Reef Pi Attempt #2!

Schreiber

Leviathan
View Badges
Joined
Jun 28, 2016
Messages
468
Reaction score
593
Location
Knoxville
Rating - 0%
0   0   0
My Auto-Aquarium Adventures

The only thing more frustrating than unknown problems in a project is having nobody to share them with. I’ll be chronicling the adventures of my Reef-Pi troubleshooting as I dive into what could wind up being a multi-week undertaking. If you're already familiar with Reef Pi, go ahead & skip to post 2 to dive into the details of my build. If you're not already familiar with Reef Pi, read on!

For those of you unaware what “Reef-Pi” is, it’s an open source aquarium automation platform created & maintained by @Ranjib. It runs on a single-board computer called a Raspberry Pi & can control nearly every aspect of an aquarium. It will monitor & control temperature & pH, schedule light & pump activity, & much more! Sound great, right?

The only problem is, I can’t get mine to work. My first Reef-Pi setup was a simple equipment controller with none of the fancy sensors or extra features. It worked great, so once I found the time, I decided to include nearly all the features in build 2.0. This one….. hasn’t been so great, for reasons I hope to solve & report here!


Wires and Solder and Amps, Oh My!

Full-disclosure, I’m not an Electrical Engineer by any means. I have enough electrical knowledge to be dangerous, but not enough to magic together a working prototype overnight. Unfortunately, I tried to magic together a working prototype overnight & wound up with the results you would expect.

1565783812455.png


It’s not *quite* as bad as it looks, but it certainly does not work as intended. A few ideas I have on why: I wound up connecting some contacts together that aren’t happy about it, something in my configuration isn’t correct, or one of the changes I made in my pinout wound up not working. That being said, all is not lost! When I power up my tangle of wires & hopes, my relays power on with a burst of mechanical clicks in rapid succession.

20190811_152529[1].jpg



Digging In

The reason I’m bothering to type out all of this is two-fold. First & foremost, is for selfish reasons. If you can’t tell from the picture of the current state of my project above, I tend to jump head first into things without much planning. Occasionally, it works out & I save some time. More often, it winds up biting me & I spend twice the time rooting out problems. I figure if I take the time to plan my troubleshooting process out, I’ll be less likely to miss something obvious.

The second reason is for public knowledge. Ranjiib making this project open to the public free of charge is incredible. The community support for people building their own is great, but it sometimes requires searching through literally thousands of unrelated posts interweaved together in one thread. If anyone else runs into the issue I (hopefully) solve in this process, it would be great to have a clear solution to point them to.

My next post will cover my plan of attack-what I think the issues may be & how to solve them. Once I have that laid out, I can go ahead & start trying to make this thing work!
 
OP
OP
Schreiber

Schreiber

Leviathan
View Badges
Joined
Jun 28, 2016
Messages
468
Reaction score
593
Location
Knoxville
Rating - 0%
0   0   0
What's Wrong?

Reef Pi allows for dosing pump control. After wiring up my system & testing it out, I don't get any response when attempting to calibrate my pump. All of the inputs on my L293D are reading 0 (or close to) 0, while the outputs are all sending out 2.7 volts, whether or not I'm running calibration. Since the pump I'm using for dosing requires 12 volts, this just won't do.


Current Configuration

Here's my wiring diagram:
1565784031683.png




Here's my list of connections:
1565784106090.png



Major notable change from Ranjiib's recommended configuration:
  • I'm using GPIO 18 & 19, rather than GPIO 18 & 10 for the dosing PWM. I'm afraid this may be the cause of my issue. If it is, I'll end up needing to unsolder GPIO 10, which is currently controlling equipment 11 & res assigning a different pin for that purpose.


Where to From Here?

First order of business is to disassemble part of my circuit so I can unsolder GPIO 10. If my theory about GPIO 10 is correct, I should be able to use GPIO 18 successfully once 10 is removed. I'll then be faced with the decision of whether to keep just one dosing pump, or to perform surgery on my setup to save the option of 2 pumps.

