Synthbox

[solderman]

Hi all
First, TNX:s for sharing this Liquids. I fell in love with the sound and had to test it.

Her is a bread board template if any one wants to test before doing it on PCB or vero etc.
Its kind of tight but verified (at least it worked for me).

I did not, How ever, get any where near that sound in the post earlier. It migt be that I did not have the specified transistors. I used 2N3904 for Q1 and Q 6. The rest is BC547C they are 470 hfe. Maybe to low. I matched Q4 and Q5. (469 and 470)
Any one have tried some other ones with a good result?


http://solderman.fatabur.se/Future%20project/Synthbox-bom.txt

[liquids]

I'm surprised that people are having such trouble.  Except that I find wiring the shocktave was difficult to wrap my mind around at first.

But if you've done a shocktave than this should be intuitive - if not, than you might want to build a shocktave first, minus the clean blend.  I'll repeat that the schematic I drew is better when referenced with the clearer layout of the shocktave.   The shocktave uses all 2n5089s and those should be fine, as should 2N5088s just to get it fired up.

If/once you can get a shocktave to work, from there, just swap some components with the values I've changed.

And from there, all you are doing is adding some filtering and two buffers....think building blocks.  All it is is a shocktave with some optimized tweaks, if you look at it critically.   I'm bummed that such a small percentage of people have had success.       

[~arph]

I have breadboarded this three times, but not build one yet. Somehow I keep coming back to it.
I tried driving the frequency to voltage converter with an LM386 and plain opamp boosts. Still the transistor version tracks the best.

I found some great sounding mods that give you octave up and down as well. I'm sure I'll do a vero for this and one for the mods I have. because as I said..
this thing sounds so cool I keep getting back to it. I have to build one..

[gigimarga]

My Synthbox it's still dead, but I hope I will have some time next week to try to tweak it

[bluesdevil]

My Synthbox it's still dead, but I hope I will have some time next week to try to tweak it

Might be a good idea to ask the the designer of the pcb (Naz Nomad) if it's verified and working. I was REALLY anxious to get one built when the artwork was posted, but got too distracted with other projects to build it myself.
    Regardless,this one looks best to tweak on the breadboard before committing to pcb/vero it seems.

[solderman]

I'm surprised that people are having such trouble.  Except that I find wiring the shocktave was difficult to wrap my mind around at first.

But if you've done a shocktave than this should be intuitive - if not, than you might want to build a shocktave first, minus the clean blend.  I'll repeat that the schematic I drew is better when referenced with the clearer layout of the shocktave.   The shocktave uses all 2n5089s and those should be fine, as should 2N5088s just to get it fired up.

If/once you can get a shocktave to work, from there, just swap some components with the values I've changed.

And from there, all you are doing is adding some filtering and two buffers....think building blocks.  All it is is a shocktave with some optimized tweaks, if you look at it critically.   I'm bummed that such a small percentage of people have had success.        

Hi
I did not find this one hard to build. More like standard. Many breadboard errors have made me a wiser man so now I always do a bread boar template before the bread boar sees some action when it comes to +10 components. This way I don't need to trouble shoot my sloppy "from the schematics builds" as I did before.
Sound vise. I thing the Q1 and Q6 can be any low noise transistor. Q2 and Q4 probably needs 7-900 hfe  (2N5089 is 1200 max) and mine was 470. The sound I got from them was quite OK between the to choppy and octave jumping sound and the tacking was too jumpy and the decay sounded to buuzzz. I'll gonna change Q2-Q3 to 2N5089 as soon as I find some( or equal in gain) and test with different gain for Q4-Q5 as long as they at matched. I'll get back with the results.
So in the end I confident that Liquids design sounds good and I'm gonna get there as well.
Cheers

[~arph]

Allright here is an idea I have..

In order to have better tracking we filter out frequencies above ~2200Hz which also removes the harmonics for frequencies above 1100Hz.
If we filter at a lower frequency we affect the playable range of the circuit as we will be cutting frequencies of playable notes. (1100 Hz ~C# 6th oct. 21st fret on high e string)

The tracking problems mostly occur in the low frequency range. Say low E to low A which is 80 - 110 Hz. tracking above that seems pretty stable. Now what if we also apply a not so deep notch filter at the 1st harmonic of the center of the problem frequencies. In this case at 190 Hz (=2x 95 Hz). Will this give us better tracking? or will it become problematic around the notch frequency?

Also are higher order harmonics going to be a problem? ( I assume that the first order harmonic usually has the largest amplitude)
In that case we can add another notch filter at 380 Hz.


Does anyone see any errors in my assumptions?

[Naz Nomad]

My Synthbox it's still dead, but I hope I will have some time next week to try to tweak it

Might be a good idea to ask the the designer of the pcb (Naz Nomad) if it's verified and working. I was REALLY anxious to get one built when the artwork was posted, but got too distracted with other projects to build it myself.
    Regardless,this one looks best to tweak on the breadboard before committing to pcb/vero it seems.

Not verified but gigimarga built it using different trannies and had problems ... I haven't tried it myself.

[gigimarga]

Very good news: after 2-3 hours of debugging I found the best solder joint that I ever made (exasperated, I've begun to measure the voltages and I saw that the emitter and the base of one of the transistors of the oscillator are at 0V...after 15 minutes I found the solder joint)!!!
So, now it's working fabulous!!!

Instead of 2N5089 I used BC549C that had the gain over 600 and instead of MPSA18 I used two BC549C with gains around 700.

Thx a lot liquids for the schematic and Naz Nomad for the PCB!!!

I have only two observations to make:

1. it's not very clear that lug 1 of the Shape pot is not connected
2. the 10nF cap from the Bass switch seems too low: the sound is thin and lifeless...I will try higher values instead

THX A LOT ALL OF YOU AGAIN!

[~arph]

Good news!

I hope you like it. Ho wis the tracking with those transistors?

[gigimarga]

Good news!

I hope you like it. Ho wis the tracking with those transistors?

Seems to be good to very good, but I must to tweak the trimpot more.

[Naz Nomad]

Sweeet ... glad you got it sorted.

I always try to quadruple-check my layouts but you can never be reeeealllly sure until someone builds a working version.  


... just need someone to cobble together a parts list and we got ourselves a project. 

[liquids]

Very good news: after 2-3 hours of debugging I found the best solder joint that I ever made (exasperated, I've begun to measure the voltages and I saw that the emitter and the base of one of the transistors of the oscillator are at 0V...after 15 minutes I found the solder joint)!!!
So, now it's working fabulous!!!

Instead of 2N5089 I used BC549C that had the gain over 600 and instead of MPSA18 I used two BC549C with gains around 700.

This is fantastic news.  And congrats for completing what is...well, seeming to be a difficult build for some.    Your report is great in that it's excellent for others to know that a variety of transistors will indeed work just fine--so long as the gains are not low and inconsistent, as I've hopefully made clear in the past.  And that tracking can be considered 'good.'  

I have only two observations to make:

1. it's not very clear that lug 1 of the Shape pot is not connected
2. the 10nF cap from the Bass switch seems too low: the sound is thin and lifeless...I will try higher values instead

I truly appreciate your input and observations.    

Re: 1) I looked back, and looking at the schem, I agree about your observation.  I wonder if this has confused others.  I will change the schematic so the shape pot looks more like the 'tracking' pot.  Even better would be someone who is good with schematic drawing programs to re-draw it to make it clearer!  Please?  Anyone?  

Re: 2) Interesting assessment.  Are you certain you used .01uF/103 and not 102 (.001uF)?  

The 10n should definitely sound tighter, not thin by any means, but it may depend on your amp.  I do have a pretty bass-heavy amp, so to those who don't, a 22nF might be a better option, if one is needed at all.  

But if you'd double check your values and let me know what amp (and speaker config) you are using, I'd appreciate it.  If this is a common complaint, I'll mod the schematic again for others to have a better starting point.  

The 1uF sounds good on it's own and has no calculated roll off (1.6hz), but could easily be subbed for 100n probably, even, without audible change (16hz).  Now, I thought I tried the 10n and liked it...and the R-C calculator I used estimates the the roll off point for the 10nf and the 100k should be about 159hz...but my memory of all this is getting dim.

The switch is because I felt an option for something with a little trimming of the bass would be good 'in the mix' live or otherwise, but the cap value should be 'to taste.'  If indeed 10nF is too think, 22nF should do. If not, than maybe just eliminate the switch completely, as, on paper the differences shouldn't be that significant audible between 47n and 100n...which is why I want to confirm you are using 10n/103.  

Anyhow, enjoy experimenting!  Now that you have it working, mod away    Last note - many people have commented on finding the tracking pot subtle, or having no audible effect.  This is, to me, in some ways a good thing (a compliment to the EHX-based filter), as it means that the other stages probably do well enough on their own that the minor tweak of high frequencies at the input is a minimal tweak...but if you find it to make a difference in tracking, especially with a bridge pickup, please let me know.  

[gigimarga]

Re: 1) I looked back, and looking at the schem, I agree about your observation.  I wonder if this has confused others.  I will change the schematic so the shape pot looks more like the 'tracking' pot.  Even better would be someone who is good with schematic drawing programs to re-draw it to make it clearer!  Please?  Anyone? 

When I've finished it I wired the lug 1 to the ground ...but it was easy to find the mistake because it had acted like a second volume pot

Re: 2) Interesting assessment.  Are you certain you used .01uF/103 and not 102 (.001uF)? 

I am very sure that it's a 10nF because it's very easy: I use only some MKT yellow capacitors on which is written very clear his value
I forgot to mention that it's a big volume drop when I switch to 10nF...I think I must to test the switch, too.

But if you'd double check your values and let me know what amp (and speaker config) you are using, I'd appreciate it.  If this is a common complaint, I'll mod the schematic again for others to have a better starting point. 

My amp it's a 20W full-tube romanian old amp which (with a no-name speaker) was modified by a friend of mine to be more fenderish (he modified the tonestack adding a mid pot and a bright switch). It has only one clean channel, but I am very happy with it. I play only at home, for fun, so it's enough. Anyway, it has enough bass

Anyhow, enjoy experimenting!  Now that you have it working, mod away    Last note - many people have commented on finding the tracking pot subtle, or having no audible effect.  This is, to me, in some ways a good thing (a compliment to the EHX-based filter), as it means that the other stages probably do well enough on their own that the minor tweak of high frequencies at the input is a minimal tweak...but if you find it to make a difference in tracking, especially with a bridge pickup, please let me know. 

Yes, the tracking pot has no audible effect because it tweak the tracking. I have no time to tweak it seriously, but when I rotated it, I heard a better tracking (but not perfect, yet).

Thx a lot!

[liquids]

I am very sure that it's a 10nF because it's very easy: I use only some MKT yellow capacitors on which is written very clear his value
I forgot to mention that it's a big volume drop when I switch to 10nF...I think I must to test the switch, too.

Fair enough.  You will loose 'volume' with a bass roll off of any kind, so that is not unusual, the switch is probably fine. 

One thing that you or others might want to do is get to know the R-C cutoff and frequency of what you experiment with: http://www.muzique.com/schem/filter.htm

There's lots of possibilities.  Those caps work in conjunction with the 100k resistor for  high pass filtering.  The 100k may be too small to some anyhow, so I'm considering increasing it on the schematic once I get back to it (someday I will hopefully soon...  ), in conjunction with tweaking cap values, especially since 1uF caps are typically electrolytic or physically large, so using a 100n might be a better option.

You can see that making the 100k resistor ~200k will do the same thing as swapping the 10nf with a 22nF cap.  The side affect would be that the 1uF would likewise be the equivalent of modding it to a 2uF.  So off the top of my head, leaving the 10n in there, but making the 100k to 220k, and 1uF to 100n might be a better starting point.  Then you can mod the resistor (increase) the value for your preferred frequency response rather than mod the caps.  So long as you only make the resistor larger than the current 100k (for more bass) the 100n cap side of the switch will keep it's humongous bass response and have the same response into the sub-audio range.

At least, I think that's how it works...   

[gigimarga]

I have almost no theoretically knowledges in electronics, but I am curious if it possible to obtain something like an octave up from Synthbox (or Shocktave)?
More clearly, it's possible to change the caps in the oscillator to do what I want?

[liquids]

Read the full thread...this has been discussed.   

[gigimarga]

Read the full thread...this has been discussed.   

OK...thx!

[gigimarga]

I've read all the thread, but I didn't found the answer...

[liquids]

I've read all the thread, but I didn't found the answer...

PM'ed.