The Planted Tank Forum banner

1 - 20 of 34 Posts

37 Posts
Discussion Starter #1
I was looking up at different filteration techniques for fresh water aquariums and found this on the net:

Simple schematics of how the filter works with a power filter :

with an air pump:

Have any of you ever used/seen something like this in your tanks, and what are your comments about this method.

9,600 Posts
And if it has been around since the 1960's it's really not "new". As I recall either TFH or AFM have had an article on this in the past couple of years.

714 Posts
Not ugly in any way if you put it in the corner. The plate darkens, goes dark brown/black when it ages and you can attach ferns and mosses to it. And it works.

Here's one of mine, the part above water is blue:

Bad picture and bad lighting, but the filter is on the left corner of the tank, no other filtration:

And here's a topic where we discussed HMF a bit:

Oh, and I forgot to add this, check these pictures, the whole back wall is a big HMF, the blue part is new and the dark part is old and you can see how, well, you can't see the plate. At least check the first picture of the seventh row from top.

Are these real?
15,692 Posts
I have been using one of them for about 2 years now, the tank (43 gal) is totally understocked, and I have cleaned the mats only twice, more out of curiosity than necessity. I guess things collecting in the foam are broken down by microorg activity, so it stays clean. I don't think it is ugly, looks kinda natural after a while (unless you use blue :tongue: ).

With a higher stocking level I would not recommend it, since it would clog quickly and pulling it out and putting it back in on a regular basis would probably not work too well. But it's a great way to hide equipment (heater, temp sensor etc) and very safe since it is inside the tank, no spilling possible.

It's sort of like a simplified internal filter. That tank where I have it in sits in the garage, so not having water circulate through an un-insulated external box helps to keep the temp up in cold winter nights.

1,614 Posts
biological not mechanical

It ought to give great biological filteration, but at low flow, not good mechanical.

IME, in a planted tank, you need mechanical filtration to catch the tiny plant bits that get loose.

I have thought that the Mattenfilter would be great in a discus or goldfish tank, particularly where you have a barebottom tank and do frequent siphoning of solid wastes. I actually did the calcs to build one, but never got around to it. I'd think a breeder would like one.

714 Posts
I have never needed a mechanical filter for my planted tanks, I don't know why, but there's not much floating around :)

But, there's many people using HMF here in Finland and they usually have HMF and a small inside filter to collect the floating particles, you can also clean the surface of the plate when you are doing a water change, just suck it lightly with the hose. There's also many variations of HMF and many of them are easy to remove from tank if it clogs up.

And here's my small apple snail baby tank, the filter foam plate is rather new and has darkened only a bit, heater and the circulation pump are behind the plate:

And another, bigger one, 132 gallon, here I use the plate to keep the babies in only small part of the tank (it's easier to feed them that way), there's Hydor Crystal something inside filter with the snails and Eheim Compact 1000 in the other side of the plate pumping water to the snail side, not visible in this picture:

Both of those plates are just tightly fitted and not glued in place.

And here's my betta barracs (38 gallons) :D The water is murky because I have just put water, filter foam plates and sand there:


960 Posts
what type of foam do you use?

15 Posts
Although this is a very old thread, I would like to pull it up:

what type of foam do you use?

The typical foam for Mattenfilter is reticulated polyurethane. The same material you will find in many canisterfilters (EHEIM, e.g.) as a mechanical substrate.
I will post an article aboutthe HMF and how to build the cornered version in a few days here in this forum.

714 Posts
Oh, looks like I didn't see that there was a question back then for me.

An article in English of how to build a HMF sounds great!

And seems like I didn't show the version where the circulation pump is encased in a filter foam plate box, a sort of internal filter with huge filtration area and it's easy to clean and move.

Here's an old photo, the tank is for snail hatchlings and it has sort of double filtration, the blue box on the right is the circulation pump that's sown inside 6 pieces of filter foam plate, so it's a solid box, the only hole is for the outlet tube of the circulation pump. In the middle of the tank the large filter foam plate acted as a divider and it's irrelevant to this matter.

The main thing this photo is the blue box on the right. Since the circulation is enclosed inside, there's no need to permanently fix the filter to the side or corner of the tank. And it's easy to lift out and clean if necessary.

The one in the photo is a bit small, it's better to build a bit larger one if you are dealing with a normal aquarium, the one in the photo had only about 10-20 liters of water.

From this link you'll find instructions for how to build one, the text is in Finnish, but the photos are universal. :)

15 Posts
Heipähei Satu!

I already posted it in another forum, but I think it is worth reading (and it was a lot of work translating..) I know that the HMF is no longer exotic in Europe, but what i don't understand is that it has such a slow reception in the U:S ?? Maybe you can post your experiences then, so readers see it is not only theoretical..

