NSL-32 @ 0v - 6Vdc

[petemoore]

   NSL-32
  http://www.alliedelec.com/Images/Products/Datasheets/BM/SILONEX_INC/Silonex-Inc_Actives-and-Passives_6995014.pdf
  Scratch all that stuff below...what value current limiting resistor would you put on an NSL-32's LED, @ 5vdc to get near minimum resistance with no overcurrents? My Pitch CV seems to barely reach 4.51vdc max.
  Much typing below to try to determine how to handle numbers and decimal points again, not really worth reading if you know how to I guess...OT.
  6v supply [though I think 5v is on the safe side, judgeing from what the DMM is seeing].
  Foreward voltage is 2v [16ma.]
  Desired LED current I guess at 20ma. [max. is 40] ?
  So..in the click to calculate boxes of
  http://www.theledlight.com/resistancecalculator.html
  Supply is 6v, - 2v [1 LED diode drop]
  =4v
  4 / 20 = .2
  And the LED online calculator indicates 200ohm, when I typed '6', '2' and '20' in the three boxes.
  This is before I've figured out how not to mix up a decimal point placement using "ma" "Vdc's" 'n as ingredients to derive a value point for R...
  somehow the answer feels like it should ballpark around 2k instead of 200ohms, 4k seems fairly bright at 9v.
    It turns out the resistance rises when hand nears antenna...another re-think there..
  I used a 10k pot to try to get some resistance movement at Pitch CV output in response to hand/antenna proximity and got down to 5k resistance, and way past 2meg DMM limit.
  Deep end of the pool now contains other unknowns...
  which effect:
  Would work the right way around, and have a good, usable,..sweep or speed or other control..and would be fairly easily implimented...
  What size resistor should I fix on the NSL-32's LED so that 5v {I guess, can't seem to get more than 4.5v supply for more than a millisecond] just barely doesn't kill it?
  LED says 40ma max, voltage is less than 5v as far as I can figure...
  5v - 2v = 3v
  3v / 40ma = .75
  I think the rest of the equation is probably ok, the decimal point...I guess it'd make the resistor something like about maybe 750 ohms or whatever, but that's just because I'm afraid to say 75ohms, which is 1/2 of what the calculator spit out.

[alanlan]

3V / 20mA = 150 ohms... This seems too simple an answer - I guess you're asking something a bit more complicated?

[R.G.]

Alan is right - I think you're asking the wrong question.

If what you're trying to do is to use a 0-4.5V control voltage to run an LED/LDR, you have some up-front problems.

First and biggest, you waste 2V of your control voltage potential range because the LED won't respond at all until you get to that.

Second is figuring out a resistor for limiting; that one is easy once you solve the first one.

Third is a more subtle one. Presumably since you are using an LDR for an output, what you really want is control voltage => resistance, and either linear or log taper for the changing resistance, right.

(Do I sound like I've encountered this problem before?  )

What you want is some way to make the first 2V of no-response on the LED go away. There are a couple of ways. One is to use another similar LED to do an offset. Then you're left with figuring out how to match the two LEDs. Ugh.

The simplest good way is to put the LED in the feedback path of an opamp. If you use an inverting opamp configuration with a resistor to the (-) input and the LED on the output and the + input grounded, then a + input on the end of the input resistor piles up charge on the - input pin. The opamp action is to make the output go negative and suck current through the LED away from the - input so that there is a net 0V on the - input with respect to the + input. This is linear as long as the input resistor is linear, ma for ma.

So if you had an opamp with + and - supplies, grounded + input, resistor to - input and LED cathode to output, anode to - input, then the current through the LED for a changing control voltage is I(LED) = Cv/Rin and for a 20ma current at 4.51V of Cv, then the resistor is R = 4.51/0.02= 225.5 ohms.

Yes, it's a PITA to have to make +/- supplies. So you offset your control voltage by using another opamp to make it move upwards by enough in a single supply to make this work.

There are more schemes for a voltage to current converter with adjustable offsets and scaling. This is just one.

The third question is more interesting and will bedevil you if  you get your V=>C converter working. You need to be interested in plotting the variation of the LDR resistance per ma of input current. I think it will surprise you. For the CLM6000 and NSL devices, you get most of the range you want below 1ma; at least I did. You might want to take two meters (yep, I keep spares. It's really, really handy sometimes.) and hook one up in series with a 9V battery and a pot to the LED. Take the other one and measure resistance across the LDR. Now set the first meter to 1ma. Read the resistance and write it down. 2ma. 3ma. Pretty soon you see the amount of control current you really need, and from that you get how big your control voltage => current scaling has to be.

[petemoore]

  Excellent explanation RG !! Thank you, I feel compelled to build it while this is all fresh.
Too bad it's DC, no brilliant quick transformer fixes for DC. 
...lol..make it into AC somehow then transformer it to the voltage desired.
One application I thought to be of great interest would be to use the Theremin's CV output to control an envelop filter/wah type theremin effect, getting the CV to control a resistance seemed like a good starting point.
  Another option would be get some LM13700's, or like, and build up and see what useful I could wring out of them..maybe I'm not browsing the right pages, but I'm not very familiar about what the top 'remote control' pedal types are, much less which one is 'this' that or worse 'in whose opinion'..even less about what they would be using to produce the desired effect.
  The third question is more interesting and will bedevil you if  you get your V=>C converter working. You need to be interested in plotting the variation of the LDR resistance per ma of input current. I think it will surprise you. For the CLM6000 and NSL devices, you get most of the range you want below 1ma; at least I did. You might want to take two meters (yep, I keep spares. It's really, really handy sometimes.) and hook one up in series with a 9V battery and a pot to the LED. Take the other one and measure resistance across the LDR. Now set the first meter to 1ma. Read the resistance and write it down. 2ma. 3ma. Pretty soon you see the amount of control current you really need, and from that you get how big your control voltage => current scaling has to be.
  Yes, I was able to get the LDR resistance down to 5k pretty easily [~2k using a pot from the CV output.
  That, and at this juncture I have considerations, such as PS, which means box, which means I should completely figure out the application and estimate it's chances of being tuned into a big hit, first.
  Besides, I need to read through again...
  Meanwhile the Theremax practice is excellent, w/Analog reverb and tremolo set to at and near my hand-pitch vibrato rate...spooky cool !

[Brian Marshall]

is there a specific reason you are using an NSL32.  I've had huge problems with consistency on those.

[petemoore]

  the NSL-32 Was in a Trem-Lune, slightly expensive and tested to still work.
  The rest of the Lune circuit failed the 'looks like charcoal-ey mess' test and was not tested further.
  an NSL32.  I've had huge problems with consistency on those. 
  I can trust you have, and understand now, this component may not be the best for...certain applications I'm trying to consider.