More Progress on the Voxy Build

Just a quick update; making some tweaks to the Rockette. The Trainwreck Rocket completely ignores one triode of V1, which seems wasteful, and the AC30 dumps that output straight to the phase inverter, which seems silly. So, I got a little Plexi goin', and ran the "normal channel" into the next gain stage. Basically permanently jumpered channels that you can blend with the two volumes between bright and regular. Maybe I'll split those V1 cathodes if there isn't enough difference between the two volumes...

Moved some other things around, nothing too serious. Might add zeners on the cathodes? Maybe just on one pair? Maybe not, come to think of it.

Oh yes, and the parts are ordered, so the build will start shortly!

Latest Bodie Schematic

Okay, so I've learned to stop romanticizing vintage design yet again. I significantly increased the power supply filtering and oh my goodness is Bodie better now. Punchier with better bass response and less excess treble. I'm going to take this up another couple notches when I have more parts.

I also tweaked the bass half of a James tone stack for my bass control instead of the dual-coupling cap idea. The mid and high attenuation is very helpful, and now the possibility of bass "boost" can kick up a heavier butt for solo playing.

This is getting close to as good as it's going to get. I'm still tempted to tear apart the front end so that the second and third triode stages will be in parallel instead of the first and second. And I'm going to increase the power supply filtering a little more too, when I get some more power resistors. Knock these crazy voltages down a bit more.

Seriously, AES?

A COMPARISON OF CURRENT PRODUCTION 6 L 6 GC TUBES

CE Distribution / Antique Electronic Supply / Amplified Parts just put out this PDF where they plotted the frequency response of a single tube of each of the 6L6 family tubes they sell. 

 Okay, so this looks like a really useful document initially. They put a bunch of tubes on a tube tester, picked the average one, plugged it into an amp, plotted its frequency response, and let a guitarist subjectively describe the tone.

That last bit is a little like asking a mechanic what oil you should put in your car and then asking a guy who likes to drive how the different oils taste on bread, so I'm going to try to avoid commenting on "sizzling leads" versus "balanced and tight." If all you want is lurid prose about how many orgasms each tube brought the guitarist, there's no need to bring a tube tester or frequency plotter into this.

So ignoring the poet, we have the manufacturer's rated specs on the left. These are pretty much crap, as modern manufacturers largely just copy whatever was on the original data sheets. The maximum plate voltage is just a suggestion, let's be honest. Dissipation kills tubes, not voltage. And do we expect guitarists to know what their screen voltage is? Considering how common screen grid failure is, it would've been nice to include maximum screen dissipation buuut it doesn't matter anyway because the manufacturer's numbers are junk.

Back In The Day(tm), for a tube to be sold as a "6L6GC" it had to meet certain specs for plate dissipation, transconductance, envelope size, etc. No one cares about this any more, so it's understandable that there's some variation in rating and size from one manufacturer to another. Not to mention some of these tubes aren't even 6L6GCs, but Russian military tubes which were designed to be clones of 6L6GCs. That's another whole bag of worms, and it's fairly irrelevant, but it is interesting that some New Sensor 6L6GCs can take up to 40W on the plate comfortably.

What would really be interesting would be a comparison of transconductances, but considering a lot of these tubes are "close enough" to a 6L6GC I suspect there wouldn't be too many surprises. Specs on those Russian tubes (6P3S-E) are hard to track down, so it'd be nice to finally get some solid data. Anyway.

There are handy bar graphs of "lows," "mids," and "highs." This was kind of a mistake, or rather, how they define "low" and "high" is problematic. The open low E is 82 Hz, but this chart defines "lows" as 50 Hz. Even a baritone B string is 62 Hz, so the measurement of "lows" on these bar graphs is useless.

"Mids" are defined as 700 Hz, which is pretty reasonable.

"Highs" are defined as 6 kHz, which is not pretty reasonable. Most guitar speakers have already started rolling off around 5 kHz. 3kHz is around the "icepick" region and probably would've been a better choice.

To generate the frequency response graphs (and the guitarist gibberish,) they built a little single-ended guitar amp to put the tubes in. I take issue with a couple things here. First off, they didn't make a neutral Hi-Fi amp, they made a guitar amp. They describe the preamp as being a Blackface Fender design, and that's good for frequency plotting because there's certainly no limit to the low end, but unfortunately there's also a freaking tone stack! Yes, you can get a Blackface tonestack to be almost linear by turning up the mids to 10 and turning the bass and treble to zero, but did they do this for the frequency response plotting? Did they instead take the tone stack out of the circuit? These details aren't provided.

Anyway, these are graphs of just individual tubes so buying tube X and actually getting frequency response X is probably a crapshoot. Those little bumps are the result of manufacturing variances; if they averaged 10 of the same tube the responses would look much more similar brand-to-brand. Plotting dB on a linear scale instead of a logarithmic one is a little disingenuous too; even speaker manufacturers don't try that. Anyway, the biggest difference they show is roughly 2dB which is just under the threshold of what the human ear can discern as a difference in volume.

There is also the sizeable mid-hump which all the tested tubes exhibit. In the guitar community, 6L6s are widely considered to be somewhat "mid scooped" and the swap to EL34s will return these missing mids. At the very least, we can now clearly show everyone that the mid scoop comes from the circuit around the tubes, not the power tubes themselves. I suspect this bump in the frequencies is more related to the circuit, than a characteristic response of a 6L6.

In this case, the circuit these power tubes are plugged in is quite simple. Like I mentioned above, we have to assume they took the tone stack out of the circuit before measuring the frequency response of each tube. The circuit itself is single-ended, with no negative feedback. This is potentially a huge oversight - most guitar amps use a global NFB loop around the power tubes and phase inverter. I honestly can't think of a single 6L6-based amp that doesn't. I'm sure they're out there, of course.

The point being that NFB reduces distortion - the output is fed back to the input out of phase, so if the tube generates a boost at the output, that gets fed back to an earlier stage as a notch, and the result is a flat response. So, it makes sense that you would remove NFB if you wanted to plot a theoretical response of each tube to see if there really are significant differences from brand to brand.

But what happens when you put these tubes in your amp?

Even without NFB, a push-pull amp will cancel a noticeable amount of (even ordered) distortion. As the fluctuations in frequency response are largely sporadic, there would be significant cancellation of some deviations from linearity. Some would reinforce each other, so I suspect the overall response would look less choppy, though approximately the same.

