Univox - Superfuzz  [schematic]

Discussion regarding early stompbox technology: 1960-1975 Please keep discussion focused and contribute what info you have...
User avatar
Jan1966
Breadboard Brother
Information
Posts: 66
Joined: 15 Dec 2014, 19:17
Has thanked: 14 times
Been thanked: 14 times

Post by Jan1966 »

Dibya wrote:2sc828 are rare gens , can I use something else?
Yes. Try experimenting with what you have got.

User avatar
Nocentelli
Tube Twister
Information
Posts: 2222
Joined: 09 Apr 2009, 07:06
Location: Leeds, UK
Has thanked: 1152 times
Been thanked: 954 times

Post by Nocentelli »

Pretty sure any Si BJT in a similar hFe range (200-300ish) would do, 2N3904 are probably the most common/easily available.
modman wrote: Let's hope it's not a hit, because soldering up the same pedal everyday, is a sad life. It's that same ole devilish double bind again...

User avatar
digi2t
Degoop Doctor
Information
Posts: 284
Joined: 09 Feb 2011, 15:10
Has thanked: 54 times
Been thanked: 385 times

Post by digi2t »

2N3904 tend to be too hot. 2N2222 works better. Gains in the 150 to 180 range, with Q4 and Q5 matched. If you can find 2N2219's, even better.
No matter how many times I cut it, it`s STILL too short!

https://www.deadendfx.com/

User avatar
jalmonsalmon
Solder Soldier
Information
Posts: 212
Joined: 14 Sep 2012, 22:49
Has thanked: 97 times
Been thanked: 30 times

Post by jalmonsalmon »

digi2t wrote:2N3904 tend to be too hot. 2N2222 works better. Gains in the 150 to 180 range, with Q4 and Q5 matched. If you can find 2N2219's, even better.
matched pair for Q4 and Q5 cause more of the upper octave thing?
I made one of these years ago, seems to work good but I would like to revisit this pedal and get it the best it can be

User avatar
aion
Solder Soldier
Information
Posts: 230
Joined: 22 Mar 2012, 00:12
Location: Des Moines, IA
Has thanked: 9 times
Been thanked: 472 times
Contact:

Post by aion »

jalmonsalmon wrote:
digi2t wrote:2N3904 tend to be too hot. 2N2222 works better. Gains in the 150 to 180 range, with Q4 and Q5 matched. If you can find 2N2219's, even better.
matched pair for Q4 and Q5 cause more of the upper octave thing?
I made one of these years ago, seems to work good but I would like to revisit this pedal and get it the best it can be
Correct. The octave trimmer (found on most Superfuzz layouts) can compensate somewhat for this, but you do want them to be as close as possible.
digi2t wrote:2N3904 tend to be too hot. 2N2222 works better. Gains in the 150 to 180 range, with Q4 and Q5 matched. If you can find 2N2219's, even better.
An original Superfuzz that I measured had the following transistor gains:
Q1: 124 hFE
Q2: 95 hFE
Q3: 143 hFE
Q4: 62 hFE
Q5: 50 hFE
Q6: 52 hFE

I found my clone sounded dead-on to the original with 2N3904 in Q1-3 (measuring the to make sure they are in the sub-180 range) and 2N3903 for Q4-6, with Q4 and 5 being matched. But any of the types Dino mentioned will work as well. The 3903 was readily available (drying up now though) and much cheaper than the 2N2222 when I did it last year.

The 3046 transistor array was hit or miss. I tested a few that were in the low 40's for gain - that's too low for the Superfuzz and it doesn't sound right. If you know the transistors are in the 60-100 range then it'll work, but it's not necessarily a drop-in solution.

User avatar
Manfred
Tube Twister
Information
Posts: 1937
Joined: 04 Apr 2009, 23:42
Has thanked: 1671 times
Been thanked: 1344 times

Post by Manfred »

An original Superfuzz that I measured had the following transistor gains:
What is the measuring collector current of HFE meter?

User avatar
digitalzombie
Breadboard Brother
Information
Posts: 77
Joined: 06 Feb 2012, 18:20
Has thanked: 7 times
Been thanked: 13 times

Post by digitalzombie »

So are we still going off the schematic from page 1 of this thread, or is there an updated schematic in one of the many broken image links?

