The ACB will be a small piggyback boardlette which can be mounted on top of the main engine board which, using a PIC will extrapolate it's own sync pulse when the sequencer is running in ASYNC VC mode. Click here to see a schematic of the ACB Click here to see an image of the ACB Costs: $22 before APRIL 6, 2004 - introductionary price (includes ACB PIC, ACB Boardlette, instructions, shipping within contionental US) $30 after APRIL 6, 2004 (includes SCB PIC, ACB Boardlette, instructions, shipping within contionental US) PICs can be purchased separately for $13 paypal ordes only, sent to peter@buzzclick-music.com Operational conventions: The ACB board was designed specificall for the Milton sequencer, but there are a host of other applicatiojns as well not related to MIlton (see below - OTHER APPLICATIONS') The Milton sequencer has two modes of VC operation: Synchronous (to the incoming clock) and Asynchronous. Synchronous VC mode has it's benefits for sure, but by nature of it's design behaves like a sample and hold - it can be set up 'just so' (meaning perfectly) at a certain clock speed, but if your application requires that the clock speed change, then possibly (like all sample and holds), you're now going to be skipping stages you may not want to skip. Firmware operation: Click here for an operational flowchart The voltage present at the VC input is continuously read by an A to D and converted from voltage to a four bit binary word. These words are then fed into a latch, which passes this information as an override to the main counting chip (the 4516), either synchronously or asynchronously to the clock driving the sequencer. Being four bit, there are 16 different numbers available (0000 thru 1111) which the incoming VC can be 'rounded off' to which obviously corrospond to Milton's steps 1 thru 16. 0000 (binary 0) will send the sequencer to stage 1. 1111 (binary 15) will send the sequencer to stage 16 and any number in between will be assigned it's proper stage in either .18 or .315 volt increments, depending on the sensitivity setting of the VC input's STD/KEYBOARD mode switch. When the ACB board's PIC senses a change in state at the binary outputs of the 4042 latch, three things happen: It sends a 2 millisecond low to its output pin which is inverted externally to create the sync pulse (the flowchart lists 50ms, this is in error) -AND- it updates its internal register with this new four bit state for future comparisons. Then it starts at the beginning of the process again and again, and again - continually whether the sequencer is in async operation or not. This whole affair takes 2 PIC cycles (about 4 microseconds). Mounting the ACB: Click here to see a drawing of the ACB boardlette While the final size of the ACB is not yet nailed down, it will be made with three mounting holes to allow for easy mechanical integration onto the main engine board in the corner where the "Cynthia' logo is printed. The corner mounting hole (labeled 'mounting hole 1' in the attachment above) will accept a stackable standoff into the one which currently holds the engine board at that location, and the other two will accept conventional (non stackable) standoff which can be easily hot-glued onto the engine board. The locations of the second two mounting holes have been selected intentionally so not to interfere either mechanically or electriclly to the unit's operation. Connecting the ACB: As far as electrical connections, the ACB installation is about as easy as upgrading the RAM in your computer. It requires a single additonal ribbon cable, a single four wire power cable AND a single wire going to the faceplate. It has two four pin power connectors. Once the ACB is installed, the main power going to Milton will now be connected to the ACB and a second cable will be required to take that power to the Engine board and the rest of the unit. Further to that, the ACB requires four signals from the engine board, taken from the engine board's JM3 jumper block. These jumpers were arranged on the engine board so that a ten pin dual in line ribbon connector can be installed and that layout is also duplicated on the ACB so that a single ribbon connector will all that will be required to make full electrical connection to the engine board. The output of the ACB is routed to the faceplate SYNC/ASYNC switch as shown in the preliminary schematic (see link above). This switch, which before only required a SPST switch will now have to be replaced with a DPST switch to support this secondary signal routing. DEMO THE ACB BOARD The following mp3 sound samples were all created using the identical patch, the only variance being a change in the VC mode to milton. Sample 1 is using SYNC VC mode, sample 2 and 3 using ASYNC mode. The patch consists or two VCOs (somewhat) tuned to unison (detuned to unison?) whose frequency is being controlled by Milton, which in turn is being controlled via an envelope generator in it's VC direction input for samples 1 and 2 and a manually adjusted pot in sample 3. The pitch manipulation is subtle by design here, as I wanted to point out more the variance in triggering between the SYNC and ASYNC modes using the ACB boardlette. The envelope generator shaping these sounds (as opposed to another envelope gen voltage controlling Milton) is being triggered by Milton's Obedient Clock Output. The settings of the envelope generator controlling Milton are the same in samples 1 and 2, the only difference being the sync/async switch setting on the VC input. Sample 1: This is Milton in SYNC VC mode. The ACB board is not switched in at this point. Because of this, the VC acts like a sample and hold, with changes occuring only when a new input clock has been received by the sequencer. Therefor some of the resolution of the control signal is lost between clock phases. Click here Sample 2: This is the identical patch heard in sample 1, the ONLY difference being Milton switched to ASYNC VC mode. The ACB board is now creating the triggers which fire each sound event. Notice