With NFB though, all of that choppiness is going to cancel, so the only difference you're going to see is the overall y-axis offset - also known as the difference in emission & transconductance from tube to tube.

So to wrap this whole thing up:

• All these tubes are basically identical.
• There are some apparent deviations in frequency response at around +/- 2dB. 
• These differences are inaudible.
• Even if they were audible, most amps have NFB, so these deviations would cancel out.
• Only one tube of each brand was tested, so we have no idea if there are any trends among brands, or if CE Distribution just happened to select tubes that were basically identical. 
• This document is just a piece of marketing material, and shouldn't be viewed as anything but an advertisement. 

Bodie with zener-assisted cathode bias

I finally got around to trying the zener trick I mentioned a few months ago. Bodie's idle bias voltage was around 21V, and that surged up to 31V under heavy overdrive. A pair of 12V 5W zener diodes in series now clamp the cathode voltage at 24V. Maximum clean output power has risen from 19W to 23W. On the 'scope, crossover distortion is now only barely present during overdrive and the overdriven tone has increased rather dramatically. I think next time I'm going to have to try just straight fixed bias; 7591s lose so much in cathode bias. Not that I really need another 10-15W of output, but the tone is much improved. I can probably lower the plate voltages to something more sane than, say, 480V to keep the output level reasonable.

I'm going to be tweaking this amp for years, I just feel it.

For now, I still need to come up with a decent bass control. While I really like the "independent tone controls scattered throughout the preamp" approach, there aren't many very good one-knob bass controls. Sure, you can take the baxandall stack apart and just use the bass control from that, but it didn't work too well. It may be time to try a flat tilt control in place of just a dedicated bass control. That would give the user the option of cutting treble early and/or late, which would open up the preamp distortion characteristics a bit.

I'm tempted to start looking at FFTs of the frequency response to see if the NFB loop is causing any high-frequency strangeness. Given the (deliberately) limited bandwidth of the amp, square wave analysis is somewhat problematic, but it seems to indicate there's some unwanted phase shifting at high frequencies resulting in excess treble.

Then again, I might not notice that if I weren't using these vintage EV SRO speakers. It might even be good with a particularly dark speaker, but I feel a well-designed amp should work well with any speaker, though that may be something of an impossible goal.

Anyway.

It seems I lucked out picking a voltage for the zeners to latch the cathodes to; the plates are happy at the dissipation they're subjected to under heavy overdrive. I wonder if I could go a little colder though, get a little more squish out of the output section.

And I'm still thinking about the power supply. It's working fine, but could it be better...?

Yep, years of tweaking ahead.

New Crate Plans

Alright, so, I've decided to take the Crate rebuild in a different direction - Vox land. I was wondering if anyone would be interested in a single channel AC30 with no reverb or tremolo and it turns out yes, they call it a Trainwreck Rocket!

Now, of course, I'm going to have to deal with heat. I suppose I would have anyway (seeing how I don't work for Crate and thus am not comfortable hanging tubes below a completely sealed chassis) but cathode bias ups the heat load I'm going to have to dissipate. Also I'm going to put in (switchable) sag resistors to emulate tube rectification. Needs moar recs!

Here's what I've come up with so far. Still a lot of tweaking to be done. It's based on what people assume is in a Trainwreck Rocket, as well as the original AC30 top boost channel. I've added a handful of tweaks throughout; we'll see how they work.

• Parallel first triode. Why the hell leave it unused. Same gain, lower noise. 
• Arc protection on the cathode follower to help the tube survive switch-on.
• Grid stopper on the cathode follower. Just something I feel like trying. Figured I'd split the difference between 10k and 1M for starters. 
• Reduced grid leaks on the LTP to reduce noise. Reduced cap across the second input to help recovery from blocking distortion.
• Lar/Mar PPIMV
• Sizeable grid stoppers on the power tubes for better distortion.
• Power tubes biased in pairs, partially to spread the heat around, partially so one pair can be pulled. I might do individuals, we'll see.
• Zener-assisted bias for the power tubes. Cathode-biased 6V6s seem to distort nicely anyway, but this might be worth looking into.
• Switchable sag resistor for a "tube rectified" feel. Might make these bigger.
• Heaters referenced to power tube cathodes for free elevation. Humdinger to get the last of the noise out.
• 1k screen grid stoppers. 
• Grid stoppers everywhere. 10k for starters. 

Layout is going to be a big pain. All the tubes are in the center of the chassis, so I'm going to have to get clever. I'm tempted to start drawing it up in DIY Layout Creator or sort of CAD program. I may just use paper. It's going to have to be all point-to-point, but since this isn't going to be as much of a ground-up as Bodie was I'll be able to plan better.

Welp, I've been spending even less time in front of the computer lately, but I'll post again when I've got more to go on.

Bias Supply

This is how crazy easy this is. I bought a Triad VPL36-140, which is a tiny transformer that you can wire for either 18VAC or 36VAC.

Wired for 18VAC, I get ~24VDC. I'm not accurately simulating the load current, but 18*Sqrt(2) minus a couple volts for drop across the diode = yep.

Wired for 36VAC, I get ~49VDC.

Since I need only around -21V to bias the 7591s in Bodie, I'm thinking the 18V route is the way to go. Sure, even though I calculated the source impedances for both configurations (secondaries in series or parallel) I'm not sure I trust PSU Designer to accurately predict the load regulation on a tiny little 5VA transformer, especially considering the load current is going to be practically nil. Even assuming ~20k for a bias pot and a resistor to make sure the pot can't dump all the V- to ground, I'm still getting ~23VDC.

Merlin has R1 as 100k in his power supply book, which is probably more realistic for ripple reduction. I'll have to balance that against how long it'll take the bias supply to reach a stable voltage. Considering this is a full-wave bridge rectified supply, the ripple will be at 120 Hz which is much easier to filter than the standard half-wave rectifier in most guitar amps which pumps out 60 Hz ripple.

And hey, finally a little circuit where I can buy almost all the components at Rat Shack!

