Marshall Super Lead build, sort of

[Chuck H]

IIRC the feedback was supposedly 100k off the 4 ohm tap. Also, the big cap on V2a cathode 470uF(or thereabouts) does seem to be a mod. Also, probably used a hotter pickup (like a Super Distortion or something like that--if memory serves somebody mentioned a Might Mite clone of the SD). Additionally, in an early interview he mentioned plugging into a different setup for Eruption (maybe a higher gain one?), so possibly a source of confusion (by people who swear he was using a secret pedal or had gain mods to the amp--I don't think the big cap makes a huge difference). Also, possibly one less filter stage (there but not hooked up). Also the phase inverter (sorry...) output coupling caps (maybe?) higher (I've actually seen a stock Lead (split cathode) Trem 100 like this with 100nF instead of the usual 22n). (IIRC) according to Dave Friedman the cap across the mix R was a red ceramic (10% tol.?), the treble cap was a mica,..

My "research" (if you can call any info from the web valid) shows that the 100k NFB resistor was never confirmed and Dave Friedman reports it was a 47k on the 4 ohm tap. But then, Dave did not see the amp before others had worked on it. But if someone else DID change the NFB resistor to the correct value why would they not have also changed it's connection to the correct secondary tap? So I think it was likely always a 47k off the 4 ohm tap.

The "fat cap" mod, that is the 250uF to 470uF cap bridged across the V2A cathode resistor is confirmed by Dave but didn't come into use until much later and the amp may not even have ever been recorded in this incarnation. Eddie said "All these techs that have worked on it really f#@%ed it up". The "fat cap was added at some point to fatten the tone to where Eddie believed it sounded more correct. I might speculate that the amp may have had .022uf coupling caps for both channels once upon a time and some tech may have "corrected" that. Later, the fat cap mod restored a little of that lost bottom end. But no later tech ever thought to change the bright channel coupling cap from what appeared to be stock. But this is just speculation. I don't put much importance on the fat cap because it was a very late addition to the amp when Eddie was trying to get it to sound right again.

Your info on the pickup may well be correct for VH1 and I don't know what all else later. I've read similar as well.

The PI coupling caps are reported to be .022uf by Dave. I can buy this because some Marshalls "swirl" more than others and I've sometimes suspected that this may be due to duty cycle shift caused by grid loading on amps that may have been fitted with .1uF caps randomly (at times when .022uf caps were in short supply). I don't recall EVH's tone being real swirly.

As to other stuff... Extra filter stage, etc. I never saw anything about that. Very early on the amp DID have a pot on the back panel. It was removed later. The popular speculation is that this was a Jose installed master volume that was put in experimentally because when the band was not yet famous Eddie needed to manage volume for different sized gigs. When asked about the mystery pot in an interview Eddie did not reveal what it was and only said "I always just keep that all the way up". It was removed and the hole was plugged with a slot drive screw in the very early days. This clearly demonstrated that the amp WAS modded after all. Though just how much so and whether it was modded in any way that matters will always remain up for debate.

Also, Eddie was notorious for just saying things to perpetuate intrigue. He got this instruction from David Lee early on after a lack luster radio interview and ran with it. So pretty much anything Eddie ever said about the amp should be taken with a grain of salt. I didn't see anything about unmatched tubes or his amp red plating in my own (shallow) research. I have to wonder about that anyway because it seems to me that if voltages were adjusted one way or the other with a Variac then the bias voltage would have been analogous to that adjustment. Less than perfect but probably not that far off to cause red plating.?. Unless the amp was already incorrectly biased.?.

I saw those squashed green caps in one amp and wondered what the hell they were. I didn't know they were OEM.

[Chuck H]

This point I don't understand.
All the early Marshalls I've seen had a multi-tap primary PT and used a voltage selector switch accommodating line voltages of 110V, 120V, 200V, 220V and 240V.
Also showing on original schematics.

Now assuming the standard US export version had a dedicated 120V PT, what would Eddie's model have used?

You said "corrected" the voltage for his amp. That would indicate a lower than 120V PT, maybe a dedicated 100V version for the Japanese market?
In that case the voltage reduction by the Variac would be only 10%.

