An Engineer's first struggle with Life Sciences - BELATED UPDATE! Corals!

[Kijho]

Good looking fusion!

[mooker]

Just ordered parts for a protein skimmer, too! Looks like I'll be busy this weekend

[mooker]

CORALS ARRIVED! Boss gave me leave to work from home for the rest of the day, so I'm headed there now to pick them up.

[Astinus]

AWESOME!

PICS! When u can!

[mooker]

AWESOME!

PICS! When u can!

 

Everything's glued in place and looking relatively happy! Will post pics ASAP -- should probably actually do some of that work from home that I said I'd do...

[mooker]

Pictures!

 

FTS first:

 

 

It's a little hard to see, but there area few zoa heads and a ric in that photo! There should be two rics, but one of them refused to stay glued to his rock (didn't come with and sand/rock on his foot for the glue to actually stick to...)

 

But if we look around the corner:

 

 

I guess he's shy!

 

The other side:

 

 

And two closeups of the guy higher on the big rock:

 

 

 

The zoas aren't very photogenic yet, I'll post some shots when they open up more!

 

I'm very excited to finally have some corals in my tank.

[teenyreef]

For your ric who floated away, stick a couple little pieces of rubble rock around/underneath him. As long as he's in a low flow area, he'll glue himself to one of the rocks in a week or two, and then you can glue the rock somewhere else where he'll be more visible.

[mooker]

An aside -- any recommendations for protein skimmer pumps? I just got a Tetra "Whisper" 10G, and that's some very loud whispering...

 

My tank is in my bedroom, so I'd like to go as quiet as possible. Is the Tetra especially loud, or am I SoL and there aren't any truly quiet air pumps?

[SantaMonica]

Watching your first animals move around is an amazing feeling

[teenyreef]

An aside -- any recommendations for protein skimmer pumps? I just got a Tetra "Whisper" 10G, and that's some very loud whispering...

 

My tank is in my bedroom, so I'd like to go as quiet as possible. Is the Tetra especially loud, or am I SoL and there aren't any truly quiet air pumps?

What kind of skimmer is it? I've tried about four different air pumps on mine, and I can tell you that w whisper 10 definitely won't work in the long run

 

I finally settled on the Coralife Luft Pump: http://www.amazon.com/Coralife-Energy-Savers-ACL01655-Luft/dp/B000BJM95Y

 

I found it's really the only one that can produce the pressure needed to keep a wood air stone going for more than a few weeks at a time. I put a cheap air valve on it. When I put in a new air stone, I close the valve almost all the way and there's plenty of bubbles. After a week or two, I have to open it up a tiny bit to keep it going. I keep doing this for a couple months until it's wide open and the bubbles aren't lifting enough, then I pull out the stone and scrape it clean. It will last at least a couple more weeks then.

 

Before I got this pump, I was changing the stone aver three to four weeks.

[mooker]

What kind of skimmer is it? I've tried about four different air pumps on mine, and I can tell you that w whisper 10 definitely won't work in the long run

 

I finally settled on the Coralife Luft Pump: http://www.amazon.com/Coralife-Energy-Savers-ACL01655-Luft/dp/B000BJM95Y

 

I found it's really the only one that can produce the pressure needed to keep a wood air stone going for more than a few weeks at a time. I put a cheap air valve on it. When I put in a new air stone, I close the valve almost all the way and there's plenty of bubbles. After a week or two, I have to open it up a tiny bit to keep it going. I keep doing this for a couple months until it's wide open and the bubbles aren't lifting enough, then I pull out the stone and scrape it clean. It will last at least a couple more weeks then.

 

Before I got this pump, I was changing the stone aver three to four weeks.

 

Thanks for the advice! I'm building a cheap MAME-style skimmer (https://saltwater-conversion.com/products/mame-skimmer) using a polypropylene graduated cylinder I liberated from work. I'll post process pictures when I get the chance -- busy weekend!

 

I was hoping I wouldn't have to buy an expensive air pump, but I think you're right. The Tetra 10 is pretty anemic... I'm hoping it'll be enough for me to dial in my design, and then I'll grab a Coralife Luft once I've got the skimmer working. How loud is it?