Marshall Class 5

This is a weird little amp. Unlike the Epiphone Valve Jr or any of its ilk (or even the good ol' Fender Champ) this guy has *two* preamp tubes. Only having one preamp tube (and thus two stages) seriously hinders a lot of Valve Jr mods. Sure, you can make a Champ (sort of; at least a Champ with an EL84) but if you want any more distortion than that without making the amp sound like an asthmatic 8-bit chainsaw, you're going to need to add at least one more gain stage.

A whole lot of amps make great sounds out of only 2 preamp tubes. Blackface Fenders, Tweed Fenders, classic Marshalls - even the JCM800 preamp, you only need 2 12AX7s. If you wanted a clean preamp, you could just follow the standard Blackface fender preamp and make yourself a nice little EL84 Champ. Let's ignore that one for now; this is a Marshall, remember?

 Okay, so let's look at some classic Marshall schematics.

So say you want a JTM45/Bluesbreaker:

So you have two channels, each gets one gain stage, then the signals combine and go through another gain stage and lastly a cathode follower pushes the signal through the tone stack and off to the power amp. Two preamp tubes.

How about a plexi? How about any of em?

Same deal - two channels, mix 'em together, slap a cathode follower and tone stack at the end, bam.

Since we really don't need two channels in this amp and we're keeping the power low so it'll be a sweet little distortion factory, how about the JCM800?

Slightly different! Only one channel, but two different paths. Low goes through one gain stage, then another, then the cathode follower & tone stack. High goes through one gain stage, and then right into the Low input. Kinda clever, yeah?

Unfortunately Marshall went out of business after the JCM800 was produced and they never went on to make channel-switching nu-metal fuel.

Okay, but we're seeing a pattern. It's a pattern that a whole lot of people like, and you could even go so far as to call it the crux of the Marshall.

• So you got your first gain stage. This just amplifies the signal enough for it to mess with other preamp tubes. There's a basketball term for this but I forget. Anyway, it just gets you started. 
• Then you have a volume pot, and then a second gain stage. When that volume pot is all the way up, that second gain stage is going to start distorting.
• Then we have a cathode follower, and this sucker started distorting a while ago. This'll be the first stage to break up.
• Now, after we've generated all that distortion, we shape it with the tone stack and send it on its merry way to the power amp. Well, to the phase inverter, but single-ended amps don't have those so it's moot. Because we're using a cathode follower to push the tone stack, the tone stack doesn't cost us much gain.

Compare that to a Blackface Fender preamp:

This is an AA165 Blackface Bassman and technically it uses 2 and a half tubes for the preamp, but if we were to make this a one-channel jobber we'd only need one and a half preamp tubes. We'd probably use two because most tubes stop working when you cut them in half.

• We go into a gain stage. Gotta start somewhere.
• SMACK dab into a tone stack. This one isn't driven by a cathode follower, so it's going to cost us a lot of gain. It's a fair tradeoff though: you can cut or "boost" the bass/treble much more this way. 
• Then through a volume control and finally a second gain stage. We're almost back to square one here. Not quite, but it'll be hard to overdrive this stage.
• Then into another gain stage that acts like a mixer. I picked the AA165 because the later AB165 uses a local NFB loop here to minimize any distortion this stage might generate. 
• Then we're off to the phase inverter & power amp.

So, this design makes sense for "Loud'n'Clean" Leo. Sure, you'll get some preamp distortion at the upper range of the volume pot, but the distortion will be generated after the tone controls, so we have much less control over how the distortion sounds. Distortion, as everyone intuitively knows, generates a lot of treble content, and wouldn't it be nice to tame that a little bit? Too damn bad!

Now, of course, there are other amps that are built for distortion which use a tone stack early in the preamp (though they often have a few more gain stages afterwards), and there are many other factors that make BF Fenders less than ideal distortion machines. My point is, Marshall has a recipe, and deviation from that recipe prevents you from getting a traditional Marshall sound.


So, now we finally get to the Class 5. You can find the whole schematic here.

Here's what Marshall decided to do with their two preamp tubes:

Where to begin...

• So we have our first gain stage. Fair enough. But wait! That grid-leak resistor is only 470k instead of the usual 1M! Kill it! Well, okay. It's not terrible at 470k, but compared to a 1M your guitar is going to start losing highs when you turn the guitar's volume knob down. It might also not play nicely with every effects pedal ever made. Reducing this resistor makes sense when you're getting close to the maximum grid circuit resistance for a preamp tube and you don't want to send DC through the guitar (6SJ7 I'm looking at you) but I honestly can't see a reason that this should be 470k. 
• Then we have a 47p capacitor to ground. Not sure why they put this before the grid stopper. After the grid stopper makes more sense; you could lower the grid stopper to reduce noise and keep RF interference from getting into the amp. It's not doing anything the Miller capacitance isn't already doing and it's so small that even with the guitar's volume control turned down almost all the way it's going to be rolling off treble that the guitar can't produce. 
• The first gainstage itself is fine. It's going to amplify a bunch of bass, which is okay in the first stage but usually not ideal.
• Then we have a gain control and immediately following it another coupling cap and voltage divider. I have no idea what they were trying to accomplish here. Grid stopper for cathode follower = good. DC path for grid pulldown = good. It would've been easier and cheaper to just put that 220k resistor between C9 and the volume pot and gotten rid of R7 and C13, so there must have been a reason, but I'm guessing it came after a few rounds of tweaking. 
• Then we have a cathode follower pushing a tone stack. So early in the preamp? We're not losing much gain, thanks to the cathode follower, but we also don't have a huge adjustment range for the controls, and since it's so early in the preamp it's not going to have that much influence on the distorted tone of the amp. Kind of the worst of both worlds. Cathode followers distort really interestingly in the traditional Marshall schematic, because they're DC coupled and biased quite warm so they're always tugging on the gain stage they're coupled to. Merlin goes on for a handful of pages about this in both editions of his preamp book and he does a much better job of it than I could, but because this is an AC coupled cathodyne, you're not getting that signature smooth distortion out of it. It's kind of a wasted stage here.
• Then the signal is attenuated in half and sent to a gain stage. Fair enough. The top half of that voltage divider is also acting as a big grid stopper, which minimizes blocking distortion and also rolls off some high treble, which we need to do because there's no tone stack coming later.
• And again, the signal is attenuated in half and sent to a gain stage. This one's a high-gain stage though, and since we're at the end of the preamp it's clear they're trying to get the power tube to break up before the preamp, and that's a noble goal. There's a whole ton of grid stopperage (two 470k resistors in series) which, in addition to the higher gain (notice the plate resistor is 220k) is going to roll off a good chunk of treble. I don't feel like doing the math right now, but this seems a little extreme, a little sloppy, and excessively noisy.
• Speaking of excessively noisy ways to cut treble, they put a cap across the anode resistor. This means any AC ripple is going to be injected right into that high-gain stage. The power supply seems pretty well designed, but this is a single-ended design so we really should be extra careful about that and put it after C5 if we really need to cut even more treble. You know, we could've just put the tone stack at the end of the preamp instead. Just sayin'...
• Then finally there's one last voltage divider and then the power tube.

