My Neptune Apex web interface compatible DIY reef controller
- Thread starter d0ughb0y
- Start date
d0ughb0y
Active member
Is there need for more than 8 outlets?
The reason I ask is I found an issue with the latest chauvet v2 circuit, and it will greatly simplify the board if it only has 8 outlets instead of 16.
The circuit drives the outlet relays via I2C expander to 16 lines, but 4 of which are shared with 4 dosers. (So you either have 16 outlets and 0 dosers or 12 outlets and 4 dosers or some combination in between).
But now, I found that I need to drive Jebao pumps using 5V pwm, so I could use the second ULN2803 for 4 dosers and 4 jebao PWM and the first ULN for the 8 outlets.
I could probably pack in a 3rd ULN to handle the dosers and pwm and have full 16 outlets. So that is what I am trying to decide.
If this was for my own use, I do not need more than 8 outlets. Jebao pumps can be plugged in to a regular (non controlled) outlet, so that will free up controlled outlet spots. Dosers are 12v so they don't need outlet.
If I go with 8 outlets, it is still possible to expand since I break out the I2C signals so adding another 8 or 16 should not be a problem, it just won't be built in to the board.
The outlets I use are for return pump, heater, fan, skimmer, ATO pump, extra pump at bottom of overflow. 6 total.
The reason I ask is I found an issue with the latest chauvet v2 circuit, and it will greatly simplify the board if it only has 8 outlets instead of 16.
The circuit drives the outlet relays via I2C expander to 16 lines, but 4 of which are shared with 4 dosers. (So you either have 16 outlets and 0 dosers or 12 outlets and 4 dosers or some combination in between).
But now, I found that I need to drive Jebao pumps using 5V pwm, so I could use the second ULN2803 for 4 dosers and 4 jebao PWM and the first ULN for the 8 outlets.
I could probably pack in a 3rd ULN to handle the dosers and pwm and have full 16 outlets. So that is what I am trying to decide.
If this was for my own use, I do not need more than 8 outlets. Jebao pumps can be plugged in to a regular (non controlled) outlet, so that will free up controlled outlet spots. Dosers are 12v so they don't need outlet.
If I go with 8 outlets, it is still possible to expand since I break out the I2C signals so adding another 8 or 16 should not be a problem, it just won't be built in to the board.
The outlets I use are for return pump, heater, fan, skimmer, ATO pump, extra pump at bottom of overflow. 6 total.
d0ughb0y
Active member
come to think of it, there is not enough digital lines to have 16 outlets plus 4 dosers. The 4 dosers really have to share lines with the outlets. So I think I'll just leave it at 16 outlets with 4 shared with dosers, and add another chip to handle the 4 lines to jebao pumps.
If you only use 2 dosers, then you have 14 outlets, which I think is plenty. Or if you use all 4 (like in triton method dosing), then you have 12 outlets available.
If you only use 2 dosers, then you have 14 outlets, which I think is plenty. Or if you use all 4 (like in triton method dosing), then you have 12 outlets available.
Have you seen this:
http://www.wifimcu.com/
In this case all code could be written on one controller including 11 PWM GPIO etc.
And this module is already availabile for 11us$ as:
http://www.banggood.com/EMW3165-WiFiMCU-Wireless-WiFi-Development-Board-Using-Lua-p-1007187.html
http://www.wifimcu.com/
In this case all code could be written on one controller including 11 PWM GPIO etc.
And this module is already availabile for 11us$ as:
http://www.banggood.com/EMW3165-WiFiMCU-Wireless-WiFi-Development-Board-Using-Lua-p-1007187.html
d0ughb0y
Active member
I have not looked at that, but I tried the $9 computer chip that runs linux and for that one, I think it is too new and unstable and learning curve is steep. The main reason I still went for arduino based board is to use the stable webserver and sd card libraries. I took a quick look at the the webserver library of wifimcu and it is quite basic and probably needs a lot more work to support what I need to do.