And seems like I didn't show the version where the circulation pump is encased in a filter foam plate box, a sort of internal filter with huge filtration area and it's easy to clean and move.
You know how this is called here: The Finnish version....honestly :proud: :proud:



714 Posts
Servus, Jörg! Cool, a Finnish version, I have to tell that to my friends too. :)

HMF really should be more popular around the world, it's a dream project for all the DIY people, it's fast, cheap and it works. The ones you glue to the tank take a bit more work since you need to be more careful, but even so it's easy.

The filter foam does need to be the right type, if it's too fine, it will clog fast. But I would think it's easy to find the right type from US too. Here's some examples of the filter foam plates (I think I've used PPI 30 which is the finest one available here, but all of those, from PPI 10 to PPI 30 are good, the coarser ones might be better if there's lots of debris and stuff floating around in the tank, since they can handle it better than the fine 30 PPI): - PPI Filter Foam Plates.

I have used HMF in planted tanks, betta tanks, snail growout tanks (which produce huge amount of poop) and I myself haven't had any trouble with it. And in my old 38G planted tank the back wall was a filter foam plate. There was no circulation pump behind it since I used it only as a back wall and support for some plants growing on it and on top of it (both underwater and above water). But I could have easily added a circulation pump in there too. Even without the pump it does offer a lot of area for the good bacteria, as does the substrate.

Here's the tank with the filter foam plate background
and the thread. Here's some plants growing on the filter foam plate above the water level (Downoi, Riccia etc.). You can see that the originally blue filter foam plate has turned dark brown/black in time, I helped the process and sprinkled some peat on the part that was above the water. Here's the blue filter foam plate right after I added it to the tank.

15 Posts
Hei, I agree totally, 100%... In my eyes it is one of the best "inventions" for DIYers. I think that the big aquarium companies, esp. in the german speaking countries, will think differently about it, i could imagine that this caused a signif.loss in the sales of canister filters and supply parts. Whenever I am in a petshop and see the prices they charge for them i must smile.... With that money I can build at least 5 HMF!

I once used a very fine mat, originally designed as seat cushion,with ~50 PPI, which turned out to be too dense and caused some problems. But good to see that onlineshops sell and send the filterpads overseas aswell.
Interesting that your filter pad at the back wall worked well, cause I know from Olaf (Deters, you get the link on one of the FIN-calculation pages that his setup did not work, too much NO2, but this depends alot on circulation and micro watermovement on its surface. i think he used it in a very special Cory -river tank...But the two typical standard situations, with a filter mat/pad either on one side of the tank or the "cornered" HMF are "foolproof" and will definitely work.
In FIN, can you buy the filtermat at your local fishshop,or Bauhaus or do you have to order it online (from GER) ?

If you can read some german you should see this forum.
which is THE HMF-info source....



714 Posts
Oh, that is interesting indeed. Mine didn't have much circulation, so maybe there wasn't that much stuff getting towards the plate and in his tank there was. I had a lot of Crystal Red and White Pearl shrimp in the tank and they were constantly eating stuff from the surface of the filter foam plate, so maybe they had something to do with it too.

Jukka Järvi (the owner of the website) has a similar kind of back wall with lots of mosses and other plants attached to the surface of the plate, but unlike me, he also uses it as a filter and not just a back wall.

I think most of the people here do order their filter foam plates from Germany ( and have such nice prices), but I think they are also available in at least one Finnish online store and in some actual aquarium stores (but the price is, of course, much higher in aquarium stores).

Gute Nacht!

13,609 Posts
About 20 years ago and reported on the AP maybe 10, I used open cell foam 1/2" thick wrapped layered around a perforated 1/2" PCV pipe with a cap at the bottom and bubbler or powerhead at the top, it's not much different than this and takes up less space.

You use large rubber bands to hold the roll together on the pipe.
Cost: free if you ask the produce guys at the grocery store for the Vestax foam liner that use for the refrigeration cases or meat counters. Rubber bands are cheap.

These can be made to any size or layers thick you want or desire.
I'm not sure why a an old method is now some hot "new low tech idea" though.

I have references for DIY yeast CO2 from 1962 in hobbyist magazines for increasing aquatic plant growth as well:proud:

These filters are very popular with the Killi fish and Guppy folks here

Tom Barr

15 Posts
Hello Tom!

Of course the idea behind the Mattenfilter is much older than the articles posted anywhere in the web, but maybe one should be more precise and discern new and unknown. While it is definitely not new, it it still unknown to many fishkeepers, anf given the fact that it is very effective as a biological filter and quite fast to DIY and rather cheap I think that it is worth discussing.
The same applies to your E.I. dosing scheme. The idea behind must be older than the articles and nevertheless it is still "new"/"unknown" to many, despite its many advantages. That's why I decided to give it a try, and later,convinced of its simplicity and effectiveness:proud: , did a translation engl->german for those who were afraid from so much data in a foreign language. All I want to achieve is to give people access to information they could not get before due to language barriers, on both sides of the big pond;-) Nobody is forced to try it nor do I say that other AQ- filters or other dosing regimes do not work. And some tanks, with a very, very low load of fish and lots of plants do not even require a filter sensu stricto, just circulation.