I got to work on one of these recently, and the customer wanted these mods: Some Mods for Marshall Class 5, which I highly recommend as an easy and cost-efficient way to make this amp a whole lot better. The guy clearly knows his stuff, and while I think I would've tried some different things, I can only imagine he already did and these mods sound better.

It's not really a bad design, but considering the MSRP on these when they came out I expected a little better. Really going to town on one is pretty cost-prohibitive - who wants to dump another couple hundred rebuilding an amp that only cost a couple hundred to begin with?

I was tempted to come up with my own schematic, something I'd do if I were using a Class 5 as a donor amp and had more money than sense. Well, I guess I already meet that last criterion - based on the length of this post alone, I could be homeless and still have more money than sense. But really, if you were to gut this thing and build something inside of it, just build a Plexi or a JCM800 and call it a day. Maybe add a "preamp out" jack and run that to a beefy power amp. Or if you want something a little more unique, build a High-Octane inside. If you were really nuts and didn't mind buying all new iron, you could go for a Single-Ended Lead and really blow minds.

Alright, that's enough. I can think of a few more mods you could try on this thing - bootstrap the cathode follower to max out the gain of V1, add a freaking NFB loop for the love of god to get a crunchier overdrive and better cleans, DC-couple the cathode follower and move the volume pot to after it ( in place of jumper LK 21 perhaps), change C12 to a 1F cap for good luck... but I wouldn't have thought of the headphone mod. That's some cleverness right there.

Crate Stealth Plexi

Okay, so I'm walking back my previous schematic for turning the Crate Stealth combo into some crazily modded JCM800; I'm going to go for a lightly modded Plexi instead. It should be weird enough to have a ~30-40W plexi in an open-backed 112, with either a pair of EL34s or a quad of 6V6's.

Since the Crate's OT is fine I'm just going to use it. I have to double-check the turns ratio but if it's a 4k:4/8/16 I'll have a really hard time justifying the cost of buying a new Edcor or ClassicTone.

For the power transformer, I've decided on the Antek AS-1T275 partially for the cost, partially for the insane performance boost you get from using a toroidal transformer. Completely unloaded, it'll put out ~400VDC through a bridge rectifier. If I somehow manage to completely load it (e.g. power tubes completely overdriven to hell... and I manage to throw away another 50mA somehow) the plate voltage will only sag to ~370VDC. And, of course, it can handle the heater current without breaking a sweat.

The only downside is I'll have to use a separate bias transformer, though people have been known to find bias taps on these that aren't specified on the data sheet. In that case I'll have to use the more traditional full-wave rectifier instead of the full-wave bridge, but it's got two identical secondary windings with the phases indicated so that shouldn't be a problem.

Actually building this thing will be a little tricky. It would be great if Crate hadn't put the tubes right smack dab in the center of the chassis. This means my options are limited:

•  I could drill all new holes in the chassis for the tube sockets so that I have a nice big area in which to mount a turret board, but without buying chassis punches I'm thinking that would come out poorly. 
• I could leave the sockets where they are mount a turretboard above the sockets. I'm not sure if I have the vertical clearance to be able to do this, and since the tubes hang down they're going to be pouring heat right into the turret board. Not ideal.
• Or lastly I can get some skinny turret boards or even single-row turret strips, ignore the factory layout, and do something a little closer to point-to-point. 

I'm leaning towards the last one. Keystone makes cheap turret boards and strips in a variety of dimensions out of G-10 Garolite (glass epoxy) which I have yet to work with though I've seen it recommended highly for amp building.

This will also force me (allow me?) to disregard the original layout. That's fine; I don't particularly feel like buying some expensive cap cans anyway, and I can also design a quieter ground scheme.

For the bias supply: I'm going to try a Triad VPL36-140 which should be more than sufficient. Way more current than I need for a bias supply, but they're cheap. I've already purchased one of these for Bodie - running those 7591s in cathode bias has been a fun experiment but it kinda kills me to throw away ~15W of output power. Plus I have to bias them rather warm at idle to keep the rising cathode voltage from pushing them so far into cutoff, and even then the distortion tone suffers.

Of course, I could just follow the original schematics (in both cases) and build capacitor-coupled bias supplies. I've already had one bad run-in with those when I first started rebuilding Bodie - if they're not sufficiently loaded (say, you fire it up without the tubes in) the bias supply instead becomes a charge pump and all your filter caps see a rapidly-rising voltage. Props to whatever capacitors I was using back then for failing gracefully (instead of exploding) when subjected to 1000VDC. There have also been some rants against capacitor-coupled bias supplies over at MEF regarding their long-term reliability, so I think I'll just go with the $10 bias transformer and a half-dozen components, thanks.

Aah yes, and the filter caps. I'm keeping the plate voltages low (at least for a plexi) so standard 450V caps should be fine. Keeping the stock 450 and up plate voltages would mean needing to totem-pole some caps which takes up a whole bunch of room. Also I'm building this with the idea of using a quad of 6V6s (which is what was originally in the amp) and I know modern 6V6GTs can take that kind of plate voltage but why push it? So the amp'll break up a little earlier, oh no, not in a Plexi!

I still have to decide on whether or not to use a choke. I think most people would expect one in this sort of build, though they'd probably assume the bias transformer was one if I didn't tell them otherwise. And of course, this amp is only going to resemble a Plexi, not be a sincere attempt to clone one. The power supply is already going to be rather stiff, even if I go with the 68/69 filter cap values. Headroom would be improved, and the amp would be a touch louder.