 

 

Quick update on the wayward ric:

 

After I took that previous set of pictures, I noticed him getting pushed around by some snails, so I moved him to a nook higher on the big rock. After Teeny's post about getting him to grab onto some gravel, I tried to move him but it seems he grabbed on pretty tight so I guess he likes where he is

 

 

I'm also almost done building my ATO! so updates on that soon to, hopefully

[teenyreef]

Yeah, if a ric is firmly attached, that's his way of telling you he likes it right where he is. Don't worry, he'll let you know if he wants to be moved, usually by floating away never to be seen again

 

The luft pump is pretty quiet, especially for such a strong pump. But it's not silent. I keep mine on the carpeted floor under the desk, behind the filing cabinet, and I don't really notice it unless I turn it back on after it's been off.

[mooker]

Yeah, if a ric is firmly attached, that's his way of telling you he likes it right where he is. Don't worry, he'll let you know if he wants to be moved, usually by floating away never to be seen again

 

The luft pump is pretty quiet, especially for such a strong pump. But it's not silent. I keep mine on the carpeted floor under the desk, behind the filing cabinet, and I don't really notice it unless I turn it back on after it's been off.

 

Yeah, I think the idea of a totally silent protein skimmer was a bit of a pipe dream. Right now I'm thinking I'll just run it during the day -- no harm in that, right?

[teenyreef]

Sure you can turn it off at night, you just won't skim as much from the tank.

[mooker]

V1 of the DIY ATO is installed and working!

 

I'm going to include a medium-detailed writeup, in case someone needs to use it as reference. This is all synthesized from various sources around the internet, but I figured it'd be nice to have it all in one place.

 

First, the circuit schematic (Updated thanks to TeenyReef's sharp eye):

 

 

Basic Idea: When the water level falls, it closes a float switch which sends voltage across the relay. That closes the relay, allowing power to reach the pump, filling your tank with water. When the water level rises again, the switch opens and the power to the pump is cut.

 

Float Switches: I use two float switches for redundancy, one placed at a higher level than the other. They could both be at the same level, it doesn't really matter. The switches are wired in series, so if just one of them is open, the pump won't turn on. I sourced the float switches from amazon at about $4/switch.

 

The Relay: The relay is used to control power delivery to the the pump. I can't just use a switch, as the pump will draw (significantly) more current than the switches are rated for, which is very dangerous. https://en.wikipedia.org/wiki/Relay#Basic_design_and_operation is a good place to find more information about how relays work. I used a 12V automotive relay, a little under $5 on amazon.

 

The Pump: I chose a 12V brushless DC submersible pump to lift the water into my tank. Brushless pumps tend to last longer and be quieter than other options. $9 on amazon.

 

Power Source: I used a standard 12V power supply. There are a lot of options, I chose one that included an adapter to connect it to the rest of my circuit, which ended up being very useful. $7 on amazon.

 

Below is the whole circuit connected and set up for testing:

 

I glued the motor into the bottom of a large (3 gal) BPA-free Tupperware container (by far the most expensive part of this ATO, at $20). The bright pink things you see the float switches attached to are 3D printed brackets I made to attach them to my tank.

Here's a shot of the ATO reservoir under my tank:

And a shot of the float switches:

It's working great so far, but I don't trust it enough to just leave it plugged in yet. I'm going to put as much of the wiring and components into a tupperware as I can, and I'll put a RC outlet between it and the wall so I can control it with my rPI.

Let me know if you have any questions!

[teenyreef]

Looks great! I really like those pink brackets

 

Just a suggestion, based on more than one unfortunate experience I've had with ATO's - the most common problem with ATO's is not a failure to fill on demand, it's a failure to stop filling when the desired water level is reached.

 

If your fill switch fails and doesn't close, no water gets pumped, your tank water level goes down over time. After a while you'll hear the water flowing into the third chamber get loud, at which point you'll realize there's a problem, and the worst thing that will happen is your salinity will have gone up a point or two.

 

But if the fill switch fails in the closed position, the tank will overflow pretty quickly and won't stop until the ATO reservoir is emptied, which means a big, sudden drop in salinity along with a couple gallons of water on the floor. Because I guarantee the failure will occur when you're not around

 

Usually the second switch in most ATO setups is intended to prevent this overflow. It's put in upside down, normally closed, and above the normal water level, so that if the pump gets stuck on for whatever reason (like a snail getting into it and preventing it from dropping back down), it will stop the pump and prevent the overflow.

