Reef Pi Build

[That Crusso Kid]

Any led guys care to help me out here?

One of the things I'm not sure about for the t247 conversion to and end to end configuration is the led layout in the fixture. Would I be ok as far as spectrum goes to run the arrays end to end?
Basically I want to put the bottom right side array next to the top right side array doubling the length (should be between 24" to 26" long) and add a sheet of frosted acrylic for a diffuser that runs the length of the light about 2 inches below the leds. I will be keeping the lenses on.
I also want to add a 2 bulb 24 watt dimmable t5 HO retro fit kit from BRS with 2 coral plus or 2 purple plus bulbs. I like the look of a 14k MH so that's kinda what I'm looking to get from this set up. All of this will be stuffed into a floating canopy that matches the stand.

For the frosted acrylic diffueser I'm thinking an 1/8' sheet of acrylic sanded with 360 or 400 grit sand paper. I like the effect I got from doing this on my controller box and it seems to diffuse the leds inside the box nicely.

I know acrylic absorbs light differently than glass does, but how much par would I loose from light doing this? Maybe 20 or 30%?

While I could probably speak to your questions and almost sound like I knew what I was talking about, I will defer to my friend @dantimdad . He is one of the best LED DIYers around. As is @theatrus . I'll be following along to see their answers and learn more.

 

[dantimdad]

Moving the array shouldn't present and issue especially using a diffuser.

I haven't begun my diffuser testing but @saltyfilmfolks has done sanded plexi. May they can answer.

 

[saltyfilmfolks]

Basically and diffusion will cut the out put , and appearantly cut is as it spreads out.

IMO , best way to guage loss is a lux meter. Par meter works too but keep in mind as you use it you are cutting intensity and not spectrum.
Save that thick white frosted will add warm colors.

 

[Diamond1]

Moving the array shouldn't present and issue especially using a diffuser.

I haven't begun my diffuser testing but @saltyfilmfolks has done sanded plexi. May they can answer.

I just saw the diffuser thread today. I'll jump over there and see what you come up with from testing.

Basically and diffusion will cut the out put , and appearantly cut is as it spreads out.

IMO , best way to guage loss is a lux meter. Par meter works too but keep in mind as you use it you are cutting intensity and not spectrum.
Save that thick white frosted will add warm colors.

Makes sense that the diffuser would warm the colors up just by spreading the warmer colors in the array around a bit more.

 

[Sebastian Stankiewicz]

I am new to this forum and have gotten my PH and temperature sensor hooked up to reef-pi, but I cannot get the dosing pumps to communicate with the software. I am currently using a raspirobot v3 with a 12V external power supply, it mounts on top of the raspi gpio pins. Currently on their github, they show what pins are being used by the raspirobot. Even when I supply these pin numbers to the UI (https://github.com/simonmonk/raspirobotboard3), I have not been getting any movement from the pumps. I enabled the PCA9685 driver, but still no luck. Any suggestions? Has anyone used the raspirobot with dosing pumps and the reef-pi?

 

[b4tn]

If you haven’t already here is the link to the guid for setting up doser motors in reef pi

https://learn.adafruit.com/reef-pi-guide-5-dosing-controller/overview

Did you create the virtual equipment? Turning one off and the other on would determine the polarity of the output.

Also have you turned pwm on by adding the below to the config file?

1. dtoverlay=pwm-2chan

 

[Diamond1]

I am new to this forum and have gotten my PH and temperature sensor hooked up to reef-pi, but I cannot get the dosing pumps to communicate with the software. I am currently using a raspirobot v3 with a 12V external power supply, it mounts on top of the raspi gpio pins. Currently on their github, they show what pins are being used by the raspirobot. Even when I supply these pin numbers to the UI (https://github.com/simonmonk/raspirobotboard3), I have not been getting any movement from the pumps. I enabled the PCA9685 driver, but still no luck. Any suggestions? Has anyone used the raspirobot with dosing pumps and the reef-pi?

Welcome to reef2reef!
I don't know of anyone using that raspirobot with reef-pi.
Is the chip your using the TB6612 motor driver like the schematic shows? If so I'm unfamiliar with running motors with that board maybe @Ranjib could give a bit of help with that.
I use an L293D motor driver (this is the chip used in the build guides) for control of dosing pumps and it requires an input from R-pi's pwm signal, specifically I'm using GPIO 18 and using timers and virtual equipment to control the direction, duration and speed of the pumps.
If your not using the PCA9685 to supply pwm to your pumps you don't need to have that driver enabled. Enable rpi to control the pwm.

I would also recommend posting a build thread in the DIY section with pictures of your circuitry so that if you have any ongoing issues or questions you will get more exposure to a lot of folks who can answer your questions.
Everyone in the reef-pi family is very dedicated to helping make other builds the best they can be.

 

[b4tn]

I just noticed this was your build thread and not the main thread. But @Diamond1 is correct, if you start your own thread with your progress its easier to track your issues. That bieng Said the tb6612 works well and is the same chip I am using, also combined with the PCA9685, mine is just separate breakouts vs being on a hat. You will need to identify which 4 GPIO pins are being used as inputs (2 for each motor) and declare them as equipment in reefpi. Something like Motor1a and Motor1b then Motor2a and Motor2b. Then you have to turn one of them off and the other on to set the polarity or direction of the motor. You also need to declare jacks in the connectors section. Determine which pin of the PCA9685 is being used and select PCA9685 as the driver.

 

[Diamond1]

I just noticed this was your build thread and not the main thread. But @Diamond1 is correct, if you start your own thread with your progress its easier to track your issues. That bieng Said the tb6612 works well and is the same chip I am using, also combined with the PCA9685, mine is just separate breakouts vs being on a hat. You will need to identify which 4 GPIO pins are being used as inputs (2 for each motor) and declare them as equipment in reefpi. Something like Motor1a and Motor1b then Motor2a and Motor2b. Then you have to turn one of them off and the other on to set the polarity or direction of the motor. You also need to declare jacks in the connectors section. Determine which pin of the PCA9685 is being used and select PCA9685 as the driver.

I missed that you were using the tb6612 chip in your build. Was there an advantage to that over the L293d?

 

[b4tn]

The big one is 1.2 A continuous, 3 A max current capability vs 600mA/1A on the L293D. But its also much more efficient than the older 293 series. It uses a MOSFET based driver and has a lower voltage drop.

 

[Diamond1]

Good to know thanks.

 

[Sebastian Stankiewicz]

Thanks guys! I hadn't added the dtoverlay to my /boot/config.txt code! The equipment is now running when I flip the switches, but the dosing pump will not respond when I try to calibrate it. Here is the pin layout for the raspirobot. how would I write the dtoverlay code so that the raspi is the driver of these pins?
RIGHT_PWM_PIN = 14
RIGHT_1_PIN = 10
RIGHT_2_PIN = 25
LEFT_PWM_PIN = 24
LEFT_1_PIN = 17
LEFT_2_PIN = 4
I have a follow up question. Is there anyway to set the dosers up so that they administer some acid if the pH is too high? Is there a way to do this in macros?

 

[Diamond1]

Thanks guys! I hadn't added the dtoverlay to my /boot/config.txt code! The equipment is now running when I flip the switches, but the dosing pump will not respond when I try to calibrate it. Here is the pin layout for the raspirobot. how would I write the dtoverlay code so that the raspi is the driver of these pins?
RIGHT_PWM_PIN = 14
RIGHT_1_PIN = 10
RIGHT_2_PIN = 25
LEFT_PWM_PIN = 24
LEFT_1_PIN = 17
LEFT_2_PIN = 4
I have a follow up question. Is there anyway to set the dosers up so that they administer some acid if the pH is too high? Is there a way to do this in macros?

Which version of reef-pi are you running? I know there was an issue with version 2.0 that the calibration tab had a glitch in it. I think @Ranjib fixed it recently. If you don't have the updated version with the fix in it you need to set the speed and duration in the doser tab to get the pumps to respond.

 

[Sebastian Stankiewicz]

I guess that I am confused. Based on what I have read only pins 12 and 18 can be used for pwm0, while 13 and 19 are for pwm1. Based on the raspirobot, the pwm pins are 14 and 24. That means that you cannot control the pwm via hardware? Is that a problem?
In addition, is the PCA9685 built into the raspirobot? I don't see anything about it. I currently don't have that board, do I need to have it to make this work? I know the gpio pins can be controlled via software, which is how I am able to control the motors via python and the GPIO library. I am running version 2.0.

 

[b4tn]

The raspibot is using a tb6612 MOSFET h-bridge as motor drivers and uses the built in rpi pwm. If you want to run additional pwm stuff you will need to get the PCA 9685 breakout board and run it from the I2C pins. Keep in mind the GPIO pin #'s and physical pin #'s are not the same.

 

[Diamond1]

I guess that I am confused. Based on what I have read only pins 12 and 18 can be used for pwm0, while 13 and 19 are for pwm1. Based on the raspirobot, the pwm pins are 14 and 24. That means that you cannot control the pwm via hardware? Is that a problem?
In addition, is the PCA9685 built into the raspirobot? I don't see anything about it. I currently don't have that board, do I need to have it to make this work? I know the gpio pins can be controlled via software, which is how I am able to control the motors via python and the GPIO library. I am running version 2.0.

Yes PWM from rpi is supplied by gpio 18 and 19. I think I read somewhere on the raspberrypi.org forums that there is a way to change the pwm to other pins, but I have never tried to do that.
My set up is using only pin 18 for pwm to both pumps with pins 12,13,14 and 15 controlling the motor direction. It controls both pumps directly from the pi pwm and eliminates the need for a PCA9685. So I use rpi for the driver.

Like i said before I'm not familiar with the raspirobot board so I don't know how the circuit is set up to pull PWM from the pi. My suggestion would be if you can't make the board you have work get one of the boards like the wykat hat and go from there. I think @Jonathan Troutt has some extra boards that can be had for the price of shipping.

 

[Jonathan Troutt]

Yes I do have extra boards. And if you need the boards populated I can do that as well for a fee.

 

[Ranjib]

I guess that I am confused. Based on what I have read only pins 12 and 18 can be used for pwm0, while 13 and 19 are for pwm1. Based on the raspirobot, the pwm pins are 14 and 24. That means that you cannot control the pwm via hardware? Is that a problem?
In addition, is the PCA9685 built into the raspirobot? I don't see anything about it. I currently don't have that board, do I need to have it to make this work? I know the gpio pins can be controlled via software, which is how I am able to control the motors via python and the GPIO library. I am running version 2.0.

It will be nice if you start a dedicated build thread, that way we don’t have to hijack this thread :-) . I need some details of the build first. Pwm 0 and 1 in rpi can be configured with any arbitrary GPIO using the dtoverlay configuration. I don’t know much about the raspirobot thing . You don’t need the pca9685 board as long as you can do away with only pwm channels .
Anyway ... let’s not derail this thread

 

[Diamond1]

Well this just showed up today! It's the Roberto's ph breakout board. Big shoutout to @Jonathan Troutt the board looks great and nicely soldered. Thanks again!

I'll be getting my probe later in the week and hopefully have it integrated into my enclosure. Then with a little support from @Ranjib this bad boy will be up and running.

 

[Bigtrout]

Well this just showed up today! It's the Roberto's ph breakout board. Big shoutout to @Jonathan Troutt the board looks great and nicely soldered. Thanks again!

I'll be getting my probe later in the week and hopefully have it integrated into my enclosure. Then with a little support from @Ranjib this bad boy will be up and running.

Nice...let us know how ot works