Clones vs originals

[shakti]

I've built a dozen or so different Marshall clones. I am extremely happy with all of them. I've spared no expenses, trying to get them as close to the originals as possible by using as many NOS parts as I could find, and the highest quality transformers (usually M e r r r e n at least for the OT, and some Marstran). I've been using mustards exclusively unless the originals used something else, NOS era correct RS and Lemco silver mica caps, sometimes vintage Erie or similar filter caps, NOS Piher and Iskra resistors of correct era and wattage. For the filter caps I've been using RIFA, F&T and ARS.

Yet, almost every time I have one of my amps side by side with a good working original amp with the same circuit, there's a tiny little something that the clones doesn't match up to. It's very close, sometimes extremely close, but having had the chance to compare a few different amps and circuits now, I am beginning to see perhaps a pattern. I've compared a '66 JTM45 (actually a vintage trem chassis vs non-trem clone, so not exactly apples to apples), a '67 JMP50, a '67 JTM100 black flag, and now lastly a '69/70 Super Lead.

I'll explain more in depth about this latest comparison, because it's very interesting in that it's not a 100% vintage amp. It started as a '70 Super PA that was not working properly and had a couple of funny mods. It belonged to a friend of mine who asked me to turn it into a 69 Super Lead. I figured it was OK since it had already been modded and was in a bad cosmetic state, but everything important was in place. I have almost completely rebuilt this amp - I cut off the part of the preamp board which had channels 3 and 4. I completely redid the grounding so it now has Larry style grounding which I use in all my builds. I converted the preamp to Super Lead specs, using the same stash of components that I've used for my clones, so NOS silver micas, NOS Iskras where other values were needed, and a NOS 0.0022uf mustard for V1a, as well as NOS WIMA red box caps for the 0.068s. Recapped it with F&Ts.
I built a '69SL clone myself a couple of years ago which has exactly the same circuit except mine has a self-contained bridge rectifier and no snubber caps. The rebuilt Super PA has regular diodes as well as snubber caps on the rectifier. My clone amp also uses NOS WIMA TFF caps for the 0.022uFs. Lastly, ARS filter caps on the clone. Apart from those component differences, they are exactly the same amp.

Now on to the differences I hear. And these differences seem to be quite consistent between all the originals and clones I\ve compared.
The originals all seem to sound a touch "tidier" and tighter, particularly in the bass. They all have less bass than the clones, even With the bass on 0, and you can turn up the bass more and it comes on much more gradually. The distortion also seems to come on slightly more gradually. The clones have a slightly grainier feel, especially from 0-5 on the volume control. From 7 and up they are more similar in distortion characteristics, but the tonal balance always seems to differ slightly in the way I described above; less bass, a "cleaner" or "tidier" sort of distortion, more of a mid peak. The clones appear almost a tiny bit scooped by comparison.
Now which one is better? I suppose that could be a matter of taste, but I suspect most people would choose tehe originals. The clones almost appear to have an extended frequency range, especially in the lower register, but it's really just in the way for guitar playing. I also think that bass response tends to clutter Things up and introduces unwanted distortion.

Mind you, I am trying to describe what I hear and feel, and it's not easy to capture these differences in words. But if I were to sum it up, I'd say the originals sound a touch cleaner and thereby warmer, and with a more balanced tonal respones that just sounds "right".

But now, the $1,000,000 question; why is it so? Even when using the exact same Components, they end up sounding, to varying degrees, different from the originals.
I am of course coloured by this latest comparison between my 69SL clone and the rebuilt Super PA. What are the differences between the two amps?
- I strongly doubt it could be because of the different brands of caps (ARS vs F&T, mustard vs WIMA TFF) although I can't rule that out.
- Is it the transformers? Could be, but it seems like M e r r e n is as close as it gets.
- Component value drift? That would be minimized by the fact that I was using a lot of new NOS Components in the rebuild. I also measured most resistors and found minimal drift, no more than with the ones that went into my new build.
- Rectifier noise? I am going to look into this one more closely. I had a problem with my JMP50 build when I used cheap N4007 diodes. When I switched to faster UF4007(?) diodes the distortion tidied up considerably. The old amp has snubber caps which would remove some of the high frequency hash which can clutter things up, and that *could* account for some of the cleaner, fatter distortion I am hearing vs my clone. I'll update you all on this one.

- Then how about the differing tonal response; leaner and tidier bass and a more pronounced warm mid peak? This is where it gets interesting...could this have a lot to do with the potentiometers? People are always on about the drifting of component values. Yet I have never seen any considerably drift at least on the resistorts used in vintage Marshall amps. However with pots there's a wide variance. When you consider the values of the pots in the tone stack, could this account for the different tonal response in a vintage vs new amp? I can't use the Duncan tone stack calculator on my Mac, but it would be interesting to plot in the values of the old pots vs new production pots in my clone and see what that brings out. Generally, the older pots measure considerably more than their nominal value. When I get a chance, I will try to swap the pots around between my clone and the original, and see what that tells me. If it turns out to be significant, then the newt step is of course; how to get the new pots to sound/work more like the old ones?

