Wednesday, February 18, 2015

Analogman Sunface Build

Here is a first build I made not so long ago, an Analogman Sunface.

Very simple layout, a classic fuzz face with some light modifications, like a bias pot that I called "woof" here for fun. 



 
Since this first picture, I improved a bit the circuit with transistors sockets that are very nice in order not to overheat germanium transistors, and to test different combinations of hfe. 

You can see the really simple NOS components here: carbon comp resistors, Mullard "tropical fish" capacitors (love this name, very accurate description!) and of course two AC128 germanium transistors. 

I got most of these compounds in old radios that I disassembled a long time ago, except for the germanium transistors that I bought online. I used the guitar fx layouts scheme to do it and I enlarged a bit because my capacitors were axial. Since, I built another fuzz face using a PCB.


If you want to be super accurate, "vintage correct" or full mojo for your build, then you should get the original components (even if - apart from the transistors - I do not think that it will change the sound a lot): 
  • NOS vintage NKT 275 Germanium Transistors
  • NOS vintage Mullard Yellow Capacitor
  • NOS vintage Philips Capacitors
  • NOS vintage Allen Bradley Resistors
If you want to re-create a proper vintage fuzz face replica, I advise you to buy the pigeon fx kit with a Dallas arbiter board and most of the vintage components listed above. Personally, I preferred to stay with a Sunface design which more practical with the bias knob.

I reverse-etched the enclosure, which was quite of a job, but I love how it finally came out! If I do another one, I will maybe readjust a bit the position of the main logo.



How does it sound?

I made a few samples with my Gibson Les Paul 1954 reissue (gorgeous guitar, the best!), on a Vox Lil' Night Train, a small practicing tube amp that I use at home.
Germanium transistors gives the fuzz a nice smooth saturation, still a bit mushy/harsh, but not as gritty as other silicon fuzz like Big Muff for instance.



How does it work?


Fuzz face is really a simple circuit design, as you can already tell by the number of components of the circuit: 4 resistors, 3 capacitors and 2 transistors! And of course two controls: fuzz + volume! As simple as that! But when we look at the circuit:


Fuzzface circuit
Fig.1 Fuzz face circuit, classic 60s original schematics
(made with Fritzing)

Not so easy to understand, isn't it?
Let divide this schematic in 3 sections:
Fuzz face circuit electronics
Fig.2 Fuzz face circuit, classic 60s original schematics
with different sections of the circuit
(made with Fritzing)


The first capacitor (C2, 2.2uF) filters the DC current at the entry of the circuit to avoid troubles with the next part of the circuit. This kind of capacitor (that you will find in a lot of circuits) is called a coupling capacitor. It also acts as a high pass filter, removing some bass frequencies. Enlarging its value will let more bass go through the circuit.

The gain amplifier section (blue part of the circuit) is a simple common emitter layout to amplify the signal. It has the maximal gain possible in this configuration: the emitter is directly connected to the mass, and the collector is directly linked to the base of the second transistor. It is a very classical layout that you will find in many many transistor based circuits, even if usually a resistor is present between the emitter of the transistor and the mass. This stabilizes the circuit, and allows a fine tuning of the transistor gain by creating a negative feedback. The second transistor is in the same layout, except for its emitter.


The main interest of the circuit is the feedback loop (in green). The emitter of the second transistor is linked directly to the base of the first one through a 100k resistor. A part of the current goes back to the first transistor, creating an amplification loop: current from Q2 comes back to Q1!
The counter reaction (therefore the gain) of the second transistor is modified by a variable resistor placed between the emitter of the transistor and the mass: the fuzz potentiometer! By diminishing its value, the quantity of current going to the mass increases whereas the amount of current going through the feedback loop is diminishing. If the value of the variable resistor increase, more current goes through the feedback loop: the base of Q1 receives more current, saturation increases.

The volume / output section (in red) is the simplest volume control that you can imagine: a 0.1uF coupling capacitor to eliminate DC current, and a volume potentiometer wired as a variable resistor. The output signal will be more or less important depending on the value assigned to the potentiometer.

Analogman Sunface circuit
Fig.3 Analog.man Sunface circuit
(made with Fritzing)

The Analog.man version of this circuit introduces some little improvements to this circuit. First of all, a trimpot (wired as a variable resistor) is present at the input of the circuit before the 2.2uF capacitor. Thus, the entry impedance can be adjusted a bit. It plays the same role as a guitar volume pot, so you can adjust the fuzz gain, which is really useful for high output pickups for instance.

Another modification to the circuit is the 8.2k resistor that is replaced with a 2.2k resistor plus a 5k linear potentiometer. I prefer to use a 10k potentiometer so the original 8.2k value is reachable. This modification allows you to set the bias of the second resistor. It is useful in order to have the maximal gain, especially as the circuit is really temperature-sensitive.

Finally, some values are changed compared to the original circuit: the coupling capacitor is 1uF and the one before the volume pot is 0,01uF. The volume potentiometer is 250k. All together, this will remove a bit of low end, making it a brighter fuzz. This is probably done in order to compensate the deeper tone of the NKT275.

Nothing difficult, but still good improvements to the circuit !



