DIY Auto-Top off with Solenoid

[christopherl]

Would someone pull their solenoid valve apart real quick and tell me what you think about the metal parts inside. I just ordered that mcmaster valve, installed it on my system and turned it on. It wouldn't hold back the water pressure so I pulled it apart thinking something could have been stuck in there or not seated right. I noticed that the plunger and spring are steel. They are threaded into the valve with brass. Finally looking inside with the plunger removed I noticed that there appears to be the end of a copper coil. These parts are all subjected to the water running through the valve. Tell me if I'm missing something here or am I crazy. Also, after checking the valve a couple of times I never did get it to hold back the water pressure from the discharge side of my Ro/Di unit.

 

[Benthic Explorer]

Quality post

Quality post

by Zephrant. I plan on building one this weekend.

 

[Zephrant]

Chris- Can you post the exact part number of the one you have? I have not taken mine apart, but the spec's said that they were all delrin where they met the water, so should be fine.

Got any pictures?

I agree that if they have copper/steel/other metal in them that we should find something else.

Zeph

 

[christopherl]

Here ya go,
It's part number is listed as 7877K53 - Polypropylene Direct. This is what is listed on the invoice that was emailed to me from mcmaster. The valve doesn't have a part number on it, but looks identical to the one in your picture.

 

[Zephrant]

Well all, Chris is right.

Here is the disassembled piston/shell:

Here is the seat of the casing:

It looks like the end of the piston has a plastic plug seated in a stainless steel pin. It has a stainless steel spring, that keeps the solenoid OFF when power is off. The casing has an end with a brass thread that mates to the plastic body. There is a small ring of brass that is in contact with the water flowing though the valve, as well as the aforementioned spring and pin.

So it looks like this particular model is not perfect for our uses. In the six months that I have used mine, you can see a little discoloration has occurred on the brass, but none on the stainless steel. There is no evidence of the brass being dissolved, in fact it appears to have a bio-slime layer on it.

I'm not sure how much of a concern this is when used on a water system. Copper/brass plumbing is used everywhere, but you are not supposed to run RO/DI water though it.

On the other hand, I run my water though this valve, then to a Kalk reactor, which is supposed to precipitate out heavy metals.

Knowns:
Small amounts of copper is bad for reef tanks.
Brass is part copper.
RO/DI water pulls copper out of brass faster than tap water does.
Kalk reactors precipitate out heavy metals.

Unknowns: Will the copper dissolve out of the brass at a significant rate for our tanks?


For my tank, I'll continue to use it for now. Eventually I'll try to find another solenoid that does not have a brass sleeve.

Another possibility would be to put a layer of epoxy over the brass, but that could be difficult to do without jamming the valve.

Looks like I should have disassembled them when I bought them.

Thanks Chris!

Zeph

 

[Zephrant]

Chris- I had the same problem getting the water to hold back when I reinstalled my unit. Turns out that there is a "right" direction, and a "wrong" direction. Reverse the plumbing and you should be fine.

You'd think they would mark these things....

Zeph

 

[christopherl]

Thank you very much Zephrant. That post was excellent. I'm not hitting a Kalk reactor with mine, but I'm gonna try it for a while until we can find a suitable replacement valve. I'll look around for a valve and if I come up with something I'll be sure to post it.

 

[JB NY]

There shouldn't be any brass in there.

From their site

A 1 3/8" overall length makes these valves the choice for small spaces. Use with water, oil, and inert gas. Body is glass-filled polypropylene and seat is nitrile. The enclosed coil is Class A. Choose valves with 1/2" male conduit and 18" double leads or valves with spade connectors.

Also does anyone know if these are safe to use continually on with no water flowing through it? I was going to use two of them with my Aquacontroller to control RO/DI water going into my nilsen. If the pH is lower than 8.4 it would always stay open, if the pH was above 8.4 I would have it shut down (power off) and another solenoid would come on that bypasses the nilsen to just put RO/DI water in the tank. Before I did this I just plugged the solenoid in and left if for about 1/2 hour and the solenoid was way to hot to even touch for more than a few seconds.

 

[Zephrant]

Chris- Thanks, Good job on finding this.

JB- Yep, that was the exact description I was going off of. Next time I'll disassemble it first.

I have never noticed mine getting hot- Maybe I should run a test on that too. I recently turned down my water flow so the solenoid needs to be on much longer to keep the water at the right level.

Zeph

 

[cali_reef]

I had my feeding my sump for a few months now; I notice it will get hot to touch after few minuets only if it is on with no water flowing through it.

I have my kold-steril unit feeding to the valve (controlled with a float switch), the water than goes through a kalk reactor (built by Zephrant). My problem is the float switch is too sensitive, turning on the valve every few minuets and making a loud clicking sound. Any suggestions?

 

[cwa46]

Re: DIY Auto-Top off with Solenoid

Here is a quick run-down on building an auto-topoff system, that utilizes a float switch and an AC solenoid. This is designed to be used in a sump of any kind. In my case, I had an internal baffle that I could install the switch in, that did not need to be water tight. This switch can also be screwed in to a Ã"šÃ‚½Ã¢â"šÂ¬Ã‚ bulkhead directly through the wall of the sump if desired.