 

Of course you can just program the pi to shut off the pump if it runs for longer than some preset time to prevent overflow as well. But the best design starts with an overflow float, and then uses the timeout just as a backup in case the overflow float fails for some reason.

[HarryPotter]

But if the fill switch fails in the closed position, the tank will overflow pretty quickly and won't stop until the ATO reservoir is emptied, which means a big, sudden drop in salinity along with a couple gallons of water on the floor. Because I guarantee the failure will occur when you're not around

 

Usually the second switch in most ATO setups is intended to prevent this overflow. It's put in upside down, normally closed, and above the normal water level, so that if the pump gets stuck on for whatever reason (like a snail getting into it and preventing it from dropping back down), it will stop the pump and prevent the overflow.

 

Of course you can just program the pi to shut off the pump if it runs for longer than some preset time to prevent overflow as well. But the best design starts with an overflow float, and then uses the timeout just as a backup in case the overflow float fails for some reason.

 

 

He has the overflow float switch, no? The upper "emergency" one?

[teenyreef]

 

 

He has the overflow float switch, no? The upper "emergency" one?

Yep, but it's not wired that way on the diagram. But I can't tell from the picture if it's mounted upside down

[mooker]

Yep, but it's not wired that way on the diagram. But I can't tell from the picture if it's mounted upside down

 

You're right! one of the switches in that diagram should be normally open, and one should be normally closed, instead of both normally open. But, neither of the switches are mounted upside-down.

 

The "normal" state of the system is (with float switch 1 being the primary and float switch 2 being the backup):

Water Level: Nominal, the water level I want my tank to be.

Float Switch 1: Open, because the nominal water level raises the float to open the switch.

Float Switch 2: Closed, because the nominal water level is not high enough to open it, it is set to open when water reaches critical levels

Pump: Off

 

The "top off needed" state of the system is:

Water Level: Below nominal, due to evaporation.

Float Switch 1: Closed, because water level has fallen below nominal

Float Switch 2: Closed because the water level is even lower than "normal" system state

Pump: On

 

Once the pump turns on in the "top off needed" state, the water level rises until it is high enough to open Float Switch 1 and return the tank to the "normal" state.

 

However, if for whatever reason the water doesn't open Float Switch 1 at the nominal level, then the pump stays on and the system enters the "Overflow" state.

 

"Overflow" system state:

Water Level: Above nominal

Float Switch 1: Closed, because it is stuck closed even though water level is above nominal

Float Switch 2: Closed because the water level has not yet reached critical levels

Pump: On

 

In the "Overflow" system state, the pump stays on and the water level continues to rise even though the water level is above nominal. This is because Float Switch 1 is somehow stuck closed. The water level is not yet critical. As the water level reaches dangerous levels, the system enters the "Failsafe" state.

 

"Failsafe" state:

Water Level: Critical

Float Switch 1: Closed, because it is stuck closed even though water level is above nominal

Float Switch 2: Open, because the water level has reached critical levels, where this switch is set to open.

Pump: Off

 

For water to actually overflow out of the system, both Float Switch 1 and Float Switch 2 should have to get stuck closed.

At least, I think that's how this should work? I should really hold Float Switch 1 closed and check that this works!

[teenyreef]

Yep, you got it! It was the wiring diagram that made me wonder. All I meant by "upside down" is that if you have a float switch that's open in the down state, and closed in the up position, you can take the same float switch and mount it upside down to work as an overflow protection switch. But it sounds like you have two different float switches with the two different states when the float is down.

 

No worries about the design - but definitely test it

[HarryPotter]

 

You're right! one of the switches in that diagram should be normally open, and one should be normally closed, instead of both normally open. But, neither of the switches are mounted upside-down.

 

The "normal" state of the system is (with float switch 1 being the primary and float switch 2 being the backup):

Water Level: Nominal, the water level I want my tank to be.

Float Switch 1: Open, because the nominal water level raises the float to open the switch.

Float Switch 2: Closed, because the nominal water level is not high enough to open it, it is set to open when water reaches critical levels

Pump: Off

 

The "top off needed" state of the system is:

Water Level: Below nominal, due to evaporation.

