Ampeg Scrambler

[grapefruit]

Howdy,

I'm new to this forum. I'm an electronics tech and former/part time sound engineer.
I've built a clone of the Ampeg Scrambler. I used MPA13 for the darlingtons and a bc547 for the other transistor. The transistors are in vaguely the same gain range as the originals and seem to be biased ok. It sounds great but there is one problem... As the note fades away the sound breaks up a bit. I'm thinking this is because the rectifier diodes are not forward biased so they are sometimes not quite breaking over when the input gets low. It almost sounds like a low frequency oscillation/modulation.

Could somebody please tell me if this is common to the scrambler design or if my parts substitution is at fault? I've tried both 1N914 and OA91 Germanium diodes.

Thanks in advance,
Stew.

[AL]

I've never built one so I can't give you any hands on "expertise" but there are schem's and build instructions here

http://www.generalguitargadgets.com/v2/index.php?Itemid=1      

and here

http://fuzzcentral.tripod.com/

AL

[Samuel]

If it's any consolation, mine does the same thing...

[Mark Hammer]

The original diodes were 1N456, which is apparently a different class of diode than mere garden variety 1N914 or other SI or GE switchng diodes.  Apparently a 1N457 can sub.

[drew]

Weirdly enough, I have a bunch of 1N456s sitting around (used them for the scrambler) because I was surplus shopping years ago and got a sack of diodes labeled as "1N914/4148 equivalent".... and they turned out to be 456s!

I believe you can get these from Steve at Small Bear if you don't want to go digging through a dirty warehouse though


drew
www.toothpastefordinner.com

[grapefruit]

Thanks for the input... I'll see if I can find some 1N456 diodes. The datasheet for them said they are low leakage silicon diodes. I don't see how being low leakage will solve the problem I'm having but it's worth a shot. I'm thinking about forward biasing the diodes, although I think this will change the sound somewhat as even using Ge diodes makes the sound a lot less harsh.

[jsleep]

You can get 1N456 diodes from mouser.  $3.00 for 100 count. They're not the pretty, old-time mult-colored ones like Small Bear sells, but I bet they are functionally the same.

JD

[grapefruit]

I've ordered a few different diodes from local suppliers to try but haven't found any 1N456 Diodes. What I would like to know is, if people have built it with original parts, or own an original, does it have the same problem breaking up with small (decaying) signals? I don't want to import diodes if this is inherent to the design. It's really only if you hold a note for about 5 seconds that you notice it.

Any advice would be greatly appreciated.

stew.

[Elektrojänis]

I just finished building my scrambler clone a few days ago. It was actually second scrambler I have built. First one was for my friend who uses it with his Yamaha YC-20 organ (crazy:-).

Both of those are bit modified... My friends one has a trimmer for setting input volume and some other stuff.

My version has an extra fet gain stage for getting a bit more volume even with the lowest settings of the blend pot, a simple tone control (that can be bypassed with a switch) and a volume control.

I tried some Ge-diodes and 1n4148's. I left the ge ones in. The difference was not big but I think the chocking/breaking problem at the end of the notes was a bit. less with the germaniums.

Another thing: I used 1n3904's as Q1, Q3 and Q4 and 2n2222 for Q2. I could not get 2n5306's and BC169B. I tried some other darlingtons for  Q1, Q3 and Q4 but the 2n3904's (not a darlington I know)  sounded better. Actually It seemed to me thet the Q1 did not matter too much (well... It's just a buffer) but Q3 and Q4 did.

I have wanted to build this for a long time... 5 year maybe... Ever since I first heard the sound samples on tone frenzy anyway. Finally I have it.

[phillip]

I've built several Scramblers for people, and I've always used either the 1N914s or 1N4148s.  The very first one that I built used the 1N456 that I got from Smallbear, but as far as I can tell, there's almost no sound difference.  

