Cascode

My booster design is still sprouting new variations. While working on a possible compressor based on my symmetric current mirror amplifier I stumbled on an aspect of the design I had not realised before: The symmetric gain stage can be made inverting, simply by altering two resistor values.
This could be a worthwhile feature in a booster design, and after some circuit doodling I ended up with a possible solution which I prototyped (photo top right). It kinda worked, but needed a dual gang potentiometer for setting the gain.
I then redesigned a bit to avoid the dual gang pot, and came up with a single potentiometer prototype (photo top left). This version also tested a couple of different topologies to increase gain. After some testing, the design was almost set; gain control and phase selection was worked out and only the best solution for implementing active bypass remained to be found.
Third prototype (photo bottom left) was fabricated to solve buffered bypass. The most elegant solution had to be discarded, much to my regret, because I couldn't get rid of a very audible click when the bypass switch was activated.
A less elegant solution, which adds four transistors to the circuit, was implemented in the fourth prototype (photo bottom right) and this solution is clickless. It does bring the total number of transistors up from nine to thirteen, but I see no way around the added complexity.
Next stage is adding a power booster to the circuit, to enable it to work with a standard 9 V supply, and also "pedalize" it by designing a circuit board in pedal format with relay switching and so on. I plan to do at least two versions: A small version with one gain range and phase switch, and a big version with two gain ranges and phase switch. I could do a true bypass version, but I really don't see the point.
I have implemented this design in my high voltage booster/buffer environment, which means that there is tremendous headroom present. The input range is 45 V peak to peak when in buffered bypass mode, five times as large as found in a 9 V pedal; at lowest gain setting in active mode the input can be as large as 8 V peak to peak before the output is saturated and start to clip the signal.
Low noise is a constant concern for me. How the circuit will perform noise wise remains to be seen, current mirrors can disappoint in that regard. A proper assesment will have to wait for a pedalized version, as a measurement without the circuit fully and properly mounted inside a metal enclosure is pointless.
The big version with an additional gain range will be able to output a staggering 18V RMS and have 45 dB of gain. That's the equivalent of a 40 W amplifier at full throttle. This could potentially fry the input stage on the following pedal or amp, so this version should be used with caution. I'm toying with the idea of putting the gain range switch under a missile launch cover, to emphasize the potential hazard.

The Camping caravan has been assembled and voiced. At least the first attempt is now complete and ready for mounting in the enclosure.
Both gain distribution and voicing has turned out to be much more finicky than expected. And I'm still unconvinced on the overall sound of this design. The overdrive is harsher than I'd like it to be, and not as tube-like as expected (or hoped for).
The transformer in the speaker simulation stage has a very weak low end response, which proved to be a hard nut to crack. Part of the motivation for doing the design was to see if a very cheap transformer could be used in a guitar pedal design, and the result is a wavering yes - but only just, and with great effort. I'll try to use the transformer as an inductor for a voicing filter in a later version.

In mid December I received the CMA5 boards I posted about in late November. As I was busy designing a board for Jesper's resonator guitar project, I postponed assembling a complete pedal using the board. But Jesper and I did a pedal test session after working on his resonator guitar, and Jesper seemed to like the sound of stacking the pedals called Cube Root and Ouzoud. I suddenly realised that I could emulate the two pedals with the CMA5 PCB, and put together a combined pedal in time for christmas.
After editing the part values in the CMA design to be like the older designs, adding a few extra parts and omitting a few other parts, the result is what you see in the pictures. I only needed four controls in addition to the main level control, so two out of the planned six controls was sacrificed in favor of the JH logo. The controls function like, and are named like, the controls on the two original pedals.
The name is based on Jesper's comparison of the sound of the stacked pedals to the sound Tony Iommi was using on a Black Sabbath album from the late seventies.
I have since realised I may have to adjust a few component values to give the pedal more gain, Jesper has installed the pedal on his pedal board and has turned all k***s, except Main Level, fully clockwise.