- Then there's a whole shady area which I hesitate to enter into, but I'll throw it out there anyway; could some of the differences be dwn to tings like the chassis (different grade of steel, perhaps?), the bard material (perhaps introducing some kind of inductance?) or even the wire (old style stranded wire vs nw top coated wire)? The mind boggles...

Sorry for my long winded post, but I am really curious to hear what you guys have found and what you think of my ramblings?

[Scumback Speakers]

No need to feel like you're rambling, shakti. As I found out in trying out different SL models, there's always some "drift" or "variables" between them. That's going to get down to things like you've mentioned above (lead dress, chassis metal, etc).

I've even compared 1969 to 69 SL's and 1970 to 70 SL's and even though they were all original, they had minor variances in tone. I've concluded you need to shoot for a tone that's within 5% or less of the tones you want and be happy if the clones hit that mark, since age, and other variables will never really, fully and completely duplicate a 45 year old amp's tone 100%.

Same thing goes for speakers. You can get it "this close!" and not be 100% exact. I've concluded that time ages the components to some degree, and there's no way to accelerate aging 100% either. You get it to within 2-3 % or the "splitting hairs" level, and you need to accept that's the best you can do with modern parts replicating 45 year old parts.

[shakti]

I try to stay away from "mojo" terms and try to figure out as much as I can that we can replicate easily without resorting to esoteric theories about dielectric orientation and molecular massaging... Old amps vary from specimen to specimen like new amps do, everything else equal. Yet there is some sort of pattern or consistency to my findings that I suspect has something to do With some of the Things I listed above. But which one(s) is it?

Furthermore, while age probably plays into it, the vintage amps sound like the vintage amps we hear on records. I suspect they sound closer to how they sounded when they were new 45-50 years ago, than the clones og today sound to how those vintage amps sounded when they were new. As close as we have come, where is that last little bit coming from? Is it the potentiometers? Or is it those other things?

[VelvetGeorge]

Several reasons:

wire

age, and it's effects

pot values and tapers

stray capacitance (grab a good meter and start measuring from everywhere to everywhere). Then, move a wire and see how it changes.

many other factors, but these listed are important. Also, every replica has more low end than any original in my experience. Old amps seem to have a bump in the lows, while replicas tend to have overall more lows.

george

[shakti]

George,

thanks for contributing. Your comments are especially valuable given your experience. And your comments about bass response do seem to mirror my own.

If you don't mind, could you elaborate a little? What is it about the wire that differs, and how does it affect the tone in your opinion?

Pot values and tapers - again what I had been suspecting. Have you been experimenting with different values? At least for the treble pot you can find 280k or 300k pots, which may make a difference?

Lastly; stray capacitance - is that something which is more prominent on vintage amps?

[neikeel]

I have got close with three amps. A 12 series with Metro trannies (jape 88 has that) a JTM45/100 Black Flag (lives in Canada) and my 2 in 1 with Marstran trannies, sounds better than my 68 JMP50 but not as good as my '67 Black Flag 50. Time does help them settle in and subtle tweaks also help. The clones are nearly always brighter using clean settings and have less of the complex harmonics when overdriven, pretty much like a lot of new off the shelf boutique amps.

There is definitely something about certain vintage amps. Not all of them have it and you have to play them to find 'the ones' that are particularly good, usually either beaten up road dogs that people cannot tour without or the occasional pampered example that people keep in the studio as it is always in demand.
There are plenty of ok vintage amps that ebb and flow out of peoples collections.
I also have some clones that sound very good and others that are average/capable but do not quite have 'it' although that maybe it is just the cocktail (speaker cab, drivers, guitar, pedal chain, which side of bed you got out of!) that does it for tone sometimes.
If you have a really good original that does have it then keep it..............

Last two smaller gigs I have been using my 18w clone (lash-up with original trannies, NOS caps and parts) mic'd up and as far as a practical rig is concerned it is all I really need (well maybe a pair, . I have three original Marshalls I am keeping ('68 1987, '68 1959 and '67 45/100 with their respective pinstripe and basketweave stacks) and the '64 AC30T. My wife is ok with that too (which counts for a lot). The clones I am attached to for different reasons. Some have to go and it is a wrench at first, but they usually go to good homes. Same goes for my metal panels which I have sold over the past few years.......not all Marshalls are great amps................ (.

My musings in the past that are in accord with yours. I attributed it to the transformers, filter caps and pots. I do use old wire from cannibalised amps when I can as well as good pots (cleaned and checked).
I think it is the whole recipe.

