2 RAs in tandem + H2O Automation

lnevo · Post by **lnevo** » Fri Jan 24, 2014 8:59 am

Hi all,

Getting ready to start the next project and wanted to outline it here, get input and give others some additional ideas. I am planning to get another RA controller to automate my RO/DI and SW mixing stations. It will also run an automatic water change (AWC) system and facilitate filling my ATO reservoir for my main tank controlled by another RA. The main challenge is that the water station is downstairs in my basement while the main tank and sump are upstairs (and on almost opposite corners of the house).

I plan to use the following features for this project

1) Multi-level water sensor (to monitor RO/DI and SW mixing containers)
2) Salinity module (to mix saltwater to proper SG)
3) Float Switch Inputs (to receive signals from the upstairs RA)

The plan is to have a 3 head peristaltic pump that I ordered on ebay. I will currently use 2 heads one to drain water from the main tank and the other head to add new water. Because both pumps are driven by the same motor, the flow rate should be identicial. You can follow the thread on RC that started this all. http://reefcentral.com/forums/showthread.php?t=2349824

I will also employ some 12V solendoids to enable/disable the RO/DI, open/close the bypass to the flow restrictor so I can automate flushing the membrane, and to direct water to either the SW bin or ATO reservoir upstairs.

The biggest challenge aside from running all the tubing (currently planning on up and over through the attic) will be having the two RA units talk to each other. I am currently planning to run a Cat-5 cable to carry the signal. I will use two pairs which will connect to the ATO High and ATO Low ports on the downstairs RA. The High ATO switch will be controlled by an outlet connected to a relay, like I do with the auto-feeder.

I will use the WL sensor in my ATO reservoir to determine when to open/close this circuit. The downstairs RA would see the circuit close and start filling the reservoir upstairs. I would let the water pressure from the RO/DI feed the system, rather than drive another pump. It should take less than 20 minutes to refill the container (less than 5g with a 75gpd RO/DI). I'll use the solenoids to flush the membrane first for a minute and then fill the container.

The ATO low will be a signal to enable/disable the AWC system. I plan on connecting these in parallel with my current float switches for high and low sump level. If either of those float switches are triggered then the AWC would be disabled. I'm not sure if that would cause any problems having two RAs share the same float switches. If anyone sees an issue with this, LMK.

For salt water mixing. I'm not "currently" looking to automate adding salt or build a complex hopper. I currently use a 20g Brute for mixing SW. I would look to add water to the 85% mark and then add enough salt to get 35ppt salinity when it's at then 90-95% level. I'll then let the salinity module drive the filling of that container until its at the right dilution. This would ensure that I'm always adding the right water salinity to the tank upstairs.

Anyway, I hope these ideas are useful for someone and would love some feedback. This is going to be a slow project as I have a lot of components to get and this isn't of the highest priority, but at least its a plan. Feel free to discuss similar setups, this thread doesn't need to be dedicated to mine

Would love to see what others might do too.

Sacohen · Post by **Sacohen** » Fri Jan 24, 2014 9:25 am

Thanks Lee.

You and I are looking at doing a lot of the same things.
I'm looking forward to seeing if and how the 2 RA's will talk to each other.

My first project is to build a new stand and move the tank before I get into all this, but I have the dual head peristaltic pump already and I ordered some solenoids today from this guy on E-bay that has a good selection at decent prices.

http://www.ebay.com/itm/1-4-Inch-Plasti ... RTM1562570

rimai · Post by **rimai** » Fri Jan 24, 2014 9:27 am

1 float 2 RAs doesn't sound like a good idea.
Ground sharing that long of a distance is not recommended.

lnevo · Post by **lnevo** » Fri Jan 24, 2014 9:47 am

Roberto,

Would it be any different using an IO module? or the outlet control triggering a relay to close the circuit? Just trying to figure out the best alternatives. What would be the drawback to the ground sharing? Forgive my ignorant electronics knowledge.

rimai · Post by **rimai** » Fri Jan 24, 2014 10:15 am

If you share grounds over long distance, you are prone to ground loop and there is also a possibility to pick up noise over the length of the cable.
The best would be to isolate the signal with opto isolators.
Your idea of the relay would work well too. They would be a mechanical isolation instead of optical.
Just make sure the relay is on the receiving RA side and not the on sender. In other words, place the relay close to the ATO port and not close to the power supply plugged into the relay box.

lnevo · Post by **lnevo** » Fri Jan 24, 2014 10:53 am

Gotcha, so I'd send a 5v/12v dc signal over the cat5 and have the relay near the remote RA. I could use the opto isolators. I'd like to not use more outlets if possible. Can you explain that a bit more? Would I be doing the same thing with that and keep the isolator close to the remote RA?

Also would it be possible to send the signal from the lowATO pins if I wanted to "simulate" the float switch being activated?

rimai · Post by **rimai** » Fri Jan 24, 2014 11:05 am

Like this:

: opto.png (2.57 KiB) Viewed 16866 times

You can also use the ATO ports or dimming ports too, but if it is very far, I'd go with a power supply.
Otherwise you would be having to use one opto in each end.

lnevo · Post by **lnevo** » Fri Jan 24, 2014 11:19 am

So that's coming from the float switches connected to the RA upstairs to the ATO port on the RA downstairs?

