Most efficient DIY needle wheel?

[scottward]

In conjunction with an AM1000, I have CO2 misting through a powerhead/spray bar in my 100g tank.

I am feeding CO2 directly into the impellor chamber (i.e. via the pump intake), and allowing it to push the mist down a spray bar and into the thickest parts of my tank. So far so good.

I was thinking of swapping the standard impellor for a 'needle wheel version', but there isn't such a thing for this brand of powerhead (Resun 3800sp).

So, I was thinking of buying another normal impellor and DIY'ing a needle wheel.

I have read the threads in the past, the hot nail trick, the 'furring' up etc.

I was wondering, what is the most efficient way of making a needle wheel - not necessarily the quickest?

I have a dremel - would perhaps cutting small slots along the top/bottom of each blade be the way to go? I think that the slots should not be in exactly the same place on adjacent blades? How many slots etc?

Is the time spend converting a standard impellor into a needle wheel version going to be worth it? My understanding is that it makes the bubbles considerably finer, and, the finer the bubbles, the easier they disolve and are therefore utilised by the plants??

[shoggoth43]

If you want a very good method for this you may want to look up the mesh mods the reefers use. They will tie material like enkamat or similar onto the impeller to cut up the bubbles. The major problems with this method are that the mesh will wear out as it tends to rub on the inside of the volute, and you will have less flow through the pump. How much less flow and how often to replace it are things you would have to find out the hard way. The size of the bubbles you'll get will be considerably smaller however so it may be worth trying. The mesh may not be easy for you to source, although I have seen "kits" available for a reasonable price ( for here anyway ).

-
S

[mi5haha]

you can use king2 needle wheel type pump directly. it is 20W power output, with good water inlet, air sucking on water inlet, and water outlet fitting, so this can be place outside the tank (dry running) if it is linked with a sponge filter. bubbles are so small.

[scottward]

The mesh mods sounds interesting. I will look into this some more. Better than modding the impellor blades?

[scottward]

I suppose it's all about the efficiency/flow rate trade off.

If I could modify my impellor such that each blade looked like the mouth of a beleen whale, sure, I'd get excellent CO2 dissolving, but at a huge cost to flow rate (and would clog up more easily). On the other hand, a few holes burned/drilled into each blade won't be quite as good at chopping up and dissolving the CO2 as the 'baleen' version, but is still better than no holes at all, yet still provides good flow and doesn't clog easily.

So for me, the decision comes down to making holes in the blades, cutting slots in the blade, or snipping the blades down the centreline.

All methods are fine for me to do, I have a dremel, very small drill bits etc.

Tom indicated that he has found snipping the blades down the centreline to be the most effective, but I have heard others say that they have had problems doing this due to the blades not staying put once they have been bent etc. I suppose different brands of impellors, different thicknesses etc, play a big part in what works/doesn't work.

So, I'm thinking the most sure fire thing for me to do is to take the impellor, and just drill (or I might melt) a series of holes down the centre line of each blade, slightly staggering them between blades. I've read enough on here to indicate that this should work and is easy to do.

I could snip the blades but there is a risk that I might not be able to bend them very well (i.e. I might be one of the people that isn't successful with this method).

I could cut slots into the blade with the dremel but this seems more fiddly and probably wouldn't be any better than putting holes in the middle.

Attaching mesh etc to the blade seems, from what I have read, to cause more trouble than it's worth, due to clogging etc. I don't want something I have to clean all the time.

So - would the general consensus be to drill holes? Easiest? Most risk free? Zero maintenace (i.e. no cleaning)?

Scott.

[Gerryd]

Hi Scott,

From my reading it appears that both methods work very well. I think it lends itself well to experimentation with fuzzing it up, etc. Different materials, etc.

I just now finished a temp install of a DIY needle wheel on my 180 gal. Here are the specs:

Current c02 method: Mazzei 584
New c02 method: Rio 1000
DIY method: Center cut and bend

I just cut down the center about 70% of each blade using a small pair of wire snips. I gently bent each one back away from each other. The 'tops' pointed one way and the 'bottoms' pointed the other. I can see where the 'bends' may straighten out over time....it will become part of the general maintenance I guess like cleaning a disc...:)

I mounted the Rio sideways in the tank on one end below the MP20. I mounted the c02 line about 1-2" BELOW the Rio intake so the c02 bubbles get scarfed (a technical term) right into the intake.

Result:

Very nice mist. Certainly comparable to what my mazzei provides. Flow seems fine but not as good as new. It DOES make a bit of a popping sound when the c02 bubbles are chopped but silent with c02 OFF. You have to be close to the tank wall to hear it...

I lowered the bubble rate as if too fast bubble build up and escape out the intake on the 'top' side....

So, I will adjust/monitor/compare using this method to the mazzei and see how it goes.

I will report my progress...

I do have several questions however.

All,

1. What seems to be the best way to introduce the c02 to the pump? Simply drill into the outer casing so I can slip the c02 tube inside? Or is my current placement/method viable?

