DIY Auto-Top off with Solenoid

[tgreene]

Originally posted by szwab
tgreene
do you have the float assembly attached to the sump via a suctioncup? If so I would think of a different way to attach the suction cup will eventually give letting the entire unit sink into the sump.

Actually, I generally have to chisel suction cups from the sides of my sump, but I am already considering a couple of more permanent & solid mounting ideas...

[Mr_Quality]

Originally posted by tgreene
[B]...The biggest difference for me, is that my system will be externally housed in a small project box, with 2 indicator lights and a SPDT toggle switch for quick shutoff when working in the tank. ...

tgreene,

I really like your version! Why an spdt switch? wouldn't an spst work? Could you show us your internal wiring?

[tgreene]

Originally posted by Mr_Quality
tgreene,

I really like your version! Why an spdt switch? wouldn't an spst work? Could you show us your internal wiring?

The way I have designed this, an SPST wouldn't work, because multiple isolated circuits are involved.

I hope you'll understand, but since I'm unemployed and thinking of manufacturing these, I would rather not show the internals...

[Mr_Quality]

Originally posted by tgreene
I hope you'll understand, but since I'm unemployed and thinking of manufacturing these, I would rather not show the internals...

That's cool (I guess). Too bad though, I thought this was a place to share DIY tips. In any event, I think I figured it out myself. Power comes in to the one center pole and two outside circuits are completed when the switch is on.

I'll hold you responsible if I electricute myself or burn my house down. (JK)

[tgreene]

Check your PM

[BeanAnimal]

I designed a similar system a few years back when I was working on my automation system. I just posted a shematic in another thread....CLICK HERE

The design is similar but also takes into consideration sump "overlfow" levels and sump "dry" conditions.

If the sump overfils, the topoff power is removed. If the sump runs dry (loss of topoff water?) then the return pump is powered off.

The only problem selling such a design would be patent issues and UL or similar ratings. I am not sure I would want to sell electronic devices with 120v relays in them! At least without having a UL or similar sticker. A simple solution would be to use the 12v solenoids I guess. As for the patent issues, all we are using are simple latching circuits to control "tank level" every city water treatment plant, refinery, industrail process uses identical designs. In that case sharing your design is not going to keep you from a patent. I do understand not wanting your secret out to other DIYers, then ya can't sell em a control box!

Good luck with your design, and good luck with finding a source of income.

BTW your fabriction of the actual control box looks great!

Bill

[tgreene]

My controller is fed directly by the RO/DI system, so there is ZERO chance of the sump ever running dry. With the use of 2 floats to create a HI/LO circuit, overflow also cannot happen.

[BeanAnimal]

I added the "overflow" protection relay and float to ensure that a stuck float or relay could not result in a flood. Floats and Relays do tend to stick in an environment such as ours. The failsafe relay removes power from the entire "top off" system.

The "sump dry" relay serves a similar purpose. If for ANY reason, the sump runs dry, the return pump power is removed. What happens if your RO/DI unit clogs? The solenoid fails closed? The water supply is turned off while you are out of town? The overflows on the tank become restricted?

It will most likely never get "tripped" but for another $4.00 in parts, it seems worthwile.

The arguement could be made that by adding 2 extra relays, and 2 extra floats, the system will have a higher failure rate! This is 100% true. BUT with 2 float logic, a single float or relay failure CAN result in disaster. This failure could be 50% of the time in regards to float failure. With the 4 float logic, a single float failure can ONLY result in a Dry floor and at worst a shutdown status of the return pump. It would take the "overlfow" float AND the "TOPOFF STOP" float or relays both failing to cause a flood with my system. Much less likely.

Bill

[tgreene]

Sure, anything could happen, but the possibility of a tank cracking and leaking is MUCH greater!

[BeanAnimal]

I am confused by your thought process on this one, it defies logic AND common knowledge.

I have never had a tank crack... I have seen dozens of float switches malfunction. I have seen plenty of relays malfunction. Solenoids tend to fail rather quickly also. They are all mechanical and have wear parts. RO/DI units do become clogged. The water company does turn off the water sometimes. You can spend more days out of town than you had planned. TANKS DON"T JUST CRACK.

Lets look at some design facts, or criteria:

A well designed solenoid for a refill application will Always FAIL CLOSED. If it does not, then it is the wrong type of solenoid.