In any case, if his PT did not have a 110V/120V primary, results with a 120V PT and Variac set to 90V will be different.

So the essential question remains, what were his resulting B+ and heater voltage values.

This aspect remains nebulous. I didn't understand the US vs Brit model implication either since I have only ever seen Marshalls with selectors on the back panel for this. But I guess that prior to '70 many Marshall amps did not have voltage selectors. Eddie reported that his amp was a 230V model and he used a Variac to increase voltages just to get the amp to work. At any rate, the oldest piece of info on how Eddie ran his Marshall, regardless of anything else, has long been held to be a Variac set for 90V when plugged into 120V AC for a US (110V/120V primary) amp. I really can't offer more than what's been reported on this matter.

EDIT: Here's an image of the back panel of Eddie's amp. Clearly no voltage selector on this one. What doesn't make sense is that the amp is reported to have been a 230V primary but Eddie was able to plug it into an American outlet.?. I don't think the plugs would have been compatible.

[dai h.]

My "research" (if you can call any info from the web valid) shows that the 100k NFB resistor was never confirmed and Dave Friedman reports it was a 47k on the 4 ohm tap. But then, Dave did not see the amp before others had worked on it. But if someone else DID change the NFB resistor to the correct value why would they not have also changed it's connection to the correct secondary tap? So I think it was likely always a 47k off the 4 ohm tap.

The "fat cap" mod, that is the 250uF to 470uF cap bridged across the V2A cathode resistor is confirmed by Dave but didn't come into use until much later and the amp may not even have ever been recorded in this incarnation. Eddie said "All these techs that have worked on it really f#@%ed it up". The "fat cap was added at some point to fatten the tone to where Eddie believed it sounded more correct. I might speculate that the amp may have had .022uf coupling caps for both channels once upon a time and some tech may have "corrected" that. Later, the fat cap mod restored a little of that lost bottom end. But no later tech ever thought to change the bright channel coupling cap from what appeared to be stock. But this is just speculation. I don't put much importance on the fat cap because it was a very late addition to the amp when Eddie was trying to get it to sound right again.

Your info on the pickup may well be correct for VH1 and I don't know what all else later. I've read similar as well.

The PI coupling caps are reported to be .022uf by Dave. I can buy this because some Marshalls "swirl" more than others and I've sometimes suspected that this may be due to duty cycle shift caused by grid loading on amps that may have been fitted with .1uF caps randomly (at times when .022uf caps were in short supply). I don't recall EVH's tone being real swirly.

As to other stuff... Extra filter stage, etc. I never saw anything about that. Very early on the amp DID have a pot on the back panel. It was removed later. The popular speculation is that this was a Jose installed master volume that was put in experimentally because when the band was not yet famous Eddie needed to manage volume for different sized gigs. When asked about the mystery pot in an interview Eddie did not reveal what it was and only said "I always just keep that all the way up". It was removed and the hole was plugged with a slot drive screw in the very early days. This clearly demonstrated that the amp WAS modded after all. Though just how much so and whether it was modded in any way that matters will always remain up for debate.

I saw those squashed green caps in one amp and wondered what the hell they were. I didn't know they were OEM.

re: the feedback R, I took a quick look at some of my info and it was indeed 47k not 100k, so sorry for the misremembered bit.

re: the "fat cap", and whether it was there for the famous recordings, I'm not sure but sound-wise it seemed to be on the right track (just from experimenting--years ago I posted some circuit values that I ended up with from experimenting (there was uncertainly on things like whether it was a bass or split cathode lead circuit, etc.), and I ended up (by ear basically trying to play rhythm along with the CD) with split cathode (470nF instead of 680nF?), fat cap (yes), 480p(x2 240p in parallel) &33k in tone stack, 22nF out of the PI, and (from rectifier out) 40-30-20-16uF for the filters--so one less filtering stage than usual). (So my conclusion: some sort of lead and not the bass circuit.) There was also a pic of one of his supposedly built by Jose with a bunch of Sprague Orange Drops and IIRC it also had the "fat cap". On the possibiilty of the amp having 22nF for the bright ch. coupler, I suppose it's not impossible (I tend to think of that as more of a later early 70s thing). Other close serial numbered amps (perf board, split cathode, axial filters, one chassis amp on top, etc.) I think they had 2n2. (My understanding is a bit vague was I think the serial numbers were not totally reliable, but I think these were ones that were similarly configured inside, so assumedly built around the same time ("similar" meaning to the couple of pictures floating around the web supposedly from his showing perfboard, "chicklet" for the 680nF, plus the features visible from the outside with the one can cap on top)).

Also as I far as techs screwing up the amp, I thought this was the guy in Europe.

re: the pickup, I remembered trying (my extreme crap condition) Kramer Japan JK1000 (with what seems to be quite hot) Brad Gillis model PJ Marx pickup (not stock) one time thru my (modded) Marshall when I saw Pete Thorn's video, and a hotter pickup made sense to me (just to get more drive out of the amp). (BTW Pete used to be an "ampager" long ago--his user name was something like, "Sinasl"(don't remember exactly--maybe a number in there somewhere?).)

re: the knob in the back, who knows what that was but partly I tend to think the master volume idea is people wanting to see what they want to see (i.e. thinking EVH had super special massive gain mods by Jose and a special secret distortion pedal (that Tone Bender housing he seemed to be using for switching is maybe the source of this notion). Also (IIRC) there was a pic of one of his striped guitars with a small toggle, and somebody concluding it was sort of secret gain booster. I could accept some secret whatever if it could be proven, but maybe (a possibly more boring explanation is) he wanted to do the "kill switch" part in "You Really Got Me" so he needed the switch(?). I find it sort like (the show) "Ancient Aliens"--can't explain (or don't know) the explanation for something = Ancient Aliens did it. Possibly also a commercial aspect to it, with the aim of promoting Jose A. gain mods (or new amps with them built-in).

With the lead circuit, "fat cap" and maybe a hotter pickup, I think you will get that gain on the recordings (and maybe even closer using Variac and re-biasing)--also some Marshall PTs sag a lot apparently. Plus the delay and "treble boost for solos"-behaving Phase 90. Another aspect is that it seemed he had a strong touch (I couldn't play the solos, but) when he does his finger tapping, it didn't sound right doing that lightly. Hammering and pulling off with a bit more power sounded more correct to me.

My old Philips/Valvo databook say Mustards were metalized polyester caps. PET and polyester means the same, namely Polyethylene terephthalate.

Yellow chicklet caps are metalized polycarbonate.

Thanks. I did know what PET stood for (not sure what the extra 'P' stands for though in 'PETP'). They just seemed a bit on the larger size for a metallized (maybe in particular the 100nF?)--not a scientific opinion, just general observation. When I saw the datasheet for the "chicklet" caps I was slightly surprised they were polycarbonate (the smaller 22nF and 680nF might have been different types?). There was an interview with the Top Hat amps guy (on Pete Thorn's Youtube ch.), and he didn't seem to have a high opinion of metalllized caps, which (IMO) sounded pretty silly.

re: the voltages on his amp: Dave Friedman shared what he wrote on a piece of paper (on a Youtube video) when he was checking out the amp.

According to what he shared:

- B+ was 463V DC@120VAC(may have been adjusted to whatever AC gave 6.3V on heater) which sags to 386V when amp is dimed (a 77V drop)
- at the PI, 315V DC@120VAC going down to 230V DC@(90VAC I guess--I've also read 89VAC(not much difference I know) and something about when the pilot light starts to go out)
- bias was 35mA@120VAC and 50mA@90VAC
- (my notes) filters (from rectifier output to preamp stage) 50uF-16uF-100uF-32uF-32uF (I think that is value stated on filters and not measured--since tolerances on alu electros can be large, also 50uF and 16uF are series connected (so double))

(info on the voltages from Tone Talk interview with Dan Tramble from about here: )

https://youtu.be/S01K9m3oTX8?t=4647

(this is the one where D.Friedman(and Jake E. Lee) talk about the 140VAC from a Variac into the 220VAC tap: )

https://youtu.be/rtB9aOntuck?t=1041

Also, there was another interesting bit where David Friedman shared that 1984 was his amp loaded down with a fixed load (16 ohm R on 8 ohm tap) into his Bradshaw rig (I'll see if I can find the bit--okay found it).

https://youtu.be/MMsazcJ9V6Y?t=1596

some links on his tuning:

https://www.youtube.com/watch?v=TEK-ZikFACc

https://youtu.be/097HgaI2EKQ?t=341

(see section, "But Those Thirds Sound So Sharp..." towards the bottom: )

http://www.doolinguitars.com/intonat...tonation5.html

[Helmholtz]

I did know what PET stood for (not sure what the extra 'P' stands for though in 'PETP').

https://en.wikipedia.org/wiki/Polyet...r_applications

[dai h.]

He mentioned in an early interview that he had multiple old Marshalls, so (judging from the interview excerpt below--apparently the last one he did with a guitar mag) the one he fed 140VAC to was a different Marshall to his main one. Also if was using the Variac for volume control, maybe it puts less credence on the back panel knob as master volume notion?

source:

https://www.musicradar.com/news/van-...-mag-interview

“The real reason I did it was I had my main baby Marshall, except it was too loud, and it would blow up on full voltage. So not only did lowering the voltage take care of the amp not blowing, it also enabled me to contain it and play anywhere between 60 to 90 volts depending on the size of the room. Everything was all the way up on the amp, and the Variac was my volume knob.

“What made me think of lowering the voltage was I bought another Marshall head not realising it was a European model. So I plug it in, and there’s a really dim pilot light. I’m waiting and waiting and there’s no fucking sound coming out! ‘Goddammit, I got ripped off!’

“So I just let it sit, and I come back to it an hour or two later. I pick up my guitar and it’s really, really quiet, but it sounded like it was all the way up. Then I realise ‘Shit! The damn thing’s set on 220 volts!’

[Chuck H]

re: the knob in the back, who knows what that was but partly I tend to think the master volume idea is people wanting to see what they want to see...

I agree about the mystery knob not likely ever having been a master volume. If only because "I" wouldn't have located it there. The location doesn't make sense.

[dai h.]

Well somebody (with Eddie's approval presumably) took the time to make a hole and put that there, so must've been for *something*. Kind of an interesting mystery I suppose. Oh, just remembered IIRC someone spotted multiple cables out the back, so signal going somewhere? Line out?

[Helmholtz]

The "fat cap" mod, that is the 250uF to 470uF cap bridged across the V2A cathode resistor ...
.

Some of these amps (probably slightly earlier ones) had no bypass cap here, others used a 680nF.

Compared with no cap the "fat cap" doesn't boost bass, rather increases gain by around 2.
Compared with 680n it gives a relative bass boost up to 6dB.

The PI coupling caps are reported to be .022uf by Dave. I can buy this because some Marshalls "swirl" more than others and I've sometimes suspected that this may be due to duty cycle shift caused by grid loading on amps that may have been fitted with .1uF caps randomly (at times when .022uf caps were in short supply). I don't recall EVH's tone being real swirly.

22nF are normal for lead amps after '66. All JTMs used 0.1µ.
I prefer 0.1µ but keep bass low.

[Helmholtz]

But I guess that prior to '70 many Marshall amps did not have voltage selectors.

My 3 Marshalls ('66, '67, '68) all have the voltage selector plug.
I live in 230V (then 220V) country and have never seen a dedicated voltage PT in a Marshall.

[dai h.]

re: the red caps:

Chuck, maybe try some of more Marshall-centric forms? (metropoulos, Marshall forum and whereever else (Facebook?--haven't really kept up))


also re: feedback R, Dave Friedman says 100k off the 4 ohm tap here:

http://forum.metropoulos.net/viewtop...2292&start=615

so...???(confusion...I suppose you'd get more gain with 100k)

(Not directly related but there is a Marshall presence scheme in a factory Marshall schematic but one I've never seen looking through pics (there are lots of amps with 680nF but not this exact setup). This is the one with a 4.7k R on the wiper (so when Presence is at 10, the 5k pot and 4.7k R are in parallel reducing the feedback--the Presence cap is 680nF). I liked that one over the usual 100nF one simply because it seemed to give a more noticable boost.)

