cutoff/rez CV in Mods, Gate in to x0x?

Discuss mods, hacks, tweaks, etc.

Moderators: altitude, adafruit_support_bill, adafruit, phono, hamburgers

Please be positive and constructive with your questions and comments.
User avatar
controlvoltage
 
Posts: 242
Joined: Tue Aug 09, 2005 12:17 am

Post by controlvoltage »

Everything is working beautifully now

I applied the inverted output coming from the collector of that transistor
to the bottom of R141 (I think it was)
I used the ring connection of the accent input jack as a switch
which disconnects the CPU's accent signal from this point on insertion of an input plug.
This works nicely, although it does mean that if you have CV input plugged in for accent, then the internal sequencer is defeated entirely as far as accents are concerned.

There is one remaining piece of troubleshooting to do:
there is an irritating "click" at the start of each envelope
is this normal for the circuit and my headphones are just bringing it out?
Or is it likely to be a fault somewhere that I could fix.

Otherwise this is an acid dream machine... I am currently running it off my MS10's CV patch panel and I'm in heaven. :)
Last edited by controlvoltage on Mon Nov 28, 2005 5:54 pm, edited 2 times in total.

User avatar
controlvoltage
 
Posts: 242
Joined: Tue Aug 09, 2005 12:17 am

Post by controlvoltage »

here is a picture of the envelope clicking... all the signal has been more or less filtered out with the VCF - I love how low its range extends with this cutoff CV-input mod.

Image

This doesn't look right to me. Perhaps a symptom of my envelope input mods? Where can I adjust DC-offset in the evelope, or what else might be happening here?

Thanks...

User avatar
subatomic
 
Posts: 523
Joined: Wed May 11, 2005 3:12 pm

Post by subatomic »

control, so could you give us a rundown of all the mods you did, and where they are (wiki, mainpage, etc.), and what changes you had to do to them? If changes, we should get them on the wiki, will you be doing that?

thanks. :) really cool work by the way.

Jonnay
 
Posts: 142
Joined: Thu Jun 02, 2005 12:28 pm

Post by Jonnay »

heh, I've been posting to the wiki as he goes, and anything I missed, he filled in.

User avatar
controlvoltage
 
Posts: 242
Joined: Tue Aug 09, 2005 12:17 am

Post by controlvoltage »

no problemo; this might help with troubleshooting too (if the mods are the culprit). I used a lot of diodes, based more on intuition than science and so perhaps they are not entirely logical... or may even be potentially problematic! If so please let me know. all diodes are 1n4148 variety.

in total I added five CV inputs as follows:

Slide - gate input; tip connection of jack is wired to the bottom leg of (I believe) R59. The bottom leg of the resistor was lifted from the PCB and a diode soldered in its place with the anode side soldered into the board and the cathode (stripe) connected to R59. The CV connection to R59 is also via a diode with the cathode to R59. The purpose of the diodes is to keep the cpu from shorting to ground via the CV input, and vice versa.

Gate - env. gate input; tip connected via diode to bottom leg of R146 and a diode between R146 and the cpu.

Pitch - Pitch CV input per the "Mods" page, using a switching jack.
Wire from the switched side of the tip connection to the top leg of R89 (junction of resistors); other (contact) side of the jack tip connected to IC11 pin 5.

Cutoff - CV input; tip connection of jack wired via a 100K resistor to the right leg of R71, which is also the base (or is it the collector?) of Q10.

Accent - This was the troublesome one, because the Accent is inactive when its control signal is high, and it is switched low to activate (opposite to the other gate signals I used). This is how I did it, with the help of a kind Guest...
The tip connection of the jack is wired to a small bit of perfboard with the following circuit:
2n3906 transistor, emitter connected to x0xb0x +5V digital supply (taken from J3-3).
100K resistor between base and +5V, another 100K resistor between base and the CV gate input.
connect collector to 100k resistor to a diode with cathode connected to the bottom leg of R141. The bottom leg of R141 was unsoldered from the PCB and connected via another diode (cathode towards R141) to the ring connection of the accent gate input jack. The other (switched) contact was wired to a diode soldered in place of the bottom leg of R141, anode on the side soldered into the board. This way with nothing in the jack the CPU can send accent signals via its original path (more or less) but when a jack is inserted, the connection to the CPU is broken. Perhaps they could have shared control in an easier fashion, but this seems to work.



So, there you have it.... in case I hadn't already gone overboard with the diodes there are also diodes intended to buffer each cv input... they are between ground and the tip connection on each jack, anode facing ground.

All these mods seem to have their intended effect... should any changed be made from an electronics standpoint, or does anything strike you as likely to cause the envelope attack offset problem that I am seeing?

Thanks all--

feel free to wiki-fy any of this, although let's see if changes need to be made first, perhaps... [edit: didn't see your post, Jonnay-- thanks man, you've been on top of this one fo' sho!]

Amos

User avatar
etaoin
 
Posts: 66
Joined: Thu Aug 25, 2005 6:08 am

Post by etaoin »

So, there you have it.... in case I hadn't already gone overboard with the diodes there are also diodes intended to buffer each cv input... they are between ground and the tip connection on each jack, anode facing ground.
You might want to use zeners there, to not only protect them from negative voltages but also from positive voltages that are too high. Have a look at these:

http://www.musicfromouterspace.com/anal ... CTCVINPUTS

Locked
Please be positive and constructive with your questions and comments.

Return to “x0xm0dz”