Tom,

1. On your 180 are you ONLY using a DIY Rio 1000 series for c02 injection? Or in conjunction with a disc?
2. How do you direct the Rio c02 flow? Out into the general tank to get picked up by the current?
3. What bubble rate do you feed into the Rio?

I had previously modded a Mag 500gph with the drill holes in the blade method. It did not work that well for me but I may also be an idiot....I have about 4-5 holes in each blade and I staggered them up/down...

Thanks in advance.

[scottward]

Hi Gerry,

Thanks for your feedback; appreciate it.

I know if I put some questions here somebody reading this might reply to my questions and miss yours. I will put a reminder at the bottom...

I have some questions for you Gerry:

- I wonder if cutting through 70%, or all the way to the root of the impellor makes much difference? With respect to both the 'chopping' ability and the possibility of the blades slowly moving back to their starting position.
- I wonder if the thinner blades would suffer from flexing and eventually snap off at the point of bend? I think holes drilled would be less likely to suffer from 'plastic fatigue'.
- Did you note what the mist output was like on the Rio 1000 before you did the mod?
- Is the mist so fine that the bubbles don't float to the top any more?
- With your Mag 500gph where the experiment with the holes didn't work out (I do remember reading your posts about it a long time ago), what was the reason it didn't work out? Was the mist not fine enough? Was it just the particular type of powerhead?

And my answer to your 'All' question
- What about connecting an airstone to the CO2 line such that rather than single noisy bubbles moving into the impellor chamber a more steady stream of smaller bubbles is introduced? I guess the downside of this is that airstones can clog up.


And I'll add another 'All' question
- holes vs snipping the blades; I think I can appreciate why the blade snipping would in theory work better, we're creating more pathways for water/bubbles inside the impellor chamber (i.e. water/bubbles stay in the impellor chamber longer hence reduced flowrate). Is that correct?

** Note Gerry's questions in the previous post, I've just 'tacked' onto his list ** :-)

[Tom Barr]

Tom,

1. On your 180 are you ONLY using a DIY Rio 1000 series for c02 injection? Or in conjunction with a disc?
2. How do you direct the Rio c02 flow? Out into the general tank to get picked up by the current?
3. What bubble rate do you feed into the Rio?

.

I use a Rio 1000 on my 180, I do this:

http://i171.photobucket.com/albums/u312/plantbrain/resizedDIYneedlewheel2.jpg
http://i171.photobucket.com/albums/u312/plantbrain/needlewheelDIY1.jpg


So I went from 5 blades to => 15 thin small blades, bending them back to get more distance between them.

This makes hardly any sound and makes a very slight haze from mist that's tough to see/notice.
This tank also does the best over time vs the disc and reactor method sI have on other tanks.
I'll likely switch to this for all the tanks since I use small rios or other supplemental flow.

I've tended to do longer term test recently and this one passes the muster.

I direct the flow in the lower corner along the back wall the length of the tank, this hides to powerhead and gives the longest fetch for the mist/gas dispersion.

The bubble rate is pretty fast, dependent on the tube sizing.
Maybe 4-5 a second? I have venier handles and use those and slow adjustment, not bubbles per second.

Impellers are cheap and easy to buy new ones for Rios, so fear not.

Regards,
Tom Barr

[Oreo]

Why do you feel this method is better then using a reactor Tom? Have you been unable to achieve 30ppm+ using reactors? Or do you think the plants actually benefit from having CO2 bubbles in the water vs. just dissolved CO2? Some other reason?

[scottward]

Hi Tom,

The Rio 1000 that you are using, is that this? http://www.aquacave.com/rio-hyperflow-8hf-br-water-pump-2042.html

So it's 1000lph? About 500gph?

And that is the only thing you are using in the 180 gallon tank - it doesn't seem like enough? And at only about 5bps you must get excellent milleage on the 15lb bottle (from memory you are using a 15lb bottle).

And no spray bar connected to the output? Just point it along the back glass and that's it?

I think it's time for me to get out the snips!

Scott.

[Gerryd]

Hi all,


So I went from 5 blades to => 15 thin small blades, bending them back to get more distance between them. The bubble rate is pretty fast, dependent on the tube sizing.
Maybe 4-5 a second? I have venier handles and use those and slow adjustment, not bubbles per second.

Tom,

Splitting the five blades comes to 10 smaller blades. Did I miss something to get to 15?

Even with different tubing and a nice needle valve, you still get a certain bubble rate no? Even if the bubble is smaller/larger based on tube size and you have a $1,000 needle valve it will still produce a bubble rate correct? Or does the nicer valve work differently somehow? What am I missing here?

Scott,

I am just using the Rio 1000+ rated at only 271 gph at 0 head.....

http://www.bigalsonline.com/BigAlsUS/ctl3684/cp48947/si1382144/cl0/rioplus1000pumppowerheadul

I think I need to get the one you linked too...I didn't realize there was more than one model 1000.

I also noticed that my impeller is smaller/thinner than the one in Tom's pic....assuming it is also a Rio impeller.

