Tuesday, 19 April 2016

Gem Toaster amp

Don't know if you enjoy occasional story or not, but here one goes anyway. A friend at work was visiting his parents during easter holidays. His father is quite handy building stuff and while he had joined a local combo of fiddlers. He built himself a mandolin which was truly beautiful workmanship. Being handy as he is, he built himself a preamp for the guitar pickup housed inside the instrument. Due to his limited understanding of electronics, he tried his hardest to make the preamp drive a speaker. As we all know, that won't work. Or it will, but not in a manner that's even remotely usable. His biggest issue was that this setup ate up a lot of batteries and quick. The son told me about this case and wondered what could be done to remedy the situation. He figured that i might have an answer of some sort, being the guitar enthusiast i am. I thought about it for a while and my first instinct was to recommend a small tube combo. But the goal was to have the mandolin heard, not necesserily loud, just a bit louder. And this was the best to achieve with a single 9V battery - and that battery should last as long as possible. Although being a true craftman the money became the next issue. So a 9V battery and a decent level of volume on a decent frequency range on a budget. Lucky for me, the pickup used was a standard single coil, so no ridiculous input impedance numbers as with some high grade piezos were needed. I had drawn myself a "FYA Bone Toaster" circuit some months ago. This was aiming to be a JFET based preamp that could perform as a clean boost with minimal distortion and thus, perfect for jazzy tones. With controls of level and tone (not completely unlike big muff tone stack made active), i figured this could very well work as a preamp. And followed by yet another Ruby Amp with one cap value upped to keep the low frequencies in tact.

Took both schems and drew a layout up. The next thing was to draw the board so it could be mounted in a hand build combo cabinet with as little hazzle as possible. So i figured the pots should be board mounted. This sets a number of limitations to the board, but i think i managed to squeeze it in rather nicely. I've had a love/hate relationship with LM/JRC386 distortion and gain settings, so to keep this thing's gain at usable levels for a village fiddler, i used a 5K trimpot for power amp gain. This way the gain and volume pots could be maxed and then the trimmer used to tweak the best balance between low distortion and high volume. It's not possible to get both at the same time, so the trimmer comes in handy here.

I printed the layout out and gave it to my friend. Helped him with ordering parts and then we wait. I wanted to try the layout myself before i send a novice on a path of becoming electronics building addict with an unverified layout (and no guarantees it'll work). Built it and was rather pleased with the results. This circuit ran into 4ohm 15" element. Whoah.

For the pots, the gain and master volume could be logarithmic. For the rest of it.. Yup. A decent practice amp. For about 6-7 dollars/euros/pounds, it doesn't get much better than this. Now, if we have a handy wood worker to build up a beautiful combo...

(Do note that the power amp chip is "upside down")


17 comments:

  1. Miro - it is very cool that you were able to make something from parts on hand that actually works. How many watts would you estimate it puts out?

    FWIW: I met a guy here in San Diego who is making higher wattage amps in pedal size (~50 watts output) based on somehow using the clock in the 555 IC chip to boost the wattage. website: http://www.sibfx.com/charger/charger.html)

    ReplyDelete
    Replies
    1. It's still a Ruby Amp, so around 0,5W would be accurate.

      I'm guessing the SIB amp is based on a TDA chip, which is also backed by the fact that true output watts are determined by load. 555 timer can be used to up the voltage. We have 555 pump layout up here on the blog that is based on the same idea.

      Anyway. On small amps.. I bought myself one of these: http://www.taurus-amp.pl/guitar/stomp-head/classic/sh-1-bl.html
      I'm using it on a pedalboard specially made for it and 6 pedals. This board is my main amp for D-Beatl:s. Flying V in a backbag and one case in hand. That's all i need for giggin'.
      +m

      Delete
    2. Uncle Miro
      where i can get schematic? just want build this on perf/pcb

      Delete
    3. There isn't one :)

      The power amp is Ruby from ROG http://runoffgroove.com/ruby.html and the preamp section is something i haven't published. Not yet at least...
      +m

      Delete
    4. Right, voltage not wattage. I have seen the layout here that goes to 18. I guess that is the start to getting even higher voltages.

