DE Filters - Use, design and implementation.

[robert]

Basically I'm splitting another thread to keep from hijacking it at the suggestion of @ca1ore

Diatomaceous Earth filters are one of the very best forms of mechanical filtration. Since they filter down to the range of 1-2 microns they can easily remove parasites such as ick and velvet (even bacteria) from the water column, but they are hard to implement and operate.

I use mine in conjunction with UV. DE for removal of Parasites/Bacteria/ and UV for Bacteria/viruses. The DE filter can also be used for rapid stripping of organics and chemicals when GAC is added to the mix.

If there is an interest - I can show how I run mine and how I've worked around most of the issues involved. I really don't want to engage in debates about efficacy against diseases and whether fallow/quarantine/copper/etc approaches are superior although understanding how and where these pathogens live in the tank helps in getting the most out of the filter.

Read" https://www.reef2reef.com/threads/diatom-filter-for-treating-external-parasites.212429/
for some background.

What I can share is appropriate (and most necessary) for keeping a big tank with sensitive species.
I can also share alternatives which could be used on smaller systems if the interest is there.

 

[saltyfilmfolks]

Good to see you back.

 

[ca1ore]

Yes, I remember that prior thread. In fact I stated back in 2015 that’d I’d be getting one .... which I never did do. I totally agree to suspend any debate about how well a DE might remove parasites and focus on how best to implement one. For example, is the suction side of a DE filter best located in the display or can it be in the sump. I’d guess the former. I was recently looking at the Hayward pool units, but unsure which model to use for a 400 net gallon tank. The EC40 proflux looked about right. Is that the one you use? I wonder if any aquarium pump can be used with it since I have a few PanWorld sitting around.

Robert, I’d be very interested if you’d be willing to share your approach so that I can avoid any of the pitfalls you’ve already navigated. Fascinating subject and no small irony that an ‘old’ technology may be relevant in ‘new’ ways. Lol.

 

[robert]

The Hayward EC40 is the one I use and it seems ideally sized for 400 gallon the way I have it plumbed in. Plumbed differently I would imagine you use it with a substantially larger volume.

Your question on whether to draw from the tank or the sump is exactly the issue. The Vortex filters and all the DIY's that I built drew water directly from the main display as that's the most effective, allowing a deep draw from the problematic areas where parasites accumulate. This filter implementation restricted their size, made them prone to leak water and DE, suck air, made them unwieldy to charge and clean, and just generally made them a pain in the rear to anyone using them... But when they worked they were awesome!

To cut to the chase, I don't draw from either. Instead I draw from what I call a mezzanine sump, a much smaller sump located between the overflow and tank sump. This fixes almost every issue at the cost of drawing from the wrong location in the tank, which I fix with flow.

The MS has two 2" bulkheads - one for the DE pump and one to return water to the sump, when the height of the water in the MS exceeds the height of the standpipe.

The mezzanine sump (MS) is open top, fed by the overflow and ends up holding something less than 30 gal. The DE filter has its own 2 speed Hayward pool pump pulling from the bottom of the MS and pushing through the filter returning its outflow to the top of the MS.

The main tank return pump runs like normal, drawing from the main sump and returning water to the tank which hits the overflow and drops into the mezzanine sump. (The MS is a kitchen trash can. Yes kinda ghetto - I should try to do something nicer but it has worked without issue, and its hidden so...I'll fix it later)

If I turn off my main pump, water drops back into the sump until the overflow is exposed like normal. At this point the MS drains down to the top of the MS standpipe. The DE pump continues to run, circulating the water held in the MS. The MS is never dry. This design accomplishes a couple of important goals. It preserves the flood proof design of a traditional sump without adding any risk of siphon, eliminates tank micro-bubbles, prevents DE spill into the main water column - and most importantly it makes the system easy to charge with DE powder.

To charge the system with DE all you do is cut power to the main pump until water quits overflowing the top of the standpipe in the MS. You add 4 large cups of DE (and carbon if you wish) to the water in the MS while the DE pump is actively circulating. In a minute or two, the water in the MS is perfectly clear and the main pump can be turned back on. System is charged.

To flush is just as easy. You don't need to turn off the main pump. Simply switch off the DE pump - hit the bounce lever a few times to knock the DE loose in the filter - open the flush outlet on the DE filter itself and drain the dirty DE powder in into a salt bucket. Repeat 1-2 times until the spent DE powder is removed.

