Running tank equipment with computer program

[aandon]

I have a somewhat complicated setup: several tanks all with their own pumps, filters, chillers, etc... and I was hoping to set up a dedicated desktop computer that would control all of it. Is there a program out there that's pretty customizable? Cheap or free would be great too. Any experience with these?

[drinkhomebrew]

i have heard of people using x10 in the past, not sure how well it worked.

[REEF-n-Chicago]

I too would be very interested in this..... Tagging along!

[BeanAnimal]

I have authored software to do just that. Many other folks have built dedicated hardware based on microcontrollers, home automation controllers, linksys routers, etc.

In all honestly, you DO NOT want a PC controlling the life support systems. Monitoring is one thing, control is another.

[stugray]

I am currently working on an unusual design.

I intend on using a PC running Labview for monitor and control.

However the electronics are a bit unusual compared to the other systems out there. I plan on using the 1-wire devices from Dallas/Maxim.

The 1-wires devices include temp sensors ( with alarms ), adressable switches ( similar to X-10 but for low voltage apps ), Analog to Digitial converters, and other devices.

I plan on using the addressable switches to control latching relays which, in turn, drive solid state relays. This allows the computer to set the state of the latching relay to turn on & off AC loads. The beauty here is that if power is cycled on the entire system, the AC power switches remain in the same state as before the power outage. This should be OK until the control system gets back in control.

The temp sensors will just be dipped in epoxy and used as temp probes.

The AtoD converters will be used for ph & orp probes if they prove to have high enough internal impedance. If not, I will need to incorporate instrumentation op amps.

The beauty of these devices is that they are very cheap ( $1-3 a piece ), and they can ALL be ganged together and controlled over only two wires. They can be "bussed" all through cat-5 enet cables or twisted shielded pairs. Most of the devices wont even need their own power, they are 'parisitically' powered from the serial lines.

The eval kit should be here any day, and I already have some labview experience. The NI website has labview VIs for driving the devices, I just need to upgrade to LV 7.1 or earlier. That should happen this week.

If I can get most of it working through labview, I would then prefer to port the controls to C code and run it on a linux box ( far more dependable ) with a dedicated UPS. I am even considering using an old XboX with the linux hack installed. All these devices can be driven from one USB port, and the XBox has multiple.

I'll start a thread if it looks promising.

Bean have you experimented with the different SSRs?
Or do you prefer the basic Digital Output Modules?

Stu

[BigEivlSquid]

I too plan on automating most of my tank. I am going to use an pic18 series microcontroller, a pc is just too unstable to trust running things.

[BeanAnimal]

Stu I am using dallas 1-wire stuff :)

[stugray]

Bean,

Which devices are you running & using what software?

I am definitely going to build the AC power station w/ adressable switches, just waiting for the eval kit ( usb version ).
Have you toyed with the 4 chan AtoD ?

Stu

[BeanAnimal]

I have built a PIC project that reads the DS18S20 temp sensors and dumps them to a serial port. I have played with the 1-wire pots to control my T5 dimmable ballasts and have also played with the ADCs and adressable switches. I have used the serial and USB 1-wire masters. I am using VB abd VB.NET

There is a good thread about using Linux (the linksys thread).

[cd77]

If you just want control (no monitoring) using ethernet, one of these will work great:

http://www.digital-loggers.com/EPC.html
http://www.digital-loggers.com/EPCR2.html

If you want to go wireless, get a patch cable (I think you'll need a crossover cable) and a wireless bridge (sometimes called an 'Ethernet Converter') and you're all set.

With a little bit of programming to access the web interface, you can setup a very advanced custom timer system.

As was mentioned earlier, you really don't want your home PC doing this. A dedicated embedded system would be better. a PIC microcontroller would be cheap. If you have to use a PC, use a very reliable OS (BSD, Linux, Solaris, etc..) that is built to cater to a flash drive system (limited writes) and try to get it as solid-state (replace fans with heatsinks, use flash memory instead of hard drives, etc..) as possible.

The time you put into building your own custom system would probably outweigh the cost of a Neptune AquaController III though...

[kydsexy]

dangerous territory. ive tried this for a study hatching smelt eggs influenced by the tide. after about 4 weeks, there was a storm and no power in the entire building. we did have 3 hatches before the problem tho.

[BeanAnimal]

My current application runs on a PC and is authored in VB.NET It is not in "live" on the aquarium and will likely never be.

I am now looking at http://www.coridiumcorp.com/ARMweb.php as a possible controller. I am a VB guy.. so it appeals to me. Most of the other platforms are C or some other language that I have no desire to fiddle with.