MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Roman 245 / 246 CV pot "dead zone" fix
MUFF WIGGLER Forum Index -> Buchla, EMS & Serge  
Author Roman 245 / 246 CV pot "dead zone" fix
Peake
This is rambling and incomplete but will put you on the right track.

Roman 245 / 246 clone note.

I've seen people discussing a dead zone at the top of the CV pot travel. I have a 245 PCB set and a finished 246 in front of me and can confirm the CV output sections on both are the same...and both share differences from the schematic which produce this issue. Dave Brown as always kicked ass on this inquiry.

There are two ways to approach the solution.

-Change two resistors per each of the four CV out stages and get a reduced dead zone but it still slows down a lot in that top range and a maximum CV out of about 10.44V

-Change two resistors per each of the four CV out stages, also lift one end of two other resistors per CV output stage and bus them together to a +24V source. Produces no dead zone and a maximum CV output of a bit over 12V. The top line on the schematic to which R140 and R141 connect is a +24V rail.

I've sourced a 15V to 24V DC/DC converter which is $13 with enough milliamperage for this project and on its side fits perfectly between the two 246 PCBs. It apparently has an "on" connection requiring positive voltage and I'll have to purchase another to make sure this is correct; I apparently fried one device. TEC3-2415WI


The two resistors which require changing in both cases per CV section are R138 and R143 in the original schematic. I haven't pulled every resistor yet so I can only name two in the section I modded.

245, unconfirmed:
R150, R152, R154, R166 should be 150K per schemo
R151, R153, R155, R167 should be 470K per schemo

For 12 to 14V+ output range, omit on PCB1:
R178, R180, R182, R184

On the 246:

R294, R296, R298, R317 should be 150K per schemo
R295, R297, R299, R318 should be 470K per schemo


With above resistors in place and +24V to appropriate resistors (Schemo R140 and R141)
CV output is from 0.182VDC to 12.38V and no dead zone at top of pot travel

With above resistors in place and +15V to where +24 should be, in other words, how the PCB already is
CV output is from 0.101VDC to 10.66V with the bit of pot turn stretching out a bit

So if 10V output is fine with you swap out the 10Ks where R294 etc. are to 150K
and all 150Ks where R295 etc. are to 470K, just two per each of four CV out stages

On the 245 PCB set the 10K and 150K resistors which must be changed are near the multipin through header. Will update with their Roman BOM identifications when I get to it.

Note that 741J were used in the +24V experiment and might have 44V total rail capability
Regular 741 these days are below that range, it’s been reduced over time to 36V.
Dave says it’s possible to use a small -5V regulator off of the -15V rail to send -5V to pin 4 of each 741
but you’d have to lift the pins from the PCB or IC sockets and bus them together. He pointed out the problem with modern parts rail variation.

http://fluxmonkey.com/historicBuchla/buchlaFiles/Buchla245pg1_daveBrow n.jpg



Bottom of a 246 showing the equivalents of schemo parts R140 and R141 lifted from the pads going to the +15V rail and brutally bus'd together for connection to +24V:
cygmu
Thanks for pulling this together Mike. Wiring in that 24V supply was more trouble than I wanted to go to on my 245 but it is clearly the way to restore proper behaviour.

You mentioned the overall output range. If you want a bit more range, you can remove the 1k/4k99 voltage divider on each output. Just taking out each of the 4k99 resistors will achieve this. On the 245 they are R178, R180, R182, R184. You could also remove the 1ks and replace them with a jumper (R177,R179, R181, R183 on the 245) but this is probably not necessary. Unless you have done the 24V mod, you will still have a bit of dead zone on each pot, but the overall range will be 20% greater. With the 24V mod done this should get you pretty close to a 0-15V range as per the original.
Peake
There's a voltage divider after the CV out sections? Damn, why...must be on the main PCB(s). There definitely aren't any in the schemo.
Peake
I lifted one end on each of the 4K99 resistors and now with +24V on the CV output range is between 0.017V and 14.3V.

With +15V instead of +24V the CV output range is 0.139 to about 14.00 or just under. Top range of the pot feels weird.
cygmu
Peake wrote:
There's a voltage divider after the CV out sections? Damn, why...must be on the main PCB(s). There definitely aren't any in the schemo.


Yes, not on the schematic -- the original design takes the output straight from a transistor's collector.

My hypothesis for how they got there is that, having replaced the 24V line with the 15V line, in testing it was noticed that the CV went up to about 12V, so this divider was added to bring the max down to 10V. Obviously this is pure speculation.

Until yesterday I was pretty cross about the way these boards had been "converted" to avoid the 24V line. The original design biases the output transistor using a voltage divider between 24V and 15V. In Roman's boards this "voltage divider" ends up being between 15V and 15V... which is silly... but it does mean that your mod is a lot easier than it could have been, because you just reattach 24V in the right place and bingo.

Great work by you and Dave.
cygmu
Peake wrote:

With +15V instead of +24V the CV output range is 0.139 to about 14.00 or just under. Top range of the pot feels weird.


This is interesting: you're saying it feels weird rather than being a dead zone? With that 4k99 gone and the other resistor changes made, there is less load on the output transistor which might explain that. Hmmmm interesting.
Peake
With or without the 4K99, if +15V is used throughout instead of +24V the output doesn't seem to change as much in the last part of the throw than the rest, or something. With the +24 in place it seems to change equally throughout the throw.
MUFF WIGGLER Forum Index -> Buchla, EMS & Serge  
Page 1 of 1
Powered by phpBB © phpBB Group