The whole operation, flush and recharge takes just a few minutes. Replace discarded water and your done. If your careful - you never touch water - Viola - DE filters are not a pain anymore.

I'll make a sketch -

Tank flow to address bottom draw is the next issue.

 

[Scrubber_steve]

Diatomaceous Earth filters are one of the very best forms of mechanical filtration. Since they filter down to the range of 1-2 microns they can easily remove parasites such as ick and velvet (even bacteria) from the water column, but they are hard to implement and operate.

Whats the benefit of a Diatomaceous Earth filter over a cartridge filter of 1-2 micron? Pretty easy to place one of those on the intake of a power head?

 

[robert]

Cost, scale and surface area - but as far as its ability to filter - none.

Put one of those on a suitably sized pump and place it on the bottom of your QT tank. Keep the fish several inches off the bottom with egg-crate and you have a medication free way of treating ich and velvet by changing the dynamic between the fishes immunity and the parasite. Dilution - the magic of mechanical filtration.

 

[robert]

To cut to the chase, I don't draw from either. Instead I draw from what I call a mezzanine sump, a much smaller sump located between the overflow and tank sump. This fixes almost every issue at the cost of drawing from the wrong location in the tank, which I fix with flow.​

The mezzanine sump has two 2" bulkheads - one for the DE pump and one to return water to the sump, when the height of the water in the MS exceeds the height of the standpipe.​

The mezzanine sump (MS) is open top, fed by the overflow and ends up holding something less than 30 gal. The DE filter has its own 2 speed Hayward pool pump pulling from the bottom of the MS and pushing through the filter returning its outflow to the top of the MS.​

The main tank return pump runs like normal, drawing from the main sump and returning water to the tank which hits the overflow and drops into the mezzanine sump. ​

If I turn off my main pump, water drops back into the sump until the overflow is exposed like normal. At this point the MS drains down to the top of the MS standpipe. The DE pump continues to run, circulating the water held in the MS. The MS is never dry. This design accomplishes a couple of important goals. It preserves the flood proof design of a traditional sump without adding any risk of siphon, eliminates tank micro-bubbles, prevents DE spill into the main water column - and most importantly it makes the system easy to charge with DE powder.​

 

[ca1ore]

OK, so let me play devils advocate ....

If you are relying on flow within the display to get ich into the overflow such that they then get filtered out by the DE unit, how is that inherently better than using an UV unit? Same flow would mean the same removal of ich to the overflow to be ‘treated’. Perhaps the argument is that a DE just does a better job of physically removing ich than a UV does in irradiating it - proper sizing assumed? While I think your use of an intermediate sump is quite clever, and mitigates the messy recharging process, I would have thought that the DE unit suction placed close to the sand in the display would remove ich more efficiently?

Regardless, thanks for sharing your approach. I’ve ordered a perflex 40.

 

[robert]

I would have thought that the DE unit suction placed close to the sand in the display would remove ich more efficiently?

Yes it would - and of my earlier DIY's based upon the vortex setup. These worked, and required a smaller pump for the same effect. But they were horrible to use, clean, charge and maintain - meaning you never used them until an outbreak was underway - if at all.

Perhaps the argument is that a DE just does a better job of physically removing ich than a UV does in irradiating it - proper sizing assumed?

Yes, that is the reason as I explained in a different thread:
https://www.reef2reef.com/threads/u...riate-uv-sterilizer-size.535723/#post-5543456

I run the Emperor Aquatics HO 120UV on a 400 gal tank - but not for control of ick or velvet.

• ultimately, you want the highest dose that the UV can offer (slowest speed through your vessel) but you are balancing your total tank volume's needs... so you go as slow as possible while still giving a good turnover through the whole tank...

This is ^^^ correct and in the end is why you really can't control an outbreak of either of these parasites with any practically sized UV - at least I can't. You need a lot of flow to ensure you suck up enough of the parasite to significantly counter the infection rate but at high flow rates the dwell time inside the largest practical UV fixture is insufficient to neutralize enough of the pathogens to significantly change the disease dynamic.

For an in-tank control of outbreak control of ick/velvet mechanical filtration is your best option. I use DE filters for this purpose (>99.99% retention in a single pass) but other forms of mechanical filtration up to 50 microns (with less efficiency) can work as well -

I run it continuously a low speed, I kick it up to high for an couple of hours as needed during or after tank maintenance or when introducing new stock.
If ich or velvet is suspected or present - then I run it on high overnight/low during the day, until the issue resolves - a few days.

The filter itself contains some stainless steel screws - these seem to be of good quality and corrosion isn't an issue. The pressure gauge contains brass and should be discarded - instead watch the flow to determine filter status. Where the pressure gauge would normally be screwed into the body, I instead install a drain valve so I can remove the head for the occasional (every six months or so) grid both cleaning without spilling water.

Installation is the key both for effectiveness as well as servicing. I can flush and recharge in just a few minutes - done right - its actually quite nice and convenient. How often you flush is a function of how much it pulls from the water - With the "bump" feature on that particular model - I flush (between 5 and 10 gallons) and recharge every couple of weeks.

I use straight pool DE powder but the nice thing is that I can also add GAC (BRS rox) and everything works exactly the same with the added benefit of very fast chemical removal.​

Theoretical sizing: This useful for all aquariums and filtration in general.

The theoretical amount of water that gets filtered after a number of turnovers can be calculated.

Assuming continuous and instantaneous mixing. The formula is 1-EXP(-Turnovers).
So 1 turnover filters 63.2% of the water,
2 turnovers filters 86.5% of the water.
3 turnovers filters 95.0% of the water.
4 turnovers filters 98.2% of the water.
5 turnovers filters 99.3% of the water.
(Obviously, water not in circulation patterns will not get circulated as quickly while water in circulation patterns gets circulated more than indicated.)​

For ciliates and dinoflagettes DE is 100% effective meaning - if it goes into the filter it isn't getting through. If they were uniformly distributed in the water column

Back of the envelope quick check to see if its possible:

For ciliates and dinoflagettes DE is 100% effective meaning - if it goes into the filter it isn't getting through. If they were uniformly distributed in the water column

We could be real paranoid and assume 9 turnovers is required .
assume a 100 gallon tank - to turn over the volume once in an hour you need a 100gph filter.
To get everything, all the freeswimming trophants in 1 hours would require a filter flow of 900 gph
To get everything, all the freeswimming trophants in 2 hours would require a filter flow of 450 gph

but to make the math a little easier - assume you only run a 400gph pump.)

in the first 15 mintes you've sucked up 63.2% of the parisites.
By the time 30 minutes have past - 86.5% are eliminated.
An hour later - 98.2%
​

But ciliates and dinoflagettes are not uniformly distributed in the water column.

No they are not. When they drop from the fish they sink at a rate of ~45 feet/min.
As the hatch out - they are mobile but the vast majority can be found within 5 cm of the sand-bed where the parasite acquires a sleeping fish. They do not disburse int the water column generally. The current fashion for the hobbyist tank emphasizes outflow from the top of the water column - this essentially makes the tank a trap in which the parasites to increase and multiply.

The challenge is to ensure bottom flow - an unimpeded sweeping current along the sand bed, delivering to the base of the overflow is what I have found works well for me. I've toyed with the idea of using siphons into the overflow. But I don't like siphons much so I employ the chaotic lift provided by my mp60 to keep the floor of the tank directly below the overflow in relative upheaval. Just enough to begin to move my gravel a little. (Sand would be blown around)

So I design my in-tank flow to deliver whatever is on the bottom of my tank to an area directly below the overflow where it is kept in constant motion util it reaches the overflow.

While setting this up I found Hikari Food Pellets to be a useful tool. I repeatedly dropped them into my system and watched where they went. You want them, when they get to the bottom, to be moving to the base of the overflow no matter where they drop from. Sure, some will get stuck and stop - but they needed to be moving in the right direction until then.
​

 

[robert]

So other good questions were asked out side the thread:

I took it as you run the filter continuously and only charge it with DE when needed...correct?

I run it continuously. Most of the time on low speed, only kicking to high speed if I've been working in the tank. High speed is used if I suspect a problem, or if I'm introducing new stock and then it's only high speed at night due to the circadian rhythms of the more common parasites as they drop and hatch at night. The filter is always charged with DE powder.

When charged how long on average do you run it or until it clogs up?

Typically every 3-5 days I'll notice a drop off in flow. I switch the DE pump off, bump the handle a few times which knocks the DE powder from the internal screens and switch the filter pump back on. You'll notice the flow increase again, This allows the life of the DE powder to be extended and can be repeated repeated several times before a flush is necessary. That's why the EC40 is called an "extended cycle" DE filter. The main return pump can be left on during this and nothing will get back into the tank. Every 2-3 weeks I flush the spent DE powder and recharge as I described above.