I'll be taking pictures of this process, so you can see how my setup is housed & fits together. Hopefully this will help some people I've seen in the main thread asking about all-in-one enclosure options.
 

Ranjib

7500 Club Member
View Badges
Joined
Apr 16, 2016
Messages
9,843
Reaction score
17,058
Location
Pleasant Hill, Concord
Rating - 0%
0   0   0
Thank you so much for writing down the detail and persevering. My first suggestion will be to test your circuit incrementally as you build it. i.e. test one doser at a time before completing the entire circuit. This will help us detect a problem before it get compounded by the entire circuit. By test I mean only testing voltage and continuity with multi meter. It will be tedious but I think it’s worth.
So, connect one gpio with the circuit with everything else disconnected and then configure reef-pi and check if you see the expected voltage when you make changes in brought reef-pi ui.
 
OP
OP
Schreiber

Schreiber

Leviathan
View Badges
Joined
Jun 28, 2016
Messages
468
Reaction score
593
Location
Knoxville
Rating - 0%
0   0   0
Thank you so much for writing down the detail and persevering. My first suggestion will be to test your circuit incrementally as you build it. i.e. test one doser at a time before completing the entire circuit. This will help us detect a problem before it get compounded by the entire circuit. By test I mean only testing voltage and continuity with multi meter. It will be tedious but I think it’s worth.
So, connect one gpio with the circuit with everything else disconnected and then configure reef-pi and check if you see the expected voltage when you make changes in brought reef-pi ui.

Is there a good resource that lists what voltage is expected at each pin of the L293D? I have no idea what I'm looking for at this point.
 
OP
OP
Schreiber

Schreiber

Leviathan
View Badges
Joined
Jun 28, 2016
Messages
468
Reaction score
593
Location
Knoxville
Rating - 0%
0   0   0
What's Wrong?

Reef Pi allows for dosing pump control. After wiring up my system & testing it out, I don't get any response when attempting to calibrate my pump. All of the inputs on my L293D are reading 0 (or close to) 0, while the outputs are all sending out 2.7 volts, whether or not I'm running calibration. Since the pump I'm using for dosing requires 12 volts, this just won't do.


Current Configuration

Here's my wiring diagram:
1565784031683.png




Here's my list of connections:
1565784106090.png



Major notable change from Ranjiib's recommended configuration:
  • I'm using GPIO 18 & 19, rather than GPIO 18 & 10 for the dosing PWM. I'm afraid this may be the cause of my issue. If it is, I'll end up needing to unsolder GPIO 10, which is currently controlling equipment 11 & res assigning a different pin for that purpose.


Where to From Here?

First order of business is to disassemble part of my circuit so I can unsolder GPIO 10. If my theory about GPIO 10 is correct, I should be able to use GPIO 18 successfully once 10 is removed. I'll then be faced with the decision of whether to keep just one dosing pump, or to perform surgery on my setup to save the option of 2 pumps.

I'll be taking pictures of this process, so you can see how my setup is housed & fits together. Hopefully this will help some people I've seen in the main thread asking about all-in-one enclosure options.

Changing to GPIO 10 did the trick :) Working on my new board now... Stay tuned!
 

Ranjib

7500 Club Member
View Badges
Joined
Apr 16, 2016
Messages
9,843
Reaction score
17,058
Location
Pleasant Hill, Concord
Rating - 0%
0   0   0

High pressure shells: Do you look for signs of stress in the invertebrates in your reef tank?

  • I regularly look for signs of invertebrate stress in my reef tank.

    Votes: 36 31.3%
  • I occasionally look for signs of invertebrate stress in my reef tank.

    Votes: 28 24.3%
  • I rarely look for signs of invertebrate stress in my reef tank.

    Votes: 21 18.3%
  • I never look for signs of invertebrate stress in my reef tank.

    Votes: 30 26.1%
  • Other.

    Votes: 0 0.0%
Back
Top