Who rates skimmers
 

Why are skimmers rated for system gallons rather than for bio-load? A 200 g system has little load compared to a fish only 100g

I wana know too. And why, are some WAY more expensive than others? Cause it's made in Germany, means it's 3x the price?

Why not ask the skimmer manufacturer? Aren't they the experts?

Is because you make something, does that instantly make you an expert? :D

The ratings are ridiculous. I think a BakPak is rated for 100 gallons or something stupid like that. It seems the lower the quality, the higher the rating. I think it's something you just need to get a feel for. Also, a lot of people think that if they have a 180 gallon tank that a skimmer that is good for 180 gallons is a perfect match. Well, that's just silly, you're running the skimmer as MAX suggested volume?? The skimmer should be bigger than the water volume.

Nah that's not what I was getting at, I'm certainly no expert. It was sarcasm, kind of an inside joke to another thread :wink:

Skimmer ratings will never be accurate, what works for one won't work for another. The ratings are general guidelines and some common discretion is required from the buyer. However for the most part skimmer ratings aren't that off for the major brands (cheap novelty brands aside). I've used a Euro Reef 180 on a ~250 gallon decently stocked reef tank and it worked well for the system. Also your skimmer dependence actually decreases with a fish only, yes you can use a larger one and it can be beneficial but the higher nutrients won't necessarily cause a problem. Many people with FOWLR tanks don't even use skimmers and do quite well.
I can't see a bio-load rating being practical, how exactly do you rate bio-load? Either way it's common sense, if you plan on having a higher than normal bio-load for your tank you always have the option of using a larger skimmer. One could simply assume the rating is based on a standard load for the given volume.

Just an off the cuff though, could you not rate a skimmer based on ml/hour of "dense" skimmate (i know dense is a relative term) so at least all skimmers are on a level playing field. You could then calculate how many ml/hr you would need based on your bioload and needs. (no I don't know what that calculation would be, but it wouldn't take much other than experimenting with a baseline skimmer, and varying bio-loads to come up with an approximation)

I wouldn't want to be that sales guy trying to sell those skimmers :wink:
Simpler is usually better and specifying specifics like that will cause huge issues as results will always vary. It's one thing to say this skimmer will typically handle 200 gallons of tank volume but to say this skimmer will produce X amount of skim mate is another story. There are certainly skimmers out there that have bad ratings, seaclones are one that come to mind, but good quality skimmers are rated fairly decent in my books, and on occasion some manufacturers do actually give a few rated volumes based on high and low bio-loads.
I think too much thought is going into this, consider all the variables involved in actually determining the right skimmer. Amount of fish and other livestock, size of fish and other livestock, type of fish and other livestock, amount of food feed, allowable nutrient level (sps-low, lps-med, fowlr-high), tank volume, and so on. I mean these ratings are just general guidelines, just like fuel ecomony ratings on cars, they aren't that accurate either but there are too many variables involved for it to be accurate for each person.

One thing I know, BOTH my skimmers suck, and Im to broke to get BK's LMAO

Tunze gives you a little more help by listing a "maximum" tank size for their skimmers which you subtract down from depending on what you are keeping in the tank (e.g. SPS, LPS, softies, FO).

little dated and yes could be problems but here:

http://www.aquariumcontrollers.com/a...p#skimmerchart

Best they can do is give you ballpark ideas and then you ask people who have experience with them I think.

I'm recalling that study that recently tried to measure WHAT was being removed by skimmers and how quickly various models removed it.

I believe that their conclusions were that the "larger" skimmer (the one the processed a larger volume of water per hour) removed the measured stuff more quickly, but that all the skimmers tested reduced the measured stuff to about the same level given time.

It struck me that a big difference between the test environment and an actual aquarium is that the aquarium continually generates new waste so a truly undersized skimmer won't ever get the chance to catch up and waste could accumulate.

Yes, I agree...it is more about bioload than actual water volume.