      Delete
  2. The 386 will fry if you give it 18V

    ReplyDelete
    Replies
    1. Second.
      http://www.ti.com/lit/ds/symlink/lm386.pdf

      Except for the LM386-4, that can handle maximum of 18 volts, but all the others have maximum rating of 12V.
      +m

      Delete
  3. I love how you managed to align the pot connections in the "right way". Once you have a schematic drawn I'd create a PCB layout for etching if you don't mind. Btw. the Taurus amps are indeed killer. I got a Stomp Head 4.SL recently and I have to say it beats my all tube hand made amp if I consider the price as well.

    ReplyDelete
    Replies
    1. So... I did the schem myself: http://drdfx.hu/private/FYA_GemToasterAmp.jpg
      Please check and let me know if it is correct. Also if itis correct I have a few questions:
      - Isn't there an input cap missing? Something around 100nF would seem to be appropriate.
      - Wouldn't it be better to replace the drain resistors with trimmers? I know they have a larger footprint, but we all know the accuracy of FETs...

      Delete
    2. Whoah! Cool!!

      The pedal version of my Bone Toaster does have an input cap, but since this is an amp, i dropped it. Many vintage amps connect the input straight the to tube (or as in this case - and many ROG/Wampler designs - straight to first JFET gate. This is why i didn't include it. The cap in Ruby, between the master volume pot and 386 input is vital to not cutting the bass. 1µ and up will be good.

      And yes, those resistors could be trimmers. But i found that my (probably fake) batch of J201s work and sound good in this circuit no matter which ones i take from the bag. In my experience - if those totally-out-of-spec transistors work with my values, then the good ones will work too... :)

      So.. Anyone interested in getting a layout for my jazzy preamp circuit, The Bone Toaster? :)



      P.S. Played my first !"ampless" gig on wednesday. That Taurus is a F**N killer. Sufficient volume levels with great tube-like tone and it even takes pedals very well.
      +m

      Delete
    3. Alright, I have created a layout for this: http://drdfx.hu/private/FYA_GemToasterAmp.pdf
      Should fit in a 1590B with the two jacks and a power jack. A mains switch might fit as well, though I'm not sure about that. In 125B or bigger everything will fit nicely. Layout is not verified yet.

      And yes: I'm for one surely interested in your preamp design ;)

      Delete
    4. Did you verify your etch layout yet?

      Delete
  4. Here's hybrid of Ruby and Noisy Cricket - http://mylkstuff.com/page35.htm#.VxyVC1R94dU
    I built it from these schematics and I'm very satisfied (for guitar). There's single op-amp as preamp, and 386 as power amp (gain, volume, brightness and other controls are optional).

    ReplyDelete
  5. This comment has been removed by the author.

    ReplyDelete
  6. Hi guys, wonder if someone might be able to advise - I've had a go at this circuit and just can't get it going. Pretty much nothing coming through unless I max all the pots, and then it's just a very quiet, muffled signal. (Adjusting the trim pot down cuts the signal to nothing, increasing it gives a weird low buzzing sound).
    I've checked the chip is working and tried 3 different sets of JFETs, but always get the same results. Have cut the tracks and am fairly sure there are no bridges.
    The chip readings are as follows:
    1 - 1.3
    2 - 0.0
    3 - 0.0
    4 - 0.0
    5 - 4.2
    6 - 8.7
    7 - 4.4
    8 - 1.3

    Anyone got any ideas what I can try?

    Anyway, thanks to all you guys who contribute to the site - I've been building a few things for a while, but am still very much a beginner. The work you all put in is very much appreciated :)

    ReplyDelete
    Replies
    1. Hmm. Is your chip "upside down" as it should be?

      You could try probing it to see where the signal dies.
      +m

      Delete
    2. Thanks Miro - yup, the chip is upside down. I'll keep on probing!

      Delete