Friday 7 July 2017

Spaceman Sputnik II

"Starting with the same germanium transistors as the much-loved original Sputnik, but with slightly different specs, we added a few slight tweaks to get very close to the sonic details of the original circuit. Many of the other parts in the Sputnik II are different but of same or similar value, including many vintage and rare parts.
While extremely similar to the original, the Sputnik II is a beast all its own. The Sputnik II takes everything we loved about the original, and adds in the things we dreamed of after its original release it:
• Full-size control knobs for every parameter
• Foot-switchable DRIFT mode, with its own status indicator
• Additional FILTER options for maximum tone-shaping
• Additional circuitry making it more friendly to active pickups, buffers, synths, pedals, etc
Within Sputnik II lies a plethora of fuzz tones. Smooth, gritty, vintage, modern, thick, thin, buzzy, bassy, splatty, sputtery, noisy, buttery, controlled or very much out-of-control. A truly endless cornucopia of fuzz."

Here is another pedal reverse-engineered by Dino (Digi2t).
He also added an extra diode reverse switch.
For all infos on how to build this pedal you can find Dino's original thread (with his own vero) and schematic on the DIY forum here.




34 comments:

  1. Wow, another beautiful looking loayout Alex. Well done!

    ReplyDelete
  2. Great work Alex. Your layouts really are masterpieces. Gonna have to buy some bits.

    ReplyDelete
  3. Nice! Can someone enlighten me on how this is pos 9v if the ge transistors are PNP? Thnx...

    ReplyDelete
    Replies
    1. PNP doesn't mean it needs positive ground. Take a look at fuzz factory or silicon fuzzmaster. It's always the question of how the transistors are wired.
      +m

      Delete
  4. man original here is loaded with some mojo!
    http://www.spacemaneffects.com/model/SputnikII/Sputnik2Guts.jpg

    ReplyDelete
  5. I should have a video of the clone out soon. I highly recommend the diode reversing switch, don't know how that one slipped by Spaceman's radar. 22K to SCAN 2 should be replaced with a 50K trimmer to tweak the diode voltage. I'll try to come up with some kind of adjustment guide for that. Depending on the diode used, "I think" the span is somewhere in the 15K to 30K range.

    ReplyDelete
    Replies
    1. Oh yeah... I used a Russian D9E diode. In a recent post by diy-tubes over at diysb, he demoed the curves of several Russian Ge and Si diodes. The D9E Vf is almost identical to the 1N34A, but with a much softer knee. The softer knee comes in handy when using DRIFT mode on the edge of the envelope. You can fire the oscillation with a sharp input strike, and it will give you a longer decaying oscillation, which is really cool. I tried a bunch of different diodes, 1N60, 1N270, 1N118, 1N58, 1N34A, some unknowns from pulls, even BAT41/8, and the D9E was the best one. A D9J MIGHT be better, it has an even softer knee, but the trade off is a slightly higher Vf. Worth trying though. As for the transistor voltages, +/-0.5 volts here or there is tolerable, you'll be in the wheelhouse sound-wise. Any low leakage transistors in the 70 to 85 range will get you there.

      Delete
  6. Call your layout verified Alex! Just built it over the weekend, what a fantastic fuzz machine. Used MP20Bs initially for GE trans but found them a bit noisy (their HFE was 88, 111 & 113) so tried IT308Bs (74, 85, 75) and sounds much nicer. Used D9E for GE diode and the rest as written. Dino, A massive thanks for the amazing work you put in to give us these beauties. Time to have some fuzz!

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete
    2. This comment has been removed by the author.

      Delete
    3. Thanks Dino! I've added the extra 100n cap. Is it correct?
      Thanks also to Gavin for verifying the layout.

      Delete
    4. Belay last!!!!
      After further noodling, the cap wasn't the best idea. Yes, it helped control the oscillation to a degree, but the problem is that it also really darkens everything, so it becomes a game of diminishing returns. By the time you get to a value that starts to kick in, you've really darkened the original tone. Not good. BUT... I did discover that using a Ge diode with a low Vf does improve things. The D9E that I had in there measured in a 0.323, but sorting through my bag, I found one that came in at 0.272. I swapped it in, and the oscillation is better than before, meaning that it's a bit easier to tune in jumping octave notes, tuned motorboating (is that a thing?), and the like, while still retaining the top end of the original tone. At the end of the day, different transistors = different tones and settings than the original. That's just the nature of the beast, and I can live with that. So, scratch the diode cap... unless you want to experiment, but I'm going without. I'm sticking a fork in this sucker, it's DONE. Muchos gracias to Gavin and Alex. Takes a village to reverse a pedal... and still retain your sanity!! ;-)

      Delete
    5. Thanks again Dino!
      I've removed it.