the difference of the same VC (EG) into it's VC input, as stage events and triggers are now created the moment that voltage reaches a stage change threshold at Milton's VC input. You can now hear the non-linearity (and expression) of the envelope generator controlling the sequencer as none of the resolution of the VC input is lost. REMEMBER: the settings (attack, decay, etc.) to the EG controlling direction in sample 1 and 2 are identical - the ONLY difference being a switch from sync to async modes on MIlton. Notice all the activity that was lost when the VC is only applied with the next phase of the incoming clock. Click here Sample 3: This is the same setup heard in sample 2, but the enevelope generator controlling MIlton's direction has been replaced by a fixed voltage pot. Stage advancement AND TRIGGERS are now created only by turning that pot. Turning it CCW causes Milton to travel in reverse with the speed of those events tracking the the pot movement, while turning it CW causes it to advance in a forward direction. Notice the change of velocity in stage advancement, caused by turning the pot faster or more slowly. Note the expression - this is what happens when triggers are produced totally by the velocity of change in an event. Click here You may think you could get the same effect by applying the VC envelope to the clock driving the sequencer. Nope! Try it at home and prove me right. VCs applied to clocks are governed by the initial speed of the clock and the phase of each cycle. The ACB doesn't do that - it creates 16 windows which that voltage will force the same outcome - a trigger. OTHER ACB APPLICATIONS Do you own a: Blacet Minwave? Modcan Miniwave?? Wiard Waveform City or Noise Ring??? Then my friends, the ACB board is FOR YOU! The ACB board can easily abapted as a modification to a Blacet or Modcan Miniwave to produce triggers whenever a change of WAVE or BANK are selected (dailed). It will run seamlessly in the background and will output a 2ms pulse each time a change occures. This of course can also be installed into a Wiard Waveform City, but you'll have a bit more of a challenge in that there's not a lot of room for two additional jacks - the obvious place being the sacrifice of the bottom row muilltiple. Please be aware that one ACB will be required for the BANK and another for the WAVE. Also be aware that this addition may void the manufacturer's warranty on the instrument you are modifiying - you MUST contact the manufacturer for more information. I cannot be held reasponsible nor reasonable for damage! The ACB can also be used in any project in which a pulse will need to be extrapulated. Think of it as a logic summer (magnitude comparator) of sorts which will produce a 2ms pulse whenever any of it's four inputs change states. THis is not to be confused with the operation of a four iput or gate which wil simply sum the states of the four inputs. The ACB outputs a pulse, not a contant summed state. ACB PARTS LIST: NOTE A - One of these two .01's are optional and only to be installed if you are experiencing noise problems associated with extraordinarily long cabling from and to the faceplate and ACB board. In most case, this second .01 wil be omitted and WILL cause a slight rounding of the output pulse waveshape. NOTE B - One of these two get installed on the Engine Board at JM3 * - There is a location for an OPTIONAL zener diode at the output stage to limit the amplitude of the output trigger. If omitted , the ACB will produce a 12 volt output trigger . If your system requires only a 6 volt trigger, install a 6 volt zener into this location. If your system requires only an 8 volt trigger, install an eight volt zener, etc. ** - IC sockets are cheap and safe. Always practrice safe IC insertion!
ASYNCHRONOUS CLOCK BOARD (ACB)
There is nothing that can be done about this by virtue of the fact that it's running in sync with the clock. The changes will not take effect until a new clock is received. If your VC only has time to advance one position in that clock, you're good. But if your VC rises four levels during that clock cycle, then you're going to skip four stages of the sequencer during that one clock phase (hertetoafter refered to as "SOL" - sh@t out of luck)
Async doesn't care if a clock has occured. As soon as the incoming CV reaches (or crosses) one of the 16 voltage thresholds in the A to D and change its digital output, that number is IMMEDIATELY fed into the counting engine. A matter of fact AN INCOMING CLOCK IS NOT REQUIRED TO RUN A SEQUENCE WHEN IN ASYNC VC MODE - advances are determined by state changes to the VC input alone The caveat is, if you are running the sequencer to a clock, by virtue of the fact that this is happening asynchronously to that stage changes will be out of phase with that clock. So if you're firing envelopes with the sequencer's pacing clock, these will not be in sync with the stage changes.
This PIC will fix this. It will create it's own pulse which will fire the moment it senses a change in the binary outputs of the A to D and will be routed to the OBEDIENT CLOCK OUT only when the VC is running in async mode. So when you switch from sync to async on the fly, you won't even know it's happened outside of the sequencers change in response to the VC Input.
The ACB PIC scans Milton's A to D four bit output states every 1 microsecond to determine if a change of state has occured. A change of state is defined as any variation in the four binary output of the 4042 latch in Milton's engine board. The way this manifests is somewhat interesting:
PART
QTY
COMMENT
y
y
y
ACB PIC
1
2N3904 NPN
6
10K 1/4 watt resistor
11
1K 1/4 watt resistor
1
100 ohm 1/4 resistor
1
7805 voltage regulator
1
10 uf radial electrolytic cap, 16 volt
3
10 Conductor DIN ribbon cable
1
0.01 uf bypass cap
2
SEE NOTE A (BELOW)
.156 inch 4 pin inline Molex connector
2
POWER I/O
.156 inch 4 pin power cable Molex
1
10 pin dual inline header connector
2
SEE NOTE B (BELOW)
ZENER DIODE (voltage varies)
1
OPTIONAL*
8 PIN DIP SOCKET
1
OPTIONAL**
1