I bet the larger flow makes a bigger difference as my pump only makes it halfway across the tank before it starts to peter out....

[Gerryd]

- I wonder if cutting through 70%, or all the way to the root of the impellor makes much difference? With respect to both the 'chopping' ability and the possibility of the blades slowly moving back to their starting position.
- I wonder if the thinner blades would suffer from flexing and eventually snap off at the point of bend? I think holes drilled would be less likely to suffer from 'plastic fatigue'.
- Did you note what the mist output was like on the Rio 1000 before you did the mod?
- Is the mist so fine that the bubbles don't float to the top any more?
- With your Mag 500gph where the experiment with the holes didn't work out (I do remember reading your posts about it a long time ago), what was the reason it didn't work out? Was the mist not fine enough? Was it just the particular type of powerhead?

And my answer to your 'All' question
- What about connecting an airstone to the CO2 line such that rather than single noisy bubbles moving into the impellor chamber a more steady stream of smaller bubbles is introduced? I guess the downside of this is that airstones can clog up.


And I'll add another 'All' question
- holes vs snipping the blades; I think I can appreciate why the blade snipping would in theory work better, we're creating more pathways for water/bubbles inside the impellor chamber (i.e. water/bubbles stay in the impellor chamber longer hence reduced flowrate). Is that correct?

Hi Scott,

1. The 70% cut was simply because I cut the first blade too far to one side and it splintered kinda like wood does. I got scared and thought that cutting to the end would make them too brittle and break...not sure if this could happen but the blades on this Rio model are very thin...not like the Mag impellers which are much heavier/larger.
2. I could agree with this.
3. Sorry I did not. Just comparing to the mazzei :)
4. Mostly very fine mist. However this gets caught in the current and flows all over. It will still float to the top even if mist...
5. Can't even remember now. I think I still have it and will experiment again and let you know....

Thanks,

[csmith]

Hey Gerryd,


1. What seems to be the best way to introduce the c02 to the pump? Simply drill into the outer casing so I can slip the c02 tube inside? Or is my current placement/method viable? You've done what I do. My CO2 line is attached to the glass 1-2" under the powerhead via suction cups. I don't lose bubbles out the back and it works well. No real reason for not inserting it into the powerhead itself, just lazy I guess.


I had previously modded a Mag 500gph with the drill holes in the blade method. It did not work that well for me but I may also be an idiot....I have about 4-5 holes in each blade and I staggered them up/down... I have a mini-jet 404, so snipping the blades was out of the question. I took a sewing needle, heated the tip and shoved it through the blades. I can't give a definitive amount of holes I made as I just went until I thought one more might be too much. The holes left from the needle were tiny and they eat the CO2 bubbles up incredibly well. Maybe try smaller holes?


Even with different tubing and a nice needle valve, you still get a certain bubble rate no? Even if the bubble is smaller/larger based on tube size and you have a $1,000 needle valve it will still produce a bubble rate correct? Or does the nicer valve work differently somehow? What am I missing here? I think the answer to your question is too many bubbles come out to count accurately. That's the answer I'd have to give about my setup, anyway.

[Gerryd]

Hi,

I apologize for monopolizing the thread...

Yes, I got the c02 line placement idea from you and others :)


I think the answer to your question is too many bubbles come out to count accurately

C,

That is exactly what I am attempting to determine. Is BPS (bubbles per second) a USEFUL term of measurement regardless of diffusion method and hardware? Or is it like watts per gallon in that it is a useless measure and comparison point among different configurations?

To me hardware can only change the size and rate of the bubble. However it still needs to come out the end of the c02 line into your diffusion method at some point. What is the rate AT THAT POINT?

Is this even comparable?

As an example if I provided even my mazzei with a 'countable' 10 bps my tank would be overrun with all types of algae and poor growth....

I need to give it 'dozens' of bbs to provide adequate c02....the stream is simply constant and not countable at all...

With the DIY mod, I am back to being able to count the bubbles. I need to start over with bubble rate as the diffusion method is different. No way I am changing that much and leave the bubble rate the same....I can live with some algae and/or poor growth while I experiment as I have done it before...

Am I making any sense?

[csmith]

Yes, you're making sense. WPG isn't useable because it's not an accurate method at all. BPS can be incredibly accurate, but I've been taught recently (http://www.barrreport.com/showthread.php/7404-Calling-Chemistry-and-Math-Pros) that sometimes absolute accuracy isn't all that necessary. In my case, coming up with an equation to figure the amount of K given from KH2PO4 isn't needed as when you add enough KNO3 you more often than not have enough K. In this case, BPS doesn't really mean much due to, as you pointed out, the diffusion method changes the bubble rate needed by quite a bit. If the answer can change by a large margin then why search for the most accurate answer? For me, dose KNO3 and KH2PO4 and don't sweat it anymore. For this, why worry about how many bubbles are coming out of the tube? How much CO2 do we need? Enough. No bubble rate can really dictate what enough is from tank to tank.

