ROSIE amp - though she be but little, she is fierce!

[jonny.reckless]

Here's another little JFET based mini guitar amplifier, this one's called the Rosie. Small, simple, pretty easy to build, low power but lots of tonal range from clean to crunch to high gain to brown sounds.
Here's the schematic:
https://drive.google.com/open?id=1dMZAvFHAK4c_JKfGeQ7KokqeplQAfuoN
Here's the PCB:
https://drive.google.com/open?id=1rTu_kDSxfQ9gOfh1zHZWlfWlKnwhzJsp
PCB Gerbers, NC drill files, bill of materials:
https://drive.google.com/drive/folders/1almm_wEa6yQQi_G-eJBSkc7cNzFrG-rD?usp=sharing
Here's a demo video of it in action:

It's tonally quite similar to the Boba FET
https://www.diystompboxes.com/smfforum/index.php?topic=118627.msg1104856#msg1104856
but with more gain, more tone variety, and less hiss.

[Kipper4]

Nice
Good job Jonny 

[Ben N]

Sounds great, jonny, really articulate.

[bluebunny]

It's a whole lotta...

   

[thetragichero]

what kind of current draw we looking at? need to make a small battery powered amp to fit into a cigar box with speaker so I'm gathering options. lm386 so far is winning because everything else wants that mondo current

[jonny.reckless]

what kind of current draw we looking at? need to make a small battery powered amp to fit into a cigar box with speaker so I'm gathering options. lm386 so far is winning because everything else wants that mondo current

50mA quiescent current in the power amplifier. Total current draw about 60mA at idle. You need 18V at 1A so not ideal for batteries.

[Ripthorn]

I really dig it! I may have to try one of these, but I would probably do two separate preamps, each with the boost and switching, so that I have loads of options!

[Ice-9]

The Rosie sounds great and thanks for sharing Jonny.   

[deadastronaut]

yeah sounds cool man...very versatile. 

[Steben]

Great.
Makes me think adding the (cathode / sag) compression of a valve power amp would make it perfect allround SS amp.

[amptramp]

That sounded good!  Amazing how a simple circuit that is entirely solid state can give you the impression of a tube amp.

[PRR]

> Amazing how a simple circuit that is entirely solid state can give you the impression of a tube amp.

I have *never* seen a single no-NFB JFET as voltage-amp in a power amplifier. (Corrections welcome.) This is not like any other SS amp out there.

Not worthy to judge "tone" here.

As an electronic nit-picker (I pick electronic nits, not.... you know):

You can't possibly need all 10k in PR1 if there's just 4.4k above center-point; even 2k here suggests you might want another JFET.

This power amp gives most of 4W in 8 ohms and 8W in 4 ohms so full-roar dissipation is not-small. It could about play a set with 8 ohms and naked TIP3_ parts, a gig at 4r is less sure. Looks like you did box-mounted TIPs but more details could help copyists.

When PR2 Bias wiper goes open bias current goes very-very high. Yes, we see this all over. Often at old-age autopsy.

I strongly endorse Cordell's book. Even though it is expensive. Even though you will ignore 95% of his "perfect Hi-Fi amp" analysis. Even though he likes 30 parts where you have 6. The book is full of practical amplifier tips.

TR11 should be ON the heatsink for tracking... ah, you are very correct, TR11 should cuddle-up with TR3 TR4.

There is zero bias-current degeneration among TR1-TR4. The bias current will come up very quickly and be hard to tame. And even with TR11 thermally-close to TR3 TR4, I would expect poor tracking. (My first amp wanted to run-away, it didn't if I trimmed it hot, but then it was "hoarse" when cold. My fix was to play a side of Bob Seeger Smokin OPs Real Loud. By the end of LoveLight it was sounding better, after Jessie James it would play all day. I may have been unlucky (all ex-military surplus parts).

_I_ would want part-ohm resistors to degenerate output stage current.

This is not an easy change to make on your very nice layout.

The Sziklai connection is notorious for 7MHz oscillation. While this may indirectly give "soft" sound, it usually heats-up and blows-out. Your conditions of 0.6mA/49.4mA may dodge the bullet, or not, depending on the wide range of things sold under the TR1-TR4 part-numbers.