I'm still deciding on the number of outlets. I might settle for 8 and connect directly to the 8 channel relay board (no ULN chip in between), and just use I2C to expand more outlets if needed. I want to keep the base controller basic, simple and stable.
I'm still deciding on the number of outlets. I might settle for 8 and connect directly to the 8 channel relay board (no ULN chip in between), and just use I2C to expand more outlets if needed. I want to keep the base controller basic, simple and stable.
Moomin1967
New member
Hi D0ughb0y,
Perhaps a backup heater, and perhaps be able to configure a duty and backup return pump. I see these as critical for tank survival if away for a few days.
M
d0ughb0y
Active member
I went back to 16 outlets and 4 doser configuration. having problem with routing. I'm already using 6 mil trace with 6 mil clearance and so far can't route 100%.
has anyone tried to use the pwm pump feature to control jebao dc return pump?
did it work?
the pwm pump feature was originally created for jebao wp and rw pumps. I never tested it on the dc return pumps.
just curious if it works as is.
curious also if it will work with cross wave pump. I've read different descriptions about this. But considering the wireless controller can control wp,rw and the cross flow pump, I am assuming it will work on the cross flow pump as well. does the cross flow pump rotate in reverse direction or does it rotate only in one direction and only the speed is variable?
has anyone tried to use the pwm pump feature to control jebao dc return pump?
did it work?
the pwm pump feature was originally created for jebao wp and rw pumps. I never tested it on the dc return pumps.
just curious if it works as is.
curious also if it will work with cross wave pump. I've read different descriptions about this. But considering the wireless controller can control wp,rw and the cross flow pump, I am assuming it will work on the cross flow pump as well. does the cross flow pump rotate in reverse direction or does it rotate only in one direction and only the speed is variable?
d0ughb0y
Active member
thanks for the confirmation.
back to led controller. It has 6 channels for LDD and 2 0-10v channels. I added the 0-10 upon someone's request and the whole time I was thinking there might be limited use for it. It turns out it would be perfect for use on kessils since they use 2 0-10v channels. If more 0-10v channels are needed, I figure I can make a small board the size of the LDD driver containing the op amp circuit and plugs in to the LDD socket and then have the 0-10v output. The current web enabled controller solution to kessil costs like $200 and supports 3 pairs of 0-10v signals. But I think most kessil users only use a pair.
I need to wrap up the chauvet v2 pcb routing so I can have the pcb made and get back to working on software for the led controller.
back to led controller. It has 6 channels for LDD and 2 0-10v channels. I added the 0-10 upon someone's request and the whole time I was thinking there might be limited use for it. It turns out it would be perfect for use on kessils since they use 2 0-10v channels. If more 0-10v channels are needed, I figure I can make a small board the size of the LDD driver containing the op amp circuit and plugs in to the LDD socket and then have the 0-10v output. The current web enabled controller solution to kessil costs like $200 and supports 3 pairs of 0-10v signals. But I think most kessil users only use a pair.
I need to wrap up the chauvet v2 pcb routing so I can have the pcb made and get back to working on software for the led controller.
karimwassef
Active member
What's the ear way to get started? Is there a simple step by step sticky for the latest iteration?
d0ughb0y
Active member
You can get ideas from existing code and circuit from my and other similar projects and build your own. You can always get help here or in arduino forum.
I will make a webpage when I have enough info.
For starters, you can get a teensy 3.2 and be familiar with it. It uses arduino ide so learning curve is not so steep, but since it uses ARM processor, there are differences in how things work. I suppose it is similar to arduino due, but in a really small footprint. Teensy requires installing teensyduino. it is a go between program between arduino ide and the teensy board. It took me a while to get used to the fact that teensy board has no reset button and you have to power cycle it to reset.
