One watt Marshallish schematic questions

[tubenit]

Not sure why the 1 watt Marshall style schematic has some of the topology in areas where I've indicated with red?  Note this schematic has a one knob tone stack.   I am also puzzled by the input components and small value of 1k grid resistor.  I am puzzled why the volume pot has the large resistor values before and after the volume pot?

Do you guys think substituting the topology for what I've drawn with more typical cathode follower and tone stack would work?

With respect, Tubenit

[shooter]

I believe the input section to V1 is a filter network, I've seen similar on audio schematics.  so the 1K is part of that overall "filter".

the .1 / 4.7M I did something similar on my last build, but my values were way different.  I used it as a Hi freq cut, so I could better balance hi's and low's, not sure that's what's going on here, I seen it on an EVH amp.

got nothing for you on the TS stuff

[jjasilli]

Ditto.  It seems Matec felt the need for filtering.   He does say the amp is quiet.   Maybe take a conservative approach & build per his schematic.  Then bypass the extra filters people your schematic and see what happens. 

[shooter]

fwiw, here's the values and config I used for input filtering, I think I stole/borrowed from a bogen?

EDIT:  also the cut filter I used on my last build.

[TIMBO]

Hey T, This might shed some light on your questions. http://www.guitargear.net.au/discussion/index.php?topic=47673.0
I thought that it would be a simple front end on some well used valves that were common in a lot of local low watt amps.

I'm thinking that its a way of controlling the gain that is produced from the normal Marshall circuit.
I found that the 6GW8s were getting hammered and causing all sorts of problems.
 

[jjasilli]

Tubenit:  Forgot to mention above.  Though you describe the tonestack as a one-knob control, I think it's more an FMV with a treble control only -- the bass and mid-values being hard-wired.  A true one-knob control would likely give more flexibility.  But a Marshall sound is mid-heavy anyway, so maybe Matec's tonestack works well in this amp.


Also, I checked out the 2nd r-c filter --  .1uF C, 4.7M R -- though I did so in an overly simplistic way.  It's complicated because this filter is interactive:  one one side with the plate impedance (63K) in series with the filter caps to ground; on the other side, with the tonestack.  I used an online RC calculator to find corner frequencies.  Plugging in the values .1uF and 4.7M gives .3Hz;  Using the slope resistor value: .1uF and 56K gives 28Hz.  From this my guess is that this filter is meant to attenuate LFO at 6dB per octave starting from a corner frequency somewhere in the deep bass range.

[tubenit]

Gentlemen,

Thank you all for the responses!  All helpful information.

With respect, Tubenit

[terminalgs]

Not sure why the 1 watt Marshall style schematic has some of the topology in areas where I've indicated with red?  Note this schematic has a one knob tone stack.   I am also puzzled by the input components and small value of 1k grid resistor.  I am puzzled why the volume pot has the large resistor values before and after the volume pot?

regarding the original (the one with the red circles you've drawn):

• I don't get the components between the input jack and V1A's grid. the 47pf might be a high cut, maybe it's meant to deal with the miller effect on V1A instead of a 22K-47K grid stopper (I don't think the math works out for that).
• On the volume pot, the 10M and 1M standalone resistors are superfluous. The 10M insignificantly lowers the resistance to ground from 1M to ~900K, and the 1M changes the volume pot's performance.  All three together with the preceding .002 are a lousy combination if you ever turn the amp all the up.,  10M||1M||1M = ~475K.   with the .002 before it,  your high freq. roll off is around 200Hz.
• The next red circle, the 4.7M seems pointless as well.  It's so big that it seem doubtful that it'd make any audible difference.

You didn't circle it, but I gotta point out that NFB looks suspect and the 56K grid leak on the cathodyne is too small.  make it 1M like any other cathodyne you are likely to come across.  and omit the 1uf and change the 2.7K to a 1.5K  (again, more standard cathodyne...)


what you've drawn looks a lot better.  here are some thoughts:

• change the 500K volume to a 1M and change the .002 to a .02 unless you detest low freq. but love high freq.
• change the 56K on the cathodyne to 1M,  and remove the preceding 470K mix resistor. 
• couple V2A directly to the cathodyne with the .022 and move the NFB input to the cathode of V2A like you see on any cathodyne with a gain stage preceding it (with NFB...). omit that .022 on the NFB circuit that is no lower needed and dial in that 1M nfb resistor to whatever make sense with the 1.5K cathode resistor on V2A and the ohm output of the OT.
• the 100K grid leaks on the 12AU7's are too low. 220K would be better.