Latest build! Here's a copper variation of the gold top design I posted about in mid December '23. I've had trouble getting the clear label, which hold the graphics, to adhere properly to the copper colored plastic foil without tiny air bubbles getting trapped between the two. Large cloudy areas screaming PLASTIC IMITATION right in your face. I find the result shown passable, but the yield in my little sticker factory is a measly 25%.
This design style can be considered a poor man's - or lazy craftman's - steam punk styling. Despite the sticker trouble, the required labor time is greatly reduced in comparison to the all-metal design I've shown recently. The lamp is made with LED filament wound round a tiny O-ring in a miniature glass bottle.
I've added some side by side pictures of the two variations for comparison. Last there's a picture with a different style copper plastic foil, non-shiny with a bit of relief pattern, which is, much to my surprise, much easier to work with than the shiny foil. I ALMOST used this instead of the shiny one, but decided that it would be too far from the gold top style.
The electronics is yet again my multi-stage (cascaded) current mirror amplifier design, which I have gained to go from a tube-like overdrive to a ripping distortion. It's not very loud, as it is very compressive at all settings. Definitely NOT an "always on" pedal

This post is about the upcoming combined cap and amp simulator pedal that I will call CAMPING. That's what I got when playing with cap and amp while trying to be smart.
So - what about the small CARAVAN addition? This is the version where you stay. The stomp switches the lead function in and out. The other version, not shown, is the one where you only stay as long as you have to, and then leave. That's the TENT version, where the stomp switches between active and bypass.
The pictures were taken before the PCBs arrived from production, so the pedal shown is an empty mock-up. As the second picture reveals, there's no electronics inside, just a tin to support the k***s.
This pedal is based on my 5-stage Sycumir circuit. I've juggled a bit with the stages to make the whole more like a vacuum tube amplifier design. Input stage, tone stage, overdrive stage, output stage. The fifth stage has been dedicated to simulate the loudspeaker cabinet, including the output transformer always found in tube amps.
Nothing is however done as a tube amp designer would do it. The input GAIN isn't a volume control, it's a real gain control. There's no traditional tone stack, the controls are separated and only the MID control is a traditional filter. The CRUNCH control is a gain control in the stage following the tone stage. The cap sim is placed before the master section, not after, as you might expect. The master section isn't a push-pull topology, or a simulation of the same, it's "just" another gain stage with BASS and TREBLE controls.
I didn't implement an effects loop, nor did I add a balanced DI output, as there's no room for the connectors needed. I'll save those features for a possible deluxe version in a much bigger enclosure.
More on this when I have assembled the PCBs and tinkered with the voicing and gain distribution.

Happy New Year!
Here's a Meenmutty pedal photographed during a steampunk cosplay to start off 2024.
It's really the same pedal electronics as the one I posted about in mid November, just dressed differently. And almost the same as the gold top pedal from my latest post. I built it as a birthday gift, and as we're now past the birthday date, I can show you the thing.
The main feature, the triangular plate carrying the light bulb is there to "hide" the buttons with the small pins, which are piercing pins, for Level, Gain and Tone. And to carry the bulb, which has LED filaments inside for function indicators. The buttons are operated by pushing the pins. As FB won't let me mix pictures and videos, I'll post a clarifying video in the comments later.
The plate has the text etched into the brass, using common PCB fabrication methods. The etched letters have then been filled with acrylic paint.
All the buttons and eyelets are from the craft shop, intended for other purposes.

A GOLD top! Bling galore! Or just an experiment in working with printable metallic foil, mixed with some parts intended for my steam punk styled enclosures. I originally intended to use a foil imitating old copper, but it proved impossible to obtain a nice result when mating the metallic foil with the transparent label containing the lettering print. The gold foil is printable and hold the lettering, with a transparent vinyl sticker added for protecting the lettering. The result is not perfect, but far better than the copper style foil.
The buttons are from the craft shop, which sell them as tassel caps. I've added a white dot and pushed them onto the pot axels with a bit of thin silicone tubing in between, to make a tight fit. The rings, or eyelets, around the k***s are "floating" in oversized holes. This makes it look like the k***s fit perfectly in the holes, which they do not. It just looks like it.
The inside is a scaled down version of the Meenmutty I posted about mid November. I have omitted the boost parts, or rather, I have omitted the controls for the boost part. The boost is always on, with a fixed setting that I have selected. I've included a photo of the inside, revealing that the pedal is free of wooly wire and germanium transistors.
I'm not sure this enclosure style will work. It can be hard to read the labels and k**b settings when light is reflected in the glossy metallic surfaces, as is clearly demonstrated in the photos, where it's next to impossible to read the GAIN label at the top. The reflected light might also be distracting or annoying in a live setting. I may have to experiment further, adding a subtle printed pattern to reduce the glare.

