DIY AUTOMATIC top-off

[Anonymous]

Two solenoids plumbed in series controlled by two float switches wired in series would be darn near bullet proof IMHO. My senerio is safe against over-fill and would do nothing to make up evaporation in case something failed. I believe that once or twice yearly routine maintanence would prevent any such failure from ever happening.

Louey

 

[holry7778]

Honestly without testing it I couldn't tell you, Mahai.

The concept of the dosing pump seems pretty easy too, as long as it holds against the vacuum pressure. it would only replace the sewer trap though (in order to keep your RO from constantly on/offing ). It would require some initial set up and tinkering but it is a lot simpler in nature.

But if that be the case why not just set up the doser to dose directly to your sump instead of through the Nurce? I know it is not as accurate but I do still beleive in K.I.S.

so LESS mech IS typically MORE better

 

[Mihai]

solenoid valve has much better reliability than any mechanical flopper with chain, lever and siphon trap, and I strongly suggest you consider other alternatives. As usually, IMHO.

I agree. I'll think about it. I really don't want to mess up this job.
Your HO is appreciated.
One more thing: by "solenoid valve" I assume you mean "electro-valve", i.e., a valve that can be controlled through an electrical current, right? Do you know where to get a cheap reliable one?


Thanks for sharing,
M.

 

[Fatal Morgana]

I used to sell them for less than ten bucks without the power supply (12 VDC).

There is a picture of my RO setup on this thread:
http://www.reefs.org/phpBB2/viewtopic.p ... 651#501651
The black thing right next to the ball valve is the solenoid.

Make sure you have one rated for the pressure. That is, 100PSI for the input. SpectraPure's LLC is good enough for the RO pressure.

 

[Anonymous]

Sorry, but I just saw this. Maybe this has already been covered, but my suggestion is to connect the output from your RO/DI system directly to your sump, and use an electronic shut off valve activated by a water level sensor that is placed in your sump, with float valve shut-off as a back-up.

SpectraPure sells these set-ups

http://www.spectrapure.com/index.htm

As you noted, connecting the your RO/DI output directly to your sump via a float valve will result in repeated, short-cycling of the membrane. This not only leads to pre-mature failure of the membrane, but also a significant loss in water quality. This is because a certain amount of TDS will pass through the membrane each time it is engaged.

The SpectraPure "water-leveling" system overcomes this problem because the water level sensor opens the valve only after the water level in the sump has dropped an inch or so, then closes it when it has raised back up the same distance. I have a large sump (and system), so this translates to automatic top-offs on the order of 4 gallons at a time.

I believe that I posted much more detail explanations of this system last spring in teh General section.

 

[Mihai]

I used to sell them for less than ten bucks without the power supply (12 VDC).

Hmmm.... you were making them? If not, where did you get them?
Do you still have them?

SpectraPure sells these set-ups
http://www.spectrapure.com/index.htm

True. They're a bit costly, but that's not the biggest problem. What bothers me is that it doesn't have any redundancy - if one thing fails, it all goes bad (and kills everything in the tank). Other than that is seems OK (except that the 1" distance is set in stone).

Regards,
Mihai

 

[Fatal Morgana]

I had a thread about group ordering auto topoff from me a while back, but there was insufficient interest when it come to money, so the project was dropped.

I may have a few in my garage, but I am sure you can pick them up locally from Grainger or somebody that sell process controll gadgets.

 

[Anonymous]

I used to sell them for less than ten bucks without the power supply (12 VDC).

Hmmm.... you were making them? If not, where did you get them?
Do you still have them?

SpectraPure sells these set-ups
http://www.spectrapure.com/index.htm

True. They're a bit costly, but that's not the biggest problem. What bothers me is that it doesn't have any redundancy - if one thing fails, it all goes bad (and kills everything in the tank). Other than that is seems OK (except that the 1" distance is set in stone).

Regards,
Mihai

Redundancy: My set-up includes a float valve in the sump that acts as a back-up should the water level sensor & electronic valve fail to shut off the supply from the RO/DI. The float valve closes off the supply line if the water level rise ~1" above the "high" setting on the sensor.

1" sensor unchangeable: True. However, you may be able to accomodate this by altering the portion of the sump that varies in water height due to evaporation. I used partitions in my sump for this purpose (as well as to stabilize the height of the water relative to my skimmer output so as to decrease skimmer sensitivity).

 

[Fatal Morgana]

BTW, the 1" delta_height is not set in stone. I can change it (with some constrainsts) for customized setup, for example, and you can do it with DIY too.

Hint: pressure and the effect of delta_height vs. surface area.

 

[Anonymous]

BTW, the 1" delta_height is not set in stone. I can change it (with some constrainsts) for customized setup, for example, and you can do it with DIY too.

Hint: pressure and the effect of delta_height vs. surface area.

Yep. I had no need to, but I can see how the effective delta could changed by altering the cross-sectional area of the sensor tube

 

[Fatal Morgana]

I always hate it when a lawyer answer my question correctly.

 

[DOGMAI]

What about this? With a few adjustments to 2 float valves you can accomplish what you want. They work together to make it happen. When the bottom valve drops it is connected to the top one via a piece of string. When it is all the way down it holds the top one closed, and it also opens its own valve draining the top tank into the sump on the bottom. When the sump gets up to its full level it will allow the top valve to open and the top tank will begin to fill again. It would take some adjustment but it is error proof as far as I can see.

Thanks,
Shane

 

[Jhong]

Hi, I've just built my auto-top off, incorporating a timer, and wanted to chime in.

If incorporated correctly, the timer does NOT dictate the amount of water that is added over a given period - it simply dictates the amount of switching your solenoid will do.

You use a simple float switch (I used three float switches, for low/high redundancy... the low float switch is also used to shut off the sump pump in the event of a low water level), connected to relay(s) to control a solenoid, via a timer. Due to the low hysteresis of the switch, the cycling frequency was high. By adding a timer set to come on, say, every hour or two (each time for 3-5 minutes), you are simply reducing the frequency. The correct amount of water is still added due to the float switches -- even if the timer is on, water will only be topped up to the required level, and will not be topped up at all if not needed.

I believe the time idea is the best (in terms of reliability) and simplest solution.

My plans are for a top-off reservoir with pump. However, the principle is the same for a solenoid and r/o unit. You would not need a second storage tank.

I am happy to post a circuit diagram for the three switch system if it would be helpful.

 

[Anonymous]

"An easy solution would be to use a float valve on the output of the filter and use it to maintain the constant level in the sump. Unfortunately this will lead to a premature death of my RO membrane due to short cycling (turning on/off several times a day). What I'm looking for is a method to automatically refill the auto-topoff container only when it empties (or almost empties). "

I think you have eliminated your best option here without much consideraion. I have run my RO to my 80 gal RO storage tank and to my sump. Both are on float valves going on 4 years now. My TDS show no change.

If you want to minimize RO on/off then after your filter, install a "drinking" water bladder storage tank. This will maintain pressure on the system and only call for new water after 1/2 the baldder tank is used.

I think you should set up your system up with a float valve in your RO container and watch your TDS. I don't think you will detect any membrane life reduction either. IF you see a problem, then go spend a bunch of money.