Holy crap, I've learned something.

[Gerryd]

C,

Glad your brain functions on Friday night :)

What you say makes sense..

BUT

Since I am now emulating THE SAME setup as Tom (that is the goal anyway), the bubble rate is more important no?

Since the pump is really independant of other plumbing as it is stand alone, I think that bubble rate comparison is more worthwhile...

I figure that if person X can use method Y on a tank of N gallons and use a bubble rate of Z and get results like Tom, than I can use their rate as a BENCHMARK of sorts.

I am never sure of this anymore as Tom told me long ago 'don't believe everything you THINK' which I think is pretty good advice..

I know about your desire to measure macros to the nth degree, but c02 is not as easily administered to have 'enough' as you well know :) More Macros can just be dumped in with little thought up to a point. Not advocating fert overdosing....

Not so with c02...

Thanks for the help.

[shoggoth43]

BPS as a measure is only useful for YOUR equipment and as a general relative measure for changes you've made. I run at 15 PSI 2nd stage, but there's no back pressure in the delivery tube as I feed right into the pump. I set it for say 4 BPS and go from there. However, if I put on a ceramic disc I have much more back pressure and now I have to crank up the amount going in to get the same BPS, but it won't deliver the same amount of CO2. Tubing size may also increase the size of the bubbles coming out depending on your equipment. There's also the question of what tubing you use and how long it is.

Basically unless what you do is very similar to what I do, chances are that my 4 BPS is not really the same as your 4 BPS. Also, how you feed your CO2 in and how you mix it in the tank may make your setup much more efficient than mine, so the 4 BPS you put in may end up in the tank better and longer than mine. Alternately maybe the fact that mine mists differently than yours may make mine more efficient for the plants and bioload that I have. Lots of variables and maybe not really the best measure of success but it's useful for when you make changes.

-
S

[Tom Barr]

While never sure, you can try and see if you can really know something, even if limited and narrow, then look elsewhere for more little nuggets knowledge.

As long as you have some sort of a relative measure, do not trust absolute ppms etc..........then you can work well with it.
I have 5 tanks, each is different, so I have 5 different relative measures.

Regards,
Tom Barr

[scottward]

I read through the above posts which were interesting to read, but I don't really have anything to add myself. ;-) This is my first go at DIY'ing a 'needle' wheel, hopefully I will have something to add soon as I experiment.

I took the impellor out of my Resun 2000lph (it is a 6 blader).

I was going to snip the blades but decided for my first go I would just drill some holes.

So, I started off by drilling 2 larger holes on each blade and then drilled smaller hole in between the two. I just guessed where to drill them, no measuring or anything.

I could easily fit more little holes onto each blade, and I probably will drill more holes, but not just yet.

I re-assembled the powerhead with the modified impellor and got it running again.

I can still see the bubbles coming out of the powerhead, but I'm pretty sure they are smaller than what I had before. The bubbles still rise to the surface but appear to be more easily influenced by the current in the tank now, such that they generally have a hard time making it to the surface.

Plenty of flow still. I think I could drill several more holes in each blade, a 'swiss cheese' look.

Attached are some photos.

Hmmm. So Tom gets his mist so fine that he can hardly see it. Do you have good eyesite Tom? ;-)

I can hear the bubbles ticking into mine, an audible bubble counter. Probably about ~4bps at the moment. AM1000 is still cranking away at the same time, can't count the bubbles going into it, too fast.

Will observer and report before making any further changes.

Gerry - how's yours looking?

Scott.

[scottward]

...and I cannot count.

It is an 8 blader! ;-)

[csmith]

Here's what I've done. First picture is for size reference of the blades, second is a close up. Lots of small holes is, personally, what I think makes this work. The "strings" hanging off are just melted plastic from when I made the holes, a razor blade can be used to remove the excess.

http://i803.photobucket.com/albums/yy316/CASDB/20%20Gallon/NeedlewheelMod2.jpg

http://i803.photobucket.com/albums/yy316/CASDB/20%20Gallon/NeedlewheelMod1.jpg

[scottward]

Cool, thanks for putting up some pics as well. I suppose as many holes as you can get in the impellor blades the better. And to counter the reduction in flow simply choose an overpowered powerhead in the first place.

I think I will get mine back out on the weekend and drill as many holes as I can into each blade. Lots of small holes I think.

I could have snipped it but I was worried about cracking it. I don't have a spare impellor.

I can still see my bubbles, so I guess I still have some work to do.

Is that really the ultimate - i.e. in answering the original reason for me creating this thread - the most efficient DIY needle wheel is the one whereby the output bubles are just so fine that they can hardly be seen (or cannot be seen at all)?

I understand through discussion with Biollante that the most useful CO2 bubbles are the ones that we cannot see with the naked eye?

The bubbles we can see are more or less useless?

Scott.

[shoggoth43]

Yes and no. The bubbles you can see are still useful if they are floating around the water column and/or stuck to the plants since they can still dissolve. If they're heading right for the surface and staying there/popping then it's more likely you are just wasting your time and CO2.