[shooter]

I don't get the components between the input jack and V1A's grid

I think the designer thought it was cool
I thought the same, the one I stole/borrowed is a variant of a Marantz, not bogen.  The amp I used it in was dead quiet, but from what, or why

[John]

Looking at it, I wondered if he was trying for some sort of notch filter to reduce feedback (vocal or harp mic). But I wouldn't think you'd bother with a small wattage amp that won't see stage use?

[tubenit]

IF I were going to attempt to build this,  ............ I think I'd give this a try?

With respect, 10thtx

[EKDENTON]

HI Tubenit, i have a couple questions.


What is the circuit on the AC in ground?


Also what PT were you planning to use?

[shooter]

What is the circuit on the AC in ground?

I believe that's a "ground lift" or isolation between chassis ground and signal ground.

[HotBluePlates]

IF I were going to attempt to build this,  ............ I think I'd give this a try?

If it were me, I'd stick with the stock schematic (except that the 0.002µF cap between V1a plate & the Volume control appears unexplainedly-wrong).  I say that mainly because the amp sound quite good stock (see video below).

Input Components:
 -  47pF across the 1MΩ is -3dB at ~3.4kHz.
 -  The 100pF from plate-to-cathode of V1a also reduces high treble response.
 -  Between the effects of the last 2 parts, a large grid stopper isn't needed to dampen treble.
 -  The 0.068µF against 1MΩ resistors has full response down to low sub-sonics.

I think the designer anticipated pedals plugged in the front of this amp, and wanted to keep any d.c. level-shifts due to pedals off the input grid.  We could speculate the designer thought the player may use super-boosted signals (from one of the many booster/treble-booster pedals out there) to slam the input stage, and a smaller cap inside the amp might help with overload recovery (compared to multi-µF output caps in a pedal).  At the same time, the grid stopper is small at 1kΩ because there's not an intent to completely avoid input overload.

I don't know why the stuff around the Volume control is drawn the way it is.  The 470kΩ (and bright cap) going into the top of the Volume control makes total sense if you expect the player to slam the input stage with a boost pedal.  The extra 1MΩ at the wiper would change the shape of the taper slightly, but I don't know if it's enough to matter in use.  As I said, the 0.002µF doesn't make sense to me, as-drawn.

I have no answer for the 0.1µF/4.7MΩ network after V1b.

I'd keep the 1-knob tone circuit as it is.  I can't believe it took me 20 years to stumble on the fact, but the standard Fender/Marshall tone stack is most guitar-frriendly when you get close to no Mid-scoop.  That happens when you keep the Bass near 0-2, Mids close to full-up, and adjust the Treble control to taste for your guitar.  Both the Bass & Mid values would occur at half-up, and I might tweak them (after building/listening) towards less-Bass and more-Mid.

I'd leave the 56kΩ grid leak of the cathodyne alone.  It forms a voltage divider with the prior 470kΩ to keep the PI and output stage from being heavily overdriven (the output stage probably only needs couple-volts to get there as opposed to usual output stage levels).  The 1MΩ feedback resistor will also be working into that 56kΩ to set the feedback level, and my mind wants to see the junction of the 3 (470kΩ, 1MΩ, 56kΩ) as a virtual-earth point for mixing signals at an opamp input (the entire output section is the "opamp").

So reiterating, I'd seek to verify that 0.002µF is correct (present in that spot in the original amp, as opposed to transcription error), but otherwise build the amp as originally drawn.

https://www.youtube.com/watch?v=a1hReI8KfIQ

[Ritchie200]

I used to see this at my fav music store when they were sold but never had time or interest to try it.  If the video is any indication, wow it needs some tweaking big time.  The cleans seem to be fairly nice if not kind of a meh, flat response.  However, the overdrive sounds like splatty crap in my very humble opinion.  Reminds me of the recent stuff Billy Gibbons has been recording - and he was using a SS amp on the verge of DC meltdown.  I've got an old Marshall SS 25th anniversary Lead 15 mini stack that sounds 100 times better than this, even with the crappy 8" speakers.  Seems like this was just an excuse to sell a low wattage anniversary "tube" amp toy to someone with deep pockets for $750 and a promise of collectability.  Maybe you could tweak it to taste, but why?  The 12AU design has been thrashed pretty good and there is a lot of info out there of better sounding examples.  I'd take a Champ any day over this - and you know how hard it is for me to say that.....

