DIY Recirculating skimmer - suggestions wanted

[spuzio]

I recently finished my 12"x36" chamber that made a 48" tall skimmer overall. I am running 3 GENX6000 pumps on it, however take a look at how the pumps are mounted on the euro-reef website. These pumps will not pump enough air at all if the output is mounted too far down. According to the measurements the pump output should be about 1/2 way up the reaction chamber and point downwards to increase contact time, and still be able to pull in a maximum amount of air.

 

[mattboy]

I'm having a hard time understanding the head pressure discussion here; a recirculating skimmer is essentially a closed loop; the input and out are submerged in the same vessel of water, the level of which is kept stable by something different. I don't see where there is any head pressure on the recirculating pump at all. In terms of pump flow, it should not matter where the pump is mounted. Now, I understand the air draw; try sucking on a hose while ten feet underwater, and you get an idea of how much suction it takes to draw air down into water, which is sort of what we're doing. If you shut the recirculating pump off but leave the skimmer filled, the venturi air hose fills up with water to the level in the skimmer, right?

Regarding contact time in a recirculating skimmer, it should have nothing to do with the recirculating pump configuration. Contact time is strictly an issue of how much flow is coming in and out of the tube (not recirculating) and the volume of the tube.

 

[spuzio]

A recirculating pump is not a closed loop because of the air pipe. As a result, the water is contantly trying to flow up the tube because of water pressure. So... the venturi must first over come this pressure before it can suck in the air.

 

[mattboy]

Right, I understand the venturi air draw situation, but in terms of water flow through the pump, it is like a closed loop.

Here's a question; let's say I wanted to incorporate the water input to the skimmer with the input of the recirculation pump, not unlike how the water into a calcium reactor is typically fed via a T connection into the input of the circulation pump. Of course we're talking about a much higher flow through than a reactor, but would this work?

 

[shelburn61]

I can't really see any benefit to that? It is better for the raw water to come in from the top so that it can be skimmed as it flows down the reaction chamber. New water in. old water out.

 

[mattboy]

If I'm not mistaken, the water needs to be turned into foam to get skimmed; that happens as it passes through the venturi and needlewheel. The benefit of combining the inputs (new water from tank and recirculating water at the pump input) is that the water from the tank immediately gets turned into foam. The benefit, as we know, of the recirc is greater contact time; we can use a big pump which creates alot of foam, but regulate the fow through with a smaller pump (or valves and gravity) so that water stays in the skimmer longer.

Now, can anyone see a problem with teeing off the pump input plumbing, maybe running a PVC pipe up to water level so there's no pressure, and pushing input water into the skimmer that way?

 

[prugs]

Shelburn61,
How does your skimmer skim?
I was looking at all your skimmer pics, trying to figure where the skimmer was getting its water supply. Does the flex hose on the right side of your skimmer go to your overflow?
Thanks Tom

 

[GROSSR]

So the pump is mounted in the middle somewhere. The input of the pump is taking water off the bottom, passing through the venturi, then the needlewheel, then into the center of the body.
The output of the skimmer draws its water from the bottom of the body(least number of bubbles down there) and exits at a point above the input.

Do I have it correct?

rich

 

[prugs]

<a href=showthread.php?s=&postid=6682494#post6682494 target=_blank>Originally posted</a> by GROSSR
So the pump is mounted in the middle somewhere. The input of the pump is taking water off the bottom, passing through the venturi, then the needlewheel, then into the center of the body.
The output of the skimmer draws its water from the bottom of the body(least number of bubbles down there) and exits at a point above the input.

Do I have it correct?

You got it, a recirculating skimmer. Tomorrow I'm on the hut for a 5gal water jug.

 

[shelburn61]

prugs,
yep water comes in from overflow through flex. The large loop of pvc there is the drain called a hartford loop. It can be raised and lowered using an expander coupling to adjust water level. much more reliable than a gate valve.

mattboy,
think about it this way: you are not turning water in to foam. the pump is just a means to create bubbles. skimming takes place by the bubbles floating to the surface up through the column of water in the body. each bubble attaches to crud as it rises to the surface. you want the "dirtiest" water to remain in the skimmer as long as possible so that the most bubbles can flow through it and extract as much as possible. This is the same concept as the old counter-current airstone skimmers and it applies whether you are using a beckett, needlewheel, or whatever

If you have a pipe entering the bottom of the skimmer you will have head pressure no matter what level you put the other end of the pipe (unless it is a true closed loop with no outlets). The whole weight of the water column in the skimmer is trying to push down and out the easiest path possible (your pipe) and to get water to flow into the pipe you have to push against all that weight.