IMHO, size matters. In order to increase the amount of water it can process per hour, the chamber needs to get bigger. Increasing the flow rate with a smaller chamber just reduces the amount of time the water is in contact with the air bubbles (dwell time). In my case, I wanted to get to 1.5x tank volume per hour (600gph) with 2 minute dwell time inside the skimmer. That means that I need to have a 20 gallon chamber....yikes!

After that, I didn't really pay attention to what the skimmer says it is rated for.

BK will tell you in a heartbeat that you can overdo skimming by buying a unit that is oversize, it simply doesn't work, or should I say doesn't work properly.

I know there is a distinct difference between having the skimmer in the tank with no sump as compared to having a sump with a 10 times turnover, it just isn't the same, the tank install though ugly will work better every time, explaining it is a different matter.

If the tank install works better every time then is it reasonable to assume a hang on the back skimmer on the main tank is also better than an in sump set up?

Palmer

In sump works better because of the overflow effect. The high specific gravity in salt tanks results in proteins/organics rising. the idea of the sump and overflow system is that the overflow constantly sends the top water from the display to the sump.

Excluding the sump
 

Hi there, so you are suggesting that the best way to operate a skimmer is to take water directly from the display tank and return it there instead of doing that from the sump?

Brad - I like your theory, it makes sense to me. I have a question though. Which of those 2 parameters is more tweak-able/forgiving - tank volume per hour or dwell time?

For example, I have ASM 5G with about 6gal cylinder capacity. Manufacturer suggested system size for this one is 700gal. Granted, real life users would not put this skimmer on anything bigger than 350-375gal. Now, applying your theory, keeping flow at 1.5x volume of the tank with 2 min. dwell time this skimmer is no good for anything bigger than 120gal. Which parameter would you tweak for bigger system? Volume per hour or dwell time? No, "Get a bigger skimmer" is not applicable answer at this time.

Thanks a lot.

Cheers.

I agree that too big of a skimmer can happen, in real life it rarely happens, unless you are rich (or used to be).:wink:

Look at others that have exceptional systems, see what they use and copy them. Don't be afraid to spend a fair chunk of change, it is a very important part of the tank. And, unlike lighting, doesn't keep whacking your wallet with additional costs (cept monthly electricity).

Yup, that is the math...If the chamber size is constant, then increasing flow rate (tank turnover) reduces dwell time and vice-versa. Assuming that you have re-circ'd the G5, then you can control the flow rate. If you haven't re-circ'd, then your flow rate is set by the amount of water drawn in by the NW pumps and you would have no control over it.

Really, controlling flow rate is one of the big reasons why one might choose a recirc skimmer over a non-recirc skimmer.

Given the limitation of the size, I would be inclined to try decreasing flow rate for increased dwell time. Playing with the flow rate a bit is certainly worthy of experimentation for sure.

Plus...if you wanted to increase the chamber size, you can have my old G6 body if you like.

Brad - yes, re-circ'd is what I have done to G5. You have PM :biggrin:
Cheers.

I am suggesting that you have the sump turnover rate at the correct volume to get maximum results from the skimmer, typically this will be a lot less than the previously suggested 10x rate.

But you can't explain why? Just cause, right?

This is somewhat dependent on your sump layout but think of this analogy:

If I drop a small piece of sinking pellet food down my overflow and I have one of two options:

a) High flow rate through my sump
b) Low flow rate through my sump

Which one do you think is most likely to allow the intake/suction of the skimmer intake to pull in the piece of food? The one ripping the current past the skimmer or the one slowly plodding along?

I'd certainly rather pull mechanical chunks of organic waste out of my tank before they have a chance to break down and then become part of the chemical soup in the water.

While I see your view and get what you're trying to say two problems exist with your example. First a skimmer removes dissolved organics which I guess can be compared with food but we don't place organics/food in our overflow but rather rely on flow to get them there. So in real life that piece of food will get to the overflow faster if flow is higher, resulting in going to the sump sooner. Dissolved organics should be looked at differently than a food pellet, its concentration is distributed throughout the tank not in one particular location.

The second problem is the assumption less flow through the sump results in more water being filtered though the skimmer before going back to the display. First these are independent, the skimmer outputs the same as it takes in. So exactly how could one possibly determine how much overflow water is actually going in the skimmer and not simply bypassing it completely? With a lower turnover the amount of water being constantly recycled through the skimmer could actually be higher meaning less overflow water is going in compared to high turnover.