Your construction idea: yes, why not.Many ways lead to Rome. I use a similar principle, "bottle-filter", in smaller tanks (<12lt). The difference I see to the traditional way (cornered HMF or at one of the sides of the tank) is that no filter mud can deposit on the bottom and work as an additional "buffer" for water-stability. It is the mud that filters, not the plastic mat or the pump. Filters constructed in that way have to be cleaned more often than the HMF, but have the advantage that you can take them out easily.

have a nice day,:icon_bigg


13,609 Posts
I have always been a fan of the DIY approaches, you can learn a lot. I'm not so sure that each project saves you much money at the end of the day, but it can be fun.:icon_lol:

Much like the DIY from 1962, getting the word out to everyone that they work well takes time and effort.

EI is nothing new, that is for sure, I have not made such a claim, I just argued in support of it. Nor is PMDD which the basic concepts came from. No one really "invents", it might be argued that all we do is take ideas from other examples and synthesize them into new ways/views/arguments/constructs etc. You can test it and try it and see. Then get an idea why the arguments of the past cannot be correct. I am amazed that folks still haggle with me over the term "excess" PO4 and NO3 leading to algae. The word has been out for over a decade, but many still have never heard of such dosing not causing algae blooms.

I think I saw another version of this filter design that added this neat little Wet/dry box above the filter using the sponge as the drip plate.
Looked nice and was used on a Killi tank system for 50 tanks all about 12 liters each. They used lighting egg crate diffusers and cut to fit them in the corn and had air bubblers driving the water up over the sponge and another bubbler sending water over the top edge of the main MF. They got both wet/dry and surface current this way, a single powerhead with a split line could also do this.

Many different design additions you could do.
I use sponge prefilters on all my canister filters also.

Tom Barr

15 Posts

So here is the translated article I posted already some years ago at Aquabotanicswetthumbforum(closed now) and at an other site. Some users here already know the concept and use HMF as an alternative to commercially available filter systems. As Tom Barr said before, it is not new, but still many don't know about it. HMF are now very widely spread in Germany, no longer an exotic system restricted to "freaks". And I would dare to say that 98 of 100 persons who tried it will never change back to standard canister filters. The following shows how to build the "cornered" vesion of a HMF, which is nowadays most widely seen.

Mattenfilter - corner version.

The principle is that one uses a reticulated opencell polyurethane foam mat(=the same blue "sponge" material in canister filters) and places it against one of the sidewalls of the tank. A standard aquarium pump is placed in the small gap(2 inches)between glass and mat.
Thus the water flows through the mat with a far lower speed but over a higher surface area than in conventionally sized canister filters (in which the current(>>5 inches/sec) usually is too high to establish a bacterial fauna. Olaf Deters first published the concept of this rather DIY -like filtration concept on his page and many aquarists "converted" to the HMF ("Hamburger Mattenfilter", to honour the region where it came from) . There is an English-language article explaining the basic principles:

"The mattenfilter in English"

which tells you more and should be read first to learn more about the idea. Here I would like to show you a way of how to implement a HMF in a display-aquarium.:

Original author: Jörn Carstens
Translation: Jörg Ortmayr
The original author gave written consent to use his article in this forum.

Construction of a cornered Mattenfilter.

The classical Mattenfilter consists of a foam mat which is jammed in between front and back wall of the tank, parallel to the one of the sidewalls. This principle, which is effective and simple, has been used in many breeding tanks, but has one disadvantage in display aquariums: It doesn't look quite well. One had to look at the front side, often coated with mudparticles, quasi directly into the "sewage plant" of the tank. Clever guys invented the cornered filter, usually placed as a quarter of a circle in one of the corners of the backside. Far away from the frontwalls it is placed far more imposingly.

But a new problem arose: How should one attach the mat ? One couldn't simply jam it between the sidewalls any longer because it tried to regain a straight form due to its elasticity. By this, water always leaked through in between mat and sidewall.

Existing ways for fixings

For stabilization small strips of glas were recommended which were siliconed to the sidewalls of the tank. Disadvantage of this methode: Not every aquarist is familiar with a glascutter und has some rests of glas, even the glazier isn't happy about such mini (?) orders. There always remains the feeling that blurry edges of glas remain. An alternative, plexiglas, available at any home depot, isn't always easy to handle either. Especially the constant glueing with the panes makes problems, some leakiness always remains. Strips of glas and plexiglas hold the mat only in one direction, so there is only a hold against the pressure of the bent mat itself.
Mechanically better it would be to fix the mat on rails. Sometime, somebody got the idea to use a cablechannel for this.
The cablechannel consists of plastic, is available at any home depot, easily to work with and gives extra fixation to the mat because of its U-like profile.

Cablechannels as holding rails

This idea is discussed here and the simple construction of a cornered Mattenfilter on holding rails made of these cable channels shall be shown on photos. Other technical devices like heater and pump are built in, too.This is shown as an example in a rather small aquarium (50*30*30 cm) . There everything is rather cramped, of course. In all larger tanks the construction is a little bit easier, the special problems of smaller tanks is explained in the text.

Original material

Plastic cable channels are available for little money in the electro department of many home depots, usually in different diameters from 2 metres lenght on. Usually they are used for the installation of cables when they are built in subsequentially, aswell for installations over the plastering or when changes in the cablenetwork have to be done often, as they have a removable cap.
The cablechannels are made of plastics and can be shortened quite easily with an all-purpose saw or a "cutter"-Knife. The cutting shall than be treated with a sharp knife to remove the sharp edge. Afterwards, the channel can be siliconed into the tank.


First on has to determine the size of the mat. You get calculationtools at Olaf's website LINK (see the example section at the end of this thread) or at JAN RIGHTER's translated site (english ))

The mattenfilter in English

For the construction of a cornered Mattenfilter one needs:

* a mat in the correct size and thickness

* a cablechannel in the necessary dimensions, length = minimum 2x the height of the tank

* silicone

and these tools:

* a yard-stick
* multi-purpose saw (with fine saw-blade)
* a knife to remove the sharp edge (or an awl)
* a pen to write on glas (e.g. Edding)

And these devices for the aquarium

* a pump with power adapted
* heater
* if desired: CO2-system and co..

First steps:

The cable channel as it is sold can be seen on fig.1. In the middle one sees a small piece consisting of the channel in strict sense, and the removed cap (left side). The cap isn't used any more. The mat can be placed in the U-like profile of the channel.
For this project, we use a mat of 3cm (= 1.2 inches) thickness which is placed in a cable channel of 30 x 15 mm dimensions (= 1.2 x 0.6 inches). For thicker mats, larger cable channels have to be used, of course.

Fig. 1: left side: cablechannel as seen in the shop, middle: channel and cup separated, right side: mat in the channel.

After the necessary calculations to find out the sizing of the mat have been done only the radius of the quarter-circle has to be calculated. This radius is the distance from the corner of the aquarium to the middle of the channel. If you forgot how to do these simple calcualations, you will find a tool at Olaf's website.
The channels are then cut to the correct lenght. The lenght should be the distance from the ground of the tank until to one of the bearing sides. Often there is a small protruding gummed hinge of silicone between sidewalls and ground, and also between sidewalls and bearing strip. So it is practical to shorten the cable channel here a little bit or to leave free a small corner. Shortening usually is simpler. It doesn't matter if the channel is shorter here for a few millimetres, this is were the sand or gravel will be later.

Now that the lenght is set one has to mark the distance to the corner of the tank (this distance = the radius of the quartercircle) in order to silicone the channels in the correct position. I use as a correct measure the middle of the cablechannel. One millimetre more or less doesn't matter, the mat is elastically enough to undo smaller errrors. The cablechannels get silicone on their backside and are glued to the position marked before. The result should look like in fig.2. As long as the silicone is still soft smaller corrections can be done, the silicone "sausage" should really be consistend (?) so that there is no leak between channel and sidewall and that no small fish might get through.

Fig. 2: The cablechannels are glued, mat is still left.

Those who want can close the contact between cablechannel and sidewall with a seam of silicone, but this is not necessary for its functionability. Now the silicon has to dry, one day should be sufficient. These seams don't have to bear that much load like the walls of the aquarium themselves.

Adding technical devices

Now comes the mounting of the devices. In our example, we want to install a pump and a heater. Especially in smaller tanks there is relatively little place left behind the mat, it is rather narrow. Then it pays to invest some time to find the correct pump or correct heater (speaking of dimensions). For tanks larger than 100 litres (> 25 G) the place behind the mat usually is large enough to fit in any standard size pumps and heaters.

