Hi all
I've found an old schematic
[img=http://img204.imageshack.us/img204/3551/minisynthlq5.th.png] (http://img204.imageshack.us/my.php?image=minisynthlq5.png)
(is it clear enough?)
I hope I'm not breaking copyright as I don't know where it cames from
The Fx will produces 3 sounds :
1st sound - the base one played by the guitar
2nd sound - 1 octave lower
3rd sound - 2 octave lower
the 3 pots will mix these sounds
I've already build 2 prototipes of it with different layouts (on PCB), and it works almost fine but there's a problem with decay/sustain. It seems that the SAJ110 is working fine as I have the 3 sounds (playing high E string (E cantino?) and with clean sound to 0 and -1 octave also to 0 it sound like a bass!!).
But I'm getting this decay/sustain problem, so after a while (as soon as string vibration start to decrease) the FX sounds disapear and I can only hear a distorted/dirty sound of the guitar string. So it seems the problem I have is to be found in the first section of the schematic, before the signal enters in the IC.
FoxFire thinks about a bias issue, but I don't know very much what's he's talking about. I've used 2n914 in one PCB and (I'm not sure) 2n5457 in another one instead of BC239 (cannot find them in Italy, I know Banzaieffect have them but too expensive for an "experiment"!).
Just to point out the problem is absolutely the same on the 2 PCB.
May be someone of you can, looking at the schematic, understand what the problem could be, I thought also I may need to amplify the guitar signal before the FX.
thx for help, and forgive my english,
Armando
The schematic is kind of small and hard to read. Can you post a bigger pic?
The original transistors are supposed to be bc239, right?
These have much bigger amplification than the ones you've got in there.
Just swap Q 1-3 to something stronger.
It seems that when the signal falls below a certain level it`s no longer triggering the effect. It might just be in the nature of the circuit or perhaps changing some resistor values might help. The schematic isn`t too clear so I can`t really help anymore than that.
Quote from: oskar on July 03, 2007, 08:51:19 AM
The original transistors are supposed to be bc239, right?
These have much bigger amplification than the ones you've got in there.
Just swap Q 1-3 to something stronger.
PS... This will extend the sustain alot but you're going to have the same decayproblem when the guitar signal fades away.
The thing you've got is something like
http://topopiccione.atspace.com/pjimages/ShinEiOctave.sch.jpg
But with an integrated divider instead of a discrete solution.
You should redo it with a 4013 chip instead of an obsolete chip, especially if you plan on
putting the thing on the net.
Check out Craig Andertons
http://www.amazon.com/Do-Yourself-Projects-Guitarists-Inexpensive/dp/087930359X/ref=sip_rec_dp_6/104-0408657-1931908
which presents...
http://hammer.ampage.org/files/rocktave.pdf
which solves the problem with decay.
thx to all for advertise/suggestions
here is a link to a different picture of the schematic, hope it's good enough (may be now it's too big!)
(http://img522.imageshack.us/img522/9058/minisynth2ir3.th.png) (http://img522.imageshack.us/my.php?image=minisynth2ir3.png)
by the way I heard it working years ago (late '70s early '80) so it would be nice to reproduce a kind of FX.
thx oskar for those links, very useful, specially when talks about how hard/soft you play a string. May be to let this FX working I should also find a different way to play the guitar. any suggestion for new Q1 and Q3.
As the chip you are using was originally intended for frequency dividing in old organs, is it possible that they are not designed to work below a certain frequency/level. As organs have a fixed level of velocity this wouldn`t have been an issue.
Just a thought.
I don't believe it will ever work as well as you would hope.
Because, it works by using the amplifying stages to square up the input signal, so it can trigger the divider chip.. so the signal needs to be high enough to do that, but if it is TOO high, then harmonics and noise will trigger the divider as well.
I'd try putting a compressor before it, and mixing some of the original guitar with the output.
I don't think there is anything wrong with the divider, just suprised you could find any!
I think the SAJ110 is similar to a CD4024, if so it needs a good loud signal to drive it, a squarewave works best. Once the signal drops below a certain voltage the chip stops triggering.
If you haven't already seen it my Slacktave design (http://www.diystompboxes.com/smfforum/index.php?topic=55644.0) is a similar effect, but using a CD4024. You could try using the first 3 opamp stages off my design as they give a good strong squarewave with decent sustain.
I CAN'T BELIEVE!!!
Are you guys talking about this little beast?
(http://i195.photobucket.com/albums/z210/alex_frias/M3MiniSynthEmThree.jpg)
(http://i195.photobucket.com/albums/z210/alex_frias/DSC03352.jpg)
I love this thing, It sports a bell clip design.
I use it in the loop of my McMeat (MeatBalls autofilter clone) and it mimics a very strong analog synth. Big and fat!
I built a MXR BlueBox, but the "tracking" is not that good.
Nothing beats this DIYstompboxes forum.
Any questions regarding it, pics, sounds, please, feel free for asking!
If I may be of any help I would be happy to do it.
Looks interesting. It`s the same chip alright. I`d love to hear a clip of it. Does it exhibit the cutoff problem that arma61 was talking about?
I can't see the pics or anything, this is frustrating! If it's the one with the article that talks about whistling into it or something, i'd be very interested to hear a sound sample of it. In fact, can we hear some anyway? Anything synth'y is good by me!
Quote from: arma61 on July 03, 2007, 04:12:37 PM
1.by the way I heard it working years ago (late '70s early '80) so it would be nice to reproduce a kind of FX...
2.May be to let this FX working I should also find a different way to play the guitar.
3. any suggestion for new Q1 and Q3.
1. Well a squarewave is a squarewave is a square... Basically it's a digital effect actually,
and so is the bluebox. All that matters in these type of effects is.
Filter away them overtones, amplify like a madman, amplify some more, go schmitttrigger, drive a digital chip (not just a
frequency divider try the complete 4000- cmos series. You can make a simple ringmodulator by an XOR gate or divide it with
a 4017 to give it a non-octave subwave. go wild man!!!)
The actual chip you use to achieve the result is completely unimportant. You shouldn't bye expensive ones...
2. Yes, that's what I found out to. Heavy string action, legato style and don't wait for the decay to happen.
3. Whatever is stronger. BC549...
Oh, it's fashionably uggly to, just keeps getting better!!!! :o
This is an alternative divider for your project. It isn't pin to pin compatible but the chip is available from
just about everywhere and you can use it for a lot of things. I don't know if the subwaves will come out
the correct phase though. If they dont you just swap the outputs to pin 2 and 12. These are inverted
outputs. I've done octave dividers with this one alot. This is the same chip MXR's Bluebox uses.
(http://i194.photobucket.com/albums/z305/owallgren/4013_divider-1.png)
Quote from: Jaicen_solo on July 04, 2007, 01:59:23 PM
I can't see the pics or anything, this is frustrating! If it's the one with the article that talks about whistling into it or something, i'd be very interested to hear a sound sample of it. In fact, can we hear some anyway? Anything synth'y is good by me!
I suspect that you are thinking about the MiniSynth (there was a kit for it that went with the Dick Smith "Funway Into Electronics").
Possibly, it's the one on GGG. I've never heard of anyone building it, so I was curious. Doesn't look like tihs is the same beast at all.
Ok. I give myself a high score for optimism and what I recalled as something easy to build just gave me a sleepness night. :)
You rather easily end up triggering the divider ( not the Minisynth specifically... I haven't tried that one ) with the overtones which
actually isn't bad at all if you can controll it. Eventually it worked and I'll try the Slacktave too soon I hope.
I think optimally the machine for the job would contain a compressor and a good tracking filter but it actually is rather easy
to have serious fun with even a simple circuit.
ps. I had a live recording with Bootsy Collins with some rather heavy strange noise from the bass on one or two tracks. I actually
ended up with some rather similar noise when I drove the circuit with too strong signal. If the signal hits the roof you'll get a
S***load of overtones and it's more noise/gameboy style that actually someone could enjoy. I didn't...
nice to see all you guys so interested and helping me to get it working, I would like to post a sound sample but my soundcard stops working yesterday night (brang new Pc bought last saturday!! #@%$@%#$!!!!!!!!!!!)
yes, alex frias, that's the beast I'm talking about, if you look at the first picture posted you can see M3 at the top. just a couple of questions :
can you see which kind of transistors are used in your FX
can you compare the components value in the schematic I've posted with the FX
my SAJ110s are black not red like yours, do you think this could be a problem (different rating, max voltage..)
I'll be on vacation (finally!) for 2 weeks, when back I'll put in place some of your suggestions and post the results,
Keep DIYing
Well... No IC racism here.
I don't think they are different in anyway... Aside the "Mojo Factor".
"Look, my red one is the real Holy Grail IC of MiniSynthes"
'Scuse the brief hijack.
Quote from: Jaicen_solo on July 05, 2007, 05:49:55 AM
Possibly, it's the one on GGG. I've never heard of anyone building it, so I was curious.
I've built a couple. A fun device! This clip used to be on the internet of one trying to keep up with a singer:
http://users.rio.com/senorris/junk/MINIASIA.mp3
It's a bad played sample recorded direct to the board, then to the A/D converter. No amp simulator/modeler/emulator was used, just a bit of software reverberation.
The neck pickup was used, a humbucking one (DiMarzio PAF Pro)
Firstly It's the sound of the distorted original octave (volume knob 1), full volume on guitar, then half way backed off. Then the other sounds (knobs 2 and 3 individualy).
At last the combination of the 3 octaves, same mix volume to the three sounds.
http://rapidshare.com/files/41370927/MiniSynthyM3.mp3.html (http://rapidshare.com/files/41370927/MiniSynthyM3.mp3.html)
I think I'd like to build one of these to try on my bass. American Semiconductor has these SAJ110 chips for $12.93 but they want like $19 shipping and handling. Anybody want to go in together to split the shipping and get some of these?
Sounds well. :)
The "lo-fi gameboy style sound" comes from the circuit tracking overtones as well (I assume from
my own humble experiment with increased filtering).
But those chips again... they are way too expensive.
If I redraw it for a modern IC instead, would anyone use it? ??? Does it absolutely have to be the SAJ110?
Quote from: oskar on July 06, 2007, 12:26:10 PM
Sounds well. :)
The "lo-fi gameboy style sound" comes from the circuit tracking overtones as well (I assume from
my own humble experiment with increased filtering).
Yeah the octave jumping and weirdness is because there's too many overtones in the signal driving the chip, so it can't lock on to the fundamental properly. Doesn't sound too bad though, in my experience conditioning the signal is the hardest part, that's the bit that took me longest to get right when designing the Slacktave.
Quote
But those chips again... they are way too expensive.
Does it absolutely have to be the SAJ110?
After listening to the soundclip, if you used a CD4024 (http://www.kpsec.freeuk.com/components/cmos.htm#4024) it would sound exactly the same. The only difference would be if the SAJ110 had any sort of filtering built in to help improve tracking, then the CD4024 might not track as well. Apart from the pinouts being different you could just drop one straight in instead of the SAJ110. A CD4040 would also work just as well.
Quote from: slacker on July 06, 2007, 01:13:59 PM
After listening to the soundclip, if you used a CD4024 (http://www.kpsec.freeuk.com/components/cmos.htm#4024) it would sound exactly the same. The only difference would be if the SAJ110 had any sort of filtering built in to help improve tracking, then the CD4024 might not track as well. Apart from the pinouts being different you could just drop one straight in instead of the SAJ110. A CD4040 would also work just as well.
No, hardly. In electric organs from those dark years the main brain was a frequency generating chip (Electric Projects For Musicians use one in a project i think)
which gives out the frequencies of the top scale. Each of these clocks are then divided down by a chip like the SAJ110. Just dividing, no filters, go CMOS. 8)
In that case the CMOS chips would work exactly the same then, and they're a fraction of the price :)
Allthough I don't think they come in red :(
If you haven't tried the XOR-ringmodulator yet I think you should. The KORG MS20 synth use it and among analog freaks
it's on manys top 10-list. It use a 4011 quad NAND chip hooked up as an XOR-gate (am I officially a geek now?) ::)
oskar
Quote from: oskar on July 06, 2007, 12:26:10 PM
Sounds well. :)
The "lo-fi gameboy style sound" comes from the circuit tracking overtones as well (I assume from
my own humble experiment with increased filtering).
But those chips again... they are way too expensive.
If I redraw it for a modern IC instead, would anyone use it? ??? Does it absolutely have to be the SAJ110?
Redraw it!
Some 4024, 4013, etc. I don't think it will sound any different with them.
But you must agree, nothing like a Red IC... ;D
The Rocktave really shows a better tracking, a good answer and soon I will build one.
But I love the quasi-stability of this thing. i don't know why exactly it seems better than the BlueBox.
I had not enough time to check the circuit presented here comparing it to the real thing, but at first look it seemed pretty close to the diagram.
My eyes are not that good these days, so I couldn't recognize the transistors in it, but soon I will. Better glasses, I hope...
Maybe using a filter I saw and built (in GGG, I think) in order to use it on my GreenRinger. It is that kind of lowpass with a very strong angle at the cutoff freq.
The kind you see before and after BBD IC`s and A/D-D/A converters in order to filter out aliasing and clock noise. Lots of dB by octave.
Another experimentation is to try som MOD's suggested by Craig Anderton's Guitar Player Magazine's article you find at the Rocktave PDF doc pointed earlier in this topic.
So many projects, so little time...:-\
Quote from: oskar on July 06, 2007, 02:12:13 PM
If you haven't tried the XOR-ringmodulator yet I think you should. The KORG MS20 synth use it and among analog freaks
it's on manys top 10-list. It use a 4011 quad NAND chip hooked up as an XOR-gate (am I officially a geek now?) ::)
You mean this? http://www.analog-synth.de/synths/ringmod/digital_ringmodulator.htm (http://www.analog-synth.de/synths/ringmod/digital_ringmodulator.htm) I'll have to give it a try.
It looks promising...
A very easy way to get more options is put a GreenRinger in parallel with it.
Or even in series, after the original octave distorted sound, assuming it will mix its output with the other octaves from the CMOS.
This very one. Another funktion is frequency doubling. I've seen it both in CMOS Cookbook and here... (somewhere in the middle)
http://www.geocities.com/tpe123/folkurban/fuzz/snippets.html
You can offcourse go fancy and use something like four outputs from a 4024 and drive the gates from an XOR-package with
an LFO or... whatever, driving the free inputs...
The funniest thing I haven't tried out yet (with a guitar) would be driving a digital noisesource, semirandom noise. I've done it on a modular and it really
is so much fun and if you have a short series like 32 bits it becomes audible as a dirty tone.
Well, if you modulate a signal with itself, you can get a pretty simulation of doubling octave....
I'm building the 4024 mod just to experiment with my BlueBox alternative version. Following the Craig Anderton notes. I will soon report the results.
The possibilities of using the 4024 output as a source for a "tracking ringmodulator" looks interesting too.
Only a note: On my MiniSynthy the transistors are BC109's.
some info on saj110
(http://www.largonet.net/midiboutique/downloads/documents/saj110.jpg)
I'm going to paint all my IC's from now on... :P
saj110 datasheet... low res.
http://www.largonet.net/midiboutique/downloads/documents/saj110lq.pdf
pinout
(http://membres.lycos.fr/alaryelectronique/memory/image19.gif)
saj110 is compatible with saj210 !!!
ps. the original drawing use input 3 and outputs 11 and 12. It would be a good idea to ground the
unused inputs: 2,4,5 and 6!
And the winer is...
It´s perfect, Oskar!
Quote from: alex frias on July 07, 2007, 10:55:23 PM
Well, if you modulate a signal with itself, you can get a pretty simulation of doubling octave....
That's true for a sine wave.
But if you try to do it with a square wave..... :icon_eek:
Which has interesting implications for an up-octave device following a distortion...
XOR-octave
http://www.geocities.com/tpe123/folkurban/fuzz/dof.gif (http://www.geocities.com/tpe123/folkurban/fuzz/dof.gif)
I haven't built this precise one but the main thing with the XOR is that it's output changes state when one of the inputs do and thus works
like a frequency adder. I have built an xor-ringmodulator with a quad-pack with each gate being triggered by an LFO/oscillator
on one of the inputs and the remaining gate being triggered by.... This one
(http://i194.photobucket.com/albums/z305/owallgren/quad-xor.png)
For a practical working unit you can't have all gates switching state at the same time. You'ld either add som filtering like the Digital Octaver
Fuzz or even better... you phase-shift the LFO-signal (cmos 4018 cirquit) to each of the XOR-inputs... ;)
The frequencies out would be signal in + LFO,signal in +2 x LFO... etc.
I honestly don't remember the precise sound of it. ???
And off course you could do something like this...
(http://i194.photobucket.com/albums/z305/owallgren/pwm.png)
Has any of you tried a 4046 for frequency doubling? Actually I think it's the way to go for a stable squarewave cirquit.
oskar
has anyone figured out the original pinout so it could be transferred to 4024? really i just need to know where in the original schematic the clock, +v, gnd, and outs came from. if you look at the saj datasheet, its so low res that its impossible to read the pin numbers.
looking at the original schematic, it looks like clock is either 3 or 1 or 7? and the outs are 12 or 11?
any one do this with a 4024 and have luck??
thanks