Regardless the point I'm making is that the two are completely independent. Dissolved organics are dissolved meaning the concentration throughout the tank will be virtually constant. The amount of organics taken in by a skimmer is essentially constant and independent from the water flow moving past it. High or low turnover (in limits), a standard in sump skimmer will work the same.

Lower sump flow rates are related to lower noise and less air bubbles but not skimmer performance. The only reading I've seen relating these two suggest higher flow rates as it will "keep solids in suspension, which increases feeding opportunities by fishes and invertebrates, and improves filter/skimmer opportunities to export such matter in a well-designed system"

A good skimmer removes both DOC and solid particulate. For instance I've had shrimp molts end up in my skimmer cup. I want my skimmer to remove as much solid particulate as possible before it becomes DOC - hence the reason for not having a filter sock that requires constant changing.

I agree with the idea that skimmers should be rated by bioload, but manufacturers are saddled with magic numbers that everyone in the hobby can understand. A better way to do it would be to measure water quality and compare it to a second sample taken 6 or 12 hours later. This would give you an index of decline and allow you to access how much equipment you need to keep it in check. Unfortunately that lowest common denominator is system volume not load. Some heavily stocked tanks are well balanced and subsequently put less demand on a protein skimmer. Other methods of nutrient export, chemical filtration and mechanical filtration also take the burden off of the protein skimmer. Protein skimming is limited to the removal of 80% of the available proteins and only 25% of the available TOC. For a proper evaluation of the abilities and limitations of protein skimming look here. http://www.advancedaquarist.com/2009/1/aafeature2

Protein skimmer design isn’t rocket science, but the only scientific formulas we have in determining size, contact time and bubble size are those offered by real life rocket scientist P.R. Escobal. His formulas have not been disputed to date and are summed up well in the following link http://www.hawkfish.org/snailman/skimmer101.htm

It is widely accepted that a two minute dwell/contact time is ideal for separating the maximum number of proteins and TOC (total organic carbon including both dissolved and particulate forms). This doesn't preclude the use of a protein skimmer that fits under your roof. In the case of a 400 gallon tank with a 1.33x flow through the protein skimmer, and a two minute dwell time, the contact chamber of the device needs to be 18 gallons. It sounds big, but so is a 400 gallon display tank, and you only need that size for optimum performance.

In addition to the comparatively low throughput rates recommended by Escobal (1.33 x per hour), protein skimmer manufacturers typically recommend 250-500 GPH throughput for larger units according to system volume and respective protein skimmer size. These specs typically equal 2 or 3 x the volume of the tank per hour and are chosen by practical trials rather than mathematical formulas. So why do some people disregard the recommendations of manufacturers and the formulas of rocket scientists in order to follow the 10x the volume of the tank guestimation of some kid working in an aquarium store?

The root of the 10x rule of thumb goes back to the days when we didn’t use closed loop pumps and we relied on a one pump system to keep detritus suspended (for feeding corals and collection in mechanical filters), to provide adequate gas exchange, and to deliver unprocessed water to filtration devices. There is no reason why we can’t pre-filter the water by collecting a thinner amount right at the surface air/water interface. This thin film is rich in surface-active agents (surfactants) that migrate to the air/water interface through the same process carried out in the protein skimmer. The practice of turning over large quantities of surface water draws a thicker intake from the display tank, thus diluting the hydrophobic (repelled by water) proteins we are trying to pre-filter.

The ideal system would collect a thin film of surfactants from the display tank and deliver it directly to the protein skimmer sitting in a sump (for safety/spillage reasons), after which it would be directed to the sump where the water can be further processed by a refugium, UV sterilizer, mechanical and chemical filtration methods, all of which requiring a similar slow throughput. The idea is to process all of the water thoroughly and only once before returning it to the display tank.

If your skimmer design doesn’t allow for a gravity feed, then the sump should be divided into two sections. One section holds the protein skimmer and the new unprocessed influent water, and the other holds water that has been processed by the protein skimmer reday fro further filtration and return to the display tank via a pump with the same output as the skimmer pump (less head loss). The water entering the first compartment by gravity or siphon from the display tank at the rate of “X”, can be pumped through the protein skimmer at the rate of “X” by a small feed pump. The processed water would exit the skimmer and enter the second compartment to assure it can only be filtered again after entering the display tank or in the case of the protein skimmer pump being greater than “X”. If the protein skimmer pump is less than “X” some water will bypass the protein skimmer, but the water will still only get processed once before returning to the display tank. The flow rate through the sump is directly related to the flow rate through the protein skimmer if you have an efficient first in first out (FIFO) plumbing system for your skimmer.

Refugiums, chemical reactors and UV units require the same type of flow volume so they can all be plumbed inline. In order to get a 99.9% kill ratio with a UV sterilizer, you can run no more than 4.16 gallons per hour per watt through it. In other words, a typical 30 watt UV unit should be fed no more than 125 GPH throughput, or the kill ratio will diminish accordingly. If you look at typical display tank and protein skimmer sizes you will find that the turnover ratios are about the same. Once again, P.R. Escobal’s formulas are considered to be the gold standard when sizing and establishing throughput for UV sterilizers. His UV formulas are summarized nicely here. http://www.aquariumadvice.com/articl...ers/Page1.html

The reason why you need to make sure your protein skimmer isn’t oversized is that you will have premature bubble popping which releases the bound proteins. As the bubble rises through the contact chamber it bombards the water passing in the opposing direction (down). The smaller the bubble, the more stable it is and the more likely it is to make it to the reaction chamber in the neck. Bubble stability increases as the bubbles bind hydrophobic proteins to form a skin. An oversized protein skimmer will cause premature bubble popping due to the lack of bubble stability. Proteins are both hydrophobic and hydrophilic (attracted to water) as they are polarized at each end. If the contact chamber is too large, the bound proteins will migrate back into the water.

Protein skimmers are primarily designed as chemical filtration, not mechanical. If you want to keep detritus, or a sinking food pellet suspended, it is the duty of a closed loop system, not the throughput. A throughput ratio of 10x will certainly not easily achieve this goal anyway.

The other benefits of following the scientific formulas and directly feeding protein skimmers is quieter overflows, no micro-bubbles and salt creep caused by oversized drains that lead to a sump compartment (ater goes directly to the skimmer where bubbles are welcome), less heat and noise from return pumps, better pod production in refugiums, a higher kill ratio for UV sterilizers, and better mechanical and chemical (longer dwell time) and mechanical (less bypass and breakdown) filtration.

The biggest problem in this hobby is the lack of effort hobbyists make to fine tune their existing equipment. Keep the glass on your lighting system clean, and spend less time "upgrading" bulbs. Use a smaller return pump and filter your water more thoroughly with fewer drain problems and better surface skimming. Plumb your skimmer right into your drain and get away from a "water mover" system that offers inefficient filtration.

While this might happen from time to time I assure you protein skimmers work on a molecular level and are not effective at filtering out particles greater than a fraction of a millimeter. The fact that it collected such items is more of a coincidence than a function. It will also only filter particles that float or are basically neutrally buoyant. This will also likely cause more maintenance requirements to keep your skimmer clean and working efficiently. Skimmers will work better for longer with a mechanical pre-filter. Quite simply you're just justifying an omission.

Mr. Wilson, I'm slightly confused by your post and was hoping you could clarify.
First, only external/recirculating skimmers have an adjustable turn over rate and none are traditionally gravity fed, unless by gravity fed you simply mean with a pump because no skimmer should be run off gravity (such as straight from overflow) as it will never produce constant flow and cause the skimmer level to be either too high or too low. Also building a separate skimmer chamber which requires a pump to feed water out at the same rate of the skimmer is a new one for me, I have never seen this. The 10X turn over recommendation is based on tank/sump flow and not skimmer flow, I have never heard of someone running 10X tank turnover through a skimmer before. Today's skimmers are not co-current, air stone powered tall structures. They mostly use a single pump and do not offer flow control. Lastly sumps should be designed to handle the tanks/owners requirement for flow, this is fairly simple and eliminates things like mechanical filtration bypass and would provide adequate contact time for carbon.