[Roe]

yes, age/break in, pots, caps and transformers all matter. Remember the old pots read 30% too high (around 1m3, 32k, 325k, 6k5). I've done some a/b testing comparing clones and originals (66 jtm45 and a lot of 69-71 superleads, superbass, superpa, 1987s etc) and sometimes prefer the clones

[shakti]

In the case of my JMP50, I actually preferred that to the original same spec amp, but even on that one, the bass response is cleaner, leaner and tighter on the original.
I am extremely happy with both my RS OT JTM45, the JTM45/100, JMP50 1986 and 12-series 1959. The black flag JTM100 is also very good, but not on par with my friend's original (which was hands down the best Marshall I've played). This 69SL build though has a touch too much grain going on yet. Perhaps some of it is the WIMA TFFs, and some of it could be due to rectifier noise, so I suspect I can get it better sounding still.

Nevertheless, as my experience with this 70 Super PA proved to me, it's not really about the filter caps. I mean, filter caps do sound and feel different, but the old amp with new caps still outperforms my clone in certain areas. The tonal balance and tidier, cleaner, leaner bass is the most obvious difference. It also cleans up better, perhaps for the same reason. And that is one thing I've found consistently with the older amps. I'm just trying really hard to pin down where this balance is coming from. And the experience you guys have could help us all piece it together. I'm not content to simply say it's down to "age" because I don't believe that to be the case. Maybe the final 0.5-1%, but the EQ/tonal balance is too obviously different for me to believe in that.

[neikeel]

In the case of my JMP50, I actually preferred that to the original same spec amp, but even on that one, the bass response is cleaner, leaner and tighter on the original.
I am extremely happy with both my RS OT JTM45, the JTM45/100, JMP50 1986 and 12-series 1959. The black flag JTM100 is also very good, but not on par with my friend's original (which was hands down the best Marshall I've played). This 69SL build though has a touch too much grain going on yet. Perhaps some of it is the WIMA TFFs, and some of it could be due to rectifier noise, so I suspect I can get it better sounding still.

Nevertheless, as my experience with this 70 Super PA proved to me, it's not really about the filter caps. I mean, filter caps do sound and feel different, but the old amp with new caps still outperforms my clone in certain areas. The tonal balance and tidier, cleaner, leaner bass is the most obvious difference. It also cleans up better, perhaps for the same reason. And that is one thing I've found consistently with the older amps. I'm just trying really hard to pin down where this balance is coming from. And the experience you guys have could help us all piece it together. I'm not content to simply say it's down to "age" because I don't believe that to be the case. Maybe the final 0.5-1%, but the EQ/tonal balance is too obviously different for me to believe in that.

Now it is time to build that 18w clone with original RS trannie to see how that works out

[Roe]

I suspect that the bass response is largely due to (a) the tone-stack and potentiometers which read high and (b) the output transformer. Newer transformers often have "improved" steel quality, e.g. m6, which typically broaden the frequency response and add a lot of deep bass which you don't need for guitar.

(Edit: the 250/270/470/500/560pf caps also matter. I often like ceramic caps with high voltage ratings when I cannot find vintage caps. Edit2: I've replaced all metal oxide power resistors with AB CCs or Iskra CFs)

[VelvetGeorge]

I've spent countless hours running frequency sweep through every part of my old amp circuits. Then doing the same on my clones and tweaking until they are identical, generally within 1db from 100-10,000Hz. Noting that it's hard to get exactly accurate sweeps at lower frequencies due to harmonics.
When the sweeps say the amps are the same, they are usually very close. But, there is always some intangible differences. Something I can hear but can't measure. Getting the percussive "thunk" sound on partially muted low strings is maddening! I've spent hours repeatedly playing one of these notes and switching between the original and replica. In the end, I've concluded that everything matters. Each part can and does affect the tone and feel of the amp. The type of B+ resistor, for example. Having at least one carbon comp in the B+ line makes the replica closer to the originals.
Types of resistors matter, of course. Keep in mind that as general rule older resistors were more inductive than modern ones.

Here's a parallel thought: when I made the pc board for the Metro-Plex I shielded both sides, properly grounded everything and spaced parts to avoid interaction. The result is a very stable, predictable, noise free and somewhat boring tone! I actually added small networks in key places of the circuit to simulate the bleed of a PTP amp.

I'll ramble more about this later, I could go on for hours.... LOL

george

[VelvetGeorge]

I suspect that the bass response is largely due to (a) the tone-stack and potentiometers which read high and (b) the output transformer. Newer transformers often have "improved" steel quality, e.g. m6, which typically broaden the frequency response and add a lot of deep bass which you don't need for guitar

Steel is definitely different today. Modern M6 is higher quality and more consistent. For years I used M19 on OT's to compensate. I love that steel in the C1998 OT, but it wasn't available for a while and I had to use M6 again. After hearing it with M6 again I realized that I was missing some of the faster attack on the notes with M19. M6 is faster and more efficient. Better note separation and harmonics. This all marginal differences, of course.