Float Switch 1: Closed, because water level has fallen below nominal

Float Switch 2: Closed because the water level is even lower than "normal" system state

Pump: On

 

Once the pump turns on in the "top off needed" state, the water level rises until it is high enough to open Float Switch 1 and return the tank to the "normal" state.

 

However, if for whatever reason the water doesn't open Float Switch 1 at the nominal level, then the pump stays on and the system enters the "Overflow" state.

 

"Overflow" system state:

Water Level: Above nominal

Float Switch 1: Closed, because it is stuck closed even though water level is above nominal

Float Switch 2: Closed because the water level has not yet reached critical levels

Pump: On

 

In the "Overflow" system state, the pump stays on and the water level continues to rise even though the water level is above nominal. This is because Float Switch 1 is somehow stuck closed. The water level is not yet critical. As the water level reaches dangerous levels, the system enters the "Failsafe" state.

 

"Failsafe" state:

Water Level: Critical

Float Switch 1: Closed, because it is stuck closed even though water level is above nominal

Float Switch 2: Open, because the water level has reached critical levels, where this switch is set to open.

Pump: Off

 

For water to actually overflow out of the system, both Float Switch 1 and Float Switch 2 should have to get stuck closed.

At least, I think that's how this should work? I should really hold Float Switch 1 closed and check that this works!

Really interesting explanation! I never thought about simple circuits in "stages".

 

Mooker,you can't sound an alarm if the emergency switch opens because they're in the same circuit?

 

What about a low water level alarm?

[mooker]

Yep, you got it! It was the wiring diagram that made me wonder. All I meant by "upside down" is that if you have a float switch that's open in the down state, and closed in the up position, you can take the same float switch and mount it upside down to work as an overflow protection switch. But it sounds like you have two different float switches with the two different states when the float is down.

 

No worries about the design - but definitely test it

 

Updated the schematic!

 

Both float switches operate identically (down = closed, up = open) -- and this confused me at first, and I thought I'd have to mount one upside-down -- but the fact that the lower switch is nominally under the water line makes it normally open, not closed. So, identical switches, different normal states because of different locations!

 

I put together the below graphic for a (pending) thread for just the ATO that Harry encouraged me to make:

 

 

Really interesting explanation! I never thought about simple circuits in "stages".

 

Mooker,you can't sound an alarm if the emergency switch opens because they're in the same circuit?

 

What about a low water level alarm?

 

Thanks! And that's correct, to add an alarm I'd need another float switch mounted the same level as the emergency switch, dedicated solely to the alarm.

 

As far as a low water level alarm (for the reservoir, I assume?) -- I'm thinking I'll try to use the ultrasonic range finder for that, since it's less critical.

 

Decided to try spot-feeding my rics last night. This guy ate an entire mysis!

 

 

Also, are my zoas already sprouting new heads??

 

[teenyreef]

It cracks me up when the rics open wide to swallow the mysis

And yes, it looks like your zoa has at least one new head coming, which is a great indicator of overall good tank health.

 

Agree 100% on the way you're doing the ATO. The only other thing I would do is add some timeout logic to the pi so that you don't burn out the pump if the reservoir runs dry. Or the ultrasound level detector would work too, and be even cooler

[mooker]

 

Agree 100% on the way you're doing the ATO. The only other thing I would do is add some timeout logic to the pi so that you don't burn out the pump if the reservoir runs dry. Or the ultrasound level detector would work too, and be even cooler

 

Good call, I'll implement that. My plan is to enable the ATO between 8:00 am and 10:00 pm on week days (probably 12:00 - 10:00 on weekends) -- I could switch it on in 5 min intervals every half hour or something. That in conjunction with the ultrasonic sensor should be safe... I'm debating throwing another float sensor in the reservoir as a hard stop for critically low water levels, but at some point I'll just have to accept the level of risk and stop adding backups

[mooker]

I spent last night re-wiring the ATO to make it a little cleaner looking and a little more robust.

 

First, I soldered and added heatshrink to all the connections. Then I moved as much wiring as I could into a tupperware container:

 

 

During this process, I decided to add a JST-XH 4 pin connector (leftover from my quadcopter hobby) so that I could unplug the power supply/relay from the rest of the system. I also added some stylish purple braided sleeving (see: quadcopters ) to neaten up the wires coming from the float switches

 

 

Here's what the system looks like in situ:

 

 

Closeup of improved cabled situation: