Reducing cavitation in needlewheel pumps

metalManiac · #1 08/17/2006, 10:30 AM

Im making a DIY skimmer and of course with it a diy needlewheel pump (im just running it aspirating at the moment, but will turn it into a needlewheel via bioball mod if it would help with decreasing cavitation).

Im feeding the pump by using a ball valve to restrict waterflow (and in turn increase input of air). It works well but the interval between no bubbles and cavitation is very low, i must play around with a combination of ball valve restriction and needle valve air restriction to get the pump not to cavitate while making some nice bubbles. Is there an easier way to increase this interval (so that the damn pump does not cavitate so easily).

Also, If i were to build the mod the pump with the bioball, could i simply adjust the throughput of the pump buy slowly/ systematically removing "needle points" until the pump cannot push as much water while still maintaining an effective uptake of air.

Thanks in advance guys, hope it makes sense to you what im trying to say here...

Pbrown3701 · #2 08/17/2006, 11:17 AM

the reason the unmodified pump impeller cavitates is because it is designed to push water. The veins build higher pressure which cause any air to interfere with it's opperation. By removing the veins and replacing with needles, the pump will not cavitate so readily.

metalManiac · #3 08/17/2006, 11:20 AM

thank you very much pbrown! thats the sort of answer im looking for!. Anyone else with some other simple facts that can help me out in the long run?

RandyStacyE · #4 08/17/2006, 07:00 PM

I wouldn't use a ball valve to restrict the pump inlet. I would only adjust the volume of air if necessary using a small valve on the air line leading to the venturi. The more water = more air. The trick is to find the best ratio if the air can not be left wide open.

What pump are you using?

GROSSR · #5 08/17/2006, 07:46 PM

I do use a gate valve on the input side to restrict turbulance. If I run the OR3700 full blast and full air it won't run. Restricting the input helps me to balance the air and water and turbulance in the skimmer body.

Works for me.

rich

metalManiac · #6 08/17/2006, 07:58 PM

the pump is a hailea (basicallyOR ripoff), they are made by the same factory that makes ocean runners.

The pump is only around 300gph. The ball valve is not being restricted at the moment, im just adjusting the air intake via a needle valve.

There is no way the air can be left wide open, it cavitates at a point much sooner than the point at which the air valve would be left wide open.

BTW i made a deltec mce600 ripoff, could the fact that the pump must draw water through pvc (and hence the difficulty of getting water to the pump) be causing the premature cavitation?

heres a pic of the skimmer (note the ball valve and air intake right after it).

closeup of the pump and the way its plumbed.

RandyStacyE · #7 08/17/2006, 08:03 PM

GROSSR,

That is true valves are used by many. Considering all a venturi is, is just a tiny ammount of restriction. I once used a barbed fitting to make a venturi. I just cut off part of the barb. It created a small 'step down' to act as some minute restriction. You can see that in this pic the hose is in the wrong spot

so I show a red do where it would work better. I used too small of a hose too

my pictures are all screwed up.

I assumed it's possible that metalManiac may have an MJ or something on the small side where a valve may be not so good of an idea.

metalManiac · #8 08/17/2006, 08:24 PM

i was thinking of maybe drilling a hole right where you have that red dot randy, that way i create a semi recir while adding flow to the pump. (im tryng not to get too much flow through the whole skimmer because it becomes a microbubble factory (yes, even with all those baffles). (the skimmer is sitting on a 2ft tank (so you can get an idea of the size).

Roland Jacques · #9 08/17/2006, 09:15 PM

this might not have anything to do with your question but.

how do you get that thing started pumping? it looks to me that you would have to close your air intake to venture get the pump running.

that pump has to pull 8" of water to get going. while their is a air "leak" in the pipe? (venture) that is questionable to me. at the very least i would put that venture way under the water line(low as possable) at least that way your pump would have a chance of starting on its own and working. then again im just guessing i never try a config like tha before.

metalManiac · #10 08/17/2006, 09:31 PM

to start it pumping obviously you need water in the back of the skimmer (enough to cover the pump). Then i just close the air intake (via needle valve) and let the pump do its thing. If it still does not happen then i need to draw out the air from the pipes by sucking on the air hose!

