Any Electronic Guru's Out There??

[NEWSALT]

I have too much time on my hands. I've had this idea about an Automated Water Change System that would basically be a "push a button and watch it go" type of thing. I kind of know how it would work, but I'm not electronics savy to be able to build it with relays and time delays, etc. I was wondering if anyone out there has those skills to help me out. Here's a picture with a little description of the operation.

 

[Phyl]

Could you please provide a URL to a full size version of the drawing (or something at least 800-1000x??)? I'm having a tough time reading some of the labels.

Thanks!

 

[arvin]

I would be interested to see some responses too. Sometime back I had envisioned a project like this. I wanted to use the AC Jr to kick off pumps from sump->drain, saltwater vat->sump and ro/di->sump etc.

I am curious, How often are you planning to do the water change, once you accomplish this project?

 

[NEWSALT]

Here's a link that should allow you to look at a bigger image.

http://i231.photobucket.com/albums/ee122/igiruam/AutoWaterChange.jpg

Arvind, I figure with something like this I'd do them more often. Maybe two smaller water changes a week rather than a bigger one every other week. You have to understand my situation. Currently, I fill a 5g bucket in the laundry room where my RO/DI unit is. From there I carry the bucket into my home office where the salt is kept. I mix the salt and water in there. When it's "ready", I carry the bucket into the garage where my sump is and do the water change.

 

[roadking05]

what do you want to know? I do this for living. I'm an electrician, instrument tech for refinery in Linden :

 

[mladencovic]

If you don't mind, I would make small modifications to your design/algorythm. Instead of using timers for X minutes of flow work of the component, use float switches. That way you can precisely measure how much water goes where. The reason why I am saying that is that it would be much more difficult to time the flows through S1 and S4, then just to turn them off when they both move the same volume of water (basically empty the sum to some low level, and then fill it back to the same high level). Then, use timers only as failsafe method in case the float switches get stuck or the system is not responding properly.

Also, I would say that it would probably be much easier and cheaper if you replace the S4 gate with a simple powerhead. So you leave your recirculating pump in a simple loop and then when you want to get the water out, you power the new pump. That is, of course, if your fresh salt w. container and the sump are relatively close and on a similar level.

Assuming you have hardware components (relays and sensors), you can approach the electronic control from two sides. You can choose to have one microcontroler (PIC, Parallax, etc) that will basically monitor the sensors and power the relays on a programmed sequence. There you can go fancy (adding timers, flow measures, additional sensors and measurements, communication with other devices and monitors etc). If you know someone who has programmers and kits for those microcontrollers, it could cost you as low as 20$, if not, then ~ 100$.

Or you can go low level digital gate electronics where you will pack everything in 20-30$ but will need to play with bunch of gate logic chips.

 

[NEWSALT]

what do you want to know? I do this for living. I'm an electrician, instrument tech for refinery in Linden :

Basically I need to know everything....what type of relays and other components to buy. How to wire them up to the solenoids. If I'm using a microcontroller, which one and how to program it.

Maldencovic: My first go round on this had float switches rather than timers. I guess I was just trying to limit the number of electical components.

I don't quite follow what you mean by replacing S4 with a powerhead. S4 is meant to either recirculate water or send it to the sump using P4 pump.

Regarding cost of this, if I could keep it to $100 or slightly over that, that would be great.

I have this drawing available in AutoCAD if anyone wants to modify it.

 

[mladencovic]

...
I don't quite follow what you mean by replacing S4 with a powerhead. S4 is meant to either recirculate water or send it to the sump using P4 pump.
...

I ment: In your design you have a gate that switches the flow between two pipes, so that pupmp P4 pumps either into the sump or back into the container. I believe that the price of such gate (that would be appropriate for reef tanks) would be higher and reliability lower then of a simple powerhead. So, I was proposing to leave the P4 to pump only back into the container and add a new pump P5 that will pump the water out of container and into the sump. Then you simply turn it on when you want to move the water from container into the sump and otherwise leave it off.