-
S

[scottward]

Thanks shoggoth.

I wonder, with dissolved CO2 in the water and tiny bubble of CO2, does the plant have to use diferent mechanisms to utilise the disolved CO2 versus the bubbles of CO2 that might come to rest on the leaves? I wonder if using a reactor and misting at the same time could confuse the mechanisms of the plant in any way and actually make growth worse?

I am using my AM1000 like normal at the moment and misting throught a powerhead - do you guys think that it is ok that I am doing this?

It's too early for me to say how well this is working out, as I've only recently starting seriously misting with a DIY needle wheel.

Gees I'm chewing through the gas though, which I don't like at all.

Maybe 2 AM1000's would be just as good, much more efficient on gas I think.

[shoggoth43]

Different theories on this one. One is that the CO2 sticking to the plant is very similar to the plant being emmersed ( i.e. in air ) and thus has easier access to the CO2 so that you could use less of it by having the CO2 "stick" to the plant and not get dissolved in the water only to have it blow off at the surface. Dissolved CO2 would have to be extracted / absorbed by the plant at a slower rate due to less of it being available. At least that's the theory. I know Biollante likes the reactor types for efficiency.

Mixing and matching methods is perfectly fine. One think the misting method does get you is a direct method of getting the CO2 "right there" since you can blast the powerhead directly at something with a stream of mist to see what's going where. You can do similar with reactors but it's harder to do depending on the design and you of course can't "see" the CO2 enriched water or the flow/distribution around the tank.

CO2 is relatively cheap so after tinkering for a bit the amount you use should settle down once you stop playing around with the different methods. For some of us this can take months. :)

-
S


Thanks shoggoth.

I wonder, with dissolved CO2 in the water and tiny bubble of CO2, does the plant have to use diferent mechanisms to utilise the disolved CO2 versus the bubbles of CO2 that might come to rest on the leaves? I wonder if using a reactor and misting at the same time could confuse the mechanisms of the plant in any way and actually make growth worse?

I am using my AM1000 like normal at the moment and misting throught a powerhead - do you guys think that it is ok that I am doing this?

It's too early for me to say how well this is working out, as I've only recently starting seriously misting with a DIY needle wheel.

Gees I'm chewing through the gas though, which I don't like at all.

Maybe 2 AM1000's would be just as good, much more efficient on gas I think.

[scottward]

I've just added a bit of extra light along the back glass of my tank to encourage faster growth of my stem plants. I'm probably going to have to tweak my CO2 a bit (unless it was already plentiful). With the AM1000 and the needle wheel both running, I'm not sure whether to increase the rate into the AM1000, the needlewheel, or both! ;-) Decisions decisions.

Tonight, if I get a chance, I might pull the impellor out and drill even more holes in the blades using the smallest drill bit that I have.

Gerry, how is yours going?

CSmith - yours?

Scott.

[Gerryd]

Scott,

So far so good! I think I may get a slighly larger model or a new impeller and try the drilling method and compare....

I just need to hide it now.....

Be careful adjusting the c02 with dual diffusion methods... very easy to gas your fish....slow and easy does it...

[Tom Barr]

The split bade impeller is much better than the holes in the impeller FYI.
I also "furred up" an impeller a few times. I think finding a needle wheel impeller pre made or splitting the blades like I've showed is better though.
Some of the DIY mods in the reef skimmer forums are okay.

I just get ultra fine, well beyond mazzie, micro mist, I do not even see any mist/bubbles which was one of the few detractions.


Regards,
Tom Barr

[csmith]

CSmith - yours?

Mine works well for what I need (or so I believe). I keep a constant yellow drop checker, and I've been working on my flow so I think I'm at a suitable level. As I said, though, I have pin sized (literally) holes so I get tiny, tiny bubbles if they can even be called bubbles. It's also on a 20 gallon, so the area to cover is very small. I had to go the hole route because the impeller was a bit small to cut and I'm sure the soft plastic wouldn't have held its form if I attempted to bend them. Perhaps one of those "next best option" scenarios. If you do go the hole route, the smaller seems best.

[scottward]

So far so good! I think I may get a slighly larger model or a new impeller and try the drilling method and compare....

I just need to hide it now.....

Be careful adjusting the c02 with dual diffusion methods... very easy to gas your fish....slow and easy does it...

Cool. Thanks for the update Gerry and the advice. Yeah, I think I will tweak the AM1000 in the first instance and if it starts choking (i.e. overwhelmed) I'll push the needle wheel harder. Keep me posted on your progress.


The split bade impeller is much better than the holes in the impeller FYI.
I also "furred up" an impeller a few times. I think finding a needle wheel impeller pre made or splitting the blades like I've showed is better though.
Some of the DIY mods in the reef skimmer forums are okay.

I just get ultra fine, well beyond mazzie, micro mist, I do not even see any mist/bubbles which was one of the few detractions.

Hi Tom. Thanks for your thoughts and advice. I can see that the real efficiency thing here is all about getting the bubbles as small as possible - if they are so small that they cannot be seen then that must be the best possible outcome. What I will do is continue with my hole driling experiment to the point where it's not possible for me to put any more holes in the impellor and I will note the size of the mist. Once I have run this experiment for, say, a month or so, I will buy another impellor to suit my powerhead and split the blades as you have suggested. I will then compare the two methods and of course provide my feedback on here. The reason I went with the holes first was because there was less risk of me stuffing it up (i.e. cracking the blades or something like that).

Clearly the smaller the bubbles the better is the aim here. Cool.


Mine works well for what I need (or so I believe). I keep a constant yellow drop checker, and I've been working on my flow so I think I'm at a suitable level. As I said, though, I have pin sized (literally) holes so I get tiny, tiny bubbles if they can even be called bubbles. It's also on a 20 gallon, so the area to cover is very small. I had to go the hole route because the impeller was a bit small to cut and I'm sure the soft plastic wouldn't have held its form if I attempted to bend them. Perhaps one of those "next best option" scenarios. If you do go the hole route, the smaller seems best.

Thanks mate. Any possiblity you might get hold of a second impellor and try the cutting approach suggested by Tom? I might use a dremel to slit the blades on mine, rather than snips. Perhaps you could use something like this too? Hmmm - regarding the holding their form, that gives me a thought...

Question for Tom (+ other gurus): Once the blades have been snipped, I can understand that if they start to close back up into a single blade again, this will be counter-productive to our efforts. But - as long as they don't completely close up will they still work ok? As long as there is some kind of gap between the blades won't that be ok? As an alternative to snipping the blades, if I were to cut the blades using a dremel such that there is actually plastic removed from the blades, if they were to close up there would still be somewhat of a gap between them - this gap might still be more than enough to do the job????

[mi5haha]

I tried Tom's way on a 18W pump. Six blades.

Blades do close back, but not into a single blade again. The issue is that the blades are very strong (they will not break), even using hot air to heat the blades split, but they will still come at least 2/3 back after a while.

The result of my test on this pump is not as good as the pump equipped with a needle wheel impellor.

I think the impellor rotating speed is also relevant if cutting the blades into half.

[scottward]

I'm not 100% sure, but I think the impellor can turn in either direction? If the impellor always turned in one direction, I suppose we could leave one half the snipped blade where it is, and bend the other half so that it is bent in the direction where the water flow will encourage, rather than 'return', the bend.

Do impellors general spin in the same direction when the pump is turned on/off?

If so, is there any way to know which way the impellor will always spin?

Then we could just bend in the direction where the bends will be best maintained?

This sounds too easy, I'm guessing the direction of impellor rotation can change. ;-)

I've been thinking hard about holes vs snipping. Yeah, I'm starting to really think that the snipping is going to work better. Long, skinny paddles (i.e. the snipped blades) is going to allow the air/water mix to pass through more easily than holes in the blades, whereby it's more difficult for the air/water mix to squeeze through little holes.

If I get a chance, I will snip my 8-blader up tonight, turning it into a 16-blader. I was going to try to snip each blade twice but because I already have holes down the centre this isn't going to work (I think the middle blade would fall off!).

16 skinny blades in a 2000 litre per hour powerhead should do a pretty good job of chopping the bubbles up.

I tried to obtain a ready made needle wheel for my powerhead, but they don't exist, so I have no choice other than to DIY.

[scottward]

I snipped the impellor tonight so that it is now 16 blades. I carefully bent the blades.

Doesn't appear to have made much difference, I can still see the mist.

For all the people that have made one of these, either snipping the blades or drilling holes, Tom you are the only person who has been able to make the mist so fine that it is difficult to see. Something doesn't add up...we can't all be incompetent at making something that appears so simple. ;-)

Tom - when you say you cannot see the mist coming out of it - when can you not see the mist? During the first few hours when the CO2 comes on and the CO2 is easily dissolved due to the lower CO2 concentration in the water? What about after it's been running for a good few hours, towards the end of the photo period, can you see fine bubbles coming out of it then? Also, are you feeding bubbles of CO2 into it relatively slowly, or are you hitting it with a decent bubble rate?

[scottward]

*Bump* ;-)

Tom, I think I found the answer to one of my questions - your feeding in CO2 at about 4-5bps?

This is for your 180 gallon tank correct?

And the DIY needlewheel is the only method that you are using for CO2 for this tank correct?

Regarding the misting, can you confirm whether the misting is difficult to see at the start/end of the photoperiod? Perhaps it dissolves straight away at the start, but the mist starts to become more obvious toward the end?

I note your comment earlier about the finer the mist the better. If the mist is so fine that it cannot be seen, I assume this is because it is actually dissolving completely and there are no bubbles in the tank? Would this not defeat the purpose of having the bubbles attach themselves to the leaves of the tank? If the bubbles should ideally be contacting the leaves, isn't it possible that making the mist too fine (i.e. dissolving) could be counter-productive?

Can somebody please clarify this for me?

Scott.