How to make it?

Even if the circuit is really simple, there are some recommendations to follow in order not to mess with it:
  • Beware of the orientation of the 10 and 22 uF capacitors: electrolytic capacitors are polarized. Messing with the orientation can result (in the best case) in a non-working circuit, or in the explosion of the capacitor, which really not a nice thing to happen !
  • The transistors are the main part of the circuit, and there are some precautions to take with it. Verify that the orientation (Collector, Base, Emitter) is correct before plugging the DC power. 
Ac128 NKT275 polarity
 Fig. 3 Collector, Base and Emitter of germanium transistors
(C, B and E, respectively)
  • Do not overheat germanium transistors! These pieces are very fragile, sometimes a bit of solder can kill them... Additionally, germanium transistors are most of the time quite expensive devices, so I would really recommend to use a solderless method like transistors sockets, and to wait that everything is really cool before putting them in. So, of course, the couple of transistor is the last component to be added to the circuit.
  • Last but not least: the circuit needs a negative tension to work properly (-9V). Do not try to use your good old boss charger! 

References / to go further:
http://www.geofex.com/article_folders/fuzzface/fffram.htm 
http://fuzzcentral.ssguitar.com/fuzzface.php
http://www.electrosmash.com/fuzz-face
http://www.analogman.com/fuzzface.htm

17 comments

  1. The to images at the top dont load. teh links seem to be dead. can you fix them?

    ReplyDelete
    Replies
    1. Hello,
      Unfortunately, those images were hosted on another server that was shut down... So I lost most of them... I will put the only remaining picture I have

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

    ReplyDelete
  3. wouldn't be 470R the resistor's value where's 470K indicated?

    ReplyDelete
    Replies
    1. Yes, and also on sunface scheme 1st and 2nd pins of sundial pot must be wired together and connected to the point between R4 and C3. I made a mistake when connected the 1st pin to the ground. It's sounds awful, don't do this.

      Delete
  4. Coda, can i use this schematic with bc108?

    ReplyDelete
    Replies
    1. Hello,

      Unfortunately, no, because BC108 are NPN transistors whereas the transistors I used are PNP. The polarity of the circuit is reversed compared to BC108 (see the inverted battery)

      You can find BC108 schematic easily on guitarfxlayouts though.

      Good luck!

      Benoit

      Delete
  5. I have just finished making a ANALOGMAN NKT275 clone from guitar FX layouts , Sorry for a daft question but is the Q2 Transistor closest to the 33k resistor, and Q1 next to the 22uf Electrolytic Capacitor. Best Regards Stu.

    ReplyDelete
  6. Its ok I found the schematic I can carry on now .

    ReplyDelete
  7. Hey man, i built from your schematic, and these things may have been mentioned before, but i don't see an updated schematic. If you could update your main schematic, you could save some people the days it took me to figure out the problems with it.

    R4 is a 470 ohm, not a 470K.
    The trimmer pot's left leg should not go to ground, it should not connect to anything.
    The bias pot's left leg should not go to ground (or it will catch on fire), it should not be connected.
    Please update your schematic because you're getting people coming to your page but you haven't given them an accurate build guide yet.
    Please fix this stuff and re-vise your post to save others from the same fate.

    And thank you very much for having done so much already on this, I really appreciate it. It sounds great once you make those changes!!!!

    ReplyDelete
  8. How did you solder the AC128's without the sockets to ensure you didn't fry them? Also, what gain were the two transistors you used? Trying to determined if I should get lower, medium, or higher gain AC128's or a mix. Thanks!

    ReplyDelete
  9. I used 2N527 germaniums (cheaper, more consistent, sound killer!), not AC128's. If I remember correctly the gain on the two transistors were 68 and 110... Regarding soldering them - don't. You can use 6 PIN DIP IC SOCKET's (and break them in half) so you can plug in different transistors, or even better get DIP SIP IC SOCKETS which you can cut/break into 3 pin pieces, solder those in, and try out different transistors to your heart's content. Hope that helps.

    ReplyDelete
  10. s flynn is right on the money, but the 6 pin machined dip sockets work way better than the sip sockets sold by everyone.. the sip sockets tend to fail real easily as stuff moves around, where as the dip ones cut in half give a much more solid connection.

    ReplyDelete
  11. nice tip, Pinkster. didn't know that. i figure the sips look cooler ;-p
    and i only try out a handful of transistors before hot-gluing it to the closest cap, but i will be using the dips from now on. cheers

    ReplyDelete
  12. Companies that create printed circuit boards can create them specifically for the intention of the customer as well as their needs. PCB reverse engineering services A PCB prototype is designed to meet these requirements, which can further facilitate mass production, as required by the client.

    ReplyDelete
  13. You, my pal, ROCK! I found just the information I already searched all over the place and just could not locate it. What a perfect web site. https://royalcbd.com/product/cbd-roll-on-gel/

    ReplyDelete
  14. Will you PLEASE correct the schematic errors? Such as showing 470k instead of 470 ohms and NOT grounding the adjustment trimmer.
    Many grab the schematic and try to build it as is.
    Google search still shows the wrong schematic.
    Thanks.

    ReplyDelete