I've been documenting this project on the DIY Amp forum at TDPRI, but I thought that it might be more interesting here, because it strays from the usual build-a-Fender-5E3 content.
My dad passed on 10 years ago this past January, at the ripe old age of 83. On a recent trip back to see my mom, I came across the Heathkit integrated amp that he built in 1961, shortly before they got married. It's in pretty sad shape cosmetically, and hasn't had any service since 1980 (per the service sticker from the Heathkit factory!). It's got no value as a vintage piece, and I've got nowhere to use a 14W mono amp, so I'm converting it for guitar use.
I've done simple things before - cap replacements, simple circuit mods, etc. I know how to follow good safety practices. This is the most involved project that I've undertaken.
The original plan was pretty simple - scrap the input selector and RIAA filter, Fender-ize the front end, and add a PPI MV. And I'm still pretty much in that ballpark, with one major change; I'm changing the third gain stage (6AU6) to a cathode follower. If I leave it as a gain stage, there's just too much gain, and I may want to try alternative tone stacks that load it down more than the Bax does.
I feel like I've got a decent handle on the circuit, based on studying both theory and execution, but there are a few things that aren't entirely clear to me; I'll be asking about those along the way. I'll post the updated schematic in the next comment.
My dad passed on 10 years ago this past January, at the ripe old age of 83. On a recent trip back to see my mom, I came across the Heathkit integrated amp that he built in 1961, shortly before they got married. It's in pretty sad shape cosmetically, and hasn't had any service since 1980 (per the service sticker from the Heathkit factory!). It's got no value as a vintage piece, and I've got nowhere to use a 14W mono amp, so I'm converting it for guitar use.
I've done simple things before - cap replacements, simple circuit mods, etc. I know how to follow good safety practices. This is the most involved project that I've undertaken.
The original plan was pretty simple - scrap the input selector and RIAA filter, Fender-ize the front end, and add a PPI MV. And I'm still pretty much in that ballpark, with one major change; I'm changing the third gain stage (6AU6) to a cathode follower. If I leave it as a gain stage, there's just too much gain, and I may want to try alternative tone stacks that load it down more than the Bax does.
I feel like I've got a decent handle on the circuit, based on studying both theory and execution, but there are a few things that aren't entirely clear to me; I'll be asking about those along the way. I'll post the updated schematic in the next comment.
Moved from Tubes/Valves to Instruments and Amps per forum policy.
I recently aquired, modified and sold the same amp on ebay. Cut my finger real bad when a tie wrap snapped as I was pulling hard on it slamming my finger onto the chassis edge; cost of the stitches well exceeded any $ I made.
I modified it by placing a fixed fender style tone stack between the 1st and second 12AX7 stage, an idea I got from a Mesa (insert model I cant remember) amp schematic on line. I put the volume after the 2nd stage, figuring the guitar volume control would define the distortion on these two stages. No one wanted that on ebay, including the guy I finally sold it to; ended up ripping it out of the circuit entirely and using just 1/2 of the 12AX7 in circuit.
I'll list the problems I had with it and comments -
- I couldnt get an isolated input jack where I could wire the ground to the amp's original star location, so either a ground loop or move the star physically to the input jack.
- Leaky coupling caps confused me initially as to why the amp output tubes were red plating. It was a kit and the builder put one in backward. Replaced those with film caps -
- The output tube bias was initially designed pretty hot and I ended up cooking a few output tubes before I was able to get it stable enough to sell in good conscious. Ended up with a waaay higher value common cathode resistor than stock.
- Leaving the original Baxandall tone controls in place, there wasnt enough room on the panel to add both a pre and post volume control, while being able to plug in the guitar there. Figured guitar volume would be "pre" - no one liked it.
- I managed a switch to go from tube to SS rectification. Made a difference.
- I managed another switch to cut the signal to one of the output tubes, which made it "single ended" with the other tube idling to provide the balance current for the OPT. Also made a difference, with a surprisingly small drop in loudness in SE mode.
- I wired a grounded 3 way cord into it, so chassis was connected to earth ground.
- I wired a 1/4" speaker jack at the back panel.