User avatar
digi2t
Degoop Doctor
Information
Posts: 284
Joined: 09 Feb 2011, 15:10
Has thanked: 54 times
Been thanked: 385 times

Post by digi2t »

aionios wrote: The 3046 transistor array was hit or miss. I tested a few that were in the low 40's for gain - that's too low for the Superfuzz and it doesn't sound right. If you know the transistors are in the 60-100 range then it'll work, but it's not necessarily a drop-in solution.
The best array results that I've had are with the TPQ2222. They are also sold under the MPQ2222 moniker as well. These are quad NPN array IC's (four 2N2222's), of which I use them for Q1, 2, 3, and 6. Another quad IC's that works well is the RFT B360Dc, which I see come up on EBay quite often. The "c" suffix is the gain range, which according to the datasheet puts them in the 56 - 140 range. A tad lower than the TPQ/MPQ series, but they can be had quite cheap in some cases (especially in Europe), which means you can buy a bunch and audition them. Now... do the array IC's provide any tonal advantage over sorting out four individual transistors? Not really I suppose, but I guess I'm just really lazy, so if I can avoid the work of sorting transistors, I'll take that out any day.

As for "dualling" the Q4/5 section, using supermatch exotics such as the 2SC1583 or LM394 aren't the ticket. The gains are way too high, and the octave just doesn't bloom correctly. They tend to make chord work rather muddy and overly compressed as well. 2N2913/15/17 work very well indeed. The 2N2913 has been my "go to" for this section for a while now. Gain range for this series is 60 - 250, with the average for any I've ever bought sitting right in the wheelhouse at 150. Using the balancing trimmer is highly recommended, as you may actually wish to "detune" this section a bit. With an overly strong octave, it can be detrimental to chord work giving them too much of a ringmod'ish feel.

Three final mods that I find to be most beneficial;
1) Lifting the clipping diode pair, just a bit, from ground. I use a 10K trimmer, but I find lifting between 1K and 2K gives the fuzz a bit more meat and body, without sacrificing any of the saw edge.
2) Tone pot, rather than tone switch. This is a taste object. If you like to switch between two distinct tones, than stay with the switch. If (like myself) you prefer to "set and forget", then a tone pot is better. It allows to tune into all the in between tones that the switch system can't give you.
3) Replace the 15K Q6 base to ground resistor with a 25K trimmer. Adjusting the Q6 collector voltage via the trimmer can help finalize the tonal and output quality of the Superfuzz. With Q6 in the 125 to 175 gain range, the trimmer should give you a usable collector voltage adjustment range of 2.7V to 7.2V. I find that 5.75V is the best balance between great tone and all out roar. YMMV.

Here's my version (just right click and "open image in new tab" to get the full image);

Image
No matter how many times I cut it, it`s STILL too short!

https://www.deadendfx.com/

User avatar
snk
Resistor Ronker
Information
Posts: 280
Joined: 14 May 2013, 20:48
Has thanked: 105 times
Been thanked: 16 times

Post by snk »

Hi,
I built a SuperFuxx on veroboard using the Derringer layout, but the effect is dead silent...
I have checked 3 times the traces and cuts, the electrolytics polarity, but obviously something is wrong with my build... :scratch:
I am suspecting an issue in the wiring.
One thing i am wondering is : is the TONE 3 lug left unconnected ? I see tone1 goes to ground, and tone 2 goes to line12, but i see no indication about lug 3...

User avatar
Ichabod_Crane
Resistor Ronker
Information
Posts: 354
Joined: 26 Apr 2016, 15:11
Has thanked: 83 times
Been thanked: 36 times

Post by Ichabod_Crane »

Can you link or post the layout you used?
I don't know the one by Derringer. Are you sure is it verified?

User avatar
snk
Resistor Ronker
Information
Posts: 280
Joined: 14 May 2013, 20:48
Has thanked: 105 times
Been thanked: 16 times

Post by snk »

Thank you, Ichabod
This is the layout, it was posted and discussed earlier in the thread, and it is verified.
It adds several mods and improvements over the "stock" SuperFuzz.
SuperfuzzVero090630.jpg

User avatar
snk
Resistor Ronker
Information
Posts: 280
Joined: 14 May 2013, 20:48
Has thanked: 105 times
Been thanked: 16 times

Post by snk »