RandyStacyE · #11 08/17/2006, 09:33 PM

I think you could get away with exactly what you have, but drill a hole right before the pump and run your air line up out of the skimmer. Was that what you had in mind?

Are you drawing air from the top of the PVC through that little line? If so, I bet thats your problem right there. I'd only use that line from now on to prime the PVC and drill another one just before the pump.

metalManiac · #12 08/17/2006, 09:45 PM

thanks for the info ill give it a try now and report back on how it goes!

Roland Jacques · #13 08/17/2006, 09:47 PM

that what i thought. put your venture right befor the pump like in randys photo. you can get a Octopus venture srcew right on to that pump. it is a good venture the other end of it slips right into 3/4" pvc fittings.

metalManiac · #14 08/18/2006, 12:27 AM

ahhhhhhhh,, much better! Thanks Randy!

I leave the air hose all they way open and i have to restrict the ball valve to get more out of it. Very simple to adjust now!

Thanx for the help!

RandyStacyE · #15 08/18/2006, 10:11 AM

Quote:

Originally posted by metalManiac
ahhhhhhhh,, much better! Thanks Randy!

I leave the air hose all they way open and i have to restrict the ball valve to get more out of it. Very simple to adjust now!

Thanx for the help!

Actually no. I think that the long plumbing on the pump's suction side will create PLENTY of restriction all on its own without closing the ball valve at all. I think that if you drill a new hole and install a NEW air line 'just before the pump inlet' you would be just fine. You could also use a small valve on that new air line to adjust the amount of air. If I had to make a prediction, I would predict that you would NEED that valve on the new air line or it will suck way too much air because of the long plumbing on the suction side of the pump.

Here is what I mean:

RandyStacyE · #16 08/18/2006, 10:42 AM

I had another idea. You mentioned “semi recirc” before and this popped in my head.

Since you have such a long suction line, I bet that your pump looses some performance. You could disconnect your ‘look-alike Ocean Runner’ from the plumbing entirely, attach an aspirating venturi to the inlet of the pump and just let the pump sit there submerged as your ‘recirc pump’.

Next, install a pump to drive your skimmer from inside your sump. This can connect to the plumbing which you already have, your ball valve won’t go to waste and you can use it to adjust the volume that runs ‘through’ the skimmer.

Like This:

Eiter way would work, but I'd bet that the 'recirc' method would give you maximum performance from the Ocean Runner Look-Alike. Especially when its modified to a Needle Wheel.

GROSSR · #17 08/18/2006, 11:10 AM

I ran the airline from the pump into the collection cup of the skimmer. Lifereef does this. If the cup starts to fill up the pump starts to draw waste water and can create bubbles to skim.

rich

Barry L · #18 08/18/2006, 12:02 PM

PBrown, said that the impellor cavitates because it was designed to push water! Thats true! but why not drill small holes in the impellor to reduce the pressure, and help break up the bubbles even further?

hahnmeister · #19 08/18/2006, 12:51 PM

lol, yeah, that air valve where it once was will simply cause any syphon that the pump forms to constantly break...lol. Easy fix.

One thing I like to do to prevent that on MCE clones is to drill some holes along the intake pipe just before the venturi. Then it behaves like a recirculating skimmer and you can pack an even larger pump in there... less microbubbles. Nice design RandyStacyE. The OR type pumps have alot of velocity at the outlet Ive noticed... not calm and gentle like some others, so you might consider some sort of diffuser on the skimmer pump outlet... at least a 'T' or something like that.

RandyStacyE · #20 08/18/2006, 01:07 PM

In a way I can disprove that theory.

MaxiJet pumps have a special inlet kit to connect an air line to the pump's inlet so it can be used on skimmers. They use paddle impellers.

For the last 2 weeks I have been running a DIY skimmer using a GX-2400 pump using the standard paddle impeller and was having perfectly acceptable results:

Here’s the link to the diy page:
http://randystacye.com/diy_needle_wheel_skimmer.htm

The thing is, the whole time I though I was using the needle impeller! I wasn’t, I just forgot to put the needle impeller in there. Today I just put the needle impeller in and I’ll be able to see how it performs in comparison.

The way metalManiac appears to have his pump plumbed; it must draw air from way up on top of the stand pipe that goes up and over the skimmer and sump. By doing this it would not suck air anywhere near as well as it would if it were ‘just before the pump’. Also the way its plumbed, the pump may choke on very large pockets of air which described the intermittent surge of ‘bubbles … no bubbles … bubbles … no bubbles’.

hahnmeister · #21 08/18/2006, 07:09 PM

That is correct, the impellers on MJ pumps are very small and at 18000rpm are plenty 'mixing' for skimmers. I have been using them 'as is' for years on my smaller DIY skimmers.

metalManiac · #22 08/18/2006, 08:49 PM

Hi Guys, well because im in Australia i get this info in the morning so i cannot report back when you guys write but here it goes.

I have written this post after i did the modification you said to do. (the one you drew on the next post)

Quote:

ahhhhhhhh,, much better! Thanks Randy!

I leave the air hose all they way open and i have to restrict the ball valve to get more out of it. Very simple to adjust now!

Thanx for the help!

And while the pump does draw sufficient air, i still restrict the ball valve in order to get the pump to draw in even more air (without cavitating the pump).

Randy, as to you second post (the recirc mod), the idea that i had was to actually drill one or two small holes in the plumbing just before the inlet, this way the pump draws water from both the tank and from the skimmer chamber (i think i would haveto be carefull with the hole size because if i drilled a hole too large it would just recirc and not draw any new water from the tank). The reason i wanted to do this is because the ammount of water the pump draws at the moment (and throws through to the frag tank/refugium bit) is ridiculous with microbubbles escaping throughout + it would eliviate the pressure on the pump (from having to draw in the water through the piping).

I really did not want to add another pump (especially in the tank) as this is going as a hang on in the main tank (which is a 2ft).

hahnmeister, yes, what youre saying about the recirc design is what i was thinking!, as for the diffuser or "T", have a look at the pics

. (i did not want to build a diffuser because i have read that diffusiers are inefficient on low diameter skimmers).

As to the type of impeller running on the pump, i have modded the OR clone with the bioball method but have tried and opted for a pondmaster 800L pump with its paddlewheel (no needlewheel mods here). It seemd to be able to handle cavitation better plus the throughput through the skimmer system was better and there are less microbubbles.

I also placed a rough sponge on the outtake baffle of the skimmer (so no microbubbles enter the frag tank) and it raised the water level by 1/2". Im not too fussed as i have plenty of room to work with in terms of the collection cup (the collection cup is a mix of the remora and deltec, with the O-ring idea of adjusting height and the lid of the skimmer taken form the remora and the general look of a slope for a riser taken from the deltec).

Also, i have on purpose placed the holders of the skimmer low so that i do not haveto add another light to the frag tank (i will used the 150W halide thats on the main tank to illuminate the frag tank as well).

And one more final Question, the outtake produces bubles (not microbubbles from the skimmer but its own actual larger bubbles), if i placed a T fitting at the top instead of an elbow (aka durso style) would it help eliminate those bubbles?

Having said that i dont really know how well this thing performs in comparison to other skimmers since this is the 1st real simmer i have ever built. Do you guys think it will be enough to handle a standard 2ft tank with a mixture of SPS, LPS and softies?

(the skimmers chamber dimensions are 4.5" x 5.5" x 20")

As you can see, the background is black and when i turn the pump on and tune it in you can JUST barley make out the background color from the water (it does not look a brilliant white but a tad greyish, keep in mind its only 4.5" wide). The top of the skimmer has no turbulence at all, just microbubbles rising and popping (a large ammount of steam like material comming from this popping).

Again guys, thank you for your help and throught provoking submissions!

metalManiac · #23 08/19/2006, 09:28 PM

noone at all has any input?

RandyStacyE · #24 08/19/2006, 10:12 PM

Oh yeah, I bet a T would make a difference. What size output plumbing are you using? Lets just say it's 1". You could use a 1" x 1-1/4" T. It would do even better to help separate the air/water.