Fig. 3: Mounting of technical accessories

For the heater, there are usually only a few ways to place it. In these small tanks, one should try to position it not below the bearing sides so that the cable can go freely upward and one can adjust temperature, if necessary. So the small pump had to be placed below the bearing side. In this example the pump was mounted rather high above. When doing a water change, it will run dry (which can be accepted for a few minutes) or has to be stopped. On the other side, it can easily distribute the warm and rising water from the heater. If one installed the pump more below, as it is assigned sometimes, one would have to build a tube from the pump-outlet upwards, including two 90°-bends. There is no place for that here and it needs additional material.
Here a small garden hose is set upon the pump (it doesn't have to be a yellow rest of a garden hose like in the picture)

Fig. 4: Mount the hose on the pump outlet.

It is recommended to use a hose that is bent a little bit, so that the tank is flown through diagonally. That way one gets the best temperature dispersion. The hose should also be bent a little bit upward, so that one gets a small movement of the water surface. . One has to find out during operations wich is the best position. Now you make a crosslike cut into the mat where the hose is passed through. Then, or rather simultaneously, the mat is pressed into the two cable channels.
This looks like in Fig. 5.

Fig. 5: et voilà : the mat in front, Finished !

In these small setups one needs a little bit of skill and patience due to their narrowness, in larger tanks (and therefore larger cornered Mattenfilters) it is easier. If the dimensioning of the mat allows, a place of about 10 cm (= 3.9 inches) should remain, then one can still work rather easily behind the mat. .A view from above shows how little place is left in this "mini" tank.

Fig.6: Little place in a "mini" tank

In the space behind the mat one can mount the thermometer or the (yeast methode) CO2-outlet below the pump. In larger tanks one usually has more comfort. But one sees how simple a construction for a cornered Mattenfilter can be. As well from the materials needed as for the time. . It is recommended to fill the tank with water and let the pump work for trial. That way smaller incontaminations (like from solvents) are washed out and one can trim the devices, if necessary.

Masking of the filter.

At first sight, now one has a round box in a high intensity blue (depending on mat colour). Not everybody's favourite. But: one sees no longer the suction tube of the filter, heater, etc… in the tank because everything is hidden behind the mat.
But, no fear (?), this view will change. The bacteria which settle will turn the colour of the mat to a dark blueblack to brownblack- In the next figure one sees a 5 week old Mattenfilter in a 80 litres ( 21 USG) standard-size-tank without fish but with snails.

Fig.7: discolouration of the mat, total view

For comparison: in the upper area it is still a little bit lighter, here the waterlevel sinks due to evaporation, here bacteria do not settle, below one sees the darkened mat. During a waterchange, when the mat is dry, one sees that the colouration is from a brown organical film (fig.

Fig. 8: Colonization of the mat, detail

Those who want can place plants onto the mat. All plants wich are offerd to grow on driftwood or stones are acceptable, like javafern, javamoss or Anubias. For sure, others are fit aswell, one can test it out.
One can attach these plants with a toothpick or sew them with fishing line onto the mat. One can also place the javamoss on the edge of the mat and let it creep below. Fig, 9 shows how this looks in a 2.5 month old tank.

Fig.9 masking of a cornered Mattenfilter with plants, frontalview.

This filter is in a 60*30*30 cm tank (23.6*11.8*11.8 inches, 14.3 USG) and is overgrown with two javaferns "Windelov", a small Anubias and javamoss. Fig. 10. shows how it looks from above.

Fig.10.: masking of the cornered Mattenfilter, view from above

On this picture one also sees how the pump was attached: It was simply jammed in between the heater and the sidewall in two blocks of foamed plastic. This has the advantage that its vibrations are no longer at the sidewall, the pump works considerably fainter. If you look well, you see that the formerly blue blocks of foamed plastic have discoloured aswell, so here a bacterial fauna has established, too. With these small colonized blocks one can easily inocculate a new mat, then the bacterial fauna develops much faster than in a "nacked" tank. Especially if sand or gravel and other material in a new tank are brandnew.


One can make a cornered Mattenfilter with only little material and timely efforts. Due to is bowlike construction against waterpressure it gets a high stability. The guide rail of the two cablechannels increases this stability even more.
A mat is rather fast colonized by bacteria. The mat can be masked by epiphyts like javamoss, javafern or Anubias. So the mat is integrated optically in the aquarium within a few weeks completely. Nearly all technical devices can be hidden behind the mat. Thus, the rather large consumption of space of this variation of the Hamburger Mattenfilter (=HMF) is relativized. It can be integrated without problems in large and small aquariums alike.

End of original article
Original German language article:
Last updated October 16th, 2003.
spelling errors removed May 2006

How to use Olaf Deters calculations tools. Until now, there are no translations I know of where one can simply enter his/her aquarium data and one gets results. So here is a step-by-step-process of how to use this site even if you don’t speak german:

First: We calculate in litres and centimetres. Please convert - if necessary - your tank volume from gallons to litres.: 1 gallon=4,546 litres, 1 U.S.gallon = 3.78 litres
Write down the volume of your tank in litres .