[mi5haha]

I used the real needle wheel pump. the bubbles are very fine, drifting in the tank, following the current to the other side of the two-feet tank, making their turn and going back slowly, and they are still there. They did not dissolve immediately after they are coming out the water outlet. But they do not look 'haze" yet. You can still each bubble with naked eyes.

However, cannot take photos of these fine bubbles. Guess need to use a micro-distance lense.

[scottward]

Hmmm. So even with a real needle wheel pump it wasn't possible to make them dissapear?

Tom - you have some explaining to do! ;-)

[Gerryd]

Scott,

I am just using the rio 1k split blade approach on my 180...bubble rate is more than 3-4 maybe 12-20? You will see.

Does a GREAT job so far better than the mazzei at least as far as plant growth/health shows....

I will take some pics/video later of this is operation at several bubble rates....

Will be after work my time so hope you can wait another 10-12 hours or so.....

[scottward]

Hi Gerry.

12-20 bps? I remember you saying in other thread that your bubble rate was incredibly high. I know you can't count this fast, you just mean that the bubble rate is going nuts. ;-)

I might be confussed about what Tom is running - was he also running a Rio 1000 on a 180g like you? I'm sure, unless I'm totally misunderstaning, that he said his bubble rate was only 4-5bps (which is fairly easy to count)? Something's not adding up. You've been doing this for a while now Gerry and have a good understanding of all the principles, there is no way that Tom could have the same result as you at only 4-5bps for the same sized tank, same lighting (?). Sure every tank is difference, but within certain boundaries, things should start to form a bit of a pattern.

I must be confussed.

I shall think no more about it until I have my facts right. ;-)

I don't mind the look of the bubbles in my tank, they look cool, but I do want to get as much out of my CO2 bottle as possible (hence the reason for me creating this thread about efficiency - not ease of use, lowest cost, lowest maintence - but efficiency).

Look forward to seeing your photos Gerry.

Scott.

[Gerryd]

Hi Scott, all,

I apologize in advance for the length of this post...

Okay. Here are three videos of my Rio 1000 with a split blade config.

The first two videos simply show the CURRENT bubble rate and mist from different angles. Can you count the bubbles per
second? I think the mist it provide is pretty darn good personally, and I have been using a mazzei for awhile now.

The third video shows the operation as I INCREASE the bubble rate significantly. At the HIGHEST rate in the video, that
was LESS than I was feeding my mazzei. More on that later.

The LOWEST rate is what I would think of as 4-6 bubbles per second.

Please note:

1. The rythmic noise from the rio. I attribute this to the shoddy DIY snipping job :) No noise with c02 off, so is useful
to know it is 'on' and getting c02 lol

2. The size increase/decrease of the mist and associated bubbles as the bubble rate changes...

Very effective method of c02 diffusion...

I will be back but wanted to post the video..

Front view of current rate and operation. Note the nice mist produced.

http://www.youtube.com/watch?v=0o6pcfMJNw0

Side view of current rate and operation.

http://www.youtube.com/watch?v=p824IhYwvk8

And here I increase the rate to what the mazzei rate was, back to my current rate, and down to 3-5 bps? Please note the corresponding increase/decrease in mist vs bubbles..

http://www.youtube.com/watch?v=dgKr3OcmOs8

So, back from commercial...lol

I am pretty sure that Tom is indeed only using this method for his 180 but his lighting is different. PAR may be closer than we think
but I think I am higher PAR than he is right now...avg range 75-100 at substrate and close to 190-200 at surface...

You can see why I am always so interested in bubble rates from the posters...

I know that something was off and looks like I was right....

Even though the mazzei provided good mist that was detectable all over, I had two lingering issues:

1. Consistent plant growth and algae issues (albeit small) that could only be c02 related. Carpet plants were the hardest to keep for me but I can see the signs of poor c02 around here and there. Intermittent issues are always a good indication that c02 may be off.
2. Tremendous usage of c02. Filling a 10lb bottle once a month!

For example my new stauro would do well, than not, than okay in MOST areas but not all. Same with HC. Do well and than get algae covered. Do well and melt. You get the idea.

About 10 days ago now I did the following:

1. Installed a vortech mp20
2. Shut off my mazzei leg and directed that flow to the other legs/outlets. Much improved flow as the mazzei requires a lot.
3. Installed the Rio pump as you see above. Don't laugh it is not permanently placed and I am still 'experimenting'
so aesthetics don't matter :) This also adds a nice flow across the center carpet stauro..
4. Lowered the bubble rate SIGNIFICANTLY as per the video. The highest rate in the last video was the mazzei rate. Compare
that to the rate in the first two videos. Can you see a difference?

Since that time I have observed the following:

1. Tremendous growth in the stauro. Size of plants and leaves increased significantly.
2. Better and earlier pearling. Much more consistent with a water change..
3. Reduction in some of the cladorpha that plauges small areas of substrate and wood.
4. Better growth in all anubias which were lagging somewhat.
5. Better overall growth it seems but it also could be plant momentum that Tom speaks of as things HAVE been going well!