- I had to throw additional outboard capacitance across the original can, to get the hum down to a reasonable level - even with the reduced gain circuit.
The next time I see one of these at a yard sale, thrift store or ham radio swap meet, I'm going to leave it. Maybe if I can get it for $5, I'll be duped again to try making something like a guitar amp of it. This last time - last time - I got absolutely killed for all the $ and effort I put into it, which I thought would be fun...including the cheap thrill of perhaps reaping $150 for ebay play money. Ended up buying new output tubes in order to get it even working - unit appeared to be in great shape when I got it in Puyallup. It was haunted.
At least I can share the experience.
I modified it by placing a fixed fender style tone stack between the 1st and second 12AX7 stage, an idea I got from a Mesa (insert model I cant remember) amp schematic on line. I put the volume after the 2nd stage, figuring the guitar volume control would define the distortion on these two stages. No one wanted that on ebay, including the guy I finally sold it to; ended up ripping it out of the circuit entirely and using just 1/2 of the 12AX7 in circuit.
I'll list the problems I had with it and comments -
- I couldnt get an isolated input jack where I could wire the ground to the amp's original star location, so either a ground loop or move the star physically to the input jack.
- Leaky coupling caps confused me initially as to why the amp output tubes were red plating. It was a kit and the builder put one in backward. Replaced those with film caps -
- The output tube bias was initially designed pretty hot and I ended up cooking a few output tubes before I was able to get it stable enough to sell in good conscious. Ended up with a waaay higher value common cathode resistor than stock.
- Leaving the original Baxandall tone controls in place, there wasnt enough room on the panel to add both a pre and post volume control, while being able to plug in the guitar there. Figured guitar volume would be "pre" - no one liked it.
- I managed a switch to go from tube to SS rectification. Made a difference.
- I managed another switch to cut the signal to one of the output tubes, which made it "single ended" with the other tube idling to provide the balance current for the OPT. Also made a difference, with a surprisingly small drop in loudness in SE mode.
- I wired a grounded 3 way cord into it, so chassis was connected to earth ground.
- I wired a 1/4" speaker jack at the back panel.
- I had to throw additional outboard capacitance across the original can, to get the hum down to a reasonable level - even with the reduced gain circuit.
The next time I see one of these at a yard sale, thrift store or ham radio swap meet, I'm going to leave it. Maybe if I can get it for $5, I'll be duped again to try making something like a guitar amp of it. This last time - last time - I got absolutely killed for all the $ and effort I put into it, which I thought would be fun...including the cheap thrill of perhaps reaping $150 for ebay play money. Ended up buying new output tubes in order to get it even working - unit appeared to be in great shape when I got it in Puyallup. It was haunted.
At least I can share the experience.
Last edited:
Thanks for sharing! I'll be glad to help if I can.
Since you're already considering a cathode-follower, you might find this video clip interesting: YouTube
That clip has convinced me to build a 2204 preamp one day. I knew you could get heavy classic-rock levels of distortion from one, but it was only recently that I discovered that the 2204 / 2203 preamp can also produce lovely clean tones. To my ears, the use of more triodes in the signal chain, and better-planned bias points (than Fender), has resulted in much more attractive clean tone than I've heard from the classic Blackface Fenders.
Ouch! Sorry to hear that. I hope it's better now.
With the Coronavirus pandemic straining the medical system to its limits, I've become more aware of the need for safety in my DIY activities. Now is a bad time to need to be patched up by a doctor.
Again, sorry to hear that.
I thought we would see a surge of new builds and activity on this forum when the Coronavirus lock-downs started, but that doesn't seem to have happened. I think we all have so many new stresses to deal with that it isn't as much fun taking on additional - and unnecessary - challenges.
I'm still doing projects, but trying to pick ones that help me to feel better. Sometimes that means picking things which require more physicality, and less brain-strain. At the moment, I'm building a wooden bench to put out on the patio - both my wife and I find we're happier if we spend at least part of our day out there, where we're exposed to at least a little bit of nature.
(That increase in happiness seemed to justify the risk of using a circular saw during a pandemic.)
The bench is together now, but needs to be sanded and varnished. Much less fun than building the structure, but it needs to be done...