Second: Go to

for Q; enter the volume of your tank in litres
n is the number of times the tank should the circulated per hour. Leave it at 2
V is the flow of the water through the mat, it should be between 5 and 10 cm/minute . Leave it at 7.5
Press enter . What you will get is an area in squarecentimetres, to convert into squareinches: multiply with 0.155

Example: you have a 80 US-Gallons tank : that makes 80*3,78= 302,4 litres
Enter Q= 302.4 (mind the . )
Leave n = 2 and V = 7.5
Press enter, result is 1344 squarecentimeters
Multiply 1344 with 0,155 = 208,32 squareinches. This is the size your filtermat has to be.

[Third step: Abolish those crazy imp.units ]

If you want to find out if one of your existing pumps can be used for this method go to

For Volumen: enter again the volume of your tank in litres.
For n (the number the tank should be cycled per hour ) set 2 again

Press =

What you will get is a simple calculation V*n = Q and tells you the performance your pump should have. In our example with a 80 gallons tank (=302 litres) it is 604 litres per hour. If you now click on suche (=Search) you will get a list of pumps that have this size. If you click the two buttons to the left you can go up and down in this list.

Defining the radius of the bent mat (for the cornered Mattenfilter)
calculating the bent mat

Höhe Aqua(cm)
Pumpe (ltr/h)

Höhe Aqua = height aquarium in cm.To convert from inches multiply with 2,54
Pumpe (ltr/h) = the performance of the pump in litres per hour. If you only know the performance in gallons per hour: divide again by 4,546 (U.K.) or 3,785 (US), resp.
V in cm/minute is the speed of waterflow through the mat. Should be between 5 and 10. So leave it at 7.5

What you will get as result e is the radius of your mat .

Our 80 gallons (US) tank gave us (in the previous example) the need for a pump of about 604 litres per hour (= 159.6 gallons/hr)
The height of the tank is 20 inches. That makes 50,8 cm (enter in Höhe Aquarium)
For Pumpe (ltr/h) enter 604
V remains at 7.5
e = 16 cm that makes 6,3 inches

The distance from the corner of the tank to the middle of the cablechannel should be 6,3 inches

I would like to thank Olaf Deters and Jörn Carstens for their consent to use their articles.


German language original URL:
Date 24-11-03
Author :Olaf Deters
The author gave writen consent to use his original article

The Hamburger Mattenfilter has gained it's rightful place among aquarists. Even mail-order shops offer the mat now, even though they would prefer to sell more expensive filter constructions. Since 1996, the mattenfilter has been
explained on these pages in text, drawings and pictures.
In the course of time some questions arose on behalf of readers and users and I get emails nearly everyday about this. The answers to questions have been incorporated either into the text or into these FAQs.

Can the Filtermat be too large ?

No, but the velocity of flow might get too low. If it is too low further problems arise. On one side suction is
reduced. By this, a part of the biology supposed to take place in the filter shifts to the aquarium. In my
experiments, some considerable nitrite levels appeared if speed was reduced too much.

I read that there should be a gap of about 2cm between mat and pane, but then the pump doesn't fit in between any longer. Can I have more distance or does this change anything in the filtration effectiveness of the HMF?

No this has no effect at all on filterpower. The 2cm result because of the construction and are thought as a help in

Can the matfilter be used in seawater aquaria ?

Usually not. According to its principle the matfilter produces nitrate and this is disposed of by water changes or plant growth. In a seawater aquarium the water changes are negligibly small so that this means there is no way of nitrate disposal. Algae as well do not absorb nitrates too much. So the aquarium has to get along with all the nitrate. In a seawater aquarium one therefore installs skimmers which extract proteins etc, before turning into nitrate.

How high is the waterspeed in the mat really ?

Very much higher than in the calculation results. The calculation does not take into consideration the narrowing of the diameter of the mat and its getting choked up with mud. Therefore the real speed is much higher. In practice the recommended approximate values turned out to be good as they contain some safety reserves. For this reason there is no danger if one calculates a too low speed of waterflow.

The calculation does not prove right

An aquarium is a biological system and no exact technical construction. Accordingly, it can be adjusted rather
inaccurately. The recommended values concerning Speed of waterflow and circulations per hour are
approximative values, which resulted mostly from experience. If the speed of water flow is only 2cm /minute but the flow rate in the tank per hour is alright no problems are to be expected. The flow rate could also go down to 1x per hour or rise to the threefold. It will have no effect, if the speed of the wate rflow remains within limits. Basically, my calculations only show the lower limit of dimensioning, the filter can very well be larger .

Can I take a mat from Home Depot/Bauhaus ?