 

[shelburn61]

If you do a search (if it ever works) for "bubbleking" on this forum. There was a great discussion in a thread a while back on what makes skimmers work well.

Here's another. Spazz knows what he's doing.
http://reefcentral.com/forums/showthread.php?s=&threadid=605196&perpage=25&pagenumber=4

 

[mattboy]

shelburn;

I'm going to think about the water input for the skimmer some more, but I understand your point about the bubbles flowing through the water. Although, I still think the cleaning occurs primarily by water turning into foam, which drastically increases the air-water surface to which proteins are attracted. Certainly in the many beckett skimmer I built all the input water goes through the beckett and turns into foam, and they work really well.

I'm pretty sure about the head pressure; if you turn on the recirc pump, turn it off, whatever, the level in the tube does not change. You're simply not pushing water up to any height. The venturi pipe and the outlet pipe allow water to seek it's level in the tube, but the pump is not pushing water into the tube or out of it. It's just like a closed loop on an aquarium with an open top. Think about it this way; sure the weight of the water is pushing against the pump output, but it's also pushing against the pump input, negating the pressure against the output. I don't think the presence of the venturi pipe changes this. I'm pretty sure that the air resistance has to do with drawing air down into water; think about a weak air pump pushing air through an airstone; near the surface you'll get a certain amount of bubbles, down deeper far fewer and smaller bubbles. This is similar to the pressure dynamics that cause divers to go through more air at depth; the air pressure has to match the surrounding water pressure.

Here's another more practical question, though. I got the sedra 5000, and the fittings are just slip fit. How do you seal this for use externally? Teflon tape? Glue on a slip fit union? Just trust the slip fit? I don't know about that one.

Anyhow, I'm not trying to be argumentative here, just trying to understand this better before I make my skimmer.

 

[GROSSR]

Glue what you need on, but, you should connect the pump with unions. That allows you to change the pump or clean it .

rich

 

[shelburn61]

What exactly do you think foam is matt?

The venturi makes it an open loop = head pressure. Take the airline off the venturi and get real close so you can look directly into the venturi inlet. Then turn the pump off. Let us know what happens.

I'm not going to argue with you, but I know what I am talking about on both counts.

 

[GROSSR]

Water will go to its point of least resistance. Watch the water flow up the air line equal to the highest point. That is why on the recirc skimmer the air line is secured and goes way up high.

rich

 

[mattboy]

<a href=showthread.php?s=&postid=6685525#post6685525 target=_blank>Originally posted</a> by shelburn61
What exactly do you think foam is matt?

The venturi makes it an open loop = head pressure. Take the airline off the venturi and get real close so you can look directly into the venturi inlet. Then turn the pump off. Let us know what happens.

I'm not going to argue with you, but I know what I am talking about on both counts.

I understand if the venturi airline were not attached, all the water in the skimmer would drain out to that point, but since it is attached, the line fills up with water to the tube level when the pump is off, then the whole system is equalized as far as water level goes. I'm not arguing that the venturi has to work to pull the air down beneath the water level, and that does in fact raise the level in the tube a tiny bit, but other than that I still don't get the head pressure bit. With the pump off, the tube does not empty, which means that the pump is not responsible for filling or holding water in the tube. The introduction of air into the pump causes friction for sure, and there is the pressure imbalance of air in the water that wants to rise to the surface. But it's not the same as pumping water from one level (i.e sump) to a higher level, i.e tank.

Anyhow, we may be getting into semantics here. When I get the skimmer made, I'll have a better idea of the details of it's workings. Still trying to decide how to best introduce water into the skimmer without drilling a hole in the side of the tube; I don't want to do that, and to be honest, I'm surprised that there are $1000 (or more!) skimmers out there with multiple holes drilled in the main tube, which no matter how you do it, still represents a structural weakness IMO.

Thanks for all the help so far, keep it coming!