The Sycumir has had a facelift. After finishing the CMA5 board I told about recently and playing around with the tonal possibilities, I re-did the first Sycumir design and ended up with what you see in the photos. I added a 5th pot in the middle of the chain, and also changed some of the transistors to packages that are slightly easier to solder.
Gain, Level, Treble and Bass k***s are mainstays, the Mid k**b can be something else if I change the way I voice the gain stages. Presence, Grit, Crunch, Brilliance, things like that. The setting shown is for flattest possible frequency response. Nasty indeed, if you ask me.
The gain stage chain is set up for a lot of gain. Minimum gain is mild overdrive, maxing it will give you cut up speaker cones and output pentodes emitting bluish light. And that's before the tone controls are moved from flat response. '70'es Sabbath tone anyone?
And yes, the QC stickers are plain silly. But they hide two erroneous holes that I'm not too eager to flaunt.

After spending an afternoon with Jesper, installing a noiseless humbucker pick-up in his recently acquired resonator guitar, I felt like doing something else in the evening. I fell into the "design a pedal"-hole - again! The following morning I spent doing the board layout you see in the rendering below.
This is yet another cascaded current mirror amplifier thingy, a cross between the Vairai and the Sycumir. The Vairai stages has been expanded to include voicing parts found in the Sycumir, and the number of stages has gone up from three to five, Sycumir style. A new feature is using a boosted internal supply voltage, as that will create a larger wiggling room for the design constraints.
My hope is to obtain something that will behave like an old fashioned amplifier, with the controls typically found on those models, and also have a vintage-ish sonic character. As usual with a twist hidden behind the front panel, which is the complete lack of vintage circuit solutions. No germanium transistors, no wooly wire.
I'll ponder how the controls should be implemented before I design a board to hold the controls, and after that I'll have a set of prototype boards fabricated. Maybe I'll have something ready to assemble during the Xmas holidays.

A cousin to the big Sycumir, the smaller Sycunoc. Instead of five gain stages, this pedal has only one. It is however followed by a buffer stage, to have enhanced drive capability. It's a study of the sonic capabilities of a single stage symmetrical current mirror design.
This pedal is not very loud. Instead it excells in a very variable voicing and overdrive character. The k**b labeled GRIT is really a low cut filter in one of the current mirrors, and when it interacts with both the treble response and the overall gain, because of the symmetrical design, it has more impact on the amount of harmonics produced when the pedal is gained up for overdrive. It will reduce the lowest range when fully counterclockwise, but, by intent, not by very much.
The k**b labeled TONE affects the response all the way from the lower mids to the brightest treble. The response is never flat, and includes a shallow mid scoop voicing.
This is not a screaming loud overdrive, it's a subtle effect perhaps best described as a gritty or hairy booster/buffer. It can get very bright, but never shrill.

Another pedal in my cascade series. This is a variation of stacking two overdrives, but done inside a single pedal.
The larger k**b for Gain is for the first ovedrive. The two smaller k***s for Gain and Level, placed under the BOOST label, is for the second overdrive in the internal chain. The larger Level and Tone k***s are placed last in the chain. The Gain k***s are intended for altering the harmonic content of the signal, not the level, and you'll find that sometimes more will result in less, and vice versa.
The boost level will have less impact as the main level is increased, as a sort of disaster prevention. Leaving the boost level at minimum setting will change the behavour of the boost function to a pure sonic character changing function - there's not always a need to get louder.
Both Bypass and Boost switches has a momentary solo function besides the normal latched on/off.
Bonus info: Meenmutty is a three tier cascade (waterfall) in India.

Here's a demo for a pedal with five peculiar gain stages that I've made up. Despite all these gain stages I have kept it rather low gained overall, the idea being that the first stage provide most of the gain while the following stages provide the voicing. This should result in low noise and well behaved overdrive.
The pedal have a frequency response shaped like the response you find in an old vacuum tube guitar amplifier. Lots of lows and highs, scooped mids. But no tonestack in sight. All the shaping is done without filters in the signal path, each gain stage has its response shaped with a lowpass or a highpass characteristic to end up with a roller coaster shape from input to output.
The only real filter is the one behind the TREBLE1 k**b, which is a simple treble cut at the very end of the chain. I'll probably remove it in the next iteration. The BASS2 k**b that was initially planned has already been removed, as can be seen in the photo.
The result is, to my ears at least, a subtle tranparent overdrive with a tube amp voicing.
The name is a shortened version of the technical circuit description, which is something like Symmetrical Current Mirror Amplifier. I wonder how it should be pronounced ...

