Sunday, 29 December 2013

Ibanez Standard Fuzz

No way we're going to beat last year's number of layout posts :) We might break the verified 600 before new years though :)

Since i was up for taking on the big fuzzes.. Here's another vintage fuzz with up octave overtones. The layout turned pretty neat with some symmetry in there, but it'll once again eat up your 10µ cap supply :) Original has no polarity protection nor filter cap, so you should think about adding those yourself. The first gain stage with SK30A FET is per original, but instead the BCE pinout of the 2SC1815 used in the original, i drew the other transistors to match the pinout of your standard 2N2222 (and similar). Anything with relatively high hFE will work fine. BC550, 2N3904 MPSA18 and so on. Reportedly, lowish hFE gain transistor would work the best here. Try any Si NPN, you might get the best result with hFEs around 200. Just mind the pinout. I also left the clipping diodes on the edge of the board so one can take them easily on a switch and have his (or hers) way with them. Maybe a three position switch with LEDs, Germaniums and a diode lift in the middle? Tone change switch can be either a toggle or a stomp, which way you prefer it. If you want, you can omit the small link and the cuts from the bottom right off the board and wire the switch lug 2 directly to balance pot's lug 3.



13 comments:

  1. Oh dear lord! Awesome stuff. Gotta bookmark this.

    ReplyDelete
  2. This my next build before the Big Cheese fuzz.

    The schematic is almost identical to the Univox Superfuzz. Have we had the superfuzz here on a verified layout?

    Think I might build with with Univox superfuzz BOM.
    Has anyone ever compared the two?

    ReplyDelete
  3. Works. Although i did find out simething that i can't quite understand yet. When the tone change filter is engaged, there is a significant volume drop. The reason i don't quite get it is that the filter is exactly the same as in Univox Super-Fuzz. With the filter bypassed it sounds like it should. Upping the value for 1n cures it, but takes away a lot of low frequencies. Tried on a few values, but everything makes me want to ditch the board completely. Yes it works, but... I'm having thoughts about the 1n cap being between rows that are next to each other..

    Also. Once again, i'm not too pleased with the germanium clipping texture either :) So Doron. Yes. Build the Univox. It is better.
    +m

    ReplyDelete
    Replies
    1. Hey, i found a few thread about this subject, with of course, different approach. i haven't tried any of the build nor fixes but i own a "bruno fuzz machine" and the original has the volume drop too.
      http://www.harmonycentral.com/t5/Effects-Pedals-Multi-Effects-and/ibanez-standard-fuzz-question/td-p/2930564
      http://www.freestompboxes.org/viewtopic.php?f=13&t=15379&start=20

      Delete
    2. That's pretty strange as the Univox doesn't have that :) But it seems that the layout works 1:1 with the original.
      +m

      Delete
  4. I make my own version of this pedal and I found with using high quality electrolytics and good GE diodes (I use old stock OA90's) I have unity volume at about 1 o'clock.

    ReplyDelete
  5. I wasn't too nuts about this as a guitar pedal, but it makes a mean bass pedal.

    I also did the diodes on a switch, bat41 on one side, germanium on the other.

    ReplyDelete
  6. High hFe transistors are not necessarily the best for this circuit. The higher the gain, the more compressed it will sound. Also, it will tend to gate single note runs more as well. The original 2SC828R's, 2SC828Q's, or 2SC1815's were not that high gain. While 828R reissues are available, they are higher in gain than the original Matshushita model. From experience, exceeding 200 hFe will start to really compress the sound, and you start to lose any note clarity that may be left. In Superfuzz circuits, too high a gain also lends to that initial blast of hiss when hitting a big chord. Go high enough with the gain, and it becomes useless. Keep the gains between 100 and 175, and they sound pretty decent. The 2N2222 is very good for this circuit.

    ReplyDelete
    Replies
    1. Insert image of a thumbs up here.

      However, the 1815s are rated at minimum of 100 and maximum off 700. This being the reason for me assuming higer gain.
      +m

      Delete
    2. That's a fair assumption, but as the range states, 100 to 700. 700 doesn't necessarily mean that it will work well. Many assume that high gain is required for BMP, yet this proves otherwise;
      http://www.freestompboxes.org/viewtopic.php?f=19&t=27146&p=256467&hilit=muff+gain#p256467
      Insofar as Japanese transistor gains are concerned, I posted this a while back;
      http://www.diystompboxes.com/smfforum/index.php?topic=107901.msg981825#msg981825
      Given that most Standard Fuzz, and Superfuzz circuits I've seen use 2SC828Q or 2SC828R, you can see that the gains tend to stay on the low side of 360. All in all, Japanese transistor gain letters tended to preset much tighter tolerances. When I did the Honey Special Fuzz project, the 2SC828R's in the original all measured between 190 and 220. Seeing as how the Honey was the forefather of the Superfuzz, I think they figured out that below 200 was really the magic territory for transistors. That's why 150 to 200 sounds more open than 250 or higher.

      Delete
    3. Addendum; I just checked my datasheet for the Toshiba 2SC1815, and they came in the O, Y, BL, and GR gain ranges. I've haven't come across any BL or GR range transistors in Superfuzz variant circuits, but O and Y will cover 70 to 240, which validates the typical gain of 100 stated.

      Delete
    4. You know what they say about assuming :)

      But you are correct. I'll add a note to the post.
      +m

      Delete
    5. No problem. It's a great circuit. In passing, the Japanese JFET's also had grading where the cut off current is concerned. The 2SK30's that I've seen in these circuits all seem to be in the Y range, which is in the 1.2 to 3 mA zone. Now... did the factory test for a particular magic number? Dunno, unless some gets their hands on good sounding one, and puts it through RG's test rig... (like I can see THAT happening) ;-)

      Delete