chance, I will have it ready to fire up tonight when I get home.   Thanks for another cool idea to try!

Peace!
Arn C.

Hey Arn.  Thanks for the reply.  Let me know what you think about it.  Once I have the two stage version done I'll post that in another thread.

One thing I can tell you is this version boosts the clean channel of my practice amp like mad.  On the overdrive channel, if I turn the gain on the amp down to 5 or even 3 and kick this on, it pushes it real nice.  I was surprised to say the least.  There was some JCM800 character in there.

[Arn C.]

Erik writes:

Tweak Update #2:

I stayed up way too late playing with this thing. While messing with it I decided to add another stage to it since I had another NTE458 sitting there. So, I added it after the gain pot. 100k trim to drain, 22n cap from drain to output, and 10k source resister to ground.

You say you added another stage after the gain pot?  Do you mean after the existing volume pot? and do you have a sketch of how you did this?

Thanks!
Arn C.

[vanhansen]

Erik writes:

Tweak Update #2:

I stayed up way too late playing with this thing. While messing with it I decided to add another stage to it since I had another NTE458 sitting there. So, I added it after the gain pot. 100k trim to drain, 22n cap from drain to output, and 10k source resister to ground.

You say you added another stage after the gain pot?  Do you mean after the existing volume pot? and do you have a sketch of how you did this?

Thanks!
Arn C.

Yes, from the wiper of the existing volume pot to the gate of another JFET.  I'm not done debugging it though.  Ran in to a bias problem on the second JFET.  I thought I had it fixed then it popped up again. seanm helped me out in another thread with it.  Now I just need to implement the suggested fix tonight to make sure it fixes the problem.

Once it's all debugged and tweaked, I'll put that schematic in my gallery.

[Arn C.]

Will be waitin

Peace!
Arn C.

[Clipped]

Will be waitin

Peace!
Arn C.

.....and Buildin'


Nice work Erik.

Ron

[RLBJR65]

Thank's! looks interesting. Don't know how soon I will get around to building it, but here is a quick vero layout.
http://aronnelson.com/gallery/Richard-Boop-RLBJR65/Marshavalve

[vanhansen]

Thank's! looks interesting. Don't know how soon I will get around to building it, but here is a quick vero layout.
http://aronnelson.com/gallery/Richard-Boop-RLBJR65/Marshavalve

Hey, that's cool man.  Thanks.

[MrChameleon]

I'm interested too. I really like the ROG Fetzer Valve, and will build your Marsha soon, too.

Thanks for contributing  

[d95err]

I'm looking at building a some kind of booster and I've been looking at the Fetzer before. Since the schematic is very similar to the Fetzer, I suppose it would be possible to make a booster that is switchable between Fetzer and Marsha.

The volume pot is 1M on the Marsha and 100k on the Fetzer. Would it be possible to use the same (1M or 100k) pot for both designs, just switching the resistor/capacitor capacitor network on the source? Would it make a difference to the sound, or is just an arbitrary choice to use 100k or 1M?

[jbm222]

I'm looking at building a some kind of booster and I've been looking at the Fetzer before. Since the schematic is very similar to the Fetzer, I suppose it would be possible to make a booster that is switchable between Fetzer and Marsha.

The volume pot is 1M on the Marsha and 100k on the Fetzer. Would it be possible to use the same (1M or 100k) pot for both designs, just switching the resistor/capacitor capacitor network on the source? Would it make a difference to the sound, or is just an arbitrary choice to use 100k or 1M?

I think the only advantage to a 1M pot is that you'll get a sligtly higher max volume.  I think a 100k pot gives you better control though.  Perhaps you could split the difference (roughly) and use a 500k.

[vanhansen]

You can use 100k too.  I just happened to have a 1M pot and since this idea came from the JCM800 which has a 1M gain pot, I decided to try it.  It gives it hell of a boost with it.  Last night I put a 500k pot there and it tamed it down.  I need to get more pot values to play with.

I'm using a 5k pot with my ROG Fetzer Valve and like it alot.  It's not so much a boost that way but it does add that Fender-like character to my sound.  The same is happening with the Marsha Valve with a 5k pot, it adds that Marshall-like character, which is what I was going for.