The Planted Tank Forum banner

Bulkhead/In-line Replenishment Snag...

18K views 125 replies 27 participants last post by  scolley 
#1 ·
For a long time I've been lusting after a form of water replenishment that doesn't require overflows. Something that would provide not only top-off, but could eliminate large water changes. For instance, if my math is right, if I had something that could automatically drain then add just 1 gallon to my 75g, 7 times a day, I would be turning over water at the same rate as a weekly 50% water change - no more water changes! And auto top-off just happens as a byproduct.

Great idea IMO - but apparently tough to accomplish without some kind of overflow.

When I set up my current tank, I drilled bulkheads in the bottom, and left one unused to help with this problem. I went and bought a couple of these pressure switches to help. The idea was easy...

Leave a bulkhead open, connected to the pressure switch. The switch would be configured in a normally-open configuration, and would only close when the pressure (water level) was below the desired height (pressure). Then it's just a matter of putting a normally closed solenoid in-line, and on a timer to drain water returning to the tank for a few minutes. When that's done, have another timer power up the pressure switch, which would detect a low pressure condition, and would turn on a pump from a reservoir until the appropriate water level switched it off.

I bought two of the pressure switches so I could put them in series, so that if either of them thought the water was high enough, the new water shuts off. Redundance seemed like a good idea. :tongue:

I've started testing these switches, and they do a pretty good job. In a test setup, over many trials, their variance in shutting off at a consistent water height is about 1/4". Good enough if for me, if used redundantly.

I'm writing because I've run into a snag though...

The pressure switches detect a "full" condition very well. The problem is state change. Once they detect the "full" condition, they open the circuit, and the circuit remains open until the pressure drops much, much lower. In my case, they won't close again until half the water is out of the tank, and then they faithfully turn on until they are within 1/4" (+/-) of the desired water height.

Bummer. So much for losing a gallon at a time, and letting these things control the top off.

I've come up with a solution, but it's so complicated, I'm not even going to post it. Too prone to failure. Any ideas? I'm stumped.
 
See less See more
#3 ·
Yeah, thanks. This would would do a water change too. The amount of pressure drop to get these switches to switch state (or reset) is almost exactly the number of inches of water my tank loses on a 50% water change - by coincidence.

But if I used this just for water change, the auto-top-off capability goes away. And then you have to get involved with using Prime - so it's not really unattended. With the smaller changes there are lots of possibilities, like no prime at all due to the small amounts. Or having a small (20g?) aging tank hooked up to the house water supply.
 
#5 ·
BlueRam said:
Try a peristaltic dosing pump that can handle two lines. Set one in, one out so that the volume is conserved and volume is set by the time opperated. 50 % a day may not be bad though if your water is cheap...
That's a great idea Blueram. But you loose auto top-off. But it is devilishly simple, and tempting...

But idea about more frequent water change is an idea though. Water is cheap here. And truth be told, in the winter (when it's so dry inside due the heaters) top-off isn't needed my often than 3 times a week. I'll bet 3 50% water changes wouldn't wast much more water than using the massively wasteful python I'm using today.

Would causes massive swings in ferts though. But I could set a different timing for my dosing pumps on "water change" days. Hmmm.....
 
#6 ·
I thougt the dosing pump does "top off" service.

You might be able to rig somethink like this in the cypress stump:
http://www.bluelineaquatics.com/products/topoff_system/

My sink is lower than my tank, so I use* the python to start and then turn off the tap and let it go. *now I just mouth-syphon and dump the open end into the tub which is even lower...

Very expensive example:
http://www.kc-denmark.dk/public_html/Small pumps for deep sea/Peristaltic pumps/Peristaltic_pump.htm
Check medical surplus...
 
#7 ·
BlueRam said:
I thougt the dosing pump does "top off" service.
Not for me... and my hat's off to people trying to make that work. My evaporation rate is way to inconsistent for me to try that. All depends on the heat, and that varies significantly.



BlueRam said:
You might be able to rig somethink like this in the cypress stump:
http://www.bluelineaquatics.com/products/topoff_system/
Seen those. Thanks! What I'm trying to do now is quite similar, except I've got a water pressure sensor at the bottom of a tube that passes through the bottom of the tank, rather than something over the top with an air pressure sensing on top. I suppose if I had set it up initially, I could get one of those bulkhead airline connectors to make that work. But messing with bulkheads now means tearing down the tank. Not an option for a year or so.