There a positive and negative experiences alike. I supect that Home Depot/Bauhaus/Obi/... doesn't get supplied
for by always the same plastics producer but changes its range of product over time and has regional differences. For this reason one cannot make any general statements.. Its safer to use real filtermats. You can find them here.

How high must be the difference in water height in front of and behind the mat ?

In new mats one often finds no difference in waterheights. This does no matter. To exclude any unwanted
circulations one can, in example, check with a suspension of dry food if the water goes through the mat or if there
are some leaky parts at the margins of the mat. Maximum height difference is somewhere around 4 - 5 cm, in my
experience. Above that, the one-sided water pressure is so high that the mat bows heavily and does not fit tight to
front and backsidewalls. Then one should clean the mat more intensively, Note: There are aquaria where this never occurs.

Where is the area of calculation being measured, from upper edge to ground or to the bottom side

Correctly from upper edge to ground, but this is no exact science. One gets along well if one takes the lenght of the sidewalls. As mentioned above, the existence of the Mattenfilter is more important. It cannot be too large, but I would not size it smaller than at the levels mentionned.

Where to buy the mat ?

I bought my foamed plasticmats at specialised dealers. There one can let it have cut, aswell. Meanwhile all
specialised mail order shops feature the mat.

How thick should the mat be ?

A mat thickness of 5 cm turned out to be okay. Certainly, one can choose a thicker one, but stability does not
increase linearly. One says that the main activity happens in the first centimeters. The other centimeters are for stability.

Can I do without a pump ?

Two of my tanks have a Mat on the entire backside, made of a 2cm coarse foam mat, which has been glued with
aquarium silicone onto an extra pane of glass. The Pump sucks water at one side and pushes it from the other side
through the tank. A river aquarium for catfishes. In the mat enough organisms pile up and streaming is sufficient.
But the aquarium is oblong. I would not recommend that for standard aquaria.

How long does a Mattenfilter needs to get going ?

Oh, difficult to say, Expect four weeks, rather more. It depends decisively on water contamination. It really works well from week 12 on, or more. But then it also works for a lifetime. If the tank has been operated before with a canister filter one could and should let it run parallely for a time (ca. 2 -4 weeks). Negative results are not to be expected.

One says that it should still be squeezed out.

No. It is not in the nature of a mattenfilter to need any care. Only some occasional sucking off every few
waterchanges is necessary. In tanks that are extremely contaminated this might be the case more often and
more periodically. Oddly enough, it also happens in normal aquaria. But one does not need to and should not
squeeze it or de-mud it by principle.

Can a mattenfilter be operated externally, too ?

Principally yes, but some major advantages get lost: An internal mattenfilter develops no suction effect worth to
speak of, as the suction takes place over a large area. Therefore, a Mattenfilter "eats" no food or young fish. Now
if one uses the suction pipe of the filter to bring water to the filter then again the suction basket might get blocked up with plant parts. I do not believe this is a good idea.

"It has happened twice that a part of the water finally made it to the appartment. The reason was, besides me, the
hoses or the canisterfilter. Beside others, the BIG advantage at the mattenfilter is that it is within the tank. If one
uses the Mattenfilter externally, one also looses this advantage."
(Henry Hoch, October 1999)

Can one also use two mats ?

Of course, if the aquarium is 1.80 m or more then I would recommend that, if one takes care that there are no larger parts without circulation. The circle effect gets lost hereby (applies to canisterfilters, too). If in these tanks one installs a filter on one side only then, possibly, the last 80cm are not moved any longer. One can prevent this if one places the gushing out pipe to the other side and thereby pushes the water towards the mat.

Which filter foam is better, the fine or the coarse one ?

Doesn't matter, I prefer the fine foam. It also does not get sealed. But if one has some fish which mess in the mud
or eat a lot, e.g. large cichlids, then the coarse one will serve better. The foam is nothing but the "stock shelf" for
the mud, the supporting framework. If that is now fine or coarse does not matter for the filtration. Only the circeling might take longer in a coarse mat, as its riddle effect is not as high.

How old does a mat get ?

My oldest mat lived to be eleven. The mats I had to change had been beaten up with pushes from the hose at
multiple cleanings. No wonder, it were breeding tanks with a lot of Baby brine shrimp feeding. But even under
these conditions, they are in it for 3 - 4 years for sure and without any intensive cleaning.

How large has the mat to be cut ?

In the width it should be 5 mm broader than the inside width of the tank. Height should be chosen so that it is 10
mm higher than the waterlevel. Apart from that , the size results from the calculation

Must the mat go down to the glassbottom ?

Yes, that's recommended. Otherwise there might be a small gap between mat and substrate, where the water
might pass through unfiltered.

What about the mud behind the mat ?