 

[NEWSALT]

Ok I got you now. Thanks.

So, do you think this "thing" can be built?

 

[JohnS_323]

That looks pretty slick Mike. Good luck with it. I'd love to be able to do something like that with my system as well. I had looked into actuated ball valves but MAN are they expensive! Keep us posted if you decide to go with it.

 

[roadking05]

What you really want is small PLC. Programmable logic controller. Only thing is COST, so relays are your best bet for your applications. I have some 120vac relays and base if you need them.

 

[DaveK]

It looks like an interesting plan, but have you priced the cost of solenoid valves using PVC? They are out of sight expensive.

Get yourself a dual head diaphragm pump like this - http://www.championlighting.com/product.php?productid=17519&cat=386&page=1

Plumb one head to fill the tank from the new SW supply, Plumb the other to empty the tank. Then use a float switch to shut off the pump when the new SW bucket is empty.

These pumps only move a few gal of water per hour, so the change is slow, and not disruptive.

You'll find this a lot less expensive to build, and a lot more reliable.

You'll

 

[mladencovic]

Here is how I would have solved your problem:

Components are (ALL could be found in Radio Shack, prices aproximate from my faulty memory :-[):

• Parallax Microcontroler ($70)
• Rel 1-3 - Relays rated for 110V (4$ x 3)
• T 1-3 - Simple switching power transistors (4$ x3)
• D 1-3 - Simple diode (1$ x3)
• R 1-3 - Depends on specific transistors used for T 1-3, but I would guess that 100K would be fine (or 50K)
• FS 1-2 - Float swithces. FS2 inverted so that both are normally open ($7 x2)
• Pump 1-2 - Powerheads (MaxiJet 1200) unless more power is needed to push the water (20$ x2)
• Gate Control/Pump - This is what controls the flow of your RO/DI water to Sump (top-off) and your new water container. Personally I would disgn this in the following way: RO/DI unit fills the fresh water contaier. Two pumps then move that water into sump and new salt water container, respectively, each controled with its own float switch (FS1 coul be used for top-off as well with some additional control code in the controller and switching circuit)

Gate control circuit can control multiple devices (pumps, valves) at the same time. If you would prefer more redundancy, finer control or smaller power through relays, you can duplicate that particular circuit and add it to the new port on the microcontroler (has 16 ports). As a matter of fact, you really do not need this circuit at all (T3, D3, R3, Rel3) if you hook your RO/DI unit the way it is on your diagram. Since the pumps will pump 100 times more water then the RO/DI can supply, the addition of new fresh water during the change process/sequence will not matter on a larger scale.

Total cost is ~ 150$. Time to make it - just a couple of hours.

Code for Parallax microcontroler is as follows. This is a very basic set of actions withot whistles and bells (like additional monitoring and safeguards for wrong parameters - temp, salinity, etc, pump and sensor diagnostics, logging and statistics etc.

DO
 IF IN1 = 1 THEN
  HIGH 6              ' Shut off the RO/DI unit supply
  PAUSE 1000            ' Wait 1s
  
  IF IN3=0 THEN          ' Pump sump out only if it has water above low level
   HIGH 5             ' Start the Pump 2

   DO WHILE IN3 = 0        ' and pump until you reach low level
    PAUSE 100

    ' You can add a timer here that will turn the pump off after some time
    ' in case there is something wrong with the sensors or the pump. Timer is
    ' implemented by a simple counter that counts how many times you waited 100ms
   LOOP
  ENDIF

  LOW 5              ' Stop the Pump 2 (disable in any event)

  PAUSE 5000            ' Wait 5s

  IF IN2=1 THEN          ' Pump into sump only if the water is below high level
   HIGH 4             ' Start the Pump 1

   DO WHILE IN2 = 1        ' and pump until you reach high level
    PAUSE 100

    ' You can add a timer here that will turn the pump off after some time
    ' in case there is something wrong with the sensors or the pump. Timer is
    ' implemented by a simple counter that counts how many times you waited 100ms
   LOOP
  ENDIF

  LOW 4              ' Stop the Pump 1 (disable in any event)

  PAUSE 5000            ' Wait 5s

  LOW 6              ' Enable the RO/DI unit supply
 ENDIF

 PAUSE 60000            ' Wait 1m until you can do the same cycle.
                   ' Modify this piece here if you want longer pause
                   ' or if this should be a sepaearete thread and you
                   ' want to free microcontroler for some other tasks
                   ' while still disabling another water change in
                   ' some small time interval
LOOP

Finally, this whole process could be 100% fully automated and you could set up a schedule and completely eliminate the push button and "button pusher" ;D

 

[roadking05]

NICE, I concur with mladencovic

 

[malulu]

Mike,

i use this: SpectraPure® LiterMeter III, it is a bit pricy (~$700). I am not saying to buy this, just a reference for you to see the "overview". Click here for Full manual

the task don't need to be complicate setup, it is just using 3 pumps, with only ONE be on at a time:
- (pump-C) suck old water out from sump to drain
- (pump-B) fill new salt water to sump
- (pump-A) fill RO/DI water to sump (can be add with regualar auto-top-off switch to prevent overflow)

in theory, you can use 2 cheap timers trigger on different time slot, which control 2 LOW flow dosing pumps.
- step-1: cheap timer 1 trigger on 9:00am for 15 mins, for pump-1, suck bad water out
- step-2: cheap timer 2 trigger on 9:30am for 15 mins, for pump-2, draw new salt water in
- step-3: just any creative way of auto topoff from RO/DI
you can do the same operation as many time a days as you wish (depends on how slow you dosing pump can be) if your dosing pump can not be SLOW, it will exchange too many water at a time.

please note:
the advantage to using dosing pump is they are very accurate, so you no need to worry about the water get drain too much and too little over time.

 

[NEWSALT]

Wow! That's what I'm talking about. I'll have to print that out and take a look at it. Thanks so much!

 

[Hawkeye]

I would just like to state that I'd like to think I'm smart enough to know anything in "general discussion" - but everything in this thread is beyond "advanced discussion" in my remedial brain. :

 

[RichT]

An Aqua Controller has 2 digital inputs (2 float switches) and if you got the break out box, you could add 2 more digital inputs. Why couldn't you use the AC to activate the pumps based on the various float switch states? That's if you had an AC in the first place.

 

[arvin]

An Aqua Controller has 2 digital inputs (2 float switches) and if you got the break out box, you could add 2 more digital inputs. Why couldn't you use the AC to activate the pumps based on the various float switch states? That's if you had an AC in the first place.

Does the AC Jr has the digital inputs too?

 

[reptilemanmark]

what do you want to know? I do this for living. I'm an electrician, instrument tech for refinery in Linden :

Basically I need to know everything....what type of relays and other components to buy. How to wire them up to the solenoids. If I'm using a microcontroller, which one and how to program it.

Maldencovic: My first go round on this had float switches rather than timers. I guess I was just trying to limit the number of electical components.
It`s possible to with a plc but not for a $100! The plc alone is going to cost $150 + the cost of the valves and float switches and containers. To really do this right I would say the cost is going to be in $400 range if you do the programming your self. Where I work we use Cutler Hammer PLC`s and the software is free for download. You dont need any relays the inputs and outputs will handle small 120volt loads. Everything would be setup in ladder logic step by step. Its not a bad idea I could do it I`m just to lazy.
Mark reptilemanmark@netscape.net

I don't quite follow what you mean by replacing S4 with a powerhead. S4 is meant to either recirculate water or send it to the sump using P4 pump.

Regarding cost of this, if I could keep it to $100 or slightly over that, that would be great.

I have this drawing available in AutoCAD if anyone wants to modify it.