-Gnobuddy
Of possible interest: a thirteen year old thread on exactly the same topic: Heathkit EA-3 conversion to guitar amp
I forgot to mention in my earlier post - replacing the 6AU6 with a 12AX7 will get you the four preamp triodes needed to build the Marshall 2204/2203 preamp, if you decide to go that route.
-Gnobuddy
I forgot to mention in my earlier post - replacing the 6AU6 with a 12AX7 will get you the four preamp triodes needed to build the Marshall 2204/2203 preamp, if you decide to go that route.
-Gnobuddy
Man, that all sounds like a real ordeal. Sorry you went through all that. I really appreciate getting your experience on it. A couple comments along the way:
(Un?)luckily, this example is missing the logo plate. Which leaves two perfectly-sized holes for jacks that are already inside the chassis. So I've been able to maintain the star without incident. Since I've only ever worked on guitar amps before, the floating ground really threw me though!
I'm replacing every cap anyway. Saving the ones I take out for another project, but I figured I'm rewiring enough of the circuit that it's not a big deal to have a few more components to put in.
Do you happen to remember what value you ended up with? I was pretty happy that the original WW resistor was within .1% tolerance still (measured 210.2R) and was hoping to leave it. I've also got a fresh pair of 6P14P-EVs to put in, just in case things still run too hot. The existing tubes are a mismatched pair with what appears to be an original ca. 1961 GE, and a Japanese "Realistic Lifetime" tube from the 80s.
Since I've got a place for the jack, I was able to put the MV in the spot for the selector switch. Yeah, it's a little odd to have the controls laid out MV-V-B-T, but whatev.
Do you recall what happened with the voltages when you did that? I've been a little stumped that I'm not getting more out of the rectifier than I am. With no load, I'm only seeing 349VDC from 352/350VAC. The schematic calls for 380VDC there. Same results with both an EZ81 and a 6BW4. Did you keep the 100R series resistor?
That's interesting! Haven't heard of doing that before. What happened to the tone, since you're only amplifying one half of the waveform? Or were you biased up into class A?
The can in mine is shot - two sections read open. I'm replacing it with a new one from JJ. I could have saved a few bucks by going discrete and leaving the old one for looks, but I haven't done a can replacement before and wanted to try it.
Awesome, thanks! I'm new to the circuit design piece, so I can definitely use the help.
That's a killer-sounding amp!
Yup, one of the first threads I saw. I wish there had been more followup on it!
Only problem with that is that the chassis is punched for a 7-pin socket and I was hoping not to have to mess with that. That said, switching to diode rectification would free up the 9-pin rectifier socket, and I could use something like a 6AB4 in the 7-pin to drive a tremolo. But first I just want to get a working amp!
As promised, here's my planned circuit. Voltages are from the original schematic. The only change from the tone stack on is the addition of a Trainwreck Type 3 MV.
Oh, wait, I just had an epiphany - the rectifier output I mentioned above isn't filtered, it's just the raw output at the cathode. So of course it's got a ~1:1 ratio of VAC input to VDC output! Once that first filter cap is in place, unloaded DC potential should rise (close) to the peak of the AC (1.4*RMS).
Right?
Oh, wait, I just had an epiphany - the rectifier output I mentioned above isn't filtered, it's just the raw output at the cathode. So of course it's got a ~1:1 ratio of VAC input to VDC output! Once that first filter cap is in place, unloaded DC potential should rise (close) to the peak of the AC (1.4*RMS).
Right?
Good to hear about the Ground. Mine was missing that plate also. I used the two holes for the two switches; the tube/SS rectifier switch and the single ended / push-pull switch.
I ended up giving each output tube its own resistor. I had to make ready for sale in a hurry and that was the only solution I had on hand. The last thing I wanted was to deliver an amp that red-plated the output tubes.
Unfortunately, I cant remember the exact values I used. I mentioned it because it's something to watch for - perhaps some output tubes can take it as originally designed. Evidence that one was replaced in your amp leaves me to further believe it was set a little hot.