Here's a batch of pedals built around the simple but nice gain stage I posted about in June '23. The Tuber pedal I built around the gain stage has since been demo'ed and has been well received by both my alpha testers.
One of the best features of this circuit, apart from the nice distortion, is that it will work well without having to measure and sort the JFET or include a bias trimmer. Just grab a JFET from the drawer/bag/tape and mount it on the circuit board, and the board will work as intended and expected. That's not common for FET based circuits.
This ease of production has inspired me to make a multistage circuit, three gain stages in a row or in cascade. It's really well behaved, and produces a nice overdrive tone without the characteristic buzz associated with Tube Screamer circuits. This sound more like an amplifier in distress.
The pedal shown on the left is a simple three stage cascade with a variable gain first stage (fixed gain in second and third stages), a treble filter and a level control on the output. About as simple as it gets.
The middle pedal is a bit more complex. It's a variation of stacking two overdrives, but done inside a single pedal. The way I have chosen to implement it is to keep the controls from the right side pedal (Gain, Tone, Level) and, with the aid of some switches, make the gain of the third gain stage variable. Additionally the output level can be increased. The switches are controlled by the "Boost" stomp switch, and the k***s below the BOOST label control the additional gain and output level.
The rightmost pedal is almost the same as the middle pedal with the boost constantly engaged. Hence the two gain control k***s, late gain is the same as boost gain. I haven't been able to think of a simple way to turn the stacking on and off using the same stomp switch that controls bypass, but the means to do it are included on the PCB. For now the microcontroller will just turn the boost ON at power-on.
Do you wonder what the pedal names mean? Let me explain: The "code name" for the three stage circuit was simply Cascade. The boost enhanced variation was nicknamed SuperCascade. That's too simple and straightforward for a pedal name, so I scoured Wiki and found a list of waterfalls. Those are sometimes called cascades too. This way I ended up picking names from three tiered cascades for the pedals. So there's no reason for mean mothers to feel offended.

I've fashioned a new input amplifier design that I haven't seen elsewhere. I'll hesitate to call it an invention, it's a novelty rather, but is solves a subtle problem in an elegant way.
This guitar input amplifier is based on a JFET input device acting as a voltage to current converter feeding an op amp in an inverting amplifier configuration.
This will ensure a wider dynamic range for the input signal when using a standard TL07x OpAmp. The amplifier is non-inverting.
If you want to learn more, I've made an article (filename Article_JFET_V2I_Amplifier.pdf) describing the circuit in more detail available on my Hightail Space: https://spaces.hightail.com/space/fLB5hOlojn

Old pedal designs in new versions.
I have revised the big version of my trebmolo pedal. The main difference between small and big versions is that the big one use a voltage controlled amplifier (VCA) instead of an optical resistor to modulate the guitar signal.
This is a modern technology that I first implemented in the Troubadour multi pedal, which allow me to go really crazy with modulation waveforms without being limited by the sluggish response of the optical resistor. The tap tempo is now put on a stomp switch, and the percieved tremolo depth is the same in both treble and full range mode. I implemented the tremolo waveform selection with two toggle switches, but as you can see in the photo's right hand side I made provisions for a rotary selector. The right hand assembly is waiting for some fancy parts to arrive before it will be housed in a steam punk style enclosure.
I've also completed my Sustainer in the more versatile version with variable bypass gain. Variable gain in bypass has been implemented by user request. The second Sustainer photo is to show how the valve cap k***s looks.
Finally I have made new versions og my booster pedal - again! I'm now on version 10 and 11. Both of these are based on the Half Current Mirror Amplifier I described in June, but with some additions and recalculated component values to keep the circuit squeaky clean. Version 10 is a 60V internal supply voltage version with a +27dBu headroom, version 11 is running on 9V to keep it as noiseless as possible.

I've come up with an unusual circuit configuration while working on a completely unrelated project. It is a gritty distortion circuit ... again.
It's as ugly as a fuzz circuit, but with a few of the worst shortfalls of said circuit changed for the much better.
The input impedance is high (as it should be for electric guitar) and it doesn't change behaviour or sonic character with temperature. Both contrary to a fuzz.
I call this contraption "A Half Current Mirror Amplifier" and it measures a bit like an overdriven tube amp ... at certain settings. It can also sound almost as nasty as a fuzz.
As is revealed on the photos, I put it in a pedal enclosure to be able to demo it.

