calling all Car and Tractor specialist

[Kipper4]

I was thinking of it like a treadleless wah.
Using the foot for proximity.
But something that would work whatever the light situation. Dark stage set to sunlit room.

[thermionix]

wah probe

Do what?

Zvex Wah Probe.  Theremin-like proximity control.

[karbomusic]

I was thinking of it like a treadleless wah.
Using the foot for proximity.
But something that would work whatever the light situation. Dark stage set to sunlit room.

I've considered using some "Johnny 5 eyes" with an Arduino nano, should be simple enough...

https://www.amazon.com/Elegoo-HC-SR04-Ultrasonic-Distance-MEGA2560/dp/B01COSN7O6/

They are pretty simple in code (examples should be easy to find) and you could use its measurement to modulate a PWM on one of the Arduino output pins. I have a few of these lying around, I should dork around with one and use a LED on the PWM output for testing.

[anotherjim]

Ok, so I know this works in principle 'cause I've already done it for some other devious purpose.

Need - u/s transmitter (Tx) 40kHz small dia like 10mm if possible. Ditto 40kHz receiver (Rx).

Circuit - 40khz oscillator to drive TX - Preamp for Rx - 2 input Exclusive OR gate (XOR) - RC filter - DC buffer.

Principle - Transmitted u/s wave is reflected back from target. Amplified RX signal is phase compared by XOR. Resulting phase difference pulses are averaged by RC to give a variable control voltage proportional to distance of target. The control voltage is buffered/scaled to drive the control current of an OTA.

For really short distances, a CD4046 chip can do everything except buffer the output control volts. Use it's VCO output tuned to 40kHz to drive the Tx and Signal input direct from Rx. Use its Phase comparator type 1 (XOR) output to RC filter. For longer distances a preamp will be needed for the Rx.
Provided a regulated supply for the 4046 and good quality timing RC parts, should be stable enough 40kHz for whock wan whoal. I've had one on BB over a few days with a scope channel always measuring the VCO freq while I was messing with it, and it never drifted by enough that mattered.

For bigger distance the Tx needs amplifying. You can use a bunch of gates or inverters in parallel/bridge drive combinations. Even small audio amp 386 & similar etc for single or bridge drive. Tx don't need sine wave, just whack it with the square from the 4046 VCO. A very good sinewave always comes back from the Rx, simply because any harmonic of 40Khz is way above Rx & Tx bandwidth.

[karbomusic]

I have a few of these lying around, I should dork around with one and use a LED on the PWM output for testing.

I just did, I think the closer working distances (and/or resolution) you need might be a little small for these. It works, but using my hand isn't the best test and it would take some potential error smoothing to keep things smooth in a real world scenario me thinks.

[PRR]

> How do those little parking sensors work

They beep/listen and let the main computer figure it out.

So not a drop-in solution for you.

[Kipper4]

 Guess I can't squeeze a tractor for into a car sized box.

I'm looking for a mini solution that would fit in an enclosure along with the effect pcb.
Sorry to be a pain.

[Jdansti]

I was thinking of it like a treadleless wah.
Using the foot for proximity.
But something that would work whatever the light situation. Dark stage set to sunlit room.

Infrared LED and sensor like the ones on automatic faucets, toilets, soap dispensers, burglar alarms, etc. No problems with ambient light. The sensors are often photo transistors. See the links below.

https://www.societyofrobots.com/schematics_infraredemitdet.shtml

http://maxembedded.com/2013/08/how-to-build-an-ir-sensor/

For a wah, you'd have to figure out how to make the output variable and not on/off like a switch. Someone here should be able to tell you how.

[deadastronaut]

hi rich, heres a nifty lil circuit...

might give this a go myself at some point..


https://www.youtube.com/watch?v=AmkYwvCjXoY

[samhay]

>For a wah, you'd have to figure out how to make the output variable and not on/off like a switch. Someone here should be able to tell you how.

Light travels a bit faster than sound - about 3.0e+8 m/s (0.3 billion meters per second; or roughly 1 billion feet/s), which makes this a bit more tricky:

Lets say your toe moves at most about 15cm from heel down to heel up.
Light has to bounce off your foot and return to the sensor, so will have a difference in length of twice this = 30 cm (~1 foot).
Light will travel this distance in ~ 1 ns ( 1 billionth of a second).
Sampling an echo (by the detector and associated processing) must occur at least as fast as 1/ns = 1 GHz.
If you want some graduated effect, you have to be able to sample the light much faster than this = many GHz.
It's easy to sample this fast in the analogue domain, but if you need a computer to turn your detector into something you can pass audio through, then this is going to be rather difficult to pull off.

Edit - I forgot something. The further your foot is from the light source, the more light will be scattered away rather than bounced straight back to the detector. So, instead of listening for an echo, this should work quite well by measuring the intensity of the light that makes it back to the sensor. Using an IR light source should minimize problems caused by ambient light triggering the detector.

[deadastronaut]

dumb question, but do the RX/TX led sensors work in 'any' light conditions?.

[anotherjim]

Passive infrared sensors used in alarms, can contain 2 Rx measured in a differential manner. One can see the world though the lens, the other reacts only to ambient. Therefore it detects some new source of heat/IR no matter what the background conditions because common ambient picked up by both is cancelled out. It's an optical humbucker. But the bad news is they contains basic detection circuitry that only has a switched output.
See differential detection....
https://en.wikipedia.org/wiki/Passive_infrared_sensor

They do jitter though when the external source is moving, so you might get lucky if you treat the output as a train of pulses which you average with an RC or integrator for a CV, or maybe an LED on a LDR will smooth it out for you. You may have to cover the lens with something that cuts the IR down for close up sensing.

[duck_arse]

[/helpfull]

They beep/listen and let the main computer figure it out.

sounds like deadastronaut.

[helpfull]

[deadastronaut]

it does...

[Kipper4]

Cheers guys
The IR transistors looks like a workable solution for now.
I'll get some and have a play
Rich
Much work to do.

[amptramp]

There are a lot of slotted limit switches available, some with an unbuffered open collector output:

http://www.digikey.com/products/en/sensors-transducers/optical-sensors-photointerrupters-slot-type-logic-output/547

[Kipper4]

Thanks Ron.
So a bit like an all in one package of transmitter and reciever.
From what I can gather from the data sheet. 1st one one the page.
Interesting.
As soon as I get some free time I'll look at some of the other solutions too.
Thanks guys

[vigilante397]

This was mentioned a couple times with no apparent acknowledgement. This is exactly what you're talking about, ZVex has had it for a while.

http://www.zvex.com/products/wah-probe

[Kipper4]

I will go look for the probe schematic too thanks Nathan.

Getting out of the bath tonight, Glancing across the room. The scales caught my eye.
How does that work? Pressure pad? Induction?
Just throwing ideas out still at this stage.

[amptramp]

Scales typically use a load cell and this is a perfectly good solution.  They tend to be linear and appear electrically as a Wheatstone bridge with resistance dependent on applied force.