One can leave that mud behind the mat. Usually it does not distract there and it serves, in some sense, as a store for bacteria and buffer.

Does one safe waterchanges with a Hamburger Mattenfilter ?

No, the mattenfilter does not replace a waterchange. A real waterchange has still to be done.

So what's the point with the HMF ?

It makes a substantially very stable biological environment in the aquarium. This is due to its true and large mass of filtermud. In canisterfilters there is less and in addition, the aquarium (plants, substrate..) filters itself. In a HMF-aquarium, one can quite possibly suck off the substrate or do some large scale water changes. The MF will buffer that. The MF-aquaria are far more resistant against influences from outside. They are biologically stable.

Canisterfilters do work, why then a Mattenfilter ?

No one says that aquaria with canisterfilters have to be adapted at once. But for a new installation one should think about a filterchange. Usually, aquaria with a canisterfilter do work as a large part of the filterwork takes
place in the aquarium and not in the canister. Experiments showed that in average canisterfiltered aquaria, one could remove the filter wool and nothing changed in the aquarium. That means that the filter does not help much,
it only provides watercirculation.

Does it make sense to use a circulation pump filter as a pump for a internal MF ?

Principally yes. But one should bear in mind that, because of frictional resistance inside, the net pump power
cannot be compared any longer to the given pump power. It is reduced. This should be checked when doing the
mathematical dimensioning.
For a repetition: One assumes a twofold circulation of the tank per hour. That means that the pump power per hour must be twice as high as the tank volume (with or without substrate, does not matter). Thereby the waterflow in the mat should be between 5 - 10 cm /minute. If the Eheim pump shows 620 ltr /h and the tank has a volume of 300 ltr/h then it might get rather on the short side of 2x per hour. This is because the real pump power is only 50% to 75 % due to the filling material and its density in the canister.

Are iron fluctuations in the mat preventable or dangerous ?

No, completely normal. The iron which comes from fertilizers into the tank is held stable by chelators. These
chelators do not work forever, the the iron precipitates and gets into the substrate or into the filter. However, some traces come back from there, so that a certain iron supply from the filtermud is given.

Mattenfilter and mould film

A mould film consists of bacteria living on the surface. They are of no importance in small numbers. So if a
mould film appears, that is no reason for panic. A mould film has nothing to do with filtration, both are principally independent matters. A mould film behind the mat is absolutely not critical. I would not do anything against it. At the aquarium side, a mould film can be prevented or fought against by a handy placement of the pump outlet pipe. Otherwise the usual operations: periodical removement with blotting paper or a litre mug.

Mattenfilter and sand as substrate

Of course, wheter gravel, sand or no substrate at all like in breeding tanks, a mattenfilter will always work. This is one of the essential points of a MF. It alone shall, must and will be sufficient to hold all the microbiology for the aquarium. That's the job of a Mattenfilter.

German language original URL:

Date 24-11-03

Author :Olaf Deters

see also some pictures:


3,297 Posts
thank you for providing that illustrated translation.

one of the reasons I would not want a MattenFilter
especially one hosting a layer of Java Moss, is if your
tank hosts a lot of nano fish and shrimp that thrive
on microorganisms, they will spend nearly all their
time living on your MattenFilter surface, which is
the reason why I'm not too crazy about having any
foam filters or foam inlet strainer covers in my tank.
I much prefer using a regular strainer, covered with
a brine shrimp net to keep the fry out the canister,
while discouraging colonies of microorganisms there.
I'd much rather see my nano fish and shrimp work for
a living, and explore my entire tank looking for food.

15 Posts
they will spend nearly all their
time living on your MattenFilter surface,....

No. they will probably not spend all their time on the matsurface only. I am no big fish breeder, just every 3-4 months some tetras and corys in community tanks with lots of plants. The MOs on the matsurface provide them in theri first days with more and better food than I could provide and once they grow larger and are ready for brineshrimps or grindals they will pick occasionally for some extra food and still explore the entire tank. The same applies to shrimps: There is no higher density of shrimps on the mat surface than on wood, javamoss, stones or plantsurfaces. The MOs on the filter surface are a big pool of living "reserve" food for fry of fish and invertebrates alike and there is no risk that your valuable fry get lost if your brineshrimp net should fail. That's at least my experience. What I >read< although is that with "exceptionally" small fry(still feeding on their internal yolk) you should not have any objects in the "tank"/"bowl"/"plastic dish" as there might be the risk that fish get stuck in the mat or in gaps between mat and cable channel or any other "hard" object and suffer damage. So no filtration at all, just very massive waterchanges daily. But these fish bowls with 0,5 liter can hardly be seen as aquaria ;-) Once your nano-fish are swimming free they will profit of the filtermat. IMHO.
1 - 20 of 34 Posts