I am not a transformer designer by any means. But, I did spend a lot of time measuring old ones and news ones. I do see an important spec in the primary inductance. I suspect that is cause the for the difference in low end response. You can pull out the wooden spacer on a Heyboer OT and squeeze the coil to change the inductance. Or just whack the laminations with a mallet. That's what MM does.

george

[Scumback Speakers]

I've only got three original Marshalls here, a 69 Super Bass, a 71 Super Lead, and a 71 JMP 1987 50w (probably all I need, right?), and clones of everything else (JTM 45, JTM 50, two 1987 and four Metro 100w done as a Super Bass, Super Lead, SIR 139 clone, and one with an fx loop/master volume mod) that really matter to my clients for demo amps.

Of course, I've spared no expense to make them as authentic sounding as possible. My SL clone is a dead nuts ringer for the 71 Super Lead I own, and I've got another clone as a 69 as well, which is a hair less aggressive sounding than the 71 SL clone. I used George's amp switcher at the LA AMP Show in 2013 to demonstrate it through the same set of speakers, and the only difference was the volume of the original 100w since it had higher plate voltage, but the tone was spot on.

I gave up trying to do anything more than that once I heard it, and everyone at my room in the show was floored at how the original and the clone sounded alike through the same cab.

So you can replicate it with modern parts, lead dress, etc, you just have to source the right parts. In the build process I went through, either original Mustards or the SoZo Premium or NexGen caps were used with insanely good results. Bear in mind, the Mustards were NOS and had not been used before, so even though they had 1968/69 date codes they were essentially NEW signal caps. I'm sure there is a "break in" or "aging" effect on the signal caps for sure, as well as other parts of the circuit.

What I found was that the three most important aspects (assuming all had good lead dress, and were identical in other ways) were the signal caps used, the tubes (preamp specifically) and the transformers made the biggest changes before you got to the speakers, and the guitar used.

Of course you have to use the same guitar and speaker(s)/cabinet for comparison or the test is flawed.

Mostly it comes down to how much time, money and diligence you want to invest in making a great clone. That's why I always use Metro parts/kits for mine. I made the mistake of buyng a W*b*r kit...once. We wound up having to throw out over 1/2 of it due to poor components included.

So make sure you have a good chassis, and other proper parts (not cheap trannies/pots/caps from Asia) for your build and you'll be happy. All you have to do is follow the instructions and lead dress from George's instructions and you're all set. At least that's all I ever did, and I've been very happy with my results.

But you can't cheap out on the parts. It is the sum of the parts, not just 3 or 4 things. So you have to pay for/get the right ones or you'll be less impressed with the results.

[shakti]

Now it is time to build that 18w clone with original RS trannie to see how that works out

I know, I know. I am still missing a truly correct chassis and hopefully a vintage power transformer as well. I also have at least two old 1M pots for that amp (scavenged from the very same Super PA mentioned above), although when I think of it I'm not even sure if the 18W uses 1M.

Probably starting a new job in the new year, and have two pedalboard builds pending, so I don't really know when I will be able to get to it...

[shakti]

But you can't cheap out on the parts. It is the sum of the parts, not just 3 or 4 things. So you have to pay for/get the right ones or you'll be less impressed with the results.

I think you're preaching to the choir here, Jim. Almost all the guys on this board (and certainly the ones on this thread) use Metroamp quality parts throughout. And a lot of people, the participants here included, try to take it even one step further by using as many NOS parts as possible.

Amps need breaking in, that much we can agree on. But even my longest used amps have a slightly different tonal balance to the originals I've played. This thread is all about digging deep into the last little details to get them those 2-3% closer still.

The comments about steel quality are interesting. One of the reasons why I went with M e r r r e n output transformers is that he claims to source the most correct steel. I don't really know any more details about it (except he says the laminations are grain oriented steel for his Partridge OT clones). I do know that when it comes to a combination of technical expertise and vintage Marshall experience, he probably has few peers. Nevertheless, I can't discount that even his clones may fall short of a good original OT. I have done two A/B comparisons, swapping M e r r r e n for Marstran OT and vice versa. One was in a JTM45/100. In that one, the M e r r e n was the clear winner for me, and for precisely the same reasons why original amps sound just that tiny bit better than even my best clones; the bass response was cleaner, tidier, rounder and firmer. It just sounded muscular and together in all the right places, accentuating that "violin" tone. The other A/B comparison was in my JTM100 black flag build where I didn't really hear any clear difference between the two brands.

BTW George, please do ramble on...I love to read about your work! Have you tried to modify the taper and value of pots? WHich results did that yield?