https://en.wikipedia.org/wiki/Polyet...r_applications

Thanks. That one just mentions that PET-P is an obsolete term. According to this page:

https://aipprecision.com/product/pet-p-polyethylene-terephthalate-polyester/

PET-P stands for "Polyethylene Terephthalate Polyester". So the extra "P" at the end = Polyester.

re: 100nF outputs from PI, Dave Friedman's favorite Marshall (not stock) is a 50W lead converted bass that has 100nF (this might be the amp THD based their Marshall type amp (that still looked like a tweed Fender) :

https://www.youtube.com/watch?v=rWEBE81DWQc

(from what I can see blowing up a screenshot) the cathode bypass cap is an alu electro 680nF 20% (it looked a bit weird since it was a small alu electro). (Interestingly?) in another thread on a Bogner with an RF prob, the chassis pics show an alu electro 680nF cathode bypass for one of the stages (plus others used for cathode bypass).

[Chuck H]

FWIW I'm not planning EVH's Marshall for this build. Just typical 68 super lead values. More versatile for most players. I only opened the topic because it's interesting.

[dai h.]

Yeah that's cool. Kind of a topic where you get people blabbering on and on (as I am, lol...).

But a lot of Marshall stuff (besides the Ed VH amp) is interesting. Clapton, Hendrix, Page, Blackmore, and others. Trying to determine what they used, circuit details, etc.

re: the smaller ceramic and micas, I've wondered if some difference is there due to the type or tolerance, maybe both. Sometimes I've found ceramics in surplus with a quantity discount (so I'd buy a bunch), and measured them on a handheld LCR meter (I don't think is a totally ideal measurement--from what I vaguely understand, with non-temp. compensating ceramics you want to control the amplitude of the signal voltage and maybe even more ideally apply any DC it's going to see in circuit). (From my recollection) some temp. characteristic 'B' (a relatively more stable no-temp. compensating characteristic somewhat similar to X7R) 500V 4n7s were mostly lower, 'E' (worse than 'B') 500V 1nF were mostly higher. So another possible source of difference. John Suhr apparently doesn't like the big (5nF) pot cap in his version of a plexi, but Pete Thorn (who uses one of the amps) experimented with bright caps (in a Youtube video) and seemed to at least find the larger value could be useful, so I thought that was interesting (this is not to say anybody is absolutely right or wrong on this). Also FWIW David Friedman did some of his own testing and found he liked (IIRC) Vishay 561R series (which is a temp. compensating but not the 'best' NP0/C0G (+/-30ppm/degree C = (I think +/- 0.003%/deg. C = hardly any change over temp.) but a relatively worse N2800 (I think negative 2800ppm? so -0.28% change per degreee C?). I tend to think probably not very noticably different to mica but possibly more so (compared to non-temp. compensating ceramic) under a lot of distortion.

[Chuck H]

Yes, ceramic vs mica debate. The ceramic temp rating as well as how applied DC affects capacitance goes beyond my understanding. So many people say they can hear the differences. I do know that DC and temp rating affect ceramic cap behavior but I don't know how or why. The general consensus seems to be that ceramic caps sound "grainy" and I really don't know what that means to anyone reporting that effect.

I did see the Thorn video on bright caps. I thought his assessment of WHY the larger cap worked better for some things made sense. But it does make the bright channel a two trick pony IMO. Cranked up full or backed off just a little. With the consequence of having very little adjustment range at the low end of the volume pot and then even if you do dial in a lower gain setting it's going to be too bright. Of course my build will have that cap because it's OEM. But I remember that on another Marshall I had that I either changed that cap or removed it (don't remember exactly). But I wasn't running the amp turned up much either. If I could have it would have made a difference I'm sure.

[dai h.]

(My very vague understanding: ) ceramic caps are made of different materials called dopers and shifters (not pulling your leg, lol) and the different mix leads to different characteristics, stability, volumetric efficiency or tweaking to some particular temp. characteristic, etc. (IIRC) Steve C. stated the bright caps, treble cap, and presence were spots a difference might possibly be heard.