I think I probably will use a choke, if only because I have to consider using two different types of output tubes. While the 6V6s won't pull much screen current, the EL34s definitely will. The question is will a quad of 6V6s pull as much screen current as a pair of EL34s? I'm thinking no. So with a dropping resistor for the screen supply, the screen and preamp voltages will drop more with the EL34s than with the 6V6s. Noticeably? I'll have to do the math. That may be a good thing though, really enhance the difference in the output tube choice. Pair of EL34s, it'll break up earlier. Quad of 6V6s, a little more headroom.

But with a choke, the EL34s will probably have similar headroom to the 6V6s. While arguments can be made about the plate curves of beam tetrodes (diode line) vs true pentodes (no knee) and their effects on distortion, the whole power section is wrapped in NFB anyway. There have been amps made that run a pair of pentodes alongside a pair of beam tetrodes with a pot to allow the player to pick one, the other, or some intermediate setting (IIRC a pair of 6L6GCs and a pair of EL34s) and most people report subtle differences at best.

So, I think that using an RC filter for the screen supply instead of an LC filter would enhance the "difference" in the two tube types. Artificially, it could be argued, sure. But the tone, man, the tone! Of course people will complain. Jim used chokes because Leo used chokes and you're not using a choke so this can't be any good. Bah. Bah!

I've been thinking that I'd like to try to sell this one but maybe this would be too much for the average guitarist. Well, no, I can put external bias points on the back of the amp. And the bias might not have to shift much. And hey, if the the primary impedance of the OT is what I suspect, they can even run a pair of 6V6GTs in there if they use a different speaker tap.

Man, I haven't even gotten to my preamp tweaks yet.

So yeah, a whole lot of work goes into these firebottles.

Vacuum Tube Valley Magazine

http://www.jumpjet.info/Pioneering-Wireless/eMagazines/VTV/vtv.htm

Thanks to jaaxx at ax84.com for finding this.

Shellac

Finally moved into the new house and have a little time for projects. Okay, there are still boxes all over the place, but priorities...!

After much hand-wringing and research I decided to shellac the 5E3. There's much dispute on the webternets over how this was done back in the day - some sources just say a couple coats of clear nitro, others say a coat of clear shellac first, then nitro. I'm leaning towards the latter camp, because apparently nitro can react with pine sap that can seep through tweed, and it can also react with the brown fibers in the tweed leading to a greenish color.

A lot of people like to add a couple coats of amber shellac to give it a fake aged look. I'm not going to be doing this, for the same reasons I don't take a belt sander to my guitars or pants.

My plan right now is a couple coats of clear shellac, if only because I'm sick of keeping this amp in a box to keep it from getting dirty. A coat or two of satin nitro lacquer is in the future, but for now shellac is a good start.

You can find dozens of walkthroughs online, but this guy gives good advice: http://www.mylespaul.com/forums/2549080-post1.html

For the shellac I went with Zinsser Seal Coat, which is a clear 2# cut of wax-free shellac. The first coat's drying right now. This stuff is incredibly easy to work with - goes on easy, dries fast, and if you mess up (knock on, well, tweed) you can fix your mistakes or even take it off entirely with denatured alcohol.

I've already got a bunch of "before" pictures so I'll be sure to post some "after" pictures too.

The next one!

So, still in the process of homebuying, yeah? Brainmeats don't want to stop thinking about amps. This one popped into my head today - okay, I've been thinking about Plexis a lot and it doesn't look like building a Plexi from a kit makes a whole lot of economic sense w.r.t. resale possibilities. And I've got this Crate GT50 Stealth combo that's been begging to be de-Lee Jackson'ed into some proper Marshall thing.

So why not a Plexi? Sounds like a plan, right? Well, my stupid brainmeats got around to thinking about a couple building blocks that it really likes, a couple it really wants to try out, and somehow I convinced myself to build a new "original" design. Well, original in this case means "stolen from Merlin" but... wait, why am I telling you this?! I am a God of Mojo! I invented the pentode!


*cough*

Sorry; I'm a little delirious. I just spent three hours or so typing up this:

Okay, admittedly, the power supply is a little lacking. The rest though... ho ho ho.

So, starting out: the first two stages can be either run in parallel (think: jumpered channels) or cascaded (think: one-wire mod) at the flip of a switch. Then those slam into a small-signal pentode (think: EF86). The sound gets smoothed out by a directly coupled cathode follower and the rest is pretty much "every amp ever made" though I'm going to try to design a lower output impedance long-tailed pair phase inverter so I can ixnay blocking distortion in the FOUR 6V6 OUTPUT SECTION. 

I do have a working OT this time but I've got to pick a PT which means the power supply world is my oyster. I've already started 'funny' for a guitar amp by using a full-wave rectified bias supply.

Okay.

Well, to summarize: this amp is going to be pretty high gain but extraordinarily versatile. Cleans are going to take a back seat to this glorious preamp distortion and who knows what the final result is going to look like considering how many times I changed Bodie around, but man... this'll be interesting.

Oscilloscopes are fun

Yeah, sealing my nerd fate with that title, I know. I found some 10W resistors my dad had given me years back and threw together a quick resistive load. Unfortunately I can't seem to take decent pics of the 'scope with my phone so you'll just have to take my word for it that Bodie makes a solid 27.5W RMS before distortion. In guitar advertising speak that's probably 35, hell, 40W.

The high frequency response is, well, excessive. Like 30 kHz excessive. That's over two octaves higher than even the brightest guitar speakers will reproduce and a good half an octave higher than humans can hear. I really should've checked the low frequency response.

There's a tiny smattering of intermodulation distortion at very high volume - not surprising considering that I used small filter caps. This results in the occasional ghost note but the ghosting is faint - hard to notice unless you're looking for it and often hard to reproduce. I could chase it out with bigger reservoir caps, but the sag is delicious and I'm not designing this bad boy for high gain so I'm just going to leave that alone.

I also poked around a bit and looked at the signal as it arrives at every grid. The paralleled input stage and the second gain stage are almost exactly center biased. The third stage is just a hair cold, so everything's where I want it to be. The third stage is the first to break up, which makes sense. I might attenuate a little more here to get the 2nd and 3rd gain stages to break up closer to each other... I probably won't, but it's a good excuse to try out both channels on the 'scope.

Cathodynes get a bad rap for distorting awfully but I slammed this one and couldn't see any nipple distortion or frequency doubling at all. NONE AT ALL!