If you are running this filter continuously do you find a need for any other mechanical filtration in your system?

No, I do run a Reef Octopus 6000ext plumbed to the main tank sump but it doesn't do much. It broke a while back (impeller) and I went six months before I repaired it. I didn't miss it. It's a good skimmer, but being downstream from the DE filter, there just isn't much in the water at that point for it to remove.

Are you using the Hayward EC40 with the attached pump or the stand alone filter with a separate "saltwater" pump?

You want a 40gpm pump with a high head. I use the Hayward two speed pump (115V - 1.5 HP max) which can attach directly to the filter, but you can run what you want. You do want at least the 2 speed for the energy savings as there is no need to run this at high speed continuously. Pool pumps are inexpensive compared to a "reef" pump but there is a downside. Being designed for a pool, they really are not designed for continuous use and running it that way it will shorten their lifespan. When mine gives out I'll replace it with an aquaculture rated version. I run a snapper hybrid dart for my main and that has been a workhorse and although it provides 3600 gph of flow, the DE filter requires a higher head than would be good for that pump.

 

[robert]

Sorry about all the typos - and repeated extra words etc.
Ask if anything confuses you.

 

[DJF]

What are the dangers if some of the pool grade diatomaceous earth were to get into the tank?

 

[robert]

^^ I said 45 feet/min - I meant 45 feet/hour - they drop pretty slow.

 

[robert]

What are the dangers if some of the pool grade diatomaceous earth were to get into the tank?

DE powder is made of the skeletons of diatoms - a little will cloud the water without any harm to corals or fish - A great deal might irritate gills etc.

DE ultimately is silica - don't breathe it in - always wear a mask when handling.

 

[robert]

A lot of questions you have might be addressed inventor in a published article, unfortunately it has aged and now must be found via the wayback machine -

Dr. Lawler provides a strong recommendation for the regular use of diatom filters. This article is a must-read for the serious aquarist looking for the best filter technique to use in fish keeping. As a bonus, learn the longest word in the English language!

https://web.archive.org/web/2015070...articles/management/Lawler_Diatomfilters.html

I'm here to help you set it up and make it work effectively on your reef tank.

 

[Monkeytank]

I have seen some people who say they don't run a filter sock and just use the diatom filter. Wouldn't it make sense to still run filter socks to catch the larger particles first and then have the diatom clean up only the smaller particles that still pass through? Shouldn't that increase the useful life of the DE and decrease the interval required for replacing it? I currently run 100 micron filter socks I found on Amazon that are mesh and very easy to clean with about a minute rinse in the sink. Every three days I replace the dirty one with a clean one, rinse the dirty one out briefly and wait until I get four of them collected and then run them through the clothes washer with other towels and things I've dirtied during that time. It was only $20 for five socks, so it's pretty cheap and easy. I plan to have my diatom filter draw water from my sump after the socks in my system.

I will use a Jebao pump to run the DE filter, but I might also plumb an alternate feed into it so I can select a second input instead of the Jebao from my sump, and feed water directly from my display through it periodically. I'm thinking something like having an extra port with a ball valve and a true union after it so I could have a way to install a removable long flexible hose. Then I could periodically put a pump on it and manually move it around in the display to areas I want to reach that might not get stirred up enough. I have sand and won't be switching to a bare bottom, so my flow in the display won't be strong enough to ensure I keep stuff off the bottom moving up and out the overflow. I could also use this secondary feed to clean up the display after anything that clouds the tank like cleaning the sand, moving rocks, etc..

Does that all make sense?

 

[HomeSlizzice]

Whats the benefit of a Diatomaceous Earth filter over a cartridge filter of 1-2 micron? Pretty easy to place one of those on the intake of a power head?

@robert What about using these in a standard canister with like a 300gph pump on a small 10-20 Gallon QT? Would that be an effective alternative?

 

[Scrubber_steve]

@robert What about using these in a standard canister with like a 300gph pump on a small 10-20 Gallon QT? Would that be an effective alternative?

It should be effective, going by the hypothesis

 

[jda]

Back when people used to run these all the time, there was a difference in a DE filter for 1-2 micron living organisms vs a 1-2 micron filter to get sediment. The living things could supposedly get through a sediment filter, which is why DE was best. Not all small stuff is created equal, IIRC.

 

[moomadcow]

how long would running the DE filter take to clear an ich outbreak?