      Delete
  7. What ferrite beads are you using in this build? It looks like a good one :)

    ReplyDelete
  8. I don't remember where I bought them from, but they came with preformed leads. The pitch is the same as a 1/4w resistor. Sorry that I don't have more info, but Google is your friend.

    ReplyDelete
    Replies
    1. Oh sure I Googled them already, most of the through hole style ones are rated by impedance, frequency range and ferrite grade. I found a few that I may just order A/B them. Thanks again for your hard work Dino :)

      Delete
  9. More build info. This is from the Laika Fuzz build, but applicable nonetheless;

    Supply voltage = 8.98v. All controls at zero, filter switches in middle position, diode switch down (normal mode), no input.

    For the germs I used (measured with a Peak DCA55 / 22 deg. C.);

    - Q1 = MP42B - gain 56, leakage 0.04mA
    - Q2 = MP16B = gain 84, leakage 0.06mA
    - Q3 = 1T308B = gain 84, leakage 0.00mA

    Voltages - Q1 and Q3 collectors are adjusted via trimmers 1 and 2, original Spaceman unit values in parenthesis;

    - Q1 - E - 8.98 (8.98)
    B - 8.77 (8.86)
    C - 5.66 (5.66) adjusted

    - Q2 - E - 8.32 (8.41)
    B - 8.19 (8.19)
    C - 6.59 (6.45)

    - Q3 - E - 8.68 (8.71)
    B - 8.46 (8.50)
    C - 6.20 (6.20) adjusted

    As you can see, the Laika's voltages are pretty dang close to the original.

    Important note; For D3 (the diode that feed Q2 emitter), I ended up using a 1N965. It was a diode that I pulled from an old Farfisa organ board, it had a groovy old school look to it. I measured it on my DMM at 0.556vF, so I plunked it in there. I had no idea that it's a 15v Zener, but it works just the same. Actually, Q2 voltages seem to be even closer between the Laika and the original, than what I got on the vero unit using a 1N4148 equivalent. Might be the diode, might be the transistor. Happy accident? Sure, why not.

    The germ diode is another Farfisa pluck. Came in at 0.332vF. No idea what the knee looks like on this sucker, but it works AMAZING!!

    The 24nF cap, I had some 20nF caps (which you'll need for the PLL build), and I tried a 4.7nF cap in parallel with it. My cap meter said that it was 24.3nF, so given the tolerances, pretty darned close. You can do the same with a 22nF, and a 2.2nF. The 5nF, I just bought some off Ebay, but again, a 4.7nF is really close, or a 3.9nF with a 1nF parallel, even better.

    ReplyDelete
  10. I'm pretty new at this so please correct me if I'm wrong, I'm trying to learn as I go, but it seems like are some errors in this layout. Unless I'm mistaken here;

    Switch 3 (DPDT) is actually drift and switch 5 (3PDT) is bypass.
    Connections for Ground and LED2- on the 3PDT should be reversed to put the ground to pin 8 and the LED cathode to pin 7. This would keep the Source on the 2SK30 connected to ground and connect the LED cathode to ground only when engaging the effect.

    Does this sound right?

    ReplyDelete
  11. Nope, the drift is 3pdt and bypass is dpdt. You can use a 3pdt for the bypass but you'll have a redundant pole.
    Remember that this layout has been verified so it should work as above.

    ReplyDelete
    Replies
    1. I've wired it all as it shows and it works properly only after reversing the connections for LED and Ground on the 3PDT pins 7/8 (before swapping only one LED would come on at a time and I had no fuzz unless I manually grounded the 2SK30, after the change both LEDs operate independently as they should and the fuzz is working gloriously). The 3PDT is working as bypass with no redundant pole, and the DPDT is egnaging drift. I'm not sure how this is verified as listed or where I went wrong if it's correct... But it seems to be working as it should now with the way I have it wired so I'm going to stick with it until it potentially gives me problems one day.

      Delete
  12. Sorry if this sounds like a stupid question. If i do not have ferrite beads for this circuit can I substitute 100UH inductors instead? I have a few of the 100UH inductors if this will work.

    ReplyDelete
  13. Where do you get the PCB for this project? I'm sorry I'm a beginner at this.

    ReplyDelete
    Replies
    1. Hi Elian.

      To begin with, there is no PCB (printed circuit board) for this particular project. You'd need to get plain stripboard aka vero board and cut it to size and make trace cuts. Here's one guide for using a stripboard - https://tagboardeffects.blogspot.com/2012/04/vero-build-guide.html
      There are other methods for making the strip cuts, but that'll work.