Mr Wilson that was awesome. Thx for taking the time to share that.

+1 Mr Wilson!

The slower the tank skims the more concentrated (read dirty) the water going to the sump is, the dirtier the water the more efficiently the skimmer functions.

But would that really mean the skimmer is more efficient, or working more efficient? or just theres more there to remove.

Like vacuuming a dirty 3 year old carpet compared to a 1 month old carpet? does it mean the vacuum is not working as efficiently on the newer carpet because its not pulling up as much junk?

Dan, if you wanted to remove a film of oil from the surface of a tank of water would you slowly add water to the tank so it comes off the surface slowly and gets caught in a sponge or would you pump it quickly.

Absolutely, you cannot use the type of skimmer feeding that I am recommending with a one pump skimmer where the water feed and air production are produced with one pump. These protein skimmers are less efficient for that very reason. One pump skimmers still benefit from directing the effluent (processed) water to the return pump section of the sump as it allows new unprocessed waster to be pumped through. These types of skimmers function best if they are plumbed directly to the tank with no sump as they process water indiscriminately (no first in first out).

The most efficient and quite drain from a surface skimmer is a siphon with an emergency overflow that has a constant slow trickle. There is no "toilet flush" siphon and siphon break with a constant siphon drain so the water level will be constant. A Durso or Stockman drain will also drain steadily if it is vented properly. The only instance that could cause a variation in the drain rate would be a toilet flush of an improperly vented drain. If "X" amount of water goes into a display tank, then "X" overflows at the same rate. The water level in the protein skimmer is regulated by the back pressure of a gate valve and or the height of the drain line.


It is a simple idea that greatly improves efficiency that is over-looked.

Tank flow rates are governed by closed loop pumps or less effective powerhead systems. Moving water to and from a sump does nothing for tank flow other than a side effect. The purpose of having a sump is to create a remote location to place filtration devices. Once you establish how many GPH these devices require, then you can pick a pump that does moves that much water. Anymore flow redundant at best, but more likely to decrease the efficiency of the filtration devices through a shorter dwell time and less concentrated surface skimming. Excessive, redundant turnover also causes noisy drains, salt creep, micro-bubbles, less surface tension at the overflow, and heat & electrical costs of an oversized pump.

You are correct, most skimmers these days are counter-current. The skimmers that were used in Escobal's work were also counter-current. Almost any protein skimmer can be plumbed for a closed loop (recirculating) pump for bubble production and a secondary feed for throughput of water. Protein skimmers with wood air diffusers introduce more air flow and with smaller bubbles than needlewheel, pinwheel, and venturi aspirated skimmers. The air-driven diffuser skimmer in the skimmer test I sited gave the same results as the downdraft and needlewheel skimmers in the test.

If you base the design on what you want, then stick with 10x turnover. If you want energy efficiency, better water quality, less noise and micro-bubbles, then size the return pump according to what the filtration requires to operate best.

No, delivering more water to the sump than the filtration devices require is not simple, it's more complicated and a waste of resources. The only bypass I would recommend is an emergency bypass for the protein skimmer feed incase something gets clogged, but I would have one in place with any drain system that doesn't have an emergency overflow. Todays standards are 20-40 x the volume of the tank in total flow. I go for quality over quantity, but even with the best flow dynamics, you won't achieve this with 10x the volume of the tank.

The worst scenario I come across is what I call a "water mover system". This is a sump with one or two poorly tuned or non-functional filters and lots of water coming and going from the display tank. A lot of resources are used to move the water, but little or no positive gain is achieved. Typically these systems have old carbon that is allowed to leach out any and all TOC (total organic carbon) it has absorbed, UV sterilizers that act only as heaters because the bulbs are expended or burnt out, protein skimmers that don't collect skimmate or sludge in the neck, and refugia that are allowed to overgrow and shadow the lower portions causing a slow unnoticed die-off.

The other way to test the theory is to take a small cup and quickly dip it into the tank, then do it again but this time submerge it a millimeter bellow the surface and you will see the surface tension isn't broken as much and water & oil is pulled from a thinner sample, over a greater surface area. You will see it pull water from several feet away.

Having a thin film overflow the surface skimmer will reduce the need for a coast to coast overflow. The best way to test the efficiency of an overflow is to place some flake food at one end of the tank and see if all of it either sinks or gets skimmed within 30 seconds. Any food that gets caught up in the middle, edges or corners means you have dead spots where films will accumulate. These oils are not only bad for water quality, they also diminish lighting intensity, and gas exchange.

My coast to coast, in conjunction with the OM 4 way, no dead spots on my waters surface :D

Eliminating dead spots on the surface is another one of those details that makes a huge difference at no cost or compromise.

Good points and good information there Mr. Wilson. I agree that the best skimmer design is based on recirculating counter current but of course it is a simple option and not a requirement (just like everything else). Recirculating skimmers, on average, cost double and not everyone should consider modifying there skimmers, this is not a simple modification for most. Drilling the expensive acrylic body can often result in fracture and voided warranty.

I wouldn't personally ever setup a skimmer to be fed by a siphon over a pump, in theory the perfect siphon will deliver consistent flow but in practice it's not the case and if flow is suddenly increased a big mess will exist. But that doesn't mean it can't be done, I simply don't see the benefit and wouldn't recomend it, however everyone's thoughts are different.

Moving a lot of flow through the sump has many benefits, of course so does moving little flow. It's important to remember it's not a simple science and there are many ways to skin a cat. I won't go into too much of my reasoning again but I will say real experience and evidence trumps theory. I believe that the best resource for this can be found in the RC site where every month some of the worlds best tanks are showcased and described. There you will find almost all tanks showcased run return flows around 5-10x display volume, if not more. For me it's hard to argue with success, they got to be doing something right.

You're obviously a smart guy and you have experience in aquarium design but we definitely have some differences in opinion and what we consider simple, which of course is a good thing :wink:
I also noticed you're new to this site (WELCOME BTW :biggrin:), perhaps you could provide a link to your tank(s) or some of the work you've done. I always like to see the work of others who offer different approaches.

I know, most people would rather do their own dentistry than drill into a new protein skimmer :)

Never say never. Technology and methodology is always adapting and evolving. If a sudden increase or decrease of flow ever occurred I would spin it as "intermittent wet skimmate to remove strongly hydrophilic proteins" :)

Such as...

You haven't gone into any details other than "everyone else is doing it", and from my experience they are not all doing it. In my travels I see 3-5x the volume of the tank to be the common practice for return pumps. Apart from theory, in practice I use 3-5x the volume with a total tank flow rate of about 20x the total volume.

Dr. Stephen Spotte said it best "The successful maintenance of a seawater aquarium is mostly witchcraft mixed with a little science. In this book I have attempted to describe the science, but with the realization that understanding the witchcraft might be more useful."

Unless you care to share some of your witchcraft, we are stuck with my science :)

I contribute regularly to RC threads, but I have never bothered to read more than one or two TOTM profiles. I don't find them to be completely honest or particularly reflective.

There is no cause and effect of what you are claiming. If sure most of the tanks were glass rather than acrylic and used metallic pumps rather than magnet coupled ones, but this doesn't offer empirical evidence of anything. If you look at the TOTM historically you will see an evolution of technology and methodology. To deny this is to deny progress, and that is what I see with your opinion.

I'm more experienced than smart, but I'm getting there. I have the advantage of 30 years of mistakes to learn from. I'm done making most of them :)

I'm active on a few sites, so I decided to divert some of my time over to this forum. This forum seems more active than the other Canadian site I belong to that I won't mention :)

I don't expect anyone to do a 180 and follow my advice, and what I said last year was as different from today as today's will be from next year's. I participate on these forums to learn and share what I have learned. I get inspiration from others and new insight into old questions.

I didn't come here to pick a fight, but it looks like I landed in the middle of one. Don't take anything I have said personally. It's all in good fun.