There is, of course, some crossover distortion. Hello cathode biased output section. I'll investigate the zener trick of course, and maybe even the Paul Ruby mod, but the amp's not buzzy so it doesn't really matter. Gotta be careful not to become an oscilloscope hypochondriac.

Man, that would not make a good band name.

"...unless they some smart-ass pawns."

Seriously.

Yeah, more or less.

Doesn't look like much, but it's stable and surprisingly quiet. The next build I'll do on turret board and see if that's any more frustrating. Definitely not Gothik Ring caliber work, but I've seen much worse that worked just fine.

Unless I find another one cheap, it looks like I'm pretty much done with the Bogen CHB35A as a mod platform. Started out as $30 for the chassis and a working power transformer and after a year and a half of buying a few parts here and there (to the tune of a couple hundred bucks, not to mention how much I could've saved in shipping if I'd had enough green lying around to buy everything at once) it's finally done.

Well, "done" enough to gig with.

And speaking of, Napoleon in Exile is playing a show this Friday at The Mr. Roboto Project and I'm bringing this beauty on stage. Hate to pimp my band but this is really going to be a special show; in addition to the projector we usually have behind us, we're going to have interpretive dancers as well as the debut of a special cover song featuring vocals by my darling partner Agape. She is going to tear the place down, so I highly recommend you swing by. Plus, it's a free show and we're playing with The Hanging and The Smoke Tree.

Okay, blatant self-promotion mode off. I'm actually taking on less amp work lately because we're house hunting, but I'm always happy to consult, so if you have any questions or need a small service job, feel free to get in touch.

The Zener Trick

After a little more digging I found out that the reversed zener diode trick was actually patented back in 1963. (Found this out somewhere in the old Ampage archive, no luck finding it again since.) It's still pretty new in the guitar amp world - pretty sure no mass-production amps use it and apart from a handful of small hobby builders (Chuck H at M-E-F says he put it in a prototype amp he built for Dean Markley) it's unheard of.

Just goes to show how behind-the-times guitar tube amp builders are! Then again, I doubt the guys who patented it back in '63 had any idea it would make 'better' non-linearity, which is the other issue with tube guitar amps - it's uncharted territory, at least from an academic standpoint.

Richard Kuehnel (of ampbooks.com) in his book on power amps proposes that while we think of (nearly all) tube guitar amps as operating in class AB1, they should really be labelled differently considering the power tubes are routinely driven past the point where grid conduction would begin if only there were something connected to the grid to supply the current. There isn't a current source though, just a capacitor to charge up quickly and drain slowly leading to delicious farting sounds.

What I'm trying to get at is this whole tube guitar thing is nuts. We have mountains of information on how to build hi-fi tube amplifiers and we can do all the math, plot the load lines and figure out dissipation, harmonic content, linearity... but an equal part (even the greater part for many modders) is just playing it by ear, literally. But then, of course, the educated guesses are usually far more productive and innovative than the blind guesses, which is how an idea novel enough to be patented fifty years ago is still novel today.

Bodie Pics

Figured I'd snap off a couple pics of Bodie as no one's seen the outside yet. It did come with an additional chassis that I'll use eventually, but it's going to require a switch to long-shaft pots so I just cobbled together a base plate so I can gig with it in the meantime.

Not the original knobs, but eerily similar.

Everything I didn't need is disconnected.

Yeah, not going to be doing any chassis fabrication for quite a long time. I'll either buy them or have a friend with carpentry skills build them. Which I guess is kind of the same thing.

Sharpie is far from permanent, so relax.

That long lead for the pilot light clips to the inside of the additional chassis. I really don't feel like drilling a new hole to mount a pilot light temporarily, so it's fine where it is for now.

Gotta love old glass.

Actually, that Mullard ECC83 and the old stock Sylvania7591s are probably worth more than anything else on that chassis.

Admittedly, not much to look at, but hey, it's a homebrew salvage job. Hopefully that'll keep the theft potential low, I mean, I couldn't even scrounge up a Dymo labeler! After I track down some quality long-shaft (hopefully D-shaft) pots it's going to look much better. Here's a stock one that shows the outer chassis: (not my photo)

Bodie & 5E3 Sound Clips

Mourning the loss of my laptop (not really; rot in hell you plastic monstrosity) I wasn't sure if I could still record digital audio, but it turns out I can!

The playing's not the greatest, but here are the first sounds from Bodie and a quick run through the 5E3.

For both of these I'm playing through an SX Tele '69 Thinline clone with Bill Lawrence Wilde L200 pickups; volume and tone all the way up and switch set to the middle position. Oh, and it's currently tuned to B standard with a custom set of 16-70 strings. Yeah, I know.*

The mic is an Audio-Technica snare drum mic approximately a foot away.

Bodie is played through a vintage EV SRO-12L in an open-back 1x12 cab which is overall a neutral (LOUD) speaker with too much high end.

I start off messing with the "character" control, starting with it dimed and backing it down to zero over a few repetitions, just to show how powerful a real 'mids' control can be!

Later I try to demonstrate Bodie's touch sensitivity and I end with showing its (currently not very good) overdriven tone. Master is all the way up the whole time, which explains the OD tone somewhat, given the cathodyne PI and cathode-biased power tubes.

Still though, those cleans are killer.

You can catch a little of the "overdriven cathodyne frequency doubling" around 2:30. Kind of a neat effect, but I'm going to squash it. Who would've thought a 470k grid stopper would be too small?  

The 5E3 is in a pine 5E3 cabinet, and the speaker is a Weber Alnico Sig 12S.
The only thing holding this back is the guy holding the guitar! I'm just playing around a little bit with the touch sensitivity and the interactivity of the controls. It's a stock 5E3, so there shouldn't be any surprises here.

I usually spend more time dialing in the best tone on this one, but this was just a quick demo. The mic doesn't really capture how bassy it sounds, even with a Tele. Still, I love this little guy with a Les Paul; don't know why, the blues just flow.

* .016, .018, .028w, .038, .050w, .070w. The F# string (the second one) is a little too slinky at .018" but not bad enough to buy the gauge I wanted separately... probably wound too. In standard tuning I usually play 13s but I'm not a nut about it for tone or anything, I just prefer the feel.