[Ben N]

what kind of current draw we looking at? need to make a small battery powered amp to fit into a cigar box with speaker so I'm gathering options. lm386 so far is winning because everything else wants that mondo current

50mA quiescent current in the power amplifier. Total current draw about 60mA at idle. You need 18V at 1A so not ideal for batteries.

You'd need a pretty big cigar box, but 4-5 18650s (with associated BMS and charging circuitry) might get you there for a limited time (min 3 hours playing time for 3000maH cells). OTOH, there are plenty of other amp options that might be better suited to battery power and a small enclosure.

[Ripthorn]

what kind of current draw we looking at? need to make a small battery powered amp to fit into a cigar box with speaker so I'm gathering options. lm386 so far is winning because everything else wants that mondo current

I've designed a few very small amplifiers for things like this (onboard preamp/cab simulator/power amp, and headphone amp for playing quietly). The LM386 design is really hard to beat. Put a switchable overdrive preamp in front, a cab sim between the preamp and power amp, and the power amp last.

[alfafalfa]

She sounds really good Jonny !

Very nice open distortion and a good clean sound.

Hats off !

Alf

[jonny.reckless]

I've added some notes, bill of materials, and PCB Gerbers and NC drill files to the Google drive.
https://drive.google.com/drive/folders/1almm_wEa6yQQi_G-eJBSkc7cNzFrG-rD?usp=sharing

I have a few spare PCBs left, email me if you want one, $5 each plus postage.

The thermal stability seems OK. The idle current does drift quite a bit if you blow a hot air gun or freezer spray onto TR3, TR4 and TR11 but it returns to something sensible pretty quickly, and doesn't seem to drift much above 80mA under normal playing conditions. I put a dab of silicone heatsink grease onto TR3, TR4 and TR11 to couple them thermally, and sat them adjacent to one another on the PCB. I found that using emitter degeneration resistors, even small ones like 0.22ohms, affected the sound quite a bit since the output stage is running completely open loop. I was trying to make something as simple as possible with very few components that sounded pretty good. The output transistors TR1 and TR2 are mounted on small screw on aluminum finned heatsinks. They barely get warm to the touch even when you crank the amp pretty loud. The use of the JFET in the power amp, plus the fact it runs open loop, means the power amp clips quite nicely when you push it hard. I wouldn't claim it's like a tube amp, but it sounds nicer under clipping than most single chip audio amps, and is simple and cheap to build.

PR1 could definitely be smaller. On my prototype I needed 1240 ohms to bias it. This will vary quite a bit from JFET to JFET. I only have 1k and 10k trimpots in my component drawer, so 10k it was. If you have a 2k or 5k trimpot I am fairly sure that would work for PR1.

[Mark Hammer]

Maybe I missed it, but what sort of speaker/s and cab are we listening to there?

[amptramp]

It seems this would be a perfect candidate for operation from a laptop power supply.  This would be good for Buskerfest, a celebration we have every summer in Mississauga with musicians everywhere on Lakeshore Boulevard where street musicians gather from all over.

[Mark Hammer]

I hadn't realized you were a fellow Ontarian!  Well HOWDY!.

I made a little LM380-powered portable amp running off 8 C-cells, and putting out roughly 2W into a 6" speaker in a sealed cab.  It is surprising just how loud a few watts can be with the right speaker and cab.  More than adequate for busking, unless one happens to be busking right beside the QEW.

[jonny.reckless]

Maybe I missed it, but what sort of speaker/s and cab are we listening to there?

Closed back 1 x 12" with a Celestion Vintage 30. Two SE2200A large diaphragm condenser mics: one on the cone, one in the room, bit of added reverb in the mix. Fairly typical how I would mic up a guitar amp for recording, except if I was recording I would probably use a bit of EQ on the close mic, and maybe move it off axis a bit more. This is a pretty representative sound of how it was in the room.

I previously did a more basic video just with my cell phone, using the built in mics, in which you can just hear the room sound. The mics in my phone are pretty crappy, and they distort at high volume, but you can get the general idea.