RGBW as parking indicator

[dw89 8]

Looks like Speedy's 'possible' tutorial could be up for the prize! Lol. I just want a simple solution, more equipment = more upkeep and more hassle plus I wouldn't even know what to do with an arduino? Like an RPi for automation I assume?

Dave:)

[Joacim 11]

Well IR doesn't come close to the distance (dynamic range) you could get with ultrasonic and if he doesn get the resolution he want then he will have to put some electronics (op-amp, dc-supply and possibly some resistors) between the ir sensor and the rgbw.

I'm interested in the result however I still think it MAY be simpler with the arduino. And no arduino is no RPi they are according to hear say pretty simple to work with. (I do feel a little lost for words here as I've just realised that I do have a Arduino laying around some where that I've never used)

[Guest Kuuno]

hi and sorry from Lauri's behalf for not answering. He ditched us and went for a vacation for 2 weeks

Please login or register to see this image.

/emoticons/default_smile.png" alt=":)" srcset="https://forum.fibaro.com/uploads/emoticons/[email protected] 2x" width="20" height="20" /> I will make sure when he comes back that he will share this

 

but shortly we needed the arduino for controlling the ultrasonic sensor.

for rgbw module we use this special module to convert pwm to 0-10V signal

[dimae 16]

Well it depends on the 1-10V capability of the RGBW. (Haven't tried it with 1-10V yet)

I will get about 2V change from the sensor, ex from 3V to 5V depending on distance. (10cm = 5V to 80cm = 3V typical)

The voltage curve is also logarithmic so i will make a custom code that changes accordingly.

 

My thought is that i will start with a change of 0.1V, that gives me 20 "steps" in color at a range of 70cm (80-10 detector range)

Writing this now makes me think that its to much at close range and a 0.2V change vill be sufficient from 30cm.

(Don't want to change color to often = faster operation)

This gives about 15 steps in 70cm range, so a change in color every 5cm (if the curve would be linear) but its close enough.

 

If its not enough range with 2V ill add an OP amp or something to make the 2V range change a 1-10V range.

But i will start with "stock" 2V and see where that ends up.

awesome, thanks a lot for sharing it... Please I am kindly asking that please, share it when you get your code

Please login or register to see this image.

/emoticons/default_smile.png" alt=":)" srcset="https://forum.fibaro.com/uploads/emoticons/[email protected] 2x" width="20" height="20" />

We really appreciate it! Thanks!

[speedy 64]

I will share everything as soon as i get the sensor and get it working!

[petergebruers 1,268]

As a rule of thumb, you can say the RGBW can measure a sensor in 0.1 V steps, from 0.1 V to 10.0 V. The truth is much more difficult to explain. It can do better than that but it depends on a parameter, polling, ground reference, noise, pull down capability of the sensor (important !), rate of change you want to detect and how fast the signal changes.

[speedy 64]

 

 

 

Ah, thank you for that info, it will save som time in trial and terror mode

Please login or register to see this image.

/emoticons/default_icon_smile.gif" alt=":-)" />

0.1V steps will do fine.

Good to know about the pull down capability, that might be an issue.

 

According to the spec sheet of the sensor it changes the output voltage according to distance at about 18 times / sec (if the distance has changed)

If no change is made to the distance then the voltage of the output will be stable. (No mention of max current output unfortunately in spec sheet)

 

I will try to hack together a "configure distance sensor" LUA loop that gets the maximum and minimum values from the sensor and then auto configures everything.

But for starters i will do simple tests for a "proof of concept" solution.

 

I have also ordered this sensor for extended range, 20cm to 150cm.

Please login or register to see this link.

[speedy 64]

So, update 1.

 

I got voltage change to work with a pull down resistor connected to ground and the RGBW device.

 

But unfortunately i burnt the sensor accidentally by connecting the output of the sensor to the wrong place when trying out different ground resistors.. 

Please login or register to see this image.

/emoticons/default_icon_eek.gif" alt=":shock:" />  

I have ordered 2 more sensors and will continue testing..

 

 

 pull down capability of the sensor (important !)

 

Well this was a big thing.

What it your experience with this Peter?

Do you have any info of the current required to the inputs etc?

 

Some info:

I could not get a "clean" 0V signal unless i connected ground directly to the input.

Using only a 470 Ohms resistor results in about 1V on the input. (strange for my experiences, usually i use a 4,7kOhm resistor and its fine)

This can be solved by saying than 1V ground level is ok, but lest for arguments sake say that i want a grund level of 0V

No connected input results in 10V readout of the input.

 

Connecting the sensor without a grund resistor results in a readout of 4.5V but it does not change with distance.

Only when connecting a grund resistor the circuit works.

 

So my thought now is to add a current amp NPN transistor (or simple op-amp to not get 0,7 C-E voltage adding from the transistor) to the output of the sensor so its not limited.

Will try this with the next setup.

[petergebruers 1,268]

Burnt module, ouch! There is a 18 k (roughly) pull up to around 11.6 Volt in the module. A CA3140 opamp (or a c-mos rail-rail opamp that works with your supply voltage) can solve your problem (without the need of a transistor). Discussion, and solution with schematics in this topic:

Please login or register to see this link.