(Again, my very general non-expert understanding) the possibility of a cap being audible depends on where and how it's used (if they see a voltage drop). I think in a speaker network more likely (large signal voltage and voltage drop)? In smaller signal solid state circuitry for a coupling cap sized adequately (or more) then less likely (larger size = less voltage drop). There is a very interesting series of articles (originally in Wireless World?) where a guy (Cyril Bateman IIRC) who worked for cap companies did some super comprehensive testing (I hope I got this right) using an ultra-low distortion oscillator (having 1ppm = 0.0001%?) and measuring whatever else besides the fundamental (i.e. harmonics and intermodulation products) occured when applying a very large signal voltage (the distortion was so low a really large voltage had to be applied to read the distortion--someone who sounds like a genuine expert on another forum pointed out the testing conditions were extremely atypical for audio circuitry, so people needed to be aware of that when looking at the results). The linear ones in general were film caps, temp. compensating ceramics, plus alu electros didn't look bad (despite some people seeming to believe that their mere presence in a circuit means the sound is irrevocably harmed). The non-temp. ceramics fared poorly, also tantalum (but in the real people might find them useful in a fuzz and in old Neves tants are used). (Personal conclusion: ) there is no absolutely good or bad sounding cap. So it might be something complicated like how is the cap being used, what frequencies are affected (the ones a typical ear is more sensitive to or not), maybe distortion, and I wonder about implementation possibly making things worse (or maybe better if not going for linearity).

(Cap trivia for the day: ) In the wayback machine (archive.org) there's a page (no longer on the Murata site), where it states on the Japanese version (for multi-layer ceramics with voltage rating of 100V or less) temp. compensating can handle three times the rated voltage and non-temp. compensating could handle 2.5 times rated voltage. (My understanding) this was *not* a license to ignore the rating but more of a reassurance/prompt that a 50V cap isn't going to blow up with 51V, and to keep things like pulses and surges that could happen in the circuit in mind when choosing a cap. (Wonder if this aspect could be made use of somehow for distortion.)

One time with a SansAmp classic I found a crappy ceramic at the input sounded more correct to me for that particular situation--more "blurred" as opposed to more clear and distinct for the temp. compensating ceramics (I don't think they sound good or bad--just depends on the circumstance). (Sorry this is probably boring, but) I had lost my original SansAmp, then later bought a "classic" version but something kept bothering me about the sound which after trying a lot of silly parts swapping (resistors, alu electros, the surface mount FET for a thru-hole one) eventually arrived at the 47nF input cap (which was an axial one in the original which I'm pretty certain was just some commonly available cheap non-temp. compensating ceramic--I do happen to have temp. compensating NOS 47nF which are probably from around the same period which are much larger also). The last thing I tried was measuring the cheap disc ceramic which seemed more satisfactory and trying a parallel pair of a non temp. compensating CER and temp. compensating which (together) measured about the same value, and (my memory is) I didn't like it because there was too much clarity. Wonder if there is some sort of logic as there might be with the tuning and distortion.

re: the Marshall 5nF pot cap, I wonder what they actually measured (I'm guessing they were quite loose in tolerance). Also, sometimes that big cap could be a red "dogbone" in some amps (might be the same part in some Dallas Rangemasters). I remember there was one guy on the old defunct Plexi Palace forum who seemed to be happier with those (not real scientific process IIRC though).

re: the mustards, according to this page on Mark Huss' site, they are film and foil:

https://mhuss.com/php/pix.php?p=Mustard

there's also some pics of a dogbone ceramic:

https://mhuss.com/php/pix.php?p=MarshallVolCap1

https://mhuss.com/php/pix.php?p=MarshallVolCap2

[Helmholtz]

I looked up my old Valvo databooks again and think you're right about Mustards being film/foil.

The German description of the 311 cap translates into polyester film with metal layer. I interpreted that as metallized.
But then I realized that the full type name is KT-311.
KT is a standardized acronym for film/foil caps. Metallized would be MKT.

Sorry for confusion.

One of these days I might crack one open.