Tools needed:

• Drill
• 1/2" NPT Tap
• 3/4" Forstner bit
• Soldering Iron, solder, flux, heat shrink tube

First, go to www.mcmaster.com and order a Polypropylene solenoid- Part 7877K53, for $18.12 each.

While you are there, order an AC float switch- Part 46515K41, $15.13 each.
Zeph

What makes you believe a system with a float switch and a float valve is more reliable than just a float valve? If you utilize the switch to control your water level, the float valve is never used. What keeps it from building up material if it never closes? When the switch does fail, how do you know the valve will work.

If I were worried about reliability, I would install two switches and forget the valve. By the way, I have been using the valve alone for more than 3 years without mishap.

 

[Zephrant]

If the float valve is used with RO/DI water (as I specified) then there is nothing that would foul the valve. Occasional testing by lifting the float would help insure there was no problems.

I have since added a Kalk reactor, and removed the float valve on my system. I will be adding another fail-safe float switch some time in the near future.

Zeph

 

[Team Slug]

I just stumbled on this thread and thought I'd add a thought:

- With this setup it is definitely a good idea to have a redundant shutoff. Not absolutely necessary, but a good idea IMO. Why...because there is a chance that the contacts on your switch can weld themselves together making a permanently closed switch.

(Long and boring engineering lecture begins here...feel free to sleep ).

I think the contacts on that switch are designed for a resistive load not an inductive load which is what the solenoid coil is. The switch looks exactly like the one in the Grainger catalog that is rated at 30W resistive. Your coil is probably somewhere around 8-10W inductive when fully open. Unfortunately, a higher resistive rating doesn't mean anything to an inductive load. This coil acts as a big inductor that stores current. When the switch contacts close the current through them ramps up slowly until the solenoid is fully opened and the current reaches steady state. This is all perfectly fine. The fun stuff comes when the switch contacts try to open. Because of the stored energy when the contacts open there is a large surge of current. This surge can cause a not-so-nice little arc between the contacts. This isn't as large a problem on an AC coil as a DC coil, but it is there none the less. If the current spike is large enough it can weld the contacts together. This would be a bad thing as your solenoid would now stay open and merrily flood your sump/tank. There may be even more of a chance for this to happen with contacts that are slow to open or chattering due to waves messing with the switch. You could add a snubber in parallel with the coil that would pretty much eliminate the spike and also make the switch contacts last much longer.

(Boring engineer language over now)

That being said, I like the idea and still plan to make something very similar this weekend. That's because though it is possible it will fail, I think one of you guys would have had it happen by now if it was a high probability. I think the little snail taking a nap is more probable at this point. I'll just use a redundant switch just in case.

 

[cwa46]

I see two switches being redundant. A switch and a float valve do not necessarily make it fail-safe.

 

[Team Slug]

You are correct. There is no such thing as fail safe. You can just reduce the odds of something bad happening to acceptable levels, whatever that may be for each person.

 

[Zephrant]

The standardly accepted definition of "fail safe" is that any SINGLE component can fail, and the system does not.

With two float switches, and one solenoid, it is not fail safe as the solenoid could stick open (flooding the sump).

With one float switch and one solenoid followed by the mechanical float valve (set 1/2" higher), it becomes fail safe. Any one of the above components can fail, and the system will not flood.

Good write-up on contacts Slug-


Zeph

 

[Team Slug]

I'm going out to Grainger this afternoon to buy the parts.

I'll probably try a simple resistor and capacitor in parallel with the solenoid to try to reduce the spikes. That should help the contacts out quite a bit. I'll do some measurements before and after if I can.

I'm just using a 5 gallon bucket to hold my top off water, so I'm not really concerned with flooding. Now when I go to Vegas for a week and the 5 gallon bucket won't cut it then I'll be worried. It's doubtful that my fish will find it amusing that I am kicking butt at craps, but they are swimming in Lake Michigan. I have 4 months to tweak.

 

[cwa46]

[

 

[Team Slug]

Hey Zeph...

Does your solenoid open and close quite often? I noticed that the switch doesn't have any type of hyteresis built into it. By that I mean that the switch opens and closes at the same trip point thereby just as the water level rises high enough to open the switch contacts when the water level drops even a minute amount the switch contacts close again. This is creating a constant on/off cycle. It would be nice if it could allow the water level to rise an inch or so before turning off and drop an inch before turning on. At least that would alllow a few hours for evaporation before the solenoid trips again. This thing is driving me crazy. Every five minutes click-click-click...

The only way I can see to do it is with a second high level switch and a relay contact in parallel with the low level switch to hold it.

 

[Zephrant]

TS- I have a small surge setup on my upper tank, so the solenoid tends to open for 10-15 seconds just before a surge, then close again.

I do notice that when I have my hands in the sump, it bounces a lot and clicks open/closed on every wave. I tend to unplug it because the clicking is annoying.

I am purposely abusing this setup a little, to see where the failure point is. I agree that a little hysteresis would be great though. Perhaps a RC delay triggering a buffered relay?

A neat way to do it would be with a PIC. Setup a tiny circuit board with an input and do a clock-counting delay. Use another two inputs for high and low alarms... There are lots of neat things that could be done there.

Zeph