Jim

! No longer available

[Ritchie200]

Ya know Tubenit, it's got the first three tubes of a very special design that will fulfill any and all of your super sauce sonic heavenly needs for a lifetime.  As I mentioned a few years ago, if you build it: guys will wish they were you, women will swoon, you will have more friends (me for one!), guitar slingers will bow down to you, and most important - it has such incredible magical mystical powers, there is a chance, although slight, that it may even make your Tele almost sound like a real guitar.  I'd say THAT is pretty special!

Yer close friend and support system,
Jim

[jjasilli]

Hotblue: Other versions of the schematic, e.g.:  https://postimg.cc/image/wdiof9iit/full/


do show 2.2nF there = .0022uF which I agree seems weird.

[tubenit]

And then there is this schematic.  Which is where I got some of the ideas from. 

Note the 560k (not 56k) grid leak on the cathodyne phase invertor.

With respect, Tubenit

[HotBluePlates]

Hotblue: Other versions of the schematic, e.g.:  https://postimg.cc/image/wdiof9iit/full/

Showing the connectors explains some of the "odd" extra components.  Ask yourself, "What happens if the connector comes loose?"

 - Input 1MΩ/47pF probably mounted on the jack side (on the jack board) for convenience.
 - Grid-side 1MΩ (R29) needed to pin down the grid if Con01/Con1 came undone.
 - 10MΩ (R32) used to pull-down/load the coupling caps after V1a if the connector to the Volume pot came undone.
 - 1MΩ (R34) is grid reference for V1b for the same situation.
 - Apparently the 0.1µF (C15) want wanted if the connector to the tone circuit came undone; 4.7MΩ (R36) loads that cap.

Still dunno about the 2.2nF cap.

[HotBluePlates]

And then there is this schematic.  Which is where I got some of the ideas from. 

Note the 560k (not 56k) grid leak on the cathodyne phase invertor.

That makes me wonder again if there is a transcription error.  "By MATEC" implies the same person drew both schematics, but the story I heard was that each amp in the 1-watt series had a different designer.  It sure seems whoever designed that last one was less-concerned about tying down every connector.  Don't know how that might factor into the use of a shared output section (or if they were different among the models).

...  If the video is any indication, wow it needs some tweaking big time.  The cleans seem to be fairly nice if not kind of a meh, flat response.  However, the overdrive sounds like splatty crap in my very humble opinion.  ...

Maybe.  I haven't played one.  Some of Johan's other videos running the 1-watt amps into 4x12's sound different/better.

https://www.youtube.com/watch?v=wKb23WP-jTI

[jojokeo]

Just my opinion here but... that's just a collection of a bunch of wasted parts to make believe that the design is interesting and cause the ignorant Marshall hacks to bust out their wallets if you ask me. There's no magic in any of it what so ever. Especially for a 12au7 "power" tube. I hate those for that application anyway. Waaay better to use at least a 12bh7a or even ecc99. If it farts and barely moves a cotton ball in the forest let alone try to budge a speaker cone, does it make a sound worth hearing? Where are you PRR? On top of everything a conjunctive filter for the final tone smothering blanket no less...I wanna puke my sushi dinner.

tubenit - you could come up with a better design in your sleep.

[tubenit]

I don't have any plans to build this amp.  I just posted the schematic because I was curious about the components and topology.  Never have seen anything quite like that before?  I drew up the layout in case someone wanted to give it a try.

My idea of a low watt & low volume amp is the HoSo56 with 5879 V1 & 6BM8 tubes with a PPIMV or a VVR & using a 10 " speaker.

with respect, Tubenit

[jjasilli]

Good grief guys, Hotblue makes 2 good points.  Regardless of the quirky schematic, the amp does sound good (though not with the treble dimed as in one video).  And it's generally a better plan to build a known working amp as-is, then mod from there.  Mass produced amps sometimes have circuits which seem hard to justify.  Maybe they're just weird artifacts.  Maybe they're added to address some condition that seldom rises; but 1/100 could affect thousands of mass produced amps.

[tubenit]

HotBluePlates always makes great points, IMO!

The diversity of views on this thread has been really interesting.  Especially such a wide range of views from experienced builders and individuals whom I greatly respect.   I've never built a clone of any "manufactured" amp like a Marshall, Vox, Ampeg or Fender  etc.............   

    

With respect, Tubenit

[Ritchie200]

That's true. I'm sure some of what we are seeing may be the once bitten twice shy design from Marshall. I'm sure the warrantee department has seen it all...  From all sorts of "items" being plugged into the front end AND the back end!  I guess the problem I have is the price point driven by the toob moniker and the anniversary tag. $750 for this is crazy. I've pulled up a couple more videos and to me it just does not sound good. It's advertised as replicating the 60's Marshall crunch..Huh?  I realize Tubenits intent was to gain a perspective on the design. It's just that Marshall peeves me off sometimes, kind of like what Jojo said.


Why I remember back in my day when Marshall was a good company - hey you get off my lawn!!!


Jim

[jjasilli]

Ritchie200: One for the record books.  I can't believe I got you on defense!     The good-natured, double-team flame-throwing of you & jojokeo is always welcome and has its own point to make.


At retail that's $750 per watt!  The newly released point-to-point 5W Champ is about $1000.  At only $200 per watt, a seemingly better bargain.  But the darn thing's TOO LOUD!

[Ritchie200]

Haha! No I agree with everything you said. When you build for the masses you have to make allowances. Hmmm $750 a watt times....  Hey I didn't lie to my wife when I bought this amp and told her it was a retirement investment!!!  You da man!


Jim

[jojokeo]

Ritchie200: One for the record books.  I can't believe I got you on defense!     The good-natured, double-team flame-throwing of you & jojokeo is always welcome and has its own point to make.


At retail that's $750 per watt!  The newly released point-to-point 5W Champ is about $1000.  At only $200 per watt, a seemingly better bargain.  But the darn thing's TOO LOUD!

Indeed jj it is to be taken as "good natured" fun as Jimbo gave me a good pointy elbow to the side. I just see a lot of electronic gibberish going on which I do not condone. I've seen these tricks used before even in many a pedal build designs such as the input mumbo jumbo for example...shave of ultra highs, set the low end limit, reduce high end gain again, then pass along some high end while limiting low end, then preserve all gain that remains...this is all just before and after the first gain stage. But the crazy part of all this is ultimately end up to be showcased by a poor performing 12au7 in the end? It seems like a lot of trouble for hardly something in the end that may or most likely not make a scintilla of difference or even be heard? Then to charge an arm and a leg with some fancy advertising based on their reputation to drum up interest and there it is I guess?  Now how do I change this perception of being associated with THE class bully???!!! 

[jjasilli]

Now how do I change this perception of being associated with THE class bully???!!!   


Too late for that.  We're all Cathode Followers on this bus! 

[Ritchie200]

https://www.youtube.com/watch?v=fay1uk7wpnQ

[PRR]

> 47pF across the 1MO is -3dB at ~3.4kHz.

The impedance drops there.

The response probably does not drop. We need to know the Source Impedance. If it is a 50 Ohm tone-box, 70MHz. If it is a 500K source, 7KHz. Likely it only "does something" when the input is left open (absorbs some radio signals).

The full math around the "0.002" cap is ugly (changes with pot setting) but in general it shaves the bottom half-octave of guitar. On a 100W amp, that may be bad or good. On this 1-Watt amp, you probably do not want to squander much power way down low, put the power where you can hear it.

The 56K (or 560K) at cathodyne grid is bootstrapped to hundreds of K (or several Meg). Note that the "NFB" is wrapped around this point. Apparently. Actually I think there is gain of less than 2 (maybe not 1) inside this loop, even no-load. So the "NFB" may not do much, though the 470K does limit grid-shift in OVER-drive.

The input 1M+0.068u+1M does protect against DC bias leaking out of pedals. Minimalists can skip this. 

> $750 per watt!

Hm.

[smackoj]

I have never built one of these 1 watter amps but I surely would avoid the 12AU7 bottle since Commander JoJo says they don't sound good in that application.  Speaking of El JoJo, since you already have quite a big following of groupies and wanna-be's that like your 'class bully' persona, I say leave it alone and enjoy the respect that only fear can create.   

[7string]

FWIW, I built one that sounds really good with a 12AU7. I just used the first two tubes of the 1987 50W and the last 1 1/2 tubes of the Firefly 1 watter. I used all 50uf electrolytics and a Hammond 269EX for 265VDC B+. V1 is at around 110vdc. Output transformer is a Hammond 125A.


I did try the 12AT7 and 12AY7. Both were cool but I ended up back with the 12AU7. I used a shielded socket, so the 12BH7 doesn't fit. Need to swap that out along with a liitle larger Hammond 125C OT.


It can't push much air through a 4x12 but it is perfect for private scales and arpeggios practice.