Here's an example of how to use a secondary, or hidden, front panel for mounting multiple small circuit boards in a pedal enclosure.
Commonly used DIY rules for guitar pedals of the "many small boards in a large box"-type would suggest using either crammed up brown paper, a mess of too long interconnect wires, melt glue or screws drilled through the front of the enclosure as the preferred mounting method. I have chosen to disregard these rules.
I have instead chosen to use brass standoffs, screwed to a thin aluminium sheet with countersunk screws.All I have to do is drill holes for the standoffs, and also provide cutouts in the aluminium sheet for the potentiometers, switches and LEDs.
The aluminium sheet is held in place just behind the enclosure front by the four foot switches, which is mounted through the enclosure front anyway. Two brass standoffs are placed near the edge opposite the switches, to provide support by resting on the enclosure bottom lid when the enclosure is fully assembled.
As I lack the proper tools for sheet metal work the hidden sheet front panel featured here is anything but a beautyful product, but it does its undercover job nicely anyway.

UV printed enclosures!
A local print shop has recently started offering "UV print on anything".
So I decided to try to let them print on my pedal enclosures. And the result is great. Until now the only affordable UV print service I had access to was Tayda in Thailand. Tayda is cheap and does a nice job, but they also demand print files generated by Adobe Illustrator, which is a rather expensive software package. And then there's the international shipping.
My local shop accept standard pdf-files from any graphics SW that can generate a layered pdf file, and does the job almost as cheap, just as good and a lot faster than Tayda. And I literally just have to cross the street to bring and fetch my enclosures. So I'm pretty chuffed that I can get my enclosures printed locally from day to day for a reasonable price.
The shop is called Frederiksberg Tryk and is located on H. C. Ørsteds Vej. Webpage: frbtryk.dk.

I've made a rather silly little pedal: A Jewelry Box Booster. It's just a booster housed in a tiny jewelry box. I couldn't find a way to have a foot switch that would look nice, so I use the lid as the foot switch. That means that the lid is ajar at all times. Maybe it will tickle your curiosity to peek inside?
There's a video too, check it out!

K***s for guitar pedals, and a lot of them.
What's so special about them? Well, first it's not k***s for guitar pedals originally, they're sold as tire valve caps. So I have drilled the little indicator dimple in each and every one of them. I've also put a dab of acrylic paint in each and every dimple. Just to make them look like k***s for guitar pedals.
But why? Because I find they
- look good,
- are quite cheap compared to "proper" k***s
- are different from what is commonly available,
- have a size that I like,
- fit well with my pedal designs and
- are made of anodized aluminium.
That's why.
You might argue that they do not fit on the potentiometer axle like a proper k**b do. No, that's true, but a small piece of cheap silicone tubing with the right size will make them fit fine on all common types of potentiometer axle.
The photos show the k***s before (left) and after (right) paint is applied.

I received circuit boards today. I couldn't resist assembling an AristocRat, and it worked as it should right away.
Although it is based on the famous RAT from Proco, there's a number of differences from the original rodent. First of all I have put my signature input stage in front of the RAT part. Then I have found, as mentioned last week, an alternative amplifier chip, just as old and cranky as the original. I have changed the clipping diodes from standard 1N914 to some fancy SMT devices I have found recently. To add some new versatility, I made the clipping devices switchable, giving the user a less distorted, more overdrive style mode. I have also changed the tone control a bit, and finally I have made an active volume/level control at the output. The original has a high value potmeter right at the output, meaning that it is not able to drive a long cable without loss of the upper frequency range unless it is at maximum setting (not very practical IMO).
The implemented gain structure allows a range from almost clean unity gain buffering to very loud distortion or overdrive. The tone control can go from muffled dull to bright as a supernova.
I haven't had time to do noise measurments yet, but first impression is a very civilised noise level.

I'm preparing a couple of enclosures while I wait for new circuit boards to be produced.
The orange AristocRat is a variation over the classic Proco Rat, for which I have found an alternative OpAmp chip that I want to evaluate. As the alternative is both as old, obsolete and (generally) hard to find as the original LM308 there's a fair chance it'll be worthy. The fact that the alternative is NOT pin compatible with the original gives it a strong mojo value.
The yellow Sustainer is a new extended version of my usual Sustainer. Per user request I have added the possibility to have buffered bypass with gain different from the active gain setting. More in a week or so, when the new boards have arrived from abroad.

Here's a themed version of my new NAKAMA ultra high voltage booster. The enclosure is supposed to remind you of a box meant for military merchandise, like shovels, tarps, field rations or other weaponry. The hand gr***de shaped k**b was the real reason for choosing this theme.
Notice that the transparent label I use for putting text and graphics on the enclosure has yellow colour on a dark background. That's usually not possible to achieve with a standard laser printer, but I have acquired a white toner cartridge, which enables me to print colours mixed with white instead of black. As the white cartridge replaces the black cartridge in my printer, the yellow color is an olive nuance on my monitor, as my graphics software is unaware that the black cartridge has been replaced.