"Bodie" Bogen Conversion Project "Finished"!

Okay, so the laptop with my schematic drawing software is currently misbehaving even more than usual so I don't have access to my schematic drawing software at the moment. There are no victories.

The good news is I finally "finished" the Bogen CHB-35A conversion project!

It's not super different from the last posted "cathodynamite" schematic, just a little tweak here and there.

Namely:

• V2A has a 47k plate load resistor - slightly less gain than standard, breaks up a little sooner, and adds more 2nd order harmonic distortion than standard or higher plate resistors.
• V2B is biased with a 2.2k resistor for a hint of crisp cold-biased flavor (not really much at all) and a little more NFB.

Here's what I need to do next:

• More useful range on the bass control (more cut probably)
• More (potential) treble cut on the treble control
• Still have to add the bright switch, might reduce the V2A grid stopper and add an attenuation resistor before the gain pot
• Bias the 7591's colder - 90+% is fine currently but I want to get that down to below 70% for tube life and so I can:
• Put a reverse-biased zener diode across the power tube cathode resistor 

That last one is pretty cool - cathode biased power amps (class AB of course) draw significantly more current the harder you drive them, which generates a greater bias voltage the louder (or more overdriven) you play. Basically the tubes bias colder at higher outputs. This is neat from a dynamic perspective (tosses a little 'squish' onto the power supply 'sag' for smooth 'bloom' and other silly words for 'compression') but the downside is when you seriously overdrive them the bias gets pushed too cold and you get ugly crossover distortion.

Right now the Vg1-k is around -18V at idle, and when slammed it jumps above -23V. Not so good! Now, if you slap a 20V 5W zener diode across that bias resistor, that 'squish' stops and you've got a fixed bias output stage. This is informally known as the "Chuck H" mod - the Paul Ruby zener mod is another neat trick for improving power amp overdrive but given the size of my grid stoppers I'm not too worried about blocking distortion.

I suppose I should point out that the 7591 data sheet lists a pair in push-pull cathode-biased AB at 28W given my approximate plate voltage and OT impedance (total coincindence... er, I mean, yeah! I was shooting for datasheet values! Nailed it!) ....but in fixed bias this pair of tubes can put out 45W.

Will that be an audible difference? Probably not; there's not much room between "That's way too loud" and "that's way too loud...er." But it's neat.


SO ANYWAY


This amp is shaping up as what I think a Fender amp should be. No reverb, sadly, but the cleans are beautiful. You can get a wide range of tones out of it - it'll ape Tweed, Blackface, Marshall, even a little Vox, but it maintains its own character.

I'm going to name the tone controls differently so they aren't confused with ordinary controls - not to be a dick, but because they don't act like typical tone controls.

• The bass control is closest to a continuously variable Orange FAC control so I'll probably call it something like "contour" or "booty" or "low range extend."
• The mid control is so incredibly powerful w.r.t. changing the overall tone (and gain!) so I'll probably call it "Character" with approximate marks for "Tweed," "Blackface," and "Death Scoop"
• The treble control is more or less the same as the tweed tone control, without any boosting, so I'll probably call it "Tone" or even better "High Cut" so people know that all the way up = flat.

I'll have sound clips up soon - it's really shaped up nicely. As it is now, the cleans are 5-star and the overdriven tone is probably 2-star, and after some tweaking, 3-star or even 4-star. It certainly won't be the first amp anyone reaches for to get a high-gain lead tone, but for primarily clean playing with occasional dips into touch-sensitive overdrive, it'll be a killer.

Oh yeah, and I went with Bodie because it sounds like Bogen and also I'm a huge Wire nerd.

5E3 Pictures

Raw cab; not a scratching post!

 The cat's been unusually interested in the raw tweed fabric. I'm keeping it in a box until it gets warm enough to spray nitro.

So, I forgot to take a picture of the chassis with nothing installed.

By the awesome power of reflection, I show you my heater wiring!

 That white lead goes to a center post in the noval socket, supposedly grounding it helps reduce noise and crosstalk and such. I've never noticed a huge difference but it's cheap so what the hell. Also I alternated phases on the heaters for the preamp tubes (and kept the phases the same for the power tubes), another thing that supposedly can help to reduce noise but if it does it's not really noticeable.

Ceramic sockets.

Above the board wiring instead of the traditional "sandwich" style. 

 Yes, all these leads are way too long but I hate this solid-core nylon cloth-covered wire and have no problem wasting it. I trimmed them before soldering to their various connections. I actually made the bright channel's volume control lead too short, which is a first.

Populated board.

PT progress.

More PT progress.

Input jacks and tone/volume controls.

 Since I didn't have a brass grounding plate on which to wire up the jacks and pots outside the chassis, I mounted them to the outside of the chassis for wiring. The piece of paper protects the face from scratches, blips of solder and/or flux, and general "whoops" moments.

Closeup!

I soldered on the underside of the board going for a better mechanical connection as that's where the leads fold over the lip of the eyelets.

Output sockets almost finished.

PT almost finished.

 I went a little nuts with the zipties. Really hard to get these PT leads to hold a tight spiral; I think next time I'll just leave them straight and ziptie every inch or so. The twist itself doesn't matter, you just want to keep the corresponding pairs close together.

Controls mounted.

Jacks.

 I gotta say, it's a real pain to try to take pictures of something so shiny.

Mounted the board.

 Still waiting on my replacement pilot light holder in this shot.

DELICIOUS SPAGHETTI

 I was tempted to just box it up at this point and see if anyone would notice.

Output sockets done. I think?

Ready to go!

 Here it's got the stock tubes in it, and an old stock 5Y3. I've since put in all old stock tubes. The 6V6s I have aren't matched (the horror!) but they're both dissipating safely at 10 and 12W.

Preamp sockets.

Output sockets totally done this time.

Standby, Fuse, Power, Pilot.

 So, I don't like standby switches, but I dislike ground switches even more. The hole's already there and labeled "STANDBY" so what the hell.

Pots all wired up.

Jacks all wired up.

 Man, those two white leads connecting to the bottom jacks were A REAL TREAT, especially as I hadn't trimmed them before I mounted the board. This cloth wire really sucks too; did what I could to get rid of the fuzzies (burned them off with a lighter) but this wire was just fighting me the whole time. They also didn't send enough colors to have a consistent color scheme. They included the slightly thicker (18 AWG?) stranded white and black wire for the speaker cable but nuh-uh, I've got 16 AWG lamp cord for that. The red, green, and yellow wires are all solid-core.