A Float switch can STICK in either postion.

A Relay will most often fail leaving the NO open and the NC closed. It is rare that a relay sticks in the acutated postion.

Your design will protect from overflow if a normal relay failure occurs. However if eithe rfloat switch sticks or fails, then you will either have on overflow or sump dry condition. In other words your topoff will start and never stop, or will not start at all.

In the case of the former you have a flood, the later a burned up return pump.

WHEN Your solenoid dies, it will prevent yoru tank from filling (if it is the correct solenoid) otherwise it could allow a flood in the event of failing open.

There is nothing to prevent the return pump from running dry in the event of solenoid or water supply failure. Again, this is exponentialy more likely that your tank cracking.

My design will tolerate the failure of any SINGLE component and limit the damage to a sump shutdown. My design will also limit damage in the event of many dual component failures (very rare). It is very easy to draw up a logic table if you would like to see the results for single, dual and tripple component failure!

Bottom line is that the parts are cheap and yoru pumps and floors are not.

Do as you please, but don't kid yourself into thinking your tank is going to up and shatter before a float switch, relay, or solenoid causes a system failure.

ANY...LET ME REPEAT ANY type of automated system that does not have a failsafe is poorly designed. My design is far from truley redundant and "FAILSAFE" but it is also a far cry from two floats and a continous supply (or lack thereof) of water.

Your dishwasher, washing machine, bathtub, sink, gas pump, fuel injector....everything you own is designed with some type of failsafe to keep it from being damaged (or damaging you) Why would your aquarium be any different? Why design something only half way and hope that Murphy and his inLAWs don't come to visit? We are not talking about hundredds of dollars or hours worth of time...we are talking about $10 worth of parts and a far superior design and piece of mind.

Regards,
Bill

[H20ENG]

Well said Bill, and nice diagram on the other thread!

[BeanAnimal]

Originally posted by H20ENG
Well said Bill, and nice diagram on the other thread!

In all fairness, my design is about the same as yours, except Ihave added the extra floats and relays. I have been building relay circuits for years.

you would be amazed at what can be done with a simple BOSCH Automotive relay!

You have done the DIY Reef Community a great service with your latching relay threads. If time permits, I will finialize a few sets of schematics along with your drawings and create a godo HOW TO for the DIY links.

Thanks for the compliment, however I found your drawings much nicer than my ratsnest schematic!

Bill

[H20ENG]

LOL,
Yep, relays are fun.
As you mentioned, we arent doing anything new, just adopting industrial wiring for fun stuff
I just got a PLC and have been trying to learn its language. This is gonna be great!

[H20ENG]

Sorry for the brief hijack, Zeph

[Zephrant]

Not a problem- I'm happily reading along too.

Zeph

[adidaswood]

What DPDT Latching relay are you using from Mcmaster?

[Foster]

Polypropylene solenoid- Part 7877K53, for $18.12 each

[tang_man_montreal]

The "guts" of the Polypropylene solenoid- Part 7877K53 are made of copper.... has anyone found a substitute with no copper internal parts?

I have many uses for such a beast.... THe only issue is finding one.

[adidaswood]

it doesn't matter if you put it on the input side of your r/o unit...all of us have copper pipes in our houses...most of us anyway...

[adidaswood]

I need to know a part number for the latching relay Platapus,,not the solenoid

any one have a part number for the DPDT latching relay?

[tang_man_montreal]

Yes, I know that there are no issues putting it on the input side of the RO unit. I have it installed that way right now.

HOWEVER,

Like I mentioned, I have many uses for a solenoid that DOES NOT have copper innards. So to re-state my question, does anyone know of a solenoid without copper internal parts?

[Foster]

oops - sorry

RELAY:
#69585K45 relay, DPDT, 12vdc coil, $7.26
#7122K22 matching socket for relay, $5.35

[Foster]

You will need both.

[adidaswood]

Thanks platapus!!


Is there a concensus on wether we should be using DC or AC.

[dwall174]

Originally posted by Zephrant
I have been considering putting the solenoid on a timer so that it can't be "on" except for 15 minutes every hour or less.
Zeph

Have you thought about trying to wire a series of relays & solenoids, So that you could have the RO/DI flush the membrane & by-pass the DI for the first 5 minutes. Then switch to another timer which will run the RO/DI for an hour then re-set. Of course a float valve or switch will also be in the sump.