Tie a line from the juncture of D35 and R133 to a single pole switch, and the other end to 12V. When the switch is on, the gate will be on forever, but the sound will still die when the main envelope closes the VCA. A good place to get 12V is pin 3 of J4 (pin 1 is 5.333v, pin 2 GND and pin 3 12v)
Tie a line from the juncture of D36 and R134 to a single pole switch, and the other end to 12V. In conjunction with the Gate On Forever mod (see above), this will allow the sound to play forever.
In situations where you are using the box with the Gate On mod above either with the internal sound or and external audio input you may want to re-trigger the envelope generator (the main EG - the one for the filter/accent/sweep). You would do this by hooking up a switch connected to 12v and the junction of D37/R152.
R123 sets the vca decay. The stock value is 1.5M and gives a vca decay time of about 3 seconds. Smaller value will give shorter decay, larger value will give longer decay. Generally you would put a 1M pot there in series with a resistor for vca decay control - if you find a 1.5M or 2M pot somewhere that's even better.
Verified: I have done this on my TB-303 and my clone. Exactly what is decribe below.
Another DevilFish mod. Wire up a 1Meg Audio (log) Pot with one end to Pin 4 of IC12, the wiper and the other end to the anode (+) D28. You will have to cut the existing trace between Pin4 of IC12 and D28. If you get a pot just like the decay pot, it will be a 1Meg Audio Dual Pot. In that case, the second set of connections should just be grounded.
R46 and R119 control the depth of the accent. If you wanted a deeper accent, you could lower their values or replace them with jumpers. R119 controls the amplitude gain of the accent, and R46 controls the resonance gain.
If you have a modded TR606 with a trigger in, and you want it to work with the x0xb0xs Gate out, remove D34 and replace it with a jumper wire.
Thanks to RobinWhittle of the DevilFish for posting these mods on his site.
Remove R61 and replace it with a jumper wire. This will make it possible to turn the envelope modulation all the way down. If you want to make it switchable, leave R61 in place and solder a wire from each terminal of R61 to a switch. This will enable you to shorten out R61, much like replacing it with a jumper wire.
Add a 100K resistor in parallel with R63 (220K). Easy. [This basically reduces the resistance to ~68.75K, about 1/3 the original value of R63.]
You may desire to increase the range to only 2x instead of 3x to keep finer control over the more 'useful' end of the spectrum. This can be achieved by adding a 470K ohm resistor in parallel with R63 to achieve ~150K resistance, about 2/3 the original value of R63.
Wire the tip connection of an input jack to a 22k resistor; connect the other end of the 22k resistor to the base of Q37. In this way the internal sequencer and the gate input signal can both control the envelope gating.
Alternate method: unsolder the lower leg (CPU side) of R146. In its place, connect the anode of a 1n4148 diode to the trace leading to the CPU. Connect the cathode of the diode to the lower leg of R146. Connect the cathode of a second diode to the diode-R146 junction you just created, and wire the anode of the second 1n4148 diode to the tip connection of your gate input jack. As with other CV input mods, CvProtection (q.v.) is a good idea…
Both of these might do well with CvProtection
This is the easiest way to handle Accent CV In. Wire the tip of a jack to a 100K resistor, and then wire that to the base of Q35.
Take a PNP transistor, and attach the emitter to the +5 supply. Put a 100K Resistor between the base and the 5V supply, and another 100K resistor between the base and your CV In jack. Finally put a 100K resisitor from the collector to the base of q35.
The DevilFish combines this switch with another function to allow the filter to self oscillate, but there is no reason to limit yourself to this layout. Making the accent sweep on/off go should be as simple as wiring a switch between D24 and R46. The state of the filter would be handled by a VcfMod
The self-described accent sweep circuit in the devilfish 303 it not as simple as what I described below. There are a few things going on. First, there is a buffer circuit inserted in line between D24(+) and the junction of D24(+)/VR7-PIN2/R119 (however you can do it without the need for the opamp or any buffer). Also, depending on the mode, fast, normal, or slow there is a cap or cap+resistor placed in line with C13. Simply changing C13 does work but there is an inherent change in the overall amount of the envelope sweep signal if this change is not compensated for. For example, with the stock setup with accent all the way up and resonance all the way up and C13 at the stock value of 1uF the accent sweep may peak at 6v - giving a big change in cutoff frequency If C13 is increased to make the sweep speed slower the sweep may peak now at only 3v, for example, if level change is not also compensated for.
The DevilFish has 3 modes: Fast, Normal and Slow. The easiest way to understand this is to picture the envelopes according to each section. THe one extreme "Fast" would be a very small and tight envelope, with a fast attack, and fast decay. The other extreme would have a very slow attack and slow decay.
Perhaps by using a switch connected to different values of capacitors, this could be handled.
See TheAccent and EnvelopeGenerator for more information.
I just did this as one of the mods to my TB303. I tried using an ON-ON-ON DPDT switch such as the C&K T211-series, Philmore 30-090, & Taiway DP-6. Use one pole to add a cap in series with C13 and the other pole to add one in parallel with C13. Hook it up so that in one state the series cap is active and in the second state the parallel cap is active. In the middle state both will be active, so you should pick capacitor values such that Cs x Cp - Cp = C13 = 1uF so that when the toggle switch is in the middle position the value will always break down to 1uF, the original value of C13. Remember that the accent sweep circuit becomes more prominent as the Resonance pot is increased since the cap is hanging off pin3 of the pot (this 'is' the resonance pot).
|from VR4 Pin 3 (this is pin 5 on the physical part - the resonance pot) | | |5 / 6/ --- |4 | | | | | ------- new series cap (2uf) | ------- | | | | ------|----------- | | | | | |2 | \ | \1 | 3| | | | | -------c13 ------- new parallel cap (1uF) -------1uF ------- | | | | | | |----------- | | |to TB303 ground
NOTE: the solution above is quite elegant and has the lowest parts count BUT you save exactly ONE capacitor. Since caps are cheap I suggest to use 3 caps - this way you can try different cap values until you get a combination that you like. I used a 0.47uF cap for FAST, the stock 1uF for NORMAL and a 100uf (!!) for SLOW. The somewhat high value for the SLOW cap makes for a quite interesting effect: the cap gets charged by consecutive "accent bursts" and the accent goes higher and higher and higher with each new hit over a long timeframe (it takes ~10 seconds to reach its peak!!). Nice if you also increased the Accent range!
|Schematic||Power Supply||Voltage Controlled Oscillator||Voltage Controlled Filter||Envelope Generator||Voltage Controlled Amplifier||Headphone And Mixer||Digital Sequencer||Midi, USB and Sync|
|Fabrication||Building the Ps||Building the Vco||Building the Vcf||Building the Envelope||Building the Vca||Building the Headphone And Mixer||Building the Sequencer||Building the Midi and Sync||Finishing It Off|
|Testing||Testing the Ps||Testing the Vco||Testing the Vcf||Testing the Envelope||Testing the Vca||Testing the Headphone And Mixer||Testing the Sequencer||Testing the Midi and Sync|
|Mods||Ps Mods||Vco Mods||Vcf Mods||Envelope Mods||Vca Mods||Headphone And Mixer Mods||Sequencer Mods||Midi And Sync Mods||Finishing Mods|