The voltage values went up with the SS rectifier and the impedance went down - even the hum sounded different. I think its a worthwhile mod, though with fully replacing the filter can, there may be a more subtle difference than what I was hearing. Values were close to 400V as I recall. If MESA can do it, so can we.
I imagine the output tubes are biased pretty strongly towards class A, so when I cut just the drive signal to one, it didnt result in a 1/2 wave output like from a diode rectifier. It changed the tone to a lot thinner sound, and reduced the output perceptibly. The idea was that this sort of amp would be used as a recording / practice tone effect anyway, versus a club gig mainstay.
Another less drastic way to do it is to deliberately unbalance the cathodyne drive - some. Like make the plate resistor 50K with the cathode 100K. I can see the increase of the 2nd harmonic on a spectrum analyser under such conditions, unsure if I can hear it. I can be pretty iron-eared when it comes to tonal differences between mid-frequency instruments and amps like guitar. "Now we're going to replace the original 7027 in the 1st stage with an EH 12AX7 and I'll play the same solo - hear what that did?" "Not really..."
I tried this with one of my builds, with unsatisfactory results. I was hoping for single-ended clean guitar sounds, but didn't get them.
At the time I didn't have a 'scope (having just moved from Los Angeles to British Columbia to avoid the impending political disaster I anticipated), so it's always possible that I actually reduced the existing imbalance due to mis-matched valves and an imperfectly wound output transformer.
In the end, I did find a way to get what I wanted - I used a small-signal beam tetrode as the first stage in the power amp section, the stage before the cathodyne, and I added some fixed attenuation between that stage and the cathodyne. This let me drive that input stage enough to produce some of the same singing clean tone you get from bigger beam tetrodes like the 6V6.
EQ turned out to be part of the recipe, too. I got better results if I applied a gently rising treble boost in the 1 kHz - 5 kHz band to the signal before it entered the little beam tetrode stage.
I came up with those ideas using only my ears, but after I'd found work and had a stable income again, I got myself a 'scope. And I was surprised at how much nonlinear distortion I'd dialed in to get that nice-sounding "clean tone". Image attached - this was the signal to the cathodyne, with a clean sinewave input, just before any outright clipping occurred anywhere in the power amp section.
Musicians fool themselves all the time. As a keen amateur musician, and also a person with a solid technical background and belief in the scientific method, I constantly find myself caught between these two different ways of making sense of the world when I'm playing music.
As a musician, I *have* to be subjective and emotional and superstitious. If I don't hear angels and see rainbows when I play music, my audience certainly won't see or hear those things! My own emotional state is crucial - if I'm in a logical left-brain-dominant state, I will invariably be a crappy musician; I need to switch over to a more hazy and emotional right-brain-dominant state.
This rule applies to every musician I've ever heard. You have to be in a non-thinking, almost meditative state in order to be musical. If you think consciously while you play, you invariably produce bad music. (But thinking and logical analysis is fine - and necessary for progress - during learning and practice sessions. Just not when you're actually performing a piece of music.)
But I'm also my own sound engineer, and often, also the guy who designs my own guitar electronics. If I stay in the superstitious musician mindset when doing those things, I'll be a gullible and crappy designer, falling for superstitious nonsense like the magic sounds of red Teflon insulation, ancient and decrepit paper-in-oil capacitors, and tubes made in Nazi Germany.
So when I'm working on electronics, I need to be in a logical, left-brain-dominant state, with critical analysis faculties fully engaged. No unicorns, no rainbows, no angels, just volts, amps, Kirchoff's laws. A terrible mental state for making music, but a good one for designing electronics!
Many years ago I worked at a mid-size electronics company, and there was an annual half-day celebration, at which many of the employees would go up on stage and play music. That's when I first noticed how bad an engineer stuck in thinking mode sounds when trying to improvise a guitar solo.
But there was one engineer - a very good one - who also turned out to be a very good musician. I could literally *see* him shift mental gears as soon as he picked up his guitar, before he ever played the first note. Watch someone like Derek Trucks, and you'll see the same shift - when he's playing, he seems to be far away, in a meditative state, unaware of the audience or the technical details of the performance.
