Small Stone LFO, Max Range +

[petemoore]

  Using one of Fransisco's boards:
   http://www.tonepad.com/project.asp?id=11
 Looking at the ca3080 data sheet:
    http://www.bing.com/search?
q=ca3080+data+sheet&FORM=IEFM1&src=IE-SearchBox
12v supply [other mods work great]: Tonepad build report stated a 1meg added between the voltage divider and the pot that goes to pin5 [bias/speed in this case] caused slower speeds, that could be switched.
 One reason I'd like to be able to have the speed range elevated is so that when the R value of the pot is near zero the speed is up faster than it normally would be, perhaps fast enough that an parallel LDR with a light on it would speed things up sufficiently to make itself interesting.
 More about that voltage divider ? Comments welcome !
 And there's a 33uf cap sitting right there...
  Apparently, the cap values I need to consider are for position C12.
In this thread Mark wrote:
  http://www.diystompboxes.com/smfforum/index.php?topic=48354.0
  In this thread Mark wrote:
  I can confirm that the 33uf cap in the LFO circuit does set the LFO range.  Using a 22uf or 47uf (the next common values over from 33uf) will provide some overlap with most of the range of speeds produced using a 33uf, just not the entire range.  So, 22uf will get you all the same medium speeds, some faster speeds, but not the slowest speeds.  47uf will get you all the medium speeds, none of the fastest speeds and some much slower speeds.

 

[PRR]

The 27K+7.5K "voltage divider" gives about 2V unloaded and 6K effective impedance.

CA3080 pin 5 is a diode to pin 4, here ground, so will sit near +0.5V.

When speed pot is 1Meg, it feels about 2V-0.5V or 1.5V across 1Meg, 1.5 microAmps, flowing into pin 5 and setting the OTA max current to 1.5uA. (It is actually 1.95V, 0.55V, 1.006Meg, say 1.4uA.)

When speed pot is zero, the open-circuit 2v is shorted by the 0.5V diode, 1.5V force against 6K "divider" impedance, 0.25mA or 250uA current flows to pin 5, 250uA OTA max current. (Maybe 210uA.)

210uA-1.4uA is over 100:1 range, a good fit for an audio-taper pot. Any more would just be twitchy at the fast-end. Someone did their arithmetic here.

> an parallel LDR

You can throw a silver crowbar across the 1Meg pot. It is still 210uA max current. Same as when pot is zeroed.

You could monkey with the 27K+7.5K divider, and directly increase OTA max current. There's room. It can take 1000uA (1mA) happy, 2mA with errors which don't matter here. BUT if you put more current into pin 5, IIRC the CA3080 simply dies.

Anyway, if you want faster overall, decrease the 33uFd as much as you want. 25uFd. 1uFd.

What do you really want? Just "faster!"? Decrease the 33u. Or remote-control faster? That sorta means re-thinking the whole control scheme. The 27K+7.5K+1Meg is simple, compact, effective; but there are more elaborate ways to set OTA current and thus speed.

[petemoore]

You can throw a silver crowbar across the 1Meg pot. It is still 210uA max current. Same as when pot is zeroed.
  I would want to know if a crowbar is thrown into the plan I have, which may have to accomodate crowbars, the only ones I have are copper lead type though..
  I want a ramp rate to vary like dudes man is turning the phase speed knob CW/CCW between medium and fast mostly.
  Since I can't figure out the LFO's well enough to LFO the LFO in a way I think could be worked on with pleasing results. Matching an LFO to the LFO in the Small Stone to control the LFO rate automatically [whether or not my understanding would fascilitate it at this point, I would guess there'd be a lot of tweeking, at least decisions on what knobs I'd want].
  I don't particularly want knobs, the circuit of my idea involves an LED w/resistor, an LDR and 4 wires, the rest is mechanical, light blocking/unblocking.
  Barring putting it back into a wah case for which I have no room on the PB, I thought perhaps a simple LED controlling a variable resistor=LDR across the speed pot [a variable resistor] will control speed by amount of HE photons allowed to be in the viscinity of the Cds cell.
  Getting it tweeked for one cool trick with various R tapering shouldn't be too hard, controlling the light is already becoming a challenge.
  I suppose simple switching of 'a' resistance [say 27k] across that rate pot [something that'd make high speed sweep] would make pops in the signal.
  "Jump to fast" might be enough, especially if there could be a medium speed ramp up of the LFO Rate, such as a pressure resistor {I?IRC] or some other pressure control method.

[Baran Ismen]

May I ask what a silver crowbar is? 

Also is it possible to make a tweak in the circuit so that it can work with linear pots?

[ElectricDruid]

May I ask what a silver crowbar is? 

