Light not dimming

[Sinibotia]

When I plug in my light to my reef pi controller, it goes to max settings for strength and spectrum. The reef pi interface does not control it at all.

I've followed guide 5 close to exactly. A bit different because I'm integrating it into what will be an AIO controller. But the only differences there are that I didn't install the power switch, and the connections for power, ground and SDA and SCL are on the perma proto hat rather than the header pins directly.

Another difference is that I'm only using one light (Kessil A80), but I've set up the perma proto board like the tutorial but omiting the circuitry that would correspond to the 2nd and 3rd lights.

Ive checked to make sure that the PCA9685 is recognized by the pi and at address 40 (64 in reef pi).

The converter is confirmed putting out 10V.


At this point I'm not sure how to troubleshoot further. Any ideas?

 

[Sral]

Two things I would start with:
- Please send us a picture of the circuit board that generates the 10V PWM.
- Try setting the light to 50% and measure the signal voltage.
- If that doesn’t work, measure the voltage that the PCA9685 puts out

I’m suspecting the PCA9685 to not behave correctly. If the PCA9685 doesn’t give any signal, you end up with a constant 10V at the output of the tutorial circuit.

 

[Sinibotia]

Two things I would start with:
- Please send us a picture of the circuit board that generates the 10V PWM.
- Try setting the light to 50% and measure the signal voltage.
- If that doesn’t work, measure the voltage that the PCA9685 puts out

I’m suspecting the PCA9685 to not behave correctly. If the PCA9685 doesn’t give any signal, you end up with a constant 10V at the output of the tutorial circuit.

the pca doesn't appear to be giving out any signal as you thought. could it be a defective chip? Would be a shame as it's brand new.

 

[Sral]

Hm, interesting. For starters, you can try my old test script. It doesn't change the outputs, but it reads a value from the register, if the chip is functioning.

If that works I can try rewriting the script to actually switch outputs, so we can test that.

 

[Sral]

Also: a well lit picture of the Pi, the I2C connection, the PCA9685 and the 10V circuit is always helpful to avoid mistakes on that front.

 

[Sinibotia]

Update: spent some more time with the multimeter testing the PCA9685. What I discovered is it only seems to be pins 0 and 1 that are defective or broken somehow. I was able to successfully control the lights using pins 4 and 5. Considering I only need to control one light (maybe more in the future but certainly not 7) and two dosing pumps, i can work with that.

 

[Sral]

Update: spent some more time with the multimeter testing the PCA9685. What I discovered is it only seems to be pins 0 and 1 that are defective or broken somehow. I was able to successfully control the lights using pins 4 and 5. Considering I only need to control one light (maybe more in the future but certainly not 7) and two dosing pumps, i can work with that.

Congratulations !

In that case I would really like a picture of your circuit, as it sounds like your circuit might have fried these two channels, if those were the ones you have used. It’s a bit too much of a coincidence in my mind, although it might be ESD related and therefore it’s less likely to repeat once you stop working on it.