      However! If you are totally new to building hand crafted circuits, this project is NOT the one i'd recommend to start with. There are lots and lots of things that could go wrong as the circuit is large and it has number of components that will be a PITA, even for more experienced builders.

      I think you're after a massive and weird fuzz tones. In this case, head over to tayda electronics shop and get your materials there. And then, build this: https://tagboardeffects.blogspot.com/2012/05/devi-ever-hyperion-2.html
      I'd suggest to have A20K instead of B100K pot for te OSC and reverse the wiring for it.

      With that simple circuit you'll get a grasp of how to build the circuits and you'll end up with a very decent oscillating fuzz monster. Also, make sure to test your build before boxing it up as a finished pedal. If there is anything wrong, it will be way easier to fix without the bypass switch and all the other wiring in the mix.
      +m

      Delete
    2. Thank you for the quick response miro. Does it matter where I buy the components as long as the values are correct?

      Also what are the markings that read CBE, DGS, SGD? I recognize most of the rest.

      Delete
    3. No matter. I'd strongly advice to use reputable sellers you trust. China sellers on ebay may sell refurbished or plain fake stuff. Which may very easily lead to issues with your build. Mouser, Digikey, Tayda, Banzaimusic etc. Those have been good.

      The pinouts of transistors are marked as those. CBE means collector, base and emitter on BJTs (bipolar junction transistors). DSG means drain, source and gate on FETs (field effect transistors). One should always refer to datasheet to verify the pinout.
      +m

      Delete
  14. Just built this with 3 308b's. Hfe's were 74, 84, 84 with no leakage. Right now I have a d311 Russian diode in there, my favorite so far but I have a few more to try. Really cool pedal that pairs really well with the Saturn V in front. Thanks guys!

    ReplyDelete
  15. I just noticed this, I think the 270pf should be 27pf.

    ReplyDelete
  16. Hello, I built this and I really like the pedal, except for one thing. My only problem is that when in Drift mode, when I turn the Scan knob above 9 o'clock I get high frequency squealing, the pitch of which is affected by all the other knobs. Its really weird because the rest of the pedal works fine, and even the drift switch is just like the demos (before 9 o'clock). I used a d9e diode and mp16b for Q1 and GT308s for Q2 and Q3. Anyone can help me? Thank you :)

    ReplyDelete
  17. I've just assembled one of these from the Laika PCB. Mine acts as you describe, I think its par for the course for this pedal. I'm going to try different Germs, I've got three 2n281s from small bear in it now. Sounds awesome, tons of harmonics. It would be cool to have a fine adjust for the Scan knob.

    ReplyDelete
  18. Yeah, I switched the places of the transistors and it helped a bit (a little more range in Scan before the hiss - which reminds me a lot of the DS1 octave fuzz mod). All hFes ~ 65, but the MP16b (which has higher leakage) is now in Q2. If I get the chance to experiment with more Ge transistors I will. Maybe some other value for the scan pot will help as well.

    ReplyDelete
  19. Hiss. I haven't got any hiss. You might want to check your V with the schem. Especially the J201. I used a 2SK30A-Y.
    I've been auditioning diodes this morning. Might get less motorboating with higher forward voltage diodes, but not much. Didn't like sound of Si. Going to try resistor in series with diode, reduce signal going to J201 feedback.

    Don

    ReplyDelete
  20. Maybe hiss doesn't apply here, I must have translated poorly. I mean a squealing that changes pitch in all knobs (except volume), more evident when Range is not low - sounds very similar to oscillation in the Hyperion 2 in extreme settings. Like you said, it looks like a charateristic of this type of circuits, but it is one not evident in most demos I found.
    I also used 2SK30A instead of J201. I am very interested in the results of placing a resistor in series with the Diode (I used D9E), please report :)
    At the end of the day, I think I will box it like this, I really like the Normal mode and Drift is fun in controlled settings.

    ReplyDelete
    Replies
    1. A small resistor in series with the diode yielded no change.

      I have since discovered that the gain of the third germanium transistor makes a huge difference. I switched my 2nd(gain 81) and 3rd(gain 128) trans. and can now get my Scan pot passed noon before squeal. Even at full drive.

      Drift mode is much more usable now.

      Delete
  21. Hi Alex. this is Alex too. I am an absolute noob in terms of solderingbut I decided to build myself my own fuzz pedal. I always loved sputniks and would like to give a try. So I planned to buy a pdb, components, enclosure, knobs etc.. and start from your template. I wanted to ask you if you have a full list of components for this build? Help would be appreciated. thnx a lot, A

    ReplyDelete