BlueRam said:
My sink is lower than my tank, so I use* the python to start and then turn off the tap and let it go.
Not an option for me I'm afraid... sink's too high.



BlueRam said:
Yeah, massively expensive! The dual perstaltic idea makes me uncomfortable, due to the "some of the water that just went in a second ago, is in the water going out now" problem. Kind of clouds the question of just how much turnover you are really seeing.



Thanks for the ideas though. Any ideas on how to get what I've got workind? How to get beyond that 10" of water state-cycling problem?
 
#8 ·
Can you set one (or a pair) as high and the other low. Something like a drop of 1" turns the pump on until the high (the one you have rigged) turns it off?

Or are you saying it takes 10" for the high to sense.

You might be able to pump water in and use the high to open the bulkhead?
 
#9 ·
First, they can set to be either normally open, or normally closed. In fact, you could put two circuits on one at the same time, and when an appropriate pressure was reached, the open circuit would close, and the other the reverse. They are pretty flexible. And pretty accurate.

Here's the problem...

Let's say it knows that the pressure threshold has not been exceeded (water column not high enough) and it closed a circuit, turning on a pump to put water in the tank. The pump will remain on until the appropriate height is reached, then the switch opens the circuit, and the pump turns off.

On the next day, you drain a bit of water out of the tank, and assume the switch will close the circuit again and start the pump to replenish water again. Nope. Once that that pressure sensor opened the circuit (the day before) it will not close it again until the water drops about 10"!

So yeah, I've figured out a kludgey plumbing mechanism to fool it into thinking that has happened. But it includes solenoids, drains, and just way too much wackiness. There has to be a simpler way.

I've thought about using both of them, as you suggest, one as a high, the other as a low... but it doesn't really work, I don't think. No matter what you rig one to do, turn on or off, at a given pressure, you still have to lower the pressure about 10", where it switches to the opposite state (off to on, or on to off) to get it to do that trick again.

It could be the solution to my problem is a refund.
 
#10 ·
Float Switches!

One (or two for redundancy) wired normaly open at the top of the tank that activates when the water level drops. Instead of relying on volume and pressure the switch would activate simply upon water heighth. Auto top offs too!

Then possibly you could use your pressure switch on a timer that would drain water to a certain level. As long as you could get water out of the tank faster than you could get it in you wouldnt be wasting too much water.
 
#11 ·
JimmyYahoo said:
Float Switches!

One (or two for redundancy) wired normaly open at the top of the tank that activates when the water level drops. Instead of relying on volume and pressure the switch would activate simply upon water heighth. Auto top offs too!

Then possibly you could use your pressure switch on a timer that would drain water to a certain level. As long as you could get water out of the tank faster than you could get it in you wouldnt be wasting too much water.
Yup...gnatster has a similar method employed on one of his tanks. He has his sump setup with a float switch so when he drains it it automatically fills up with water from his RO unit until it hits the switch again.
 
#12 · (Edited)
Yup. Thanks guys. Float switches are the obvious answer. I suppose what's not so obvious to everyone is that I can't use them. I'm on an aggressive campaign to eliminate anything that hangs over the side of the tank. And I want any equipment in the tank to be invisible. That's why I'm going to all this silly pressure switch trouble.

And a float switch either has wiring hanging over the side (against the rules for me) or it could be threaded up through a bulkhead, but even then, it will be hanging out at the surface for all to see (again, against the rules for me).

Thanks for the suggestions though. It would be a lot easier if I'd back off on the aesthetics!
 
#14 ·
Thanks JimmyYahoo!


WOW! I may have just figured something out...
I could stick a "T" in the pressurized water line just over the switch. Then hook a little air-pump up to the T. Just before powering up the switch (to turn on a tank filling pump), I could turn on the air pump for minute, just long enough to fill the tube with air, and hopefully reset the switch.

Kinda kludgey, and would send a few bubbles out the manifold, but it might work!
 
#16 ·
zzapd said:
If the switch is resetting after a ten inch drop that is pretty good for a switch of that type. The 10 inch drop is only a change of .4 psi. A quality
industrial switch that costs about 300 will only get you down to a five inch
dead band.
Thanks! I was worried that I had bought something cheap. I guess I really didn't know what I was doing (again!), and am finding out along the way!

"Dead band" huh? That's what that range of pressure is between the two different spots where it switches state?