Is another power supply required on the upstairs end? or is the one on the left the RA power?

Is the 5V power supply always on? Is that just for supplying power to the optocoupler? or is that power also coming from the RA?

The run is going to be approximately 100' give or take some. I may just go with the mechanical as that would give me actual control from the RA rather than just rely on the float switch status.

Would there be any better way to communicate between two RAs over that distance? Maybe I could simulate i2c or something over the cat-5 instead. Just thinking long term...

rimai · Post by **rimai** » Fri Jan 24, 2014 11:28 am

That's too long for I2C.
I thought you were going to use a relay port.
Connect the power supply to the relay port and use the current float that is already connected to RA upstairs.
If you want a total independent solution, wire up another float switch in series with a power supply (connected to the wall outlet) and send the wires downstairs to the optocoupler.
The +5VDC can be wired from the temp sensor connector.
This is the disadvantages of using the optocoupler. You need to power them.
If you go with the relay, you don't need the +5 and can wire the contacts from the relay straight into the ATO port.
The other alternative is an XBEE or RF link, which are wireless solution.

lnevo · Post by **lnevo** » Fri Jan 24, 2014 11:43 am

I was going to use a relay port for the ATO reservoir since I will base that on the WL sensor.

For activate/de-activate the AWC, I was going to let the state of my current float switches be the indicator. Currently the high and low float switch are wired in parallel to the upstairs RA on the LowATO port. So either switch activates, the LowATO port goes on and I disable my Return+Skimmer.

If I'm understanding correctly, which I'm not sure that I am... I could run another paralell connection from the float switches to the optocoupler which would then connect to the second RA ATO port and the Optocoupler could get power from the temperature sensor?

rimai · Post by **rimai** » Fri Jan 24, 2014 12:12 pm

No, you need to either use a power supply connected on the relay box and use RA to turn on/off that port based on your current float switches or use a dedicated float that is not connected to RA to send the signal downstairs.
Optocouplers need power on both sides.
The upstairs one is getting the power from an external power supply (this way you don't have any chance of picking up noise by using RA power supply) and the downstairs can use the one from RA, since it is a short run.
What you are trying to do is avoiding sending wires that are connected to either +5 or GND in your RA through a long wire. It acts as an antenna and noise just travels through to RA, which can cause unexpected behaviors.

lnevo · Post by **lnevo** » Fri Jan 24, 2014 4:15 pm

Ok. Now I got it. Ok 2 relays it will be..I can always replace all the wall warts with the 12V power module. And that would give me an extra 5 12V ports for something

Except it might be good to have all the 12V ports with the solenoid valves downstairs. I hope taxes work out well this year...

lnevo · Post by **lnevo** » Sun Feb 02, 2014 8:57 pm

I decided what I will do. I am gong to use my computer as a proxy between the two ra's. I already query my ra once a minute. I will look for a flag to allow/disallow water change mode and a flag to enable/disable the ro filling. I will just use a timer with enough buffer to not run into issues (like 60% then fill to 80%. Why waste relays.

Plus I can just have it run just a couple of times a day.... Gotta think of all the scenarios.

lnevo · Post by **lnevo** » Tue Feb 04, 2014 5:58 pm

So I've been thinking of how I'm going to run my salt mixing station. I will use the salinity module and the water level sensor and another pump. The pump will be set as an ATO but for salinity with the water level sensor acting as the failsafe along with a timeout. I can set it to use "water change mode" as the trigger, which would also disable the automatic water change until the new water is ready. Basically, the idea is that when water gets low, I'll run downstairs, turn on water change mode and dump a few cups of salt in. The RA will then start circulating the water and adding RO until the salinity reaches the target. Once there, it will continue mixing for x amount of time before disabling the water change mode and re-enable the auto-water change with the tank upstairs.

The overall idea is just add salt once or twice a week. What do you guys think?

Sacohen · Post by **Sacohen** » Tue Feb 04, 2014 6:02 pm

It sounds good.

Sent from my HTC One VX using Tapatalk

lnevo · Post by **lnevo** » Sat Mar 01, 2014 9:51 am

Just ordered a 24V solenoid valve with 1/4" quick connects. I will tee off my RO/DI output and this will open up the connection upstairs to fill the ATO reservoir. I'll use the RO/DI pressure instead of using a seperate pump.

I decided I will be setting up my 10G QT tank with the third head on the ametek pump. Even if I'm not actively QT'ing anything, I can use it for rinse water

i figure long term it would be better for all the heads to have the same amount of wear.

I received my new ReefAngel controller for the setup. I will be setting up a 5V power supply connected to an outlet on the RA upstairs to a relay downstairs which will be connected to one of the float switch ports. If this input is not "active" i will not run the RO or the AWC.

The float switch in the salt water bin is also going to be connected to the RA so when that flips the AWC will be disabled and I'll send myself an alert.