There are plenty of mediocre musicians who are "stars" - they don't make this mental shift, as their focus is all about themselves, their audience, and their own self-importance as the performer. Those who are good at this captivate their audiences with sheer charisma and/or good looks, despite being bad musicians. Taylor Swift comes to mind as an example.
-Gnobuddy
Attachments
I'll second that request!
(It's too small to read any of the component values or text.)
-Gnobuddy
Thanks, much easier on the eyes. The first thought, why run the first 6AU6 in pentode rather than triode since you are using it as a follower anyway? And if you are using it as a follower (clean no gain) then using it as a triode dc coupled to the previous stage will clean up some parts as well as get cathode follower compression.
I've had at least three of these amps at different life points. The one I cut my finger on was late last year, just before all this corona pandemic - so no ish going to the walk-in treatment place. One was in the early 80's and I can still recall putting a complete single ended power amplifier where the 6AU6 is - loading the output transformer with a resistor, followed by the rest of the amplifier.
At the time, I had no idea there were maybe 100's of people doing such and there'd one day be this "boutique small amp" industry. I thought at the time that anything with less than 20W was just a piece of junk no one would want - that amp literally ended up on the trash heap at the local recycling station.
At that time, I also rebuilt a Altec amp which had enough tube stages that I could move the phase splitter all the way up to the input tube. Some guy I tried to sell that to remarked I "got the sound of EL84s out of 6L6's". He never came back with the money and to this day I have no idea what he heard.
I also recall building a tube amp with lead and rhythm channels, using mechanical relays to switch the signal path. Lots of inspiration, little perspiration - and absolutely no idea what I was doing beyond simply rewiring existing circuitry - I called it relacing, like tying your shoes differently.
Yes, I'm that way, naturally, like a duck takes to water. There's been momentary glimpses of that in action throughout my amateur musical career. A work assigned road trip to a meeting with a bunch of acoustic engineers from all the major computer companies, I'm at a bar with Dr Nobile (?) from IBM. I beg the bassist to let me play a song, saying I'm from out of town and here with my friend. (He actually let's me play his instrument!) I dont know what I did or even what song, but I proceeded to blow the room away. After I can tell the bass guy is a bit miffed - he's doing all these technical runs up and down the fretboard - because I'd just shown the soul of it so clearly - and literally coming out of nowhere - apparently effortlessly.
Dr Nobile later told me in an email he bought himself a bass. Ah, that's what that instance was all about!
But I'm also my own sound engineer, and often, also the guy who designs my own guitar electronics. If I stay in the superstitious musician mindset when doing those things, I'll be a gullible and crappy designer, falling for superstitious nonsense like the magic sounds of red Teflon insulation, ancient and decrepit paper-in-oil capacitors, and tubes made in Nazi Germany.
So when I'm working on electronics, I need to be in a logical, left-brain-dominant state, with critical analysis faculties fully engaged. No unicorns, no rainbows, no angels, just volts, amps, Kirchoff's laws. A terrible mental state for making music, but a good one for designing electronics!
Many years ago I worked at a mid-size electronics company, and there was an annual half-day celebration, at which many of the employees would go up on stage and play music. That's when I first noticed how bad an engineer stuck in thinking mode sounds when trying to improvise a guitar solo.
But there was one engineer - a very good one - who also turned out to be a very good musician. I could literally *see* him shift mental gears as soon as he picked up his guitar, before he ever played the first note. Watch someone like Derek Trucks, and you'll see the same shift - when he's playing, he seems to be far away, in a meditative state, unaware of the audience or the technical details of the performance.
There are plenty of mediocre musicians who are "stars" - they don't make this mental shift, as their focus is all about themselves, their audience, and their own self-importance as the performer. Those who are good at this captivate their audiences with sheer charisma and/or good looks, despite being bad musicians. Taylor Swift comes to mind as an example.
-Gnobuddy
When I was working at the 1st computer company just after college, during that career other employees would organize band showcasing events where all the bands were comprised of employees. I participated as a bassist in two. Many others were engineers - and musicians waaay above my caliber - one guitarist being a Berkley school of music graduate, another a classically trained pianist who could play ELP tunes no problem. I've so often had the fortune of getting to play with folks 10, 100X better than me. I could "hold up the band" with a bass groove, had good timing with most any drummer and played nice 'n loud, so they'd let me in. Unfortunately, being in our early 20's none of us ended up sounding like Led Zep or Yes - when they were in their early 20's. There were some good moments!