And it could be a 12" drop. I only did a quick measure once I figured out it was about 1/2 the height of my 75g.
 
#18 ·
Thanks BlueRam. I've seen those, and think they are a reasonable value solution. (I bought the better of my two peristaltic from those guys)

Problem is the "through the bulkhead" thing. Nothing else is going through a bulkhead without tearing down the tank. So I figure that's out now. And I'm kind of adverse to putting equipment in the tank anyway. That's why I find that new thread of Tom Barr's (about the CO2 microbubbles), it's all leading back to more crap in the tank. I don't like hiding stuff. I'd rather have as little as possible to hide.

Also, the range is kinda high. It's got that "dead band" that zzapd mentioned of 1.5 inches. Just like mine, it has to drop 1.5 inches before it know there's been a change (mine drops 10-12 though!). With 1/2 inch variance, that could be a range of 2.5 inches. Not terrible. But I'd like to get it tighter if possible.

But thanks for beating that horse!


I've got to try the "air pump" on a "T" that I mentioned a few posts ago. It just might work...
 
#19 ·
I could post this elsewhere, but I think the reactor is not a problem. From what I gather, the mechanism of getting CO2 dissolved changes throughout the photo period. For example, early, CO2 goes in just fine, however, at saturation, bubbles form in the reactor. Theses can either stay there (they make some noise), be blown out large bubbles (mine runs ~5* tilt. I can purge gas mid day), or the bubbles can be chopped up and run through again. I am happy with my reactor and am in the process of confirming Tom's observations. Not quite a full fledged idea yet, look for more in the proper post.

Your tank looks great!
 
#20 ·
Thanks pal. But the tank does NOT look great. It's got algae bad. What it does apparently is photograph well. I'm working real hard to remain calm on the algae issue. The BGA is back, and making me insane! My CO2 is in the forties, and most of the plants pearl everyday, so growth is good. It's just good for the algae too...

And to further the "I should post this somewhere else" theme...

I'm in full agreement about the CO2. I'm in the process of assembling an in-line device to inject those microbubbles in to my return line before getting blown into the tank. Parts on order now.

But if this idea that dissolving stops after some level of saturation is hit (30 ppm?) is not correct, then I just wasted some more money. So I look forward to your confirmation.

Now that the money is spent, it'd be nice to not waste time too.
 
#22 ·
Yeah Sean, all in-line is a major PITA. That was my compromise, everything being a PITA in order to get stuff out of the tank.

Next tank will have a durso w/sump underneath - the compromise will be always having to hide the overflow. But everything else will get MUCH easier! :proud:
 
#24 ·
Lump - I gotta give it to you... THAT'S thinking out of the box! My reactor is not clear, it's white. But even still, when the CO2 is off, there should be no bubble there. And when it's on, the size of the bubble should vary with the speed the needle valve has been set to blow bubbles. And it the rate of change will vary as the water flow through the reactor varies with equipment added/removed in-line, and with the filter gunking up.

But it's an interesting idea.

I've pondered doing something similar for a while... taking an in-line container with a pressurized air pocket in the top and putting a float valve in it. But that has a whole lot of problems. Getting it pressurized, maintaining constant pressure, the size of the pocket remaining the same (not being subject to out-gassing or absorption).

It is a path that could be pursued. Thanks for the suggestion - it was certainly innovative. But I'm just not that courageous.
 
#25 ·
Not that I would want to build this beast – but…

You could use a pressurized chamber under your tank to remotely connect the float valve.

Oh latter thought - you would ultimately be trying to build a supper sensitive pressure sensor - no bets on if it would work.


Ok, figured out a way to possibly make a more sensitive switch, do not have time to draw it out though. Same basic concept as before, but with 2 differences. Increase volume of presure chamber - but decrease the diameter of the tube where the float switch is.
 
#26 ·
I think that's BRILLIANT jgc! But I don't think it would work...

Atmospheric pressure variations outside the tank would cause the measured level inside the chamber to swing - possibly significantly. It would take someone better informed than I to figure out how much.

If that variance proved to be small, I'd build it. Then the only thing that would have to be overcome is some sort of adjustment mechanism. Something to ensure that the float was sitting at the right level...

Maybe a very sensitive gas escape valve at the top, and a drain valve at the bottom, and a clear chamber, would be enough to allow you to visually align the water level with the float valve.

Great idea! I wish I know how much atmospheric pressure would affect that. :confused:
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top