Buchla 221 Kinesthetic Input Port

I was sent a a Buchla Music Easel for calibration and repair that consisted of an early version 208 and this 221 Kinesthetic Input Port.  This module is quite interesting in that the entire module will bend (e.g. shift) side to side as a modulation source.  This module was fully functional so I did not remove it from the cabinet.

 

The bend functionality is implemented through two light dependent resistors mounted on the PCB. The illumination source for the two light dependent resistors is an incandescent bulb fastened to the bottom of the boat. It is soldered in place with a shield over the top.

 

 

I received another vintage 221 that needed repai. It is comprised of three PCBs.

The top pair of PCBs (blue and red keys) are mounted to a rigid frame which bolts to the bottom of the boat. This flex in the side supports is what gives the module the ability to "wiggle". Four ribbon cables connect this pair to the rear PCB.

 

The 221 has both digital and analog outputs. The digital outputs are on the P1 connector in the upper right. The ICs for the digital outputs are not installed on this module (empty sockets). The two round components in the upper left near the edge are the bend CdS sensors.

 

Beneath the ribbon cables are the counters, multiplexers, and passive components for the key sense circuitry.

 

The rear of the PCB has the trimmers and factory modifications along with a few of my modifications.

 

 

Operation

When I created this webpage I said "I wish Don would have done at least a 1 page datasheet explaining the controls and operation. I have to deduce all this from the schematics." Well, Rick from The Buchla Archives sent me this document.

It does a fairly good job of describing the features of the 221 but neglects to say it is setup for 2.4V/Oct. It also says it must be reset on power up and when the sensitivity control is adjusted. I didn't really notice too much about this. It would have been interesting to investigate this more.

221 Notes for Analog Applications

 

 

Controls

uP

The reset switch clears the key detection shift registers.

The Clock control is simply a 0 to 15V CV that goes to the digital port for the 300 series controller.

Key Activated Voltage Sources

Sens controls the sensitivity of the keys. Adjust for proper operation.

Bend is a level control for the amount of "wiggle".

Portamento is a level control for the glide. It can also be controlled by an external CV.

The three outputs are CV, CV+ bend, and CV+bend+portamento.

2D Voltage Source

This is an X-Y joystick control of four CVs. The and ↑are positive going from 0 to 10V while the ← and ↓are negative going 10V to 0V. This circuitry is implemented with 5% resistors so the measured voltage ranges are off a bit with 83 mV to 10.2V, Y -82 mV to 10.1V, X  9.7V to -0.4V, and Y9.9V to -0.4V.

Preset Voltage Source

Each of the four preset controls are 0 to 10V with the output to the jack above. The Preset out jack is selectable using the +0, +50, CTRL, and keys in the upper right to select presets 1 - 4. The LEDs indicate which control is active.

The switch controls what voltage source is added to the CV outputs. Preset adds the selected control, Octave adds in 0V, 2.4V, 4.8V, or 6V as indicated by the LED and None adds neither.

The keys in the red section only operate digitally for control of the 300 system.

 

Jumpers

There are four sets of jumpers on the board.

The A-B jumpers select 63-0 or 63-1 to synchronize the scanning counters.

The C-D jumpers select the digital data format of actions (key closure, key release, pressure).

The E-F jumpers select Keyboard-1 or Keyboard-2 for operating two of these in a 300 system.

The G-H jumpers select operation of the preset keys. They are either always enabled, or enabled on command (I don't know what this means. This one is always enabled.)

 

Trimmers

There are four trimmers on the board. On this module they are all mounted on the rear.

V/Oct

+ Octave

Keyboard signal sensitivity. This is set by monitoring the voltage on IC37 pin 12 and setting it between 0.5V and 2V.  I set it for reliable operation with the Sens control at about 2 o'clock.

Bend neutral offset (of course it is on the bottom so I don't know how you really adjust it in a boat other than trial and error).

 

The first thing I spotted was this module was built with 1970's Texas Instruments IC sockets which have a known reliability problem. They wipe on the edges of the pins, not the front and rear where the edge dimension varies and there is less surface contact area. The contact is also not gas tight, leading to corrosion. You can replace these sockets by carefully prying the plastic body off and removing the plastic sheet so the pins can be unsoldered individually. One socket had already been replaced.

The pulse output is a 13V/5V trigger/gate. The gate and trigger are or'd together so you can see the slight leading step where gate becomes true slightly before trigger.

 

This scope image shows the Pulse,  Pressure (cyan) and CV (magenta) outputs.

 

This image shows the delay from pulse to CV stable at about 100 µS. The leading gate step is about 40 µS.

 

This image shows the minimum portamento ~8 mS. Note that the CV isn't stable until about 6 mS after the pulse.

 

This image shows the maximum portamento ~ 3 seconds.

 

This image is showing multiple key presses while holding a key which generates multiple triggers while holding the gate true.

 

This image shows selecting presets 1-4 set to octave.

 

This image shows the ~80 mS delay from key press to the pulse output due to the filtering.

 

This image shows a keyboard slide from high to low rolling from pad to pad to not drop the gate.

 

The bend function is a bit wonky. First, it is hard to set the offset because of incidental light. In a boat this is probably best adjusted outside of the boat in a dark room. The CdS amplifier has gain of 4.3X which is then reduced to 0.05X when added to the CV. This unit has been modified with a LED and it easily drove the amplifier to the rails. I reduced the gain to unity and decreased the LED current a bit which gives a nice +/- 2 semitone range.

 

 

Modifications

I removed a number of modifications to bring this module back to original condition, including replacing some of the non-standard form factor parts. I modified it for 1.2V/Oct by paralleling R173 on the rear with a 4K99 1% resistor. This drops the output gain by 50%.

There was also ~20 mV of offset in the output which I corrected to <1 mV on all three CV outputs by adding a 2M2 resistor on IC56 between pins 6 and 8 (+15V). There really wasn't a good way to add a trimmer and have it be mechanically secure.

I added a test point on the rear of IC37 pin 12 to be able to trim the sensitivity with the module reassembled. It was easier to set this trimmer by proper operation of the key pads.

I cut the trace on the rear between R165 and R166 to decrease the gain of the bend amplifier (I didn't want to disassemble the module once more to minimize any potential damage).

It's nice to see a ~50 year old module come back to life.

 

 

back