I read Allen Holdsworth said "The only thing I do anymore is play the 1st note - then it takes off from there all on its own". I get that. That guy can still send chills through me, 35 years after his recorded performance. I never had the perspiration part of the conditions required to reach up into that level of instrument-unification enlightenment. Just many, many opportunities to get a glimpse into what could have been achieved with the "given" conditions - versus those that had to be earned...
Last edited:
In the Bassman and its various copies, the CF compression only occurs because the half-12AX7 has a high ra (anode resistance, plate resistance), doesn't have enough DC voltage between anode and cathode, and therefore cannot supply enough current to the 100k cathode load. Since it can't deliver enough current, the output waveform gets flattened on positive peaks. This was accidental, a bug that later turned into a feature.
But if you use a valve with a lower ra, the same thing won't happen. From the datasheet, a 6AU6 will happily supply 15 mA of cathode current before running out of steam; a 12AX7 has trouble supplying 3 mA under the same conditions.
So I don't think you will get any cathode follower compression from a 6AU6 DC coupled to the previous stage, not unless you go to extremes, such as using a low-value cathode resistor, and maybe lowering B+ to the 6AU6. And if you do those things, my bet is that in pentode mode, the CF compression will be abrupt, not gradual like a half-12AX7 in the same situation.
There is a drawing error in the connection from the second half-12AX7 to the 6AU6 stage - the coupling cap should be connected to the anode, not B+ (see first attached image.)
I would also suggest a larger B+ to the 12AX7. Take a look at the second image showing the load-line and operating point. With a 195V B+ and your chosen 100k/1.5k anode/cathode resistors, a half-12AX7 is rather warm-biased, and doesn't have much headroom for guitar at either its input or its output.
A few years ago I built a preamp that contains a mix of triodes and pentodes, and I ended up running all the triodes on a higher B+, and the pentodes on a lower B+, which was the only way I could keep both types of device operating happily.
-Gnobuddy
Attachments
Cool story!
It's been the same for my best musical moments as well - I don't know what I did, and it was effortless. (But I don't switch to that mental state easily, or anywhere near as often as I'd like to.)
There is a very good book on this subject, and its title is "Effortless Mastery": Kenny Werner | Effortless Mastery
Mr. Werner's thesis is essentially that the performance has to be effortless for it to be good, and the performer has to learn how to get into that mental state if he/she wants to perform well. It's either effortless, or it'll be mediocre at best...
-Gnobuddy
Cool story!
...the performance has to be effortless for it to be good, and the performer has to learn how to get into that mental state if he/she wants to perform well. It's either effortless, or it'll be mediocre at best...
-Gnobuddy
Thanks! For me, the "mental state" was a given and one of the downsides is that I've always been sort of a space-shot and dreamer for all my life. I can do stuff like "imagine", "call upon the subconscious" and "think outside the box - feel my way to a quite different solution" as if "is there any other way to be?" but when it comes to consistent hard work - like, that other half of it - I've been pretty deficient for most of this life. Result? I simply dont have the 100's of solid accomplishments corresponding to the 100's of opportunities that life has brought my way. Maybe that's like having the success of Led Zep and ELP - it's a rarity as well for most people. Dunno -
Back on topic - try putting a 6AQ5 single ended power amp with a small OPT where the 6EU6 is. There - vindicated for the highjack!
Last edited:
Good catch! Yuck. What an awful design. Who in the world came up with that monstrosity?
That master volume cuts the bass more and more heavily as you lower the volume. Exactly the opposite of what you want!
It also concerns me that there might be up to 300 volts inside that pot. If you must use that circuit, definitely use a plastic-shaft pot, to avoid the risk of shock. Also make sure the metal front panel is grounded - you don't want the pot to make the entire front panel live at 300 volts if things go bad internally.
I've seen that used in many DIY schematics, and that's the only sane way to go for a PPIMV.
-Gnobuddy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.