I realize that the improvements were due to many items not just the rio mod, but I am just pointing out I guess that you can think that
you have a great setup (mazzei and good pump) and all is working 'optimally', and things could always be better if you are having lingering issues.

I think now that I will soon have one of those 100% algae free tanks with the great growth that folks here have...

I may not be optimal just yet, but so much better I think with just some small simple improvements.

I feel confident that I will be able to lower the rate more but we will see. No fish stress at all and plants are really doing well...

[scottward]

Hi Gerry. Cool - thanks for posting the vids. I am at work at the moment and YouTube is a blocked site. ;-) I will check out your videos tonight when I get home from work (and the kids are asleep), and will let you know my thoughts then.

I suppose I should get back to work.... ;-)

Scott.

[scottward]

Hi Gerry,

I watched your videos, nice work! I could clearly see what you were explaining in the text above.

The mist coming out of my powerhead is about the same as yours actually. I should try to film mine (I will if I get a chance).

It makes sense that the rate that the bubbles are being fed into the powerhead will have an impact on bubble size. I could see that when you cranked up the bubble input rate the resultant bubbles were larger, making the mist more obvious. This makes sense. Less bubbles cluttering up the impellor chamber per unit time no doubt results in a finer mist.

Also, do you notice any difference in the mist between when the CO2 first comes on (so, low dissolved CO2) and toward the end of the CO2 enrichment period? I would expect that the mist would seem much finer at the start of the CO2 period when the gas will dissolve more readily.

So perhaps Tom is running a lower bubble rate and made the observation at the start of the CO2 enrichment period and the mystery is solved? ;-)

I feel like I'm a bit obsessed by bubbles like that fish in Finding Nemo.

I have a spray bar connected to the output of my powerhead, which blasts the mist out through my plants from the bottom of the tank. Maybe inside the spray bar some of the bubbles are colliding together forming bigger bubbles? Dunno. At least I am directing the mist where it matter most I suppose.

Hmmm. I will think some more about this and write more from work tomorrow if I get a free moment.

Thanks again Gerry.

Scott.

[nipat]

...
So perhaps Tom is running a lower bubble rate and made the observation at the start of the CO2 enrichment period and the mystery is solved? ;-)
...


I think so.
http://www.aquaticplantcentral.com/forumapc/general-aquarium-plants-discussions/1888-tag-amano-article-04-06-issue-2.html#post13895
http://aquariacentral.com/forums/showpost.php?p=34589&postcount=4

I also checked videos of Amano's tanks in Niigata and saw that the flow
was not strong at all (bubble rates in BC's for some tanks looked very fast though).
http://www.youtube.com/watch?v=1xjV6_3Xgoc

This makes me think the good flow in our impression may be too strong.
I mean the key is circulation, no dead spots, with mild flow.

Even Tom's tanks, while probably have stronger flow than Amano's,
may not have as strong flow as we have imagined? So he can hit 30PPM
with that low bubble rate.:confused:

[Tom Barr]

Depends on where this 30ppm measure is taken.

The higher current is for the fish mostly, but does help mix the CO2 and offers more wiggle room, less stress on the fish via more O2.
As far as plants are concerned, adding plenty if fine and not an issue, it's just the upper ranges for fish that are the only risk factor.

Some fish are fine at 50+ ppm of CO2........

Some are not, so it depends.
Same differences with CO2 for plants, some do well with a little, others need more.

Regards,
Tom Barr

[scottward]

Hi Tom,

How long does a CO2 bottle (10lb?) typically last you on your 100gallon tank using exclusively DIY needle wheel powerhead?

I have just recently bought a 6.8kg bottle (15lb) and I'm hoping to get 6 months out of it feeding both the DIY needle wheel powerhead and the AM1000 I also have running.

Scott.

[a-hua]

hi tom

a question beside this diy way（i broke 2 of my powerheads because i am really not good on *cut*.....oh...）

if i just use a nice work glass diffuser in the input of a normal powerhead....is it amost the same with this diy needle wheel???

a-hua

[scottward]

Hi a-hua - just my 2 cents worth - this glass diffuser - it can clog up right? If this is the case, using it as the input to your powerhead would defeat the purpose somewhat. By using the snipped impeller you are reducing maintenance as this is far less prone to clogging than disc. But if you are happy to maintain the glass diffuser, then yeah, the powerhead will simply serve as the distribution method and should work ok as far as I can see.

[scottward]

Gerry et al, how are you guys going with your DIY needle wheel?

Mine is still ticking away nicely in the background. I still have the AM1000 running at the same time.

Plant growth is fairly good.

Bit of BBA about the place but I think that's probably just because of high levels of spores floating about in the water (I've been a bit lazy with water changes).

My wife had our 3rd bubby so I haven't had time to be on here for a while and am only now starting to get a bit of time back to work on my tank - but hardly! ;-)

Scott.

[tiger lotus]

what kind of pump would everyone suggest i use for a 60 gallon tank.