I forgot to mention, buy your teensy here for $17 shipped.
http://store.oshpark.com/products/teensy-3-1
Last edited:
karimwassef
Active member
Thanks. What's the advantage of Teensy vs. Arduino? I use an Arduino Mega but don't have near the functionality you've vested in this setup.
d0ughb0y
Active member
I am reading now that DCT and DCS return pump control is different from DC pump. That it uses 50hz PWM and control betweeen 500 and 2500us (sort of like servo motor pwm control). This should not be a problem as the pwm waveform is fully controllable in software. I also read the crossflow and gyre control signal is the same, not sure what timing will make it reverse rotation. I might get a DCT pump to replace my return, but still debating if I need a crossflow or not. If the pwm signal is different, I might need to revise my circuit, unless rw and wp pumps can operate at 50 hz pwm, then I don't think I need to change the circuit. The current program use 490hz pwm for jebao pump control.
karimwassef
Active member
I have two DCT 15000 pumps and haven't figured out how to control them with my Apex. Are you saying that it's possible to generate a PWM signal to control the speed instead of a signal voltage like the DC?
d0ughb0y
Active member
You cannot control DCT and DCS pumps with Apex.
For DC series pumps, and WP and RW pumps, you can feed 5v 488hz pwm directly. The pump will see it like an analog 0-5v signal.
I just tested using 50hz pwm on WP pump and it does not work as well as at 488hz. So I have to get a DCT and possibly a gyre to reverse engineer. So I have to make more revisions to the controller circuit. Not sure yet if I'll keep 4 488hz pwm and add 2 50hz pwm, or make it 4 pwm only, and run either 50hz or 488 hz. 50hz will still work for older jebao pumps, but it does not spin as fast. I am guessing if I add the resistor-capacitor low pass filter, it will spin faster. I will test it, and if it does, I'll make the pwm run at 50hz.
For DC series pumps, and WP and RW pumps, you can feed 5v 488hz pwm directly. The pump will see it like an analog 0-5v signal.
I just tested using 50hz pwm on WP pump and it does not work as well as at 488hz. So I have to get a DCT and possibly a gyre to reverse engineer. So I have to make more revisions to the controller circuit. Not sure yet if I'll keep 4 488hz pwm and add 2 50hz pwm, or make it 4 pwm only, and run either 50hz or 488 hz. 50hz will still work for older jebao pumps, but it does not spin as fast. I am guessing if I add the resistor-capacitor low pass filter, it will spin faster. I will test it, and if it does, I'll make the pwm run at 50hz.
d0ughb0y
Active member
Does the speed control looks like it is working correctly?
When I do a search, I see a lot of postings here, on fish street and on neptune forum saying the DCT pumps are not controllable by Apex anymore.
I think I even found a post (which I cannot find anymore) that describes the control signal as a servo pwm signal and not an analog 0-5v signal.
And from fish street, the DCT/DCS electronics is changed from original DC series pump, since DC series pump leaks electricity into the aquarium.
Is the DCT pump you have the one that comes with the 10 speed controller?
If it does control it, then that is good news. That means a PWM signal controls it, but not an analog 0-5v signal.
When I do a search, I see a lot of postings here, on fish street and on neptune forum saying the DCT pumps are not controllable by Apex anymore.
I think I even found a post (which I cannot find anymore) that describes the control signal as a servo pwm signal and not an analog 0-5v signal.
And from fish street, the DCT/DCS electronics is changed from original DC series pump, since DC series pump leaks electricity into the aquarium.
Is the DCT pump you have the one that comes with the 10 speed controller?
If it does control it, then that is good news. That means a PWM signal controls it, but not an analog 0-5v signal.
Speed controls for the DCT6000 work fine. My DC6000 died, and I just bought a DCT6000, and swapped it out. Only thing I noticed, was the DCT6000 was pushing more water at the same setting, so I lowered it from 80% to 75% (Your application percentage). I could run it from 0-100 with no issues.
Yes it came with 10 Speeds.
FYI - I just sold my take over the weekend, but I did run the DCT6000 for almost a year.
Yes it came with 10 Speeds.
FYI - I just sold my take over the weekend, but I did run the DCT6000 for almost a year.