Done!

Done!

Donnnnnne!

 Oh yeah, the cathode resistor on the power tubes is now a 300R/10W because I didn't have a 5W lying around. I could (should?) probably go a little colder but it's not a priority. I'd need a matched pair of tubes but I've got a few old stock 6V6s and a half dozen of cheap recent ones so that's not going to happen any time soon.

Aw, yours goes to 11? That's cute.

So there you go! Not particularly hard but a little frustrating at times. I spread the build out over a number of nights, maybe 15 hours or so in all. 

And dear god is this a sweet amp. I knew there was a reason why so many people build these but DAAAMN. I can't put the thing down.

I should measure the output on the 'scope. These are usually quoted around 12W which means it's too loud for bedroom playing but not quite loud enough for being onstage by itself. Of course, most venues have PA systems these days so being quiet isn't a problem.

Oh, and I guess I should mention the hiss/hum/buzz noise. There isn't any. There is some slight hiss when you turn the volume up past six - it's not the quietest build I've ever done but I think it's still impressively quiet. There are other things that could be done to kill the last little bit of his - shielded cable on that huge grid lead, grid stoppers, screen grid stoppers, actually shielding the back of the amp too but it's not really necessary.

Fun, fun little amp.

5E3 Kit Build Done!

Okay, I know, I'm a bad blogger. Been weeks without an update and now an update without pictures. C'est la vie.

First the bad:

• The cab came with some glue overspray on the bare pine, locking in someone's pencil markings. I guess I could sand this off but I don't really see the point. 
• The box joints are cut too deep on one corner so they don't mount flush - you can see some gaps on the inside. The outside is smooth even through the tweed, so I'm assuming it looks fine on the outside. 
• The kit was missing the brass grounding plate. I wasn't going to use it for grounding, but it would've helped assembly somewhat. Customer service was very responsive when I was considering upgrading my cab, but I haven't been able to get them to return my emails since.  UPDATE: I have since received the brass grounding plate. Looks like these folks are just super busy.
• The pilot light holder fell apart almost immediately. The fuse holder does look cheap (the two biggest caveats of these kits) but it's holding up fine. I bought a replacement but I don't see a need to swap it in yet. 
• One of the chicken head knobs was chipped and missing its setscrew. 
• The eyelet board was shipped loose inside the chassis (despite everything else being wrapped in foam) and scraped up the inside of the chassis a bit. The scratches are under where (heh) the eyelet board is mounted, so whatever. 
• The cap over the speaker's magnet came off during shipping and was rolling around. There was no damage. 
• The sheet metal screws for mounting the eyelet board to the chassis apparently aren't self-tapping, or if they are I did something wrong. I know this is how it was done Back In The Day, but I just used #6 machine screws with nuts for a more secure connection. 
• The ceramic octal sockets aren't great. The plastic standoffs were too thick so with them installed the socket couldn't be secured to the chassis. One of the pins just up and fell out of an octal socket - one of the four you need for the 5Y3. I replaced it with an adjacent unused pin.

 Now the good:

• The chassis face is really shiny; the stainless took a good polish and the outside of the chassis is impeccable. 
• The cab is sturdy, with no rattles or vibrations.
• The chassis mounts well in the cab. 
• Everything functions as it should, the component values were all correct, and the tubes are fine. 
• The speaker sounds great. In fact, the whole amp sounds great. That's kind of a given with a proven design, but it's the most important criterion, so there you go. 
• The cost is right. If I could've swung another couple hundred I would've avoided a few headaches, sure, but at my budget I'll take the headaches. It's like I got paid to get frustrated! 

I'm sure I'll think of more as I go along. 

I did a few things differently from the stock build, but I left the circuit alone. Some things did pain me to leave out (no screen grid stoppers?! no grid stopper on the cathodyne?! freaking huge coupling & bypass caps all over the place?!) but it's a 5E3. By the time I got done with it, it would sound like a completely different amp - might not be a bad thing, but it wouldn't be a 5E3 anymore.

I did change the grounding scheme though, and I would have even if I had gotten the grounding plate. The stock grounding is just a terrible idea; sorry Leo.

Not super helpful in grey, but I can't convince Paint that this is actually a color picture. Ignore the heater wiring and ground switch.

The circuit only contacts the chassis at one point - the input jacks. Sure, they shouldn't be daisy-chained, but it's a tight fit getting three leads into those jacks' terminals. I might fix that at some point (sleeve of 1 to sleeve of 2, connect sleeves of 2 together) but it's so quiet as-is that I might not bother. Eh, it's going to bug me.

The output jacks are connected to the chassis, but they're isolated from the circuit by the output transformer, and there's no NFB. No current flow = no ground loop. If I had some isolated jacks lying around I would've used those. The main output jack is a switching jack, which helps if there's no speaker cord plugged in. Damn useless if there is a speaker cord plugged in on the amp side and disconnected on the cab side though. I'd prefer to use isolated jacks with a 470R/5W resistor across the terminals as "oh shit" protection.

Anyway -

• there are no ground currents rippling around through the chassis, so it's acting as a shield. 
• The input jacks are grounded to the first stage's cathode resistor - it would be better to have the filter cap right here, but whatcha gonna do. 
• The tone/volume controls are grounded to the next stage's cathode resistor, as they act as its grid leak.
• All the grounds on the filter caps are connected together, and the PT center tap goes right to the negative terminal of the first filter cap. 
• This chassis comes with a welded bolt intended for use as a ground point. I instead used it as the earth ground connection, which normally builders wedge between the PT and the chassis on one of the PT bolts, which is a bad idea - understandable in a vintage amp where you don't want to drill new holes, but in a new build it's a no-brainer.

In the same vein, I connected the heater center tap to the top of the 6V6s' cathode resistor for 20-some-odd volts of heater elevation. Might as well in a cathode-biased amp; free noise reduction.

The last main tweak I did is running all the wires above the eyelet board (but under the components.) It's a little uglier, sure, but it makes swapping components (and verifying connections) so much easier.

Wow would this be so much better with pictures. I'll update when I can find that stupid USB cable.