Cv inputs: how to protect them from negative or too high voltages

i always meant to add cv intputs to one or two of my shruthis for better interfacing with eurorack modular stuff. from earlier threads i have learned that they have to be protected against negative voltages and against voltages higher than 5v.

could somebody explain how exactly this is done? what parts do i need, how do i wire them?

Use Schottky diodes in this configuration (but with 5V instead of 3.3V):

Use Schottky diodes, it’s important! Normal diodes have a 0.7V drop, so they will clamp between -0.7V and +5.7V, and this is already too much for the MCU. Check in the diode datasheet, that they have a low voltage drop, below 0.3V.

To answer your question about the inputs on the side of my Shruthi… Yes they are CV Ins but they are for a “modular” controller I’m building. 4 pots / 4 sliders / pitch and mod wheel… Each with their own patch point. I’ll post in action when I complete it.

oy vey, could anybody please try and translate pichenette’s undoubtedly very helpful hints into some easy-to-follow steps for the electronically illiterate?

for example, about solution1, what makes an mcp6004 so ‘yucky’? (contrary to that other thing, the tlv274, i did find a source for the mcp6004)

and in solution2, where in that schematics is my shruthi voltage input, where is that mysterious schottky diode? am i right to ignore all the stuff on the left labeled ‘measurement equipment’, and focus on what’s happening inside the ‘device under test ic’ part of things? do i simply take two of those schottky things and wire them to the voltage input on one side, and to gnd and +5v respectively on the other? or am i totally misreading this?

which solution is more recommendable? i think i don’t need 4 cv inputs - 2 should be enough. in that case i guess the shottky thing is the simpler one?

ah, thanks for explaining, i think i’m beginning to see. guess i’ll try the schottky method first - i like the sound of that name. :slight_smile: plus it seems slighty more comprehendible than that ‘rail op-amp’ thing (although that one’s name reminds me of a railgun, which is also kind of cool in its own way).
plus i can first build the schottky circuit, connect it to my modular voltage sources, and measure the resulting output voltages before actually using it in my shruthi, so that should be safe as well.

Will this one do?

@fcd72: i don’t know - will it? the data sheets look very similar to me - but then what do i know about these things? (hint: not a lot)
that tlc274 one seems much more easily avialble than the tlv274, whatever the difference may be. even my local electronics store has them - but at 5x the reichelt price.

@whoever knows about these kinds of things: about those schottky diodes - while my local electronics store has no tlv274 or mcp6004, they do have tons of different schottky diodes on offer. what am i looking for here? would this one be a good choice, for instance?

TS 274 will not do. The op-amp needs both rail to rail input (can accept voltages close to GND / +5V). The MCP6004 is on Reichelt.

The other diode you linked to is not good. Check the forward current vs forward voltage curve. It shows that at room temperature, you have at best a 0.6V forward voltage ; so the circuit given above will clip at -0.6 V … 5.6 V, and this is too large.

I’ve just checked the datasheet of the first Schottky I found on Farnell (SD103) and the curve shows that for small currents, the forward voltage is in the 0.3V range. Much better!

ok so what i’m look for is a schottky diode with a ‘forward voltage’ <0,3v for a ‘forward current’ =1ma at room temperature, right?
then i guess this one should work (data sheet).
(the price is higher that farnell part, but farnell don’t sell anything to german end customers, only to businesses. and their german distributor has a pretty high minimum order - that’s why i keep looking at conrad’s - they are right around the corner).

Looks much better!

excellent, thanks! so i’ll get a couple of those after work and see on my breadboard where they get me with my modular voltages…

btw in adition to 2 (or maybe 3) cv inputs for the digiboard i’m also thinking of adding two purely analog cv inputs for filter cutoff and resonance. what are my cv in points here? do i go straight to pin 2 (q control) and pin 13 (f control) on the ssm, or are there better places to apply external cv for q and f?

do i have to protect these inputs too against high or low voltages? what are my margins in this case? should i add some resistors?

my input voltages will be typically between -5v and 5v, min/max -10v/10v…

SSM2044: For cutoff, through a 33k resistor to pin 9 of IC6. For resonance, try getting into pin 2 of the SSM2044 through a 6.8k resistor.


i have the same question as mic.w about adding 2 direct cv inputs to the smr-4 filter.
Anyone know the cutoff and resonance in points and resistor values?


For cutoff: look at J6 – there’s already a pad for a wire and a resistor mounted vertically (two inputs actually!). 32k will give you the Shruthi-1 standard 0.46 V/Oct tracking. Experiment if you want true V/Oct :smiley:
For resonance: 33k resistor at pin 2 of IC4.


I just noticed on the schematic there is a resistor on both pin 1and 2 of J6
if only adding 1 input what is the purpose of the wire?
again sorry about the silly question

and I assume one of the protection circuits from above would be needed for a cv in going direct to cutoff as well?

i don’t think so. afaik those diodes are only needed to protect the cv ins of the digital control board. (still haven’t got around to actually building this…)

No you don’t need CV protection here, it’s not fed into an ADC.

Near J6 you see 2 rows of 3 pins:

First row:

  • First input pad (F>)
  • First pin of resistor for first input
  • Second pin of resistor for first input

Second row:

  • Second input pad (F>)
  • First pin of resistor for second input
  • Second pin of resistor for second input