The only differences i used from this layout are these :
- As there are several GROUND and +9V inputs, i used two wires according to the layout below, and did not used the greyed out 9v and ground wires.
- Instead of trimmers, i used pots for VR1 and VR2.
Image

User avatar
Jim-Analog
Information
Posts: 16
Joined: 20 Jul 2010, 19:38
Has thanked: 43 times
Been thanked: 3 times

Post by Jim-Analog »

Greetings,

I agree here! Back in the mid '70s I had an AC Unifuzz. Somewhere along the line it got swiped from a stage or club or ??? Other than the power supply, what differences are there to the circuit vs the superfuzz? Anyone have an exact schematic for that version? I'd really like to try one again and see if the unit was really that cool or the memory was better than the reality. They seem to go for silly $ now, so building one would be the only way to find out. Thanks in advance for any clues!

Regards, Jim

drmathprog wrote:Personally, except for the fact that it's AC powered instead of DC, I've always preferred the Univox Unifuzz to the Superfuzz. It's more versatile and useful.

User avatar
snk
Resistor Ronker
Information
Posts: 280
Joined: 14 May 2013, 20:48
Has thanked: 105 times
Been thanked: 16 times

Post by snk »

Hello,

I'm replying to myself, as well as for future builders :
- I made a stupide mistake : I tried to use 2SC828 transistors (as used in the original pedal), without noticing their pinout was not the same than on the layout i used. I tried with 2N2222s, and it worked fine :)
- According to this PCB layout, tone 2 and tone3 lugs should be wired together.

I could compare the Derringer version with a vintage unit :
- The Derringer version has much, much more output volume :D
- It's good to have the tone as a pot.
- it seems that the Derringer version, letting pass more bass through, has less octave feel. This is in no way a bad thing, it is just that it seems that to get more octave, you might feed the pedal less bass, so if you want the original sound, you may consider adding two input capacitors on a switch.

User avatar
snk
Resistor Ronker
Information
Posts: 280
Joined: 14 May 2013, 20:48
Has thanked: 105 times
Been thanked: 16 times

Post by snk »

Hello,
I'm willing to use a switch to toggle between two input capacitors so i would have the choice to have more bass ("original" Derringer version) or to have a bass cut.
Which capacitor should I put on a switch ? C1 ? C2 ? C4 ?

User avatar
fauxjazz
Information
Posts: 2
Joined: 01 Feb 2014, 16:49

Post by fauxjazz »

ShortScaleMike wrote: 17 Oct 2008, 10:21 I made designed and built a stripboard layout for this. the original layout is verified (it was a paper design), unfortunately this electronic version isn't as yet.

Image
Sorry to resurrect this, but isn't the last 100k resitor before the output on this schematic missing from this layout?

Image

User avatar
Intripped
Cap Cooler
Information
Posts: 670
Joined: 25 Aug 2009, 01:03
Location: Italy
Has thanked: 989 times
Been thanked: 305 times

Post by Intripped »

yes, that resistor is missing on the layout.
Easy to add, though

User avatar
aion
Solder Soldier
Information
Posts: 230
Joined: 22 Mar 2012, 00:12
Location: Des Moines, IA
Has thanked: 9 times
Been thanked: 472 times
Contact:

Post by aion »

Manfred wrote: 22 Jul 2019, 15:05
An original Superfuzz that I measured had the following transistor gains:
What is the measuring collector current of HFE meter?
Only a year late on seeing this, but it was a Peak Atlas DCA55 which uses a test current of 2.5mA.

User avatar
Manfred
Tube Twister
Information
Posts: 1937
Joined: 04 Apr 2009, 23:42
Has thanked: 1671 times
Been thanked: 1344 times

Post by Manfred »

Only a year late on seeing this, but it was a Peak Atlas DCA55 which uses a test current of 2.5mA.
Thanks, did you also measure the voltages on the transistor terminals in the original pedal?

User avatar
aion
Solder Soldier
Information
Posts: 230
Joined: 22 Mar 2012, 00:12
Location: Des Moines, IA
Has thanked: 9 times
Been thanked: 472 times
Contact:

Post by aion »

Manfred wrote: 17 Jun 2020, 16:20
Only a year late on seeing this, but it was a Peak Atlas DCA55 which uses a test current of 2.5mA.
Thanks, did you also measure the voltages on the transistor terminals in the original pedal?
Nope, I don't think I did.

Post Reply