In the last one that I built I was out of the 2N5306 that I usually use, so I tried the MPSA13s that I've had, but they produced that terrible "farty" gated sound...not enough gain.  As soon as I got a new bag of 5306, I plugged them in and it was working perfectly.  It seems that the A13 has about half the gain as the 2N5306.  I would try MPSA14, which will work for the 5306.

Phillip

[Elektrojänis]

I've built several Scramblers for people,
In the last one that I built I was out of the 2N5306 that I usually use, so I tried the MPSA13s that I've had, but they produced that terrible "farty" gated sound...not enough gain.  As soon as I got a new bag of 5306, I plugged them in and it was working perfectly.  It seems that the A13 has about half the gain as the 2N5306.  I would try MPSA14, which will work for the 5306.
Phillip

Strange. The darlingtons I tried where BC517. They should have hfe of 30000  or more (according to datasheet).  Still I feel that the 2n3904's (hfe under 200 according to my multimeter if I remember correctly) sound better in it and the circuit seems to distort more (according to my ears). I like strange sounds, but I hear no gating effect exept maybe in the very very end of the decaying note.

I got to add that using 2n3904's was not my idea... I read it from some webpage, but I just can't remember what page was it.

Hmmm... I might try to record some soundsamples and put them online.

[B Tremblay]

Still I feel that the 2n3904's (hfe under 200 according to my multimeter if I remember correctly) sound better in it and the circuit seems to distort more (according to my ears).

I've tried 2N5306 and MPSA13 in my Scrambler, but I liked 2N5088 better than both.  However, I rarely follow the "original factory parts" movement.

[Elektrojänis]

Okay... Here is a soundclip of my scrambler: http://www.hut.fi/~pjunno/temp/scramblertest6.mp3

It's not very well thought out demo. I just hit the rec button and played. Sorry for the crappy playing. I haven't been playing too much lately and I never was any good as a guitarist anyway.

It's recorded straight to line and because of that I used the tone control I have added to my unit (a simple low pass thingy) to cut some treble (just a bit). Only control (in the scrambler) I turn on the clip is the texture. The blend was set fully to the distorted side. The guitar is some Charvel model... Strat style with one humbucker (on the bridge position) and two single coils. It's bit modified though.

BTW. Nice winter look on your sites front page B Tremblay.

-Petri Junno

[B Tremblay]

Okay... Here is a soundclip of my scrambler

Nice winter look on your sites front page B Tremblay.

Mine sounds pretty much the same as Elektrojänis' clip, if it helps.  Nice clip!

I'm glad you like the holiday 'Groove home page, too!

[Elektrojänis]

I listened the clips on tonefrenzy again and I think mine sounds weaker. It could be because the transistor selections, but I think part of it could be because of two reasons:

1. My guitars pickups are weak.  I have suspected this before.

2. I suspect that the tone of the the tonefrenzy examples is affected by the amp that was used. The amp could be even distorting a bit. I suspect this because the output level of the scrambler when the blend pot is on the more distorted side is quite big compared to output of a plain guitar.

BTW. I like to drive other distortion pedals with this too.

[Elektrojänis]

Some thoughts after toying around with my prototype.

The transistors do not seem to affect the tone too much on my units. Q3 and Q4 seem to make some difference. I still liked 2n3904 on those positions but the difference is not too big (I did not have the original transistors but I triec BC517 = darlington that has hfe of over 30000).  The Q1 and Q2  did not make too much difference.

The diodes... Silicon (1n4148) causes a gating effect to the very end of the note. It is not very serious problem though... It does not happen untill the veryvery end of the note and I think it dies away very nicely without any stuttering that I could hear. Germaniums (I forgot what I have as they do not have the numgers printed on them) cause the gating to dissapear completely as far as I can hear. They cause it to loose bit of the edge from the sound though. I now think I should have put a switch to my completed unit to swich the diodes (it has ge-diodes in it).

Last test I did was to chain my completed unit and the prototype. When I turned the blend and texture controls to the max I got a really insane sound that I liked a lot... Now I think I need to build another for myself. Argh!