For the RO I will monitor the reservoir % from my Mac. When it gets to lets say 20% I'll have a cron job to let the other RA know to open the solenoid for an hour or so. If the % goes higher than lets say 90% I'll have the upstairs controller deactivate the control signal and send me an alarm.

The relays that will control the Ametek pump and the RO solenoid will both be powered by the same 24V power supply which will also power the pump and solenoid and each will be triggered separately by a 5V analog signal from the RA. This will give me endless scheduling flexibility and the capability to change on demand.

Now that the RA is here, I can get working on the programming. I'm still waiting on a PWM motor drive and now the solenoid. I'm waiting on zachts from RC to figure out the connectors to go from the ametek to the cheap 1/4" tubing, but I think otherwise I have everything ready to go. Just need a trip to radioshack for some wire and connectors to make it all happen.

lnevo · Post by **lnevo** » Sat Mar 01, 2014 10:21 am

These are the relays I picked up, except mine are 24V

http://m.ebay.com/itm/400525119455

Solenoid:

http://m.ebay.com/itm/290836300351

PWM motor controller

http://m.ebay.com/itm/261099568424

Hope the links work...

rimai · Post by **rimai** » Sat Mar 01, 2014 10:35 am

What's the motor controller for?
I saw the link, but didn't quite get it.
What is that pot and what is POW?

Sacohen · Post by **Sacohen** » Sat Mar 01, 2014 11:00 am

It's too slow down the water change pump to a crawl. Also the slower it run the quieter it is.

Sent from my HTC One VX using Tapatalk

lnevo · Post by **lnevo** » Sat Mar 01, 2014 11:39 am

Yeah essentially it uses pwm to slow down the motor but it'll take the 24v signal and send it out as pwm at whatever frequency to get the speed you want.

These pumps are pretty loud. At the end of the day its for noise control.

lnevo · Post by **lnevo** » Sun Mar 09, 2014 2:49 pm

Ok. Today I setup the wiring for the relays that will control the automatic water change pump and the solenoid for my RO. All I have left now is to get the ReefAngel programmed and connect the 5V outputs to the trigger input on the relays. Then the main part of the electronics will be done. I do still have to get the control signal from upstairs connected and the float switch wired up for the ReefAngel's inputs, but that's just splicing wires.

https://www.youtube.com/watch?v=yvHLUKTDDv0

lnevo · Post by **lnevo** » Sun Mar 09, 2014 5:17 pm

Ok i screwed up the relays when i tried to hook up the reef angel dimming ports. Now the red light is stuck on like the trigger is active but the relay is open and the trigger no longer responds. I'm not sure whats going on. I ordered 4 more relays...will try again in a month when they arrive.

Sacohen · Post by **Sacohen** » Sun Mar 09, 2014 5:20 pm

I had a similar issue with my 12v relay and it turned out that the power supply was not putting out enough voltage.

Sent from my HTC One VX using Tapatalk

lnevo · Post by **lnevo** » Sun Mar 09, 2014 5:53 pm

Whats weird is that when it was hooked up both relays were coming on at the same time. Something with my wiring may be flawed I don't know. Will see what happens next time. I will try and document the circuit better and get some input.

Sacohen · Post by **Sacohen** » Sun Mar 09, 2014 6:04 pm

Yeah. Let me see the circuit and maybe another set of eyes will help.

Sent from my HTC One VX using Tapatalk

lnevo · Post by **lnevo** » Sun Mar 09, 2014 8:55 pm

Here's the circuit. Lmk what you think. You can watch the video to see the actual wiring job. Hopefully this makes sense.

lnevo · Post by **lnevo** » Mon Mar 10, 2014 5:29 am

So just to clarify the IN is connected to dimming ports which are set to analog / 0-5V ports. I am setting the ports to 100% for what should be the 5V trigger.

Sacohen · Post by **Sacohen** » Mon Mar 10, 2014 6:13 am

Lee;

I can see the sketch on my phone thru Tapatalk, but I can't see it here and it is too small on my phone.

Sacohen · Post by **Sacohen** » Mon Mar 10, 2014 7:31 am

It looks good to me.

The two 5v things at the top are the RA dimming ports?
Obviously it worked with the 24V triggering the relays, did it ever work with the 5V RA Signal triggering it?
I can't remember are you using the dimming channel to trigger the relay or the ATO ports?

The reason I'm asking is since the dimming channels are sending out 0-5v are you sure the relay is getting the full 5v or if it is less it may not be enough to trigger the relay.

Also (Stupid question, but need to ask) you said the relay has a high and low trigger, are you sure that you switched it over to use the low trigger.

Finally, what happens if there is no load connected, it shouldn't make a difference, but you never know.

lnevo · Post by **lnevo** » Mon Mar 10, 2014 7:59 am

Yes, I"m using dimming ports for this. I did not confirm that they are sending out the full 5V. I will be doing a bit more testing next time around. I can set them back to 0-10V and then adjust the output to get to the full 5V if thats an issue.

Reef Angel Forum