If I could afford one, I'd send you a picture! 

[Mark Hammer]

Baran asked if a linear pot could be used.
A linear taper pot of same value can ALWAYS be used in place of a log or anti-log pot in pretty much any application.  The problem is that the difference in taper will  cluster or bunch up the changes of interest into a much smaller range of rotation.

The way that many LFOs work is by varying how long it takes for a cap to charge up.  A variable resistance is used to set how much current is charging up the cap.  Smaller resistances allow more current to pass and charge up the cap faster.  And, of course, smaller-value caps will charge up faster, given the same current feeding them.

The way that human hearing and perception works, we are more sensitive to greater change than to lesser change (see Weber's Law or Stevens' Power Law).  So, we would have a hard time detecting the difference between one sweep cycle every 10 seconds, and a sweep cycle every 8 seconds, but would have a MUCH easier time detecting and perceiving the difference between one sweep every 500 milliseconds and a sweep every 400 milliseconds.  That's why, for many LFO circuits, the standard is often to use a reverse-log pot.  This taper will "spread out" the small changes in low resistance, such that it's easier to dial in the specific faster speed one wants.  Nothing really changes in the range of speeds available, only the ease with which one can find them. 

If the speed pot is being used as a simple variable resistance (i.e., only two lugs absolutely needed), a log pot can be used in place of an anti-log pot.  HOWEVER, the relationship between speed and rotation is opposite.  That is, with an anti-log pot, the smaller-resistance values are spread out at the clockwise end of rotation such that clockwise gets you faster speeds.  With a log pot, the smaller changes in resistance are spread out at the counter-clockwise end (i.e., the 7:00 position is fastest speed and rotating clockwise slows things down).

I recommend a reading of RG Keen's classic document/page "The Secret Life of Pots" ( http://www.geofex.com/article_folders/potsecrets/potscret.htm ) in which he explains, and graphs, how to modify the taper of a pot, to more closely approximate the taper you want/need, using parallel resistors.

[Baran Ismen]

Baran asked if a linear pot could be used.
A linear taper pot of same value can ALWAYS be used in place of a log or anti-log pot in pretty much any application.  The problem is that the difference in taper will  cluster or bunch up the changes of interest into a much smaller range of rotation.

The way that many LFOs work is by varying how long it takes for a cap to charge up.  A variable resistance is used to set how much current is charging up the cap.  Smaller resistances allow more current to pass and charge up the cap faster.  And, of course, smaller-value caps will charge up faster, given the same current feeding them.

The way that human hearing and perception works, we are more sensitive to greater change than to lesser change (see Weber's Law or Stevens' Power Law).  So, we would have a hard time detecting the difference between one sweep cycle every 10 seconds, and a sweep cycle every 8 seconds, but would have a MUCH easier time detecting and perceiving the difference between one sweep every 500 milliseconds and a sweep every 400 milliseconds.  That's why, for many LFO circuits, the standard is often to use a reverse-log pot.  This taper will "spread out" the small changes in low resistance, such that it's easier to dial in the specific faster speed one wants.  Nothing really changes in the range of speeds available, only the ease with which one can find them. 

If the speed pot is being used as a simple variable resistance (i.e., only two lugs absolutely needed), a log pot can be used in place of an anti-log pot.  HOWEVER, the relationship between speed and rotation is opposite.  That is, with an anti-log pot, the smaller-resistance values are spread out at the clockwise end of rotation such that clockwise gets you faster speeds.  With a log pot, the smaller changes in resistance are spread out at the counter-clockwise end (i.e., the 7:00 position is fastest speed and rotating clockwise slows things down).

I recommend a reading of RG Keen's classic document/page "The Secret Life of Pots" ( http://www.geofex.com/article_folders/potsecrets/potscret.htm ) in which he explains, and graphs, how to modify the taper of a pot, to more closely approximate the taper you want/need, using parallel resistors.

Hey Mark, thanks for your answer. I read that web-site today, also found the below one, but in practice I got this issue.

Am I too stupid to comprehend this matter or my pots are just weird?

http://web.archive.org/web/20210203023928/https://www.elby-designs.com/webtek/documents/tailoringpotentionometers.pdf

According to this PDF file's 5th Figure, I need to solder a 100k resistor between 2 active lugs of my 1m linear pot and it becomes logarithmic, right?

Mine becomes however something completely different, it reads around 80k or something. For Rev.Log, what shall I do? Add a resistor of 10m?

Thing is, the circuit I deal with right now is Small Stone and there are only 2 lugs connected. 3rd one is empty, or shorted to 2nd.

Nevermind, I assume its not possible due to being 2-terminal connection.