Hi All:

I do not believe that one can have too high a pond turnover rate. (as long as it does not produce currents within the pond that are too strong for the fish and I do agree that fairly strong currents can be good for the fish as long as they have someplace to go to relax where the currents are weaker). However, as I will show below, the flow rate through the biofilter can be either too high or too low.

It is difficult to talk about bio filters without comparing 'apples to oranges' so let me give a couple definitions before getting into the details. Only a portion of the ammonia is removed from the water passing through the biofilter. So the first definition is the 'efficiency' of a biofilter being the percentage of ammonia reduction during a single pass of water through the filter. If 0.1 ppm of ammonia was present at the inlet of the bio filter and 0.01 ppm at the outlet, by my definition, the filter would have a non-realistic 90% efficiency. The fish are continually producing ammonia and through the relationship of how much is removed by the biofilter each pass, how much the fish are producing, and how often the water passes through the filter, an equilibrium value (hopefully very low) is reached within the pond. So the second definition is the 'effectiveness' of the total pond, fish, filter, pump system as being measured by the residual ammonia in the pond water.

As result of some fairly detailed experiments, I have found that the 'efficiency' of the biofilter is dependent upon the media type, media cross sectional area, media depth, temperature, pH, and both the flow rate as well as the ammonia concentration itself. (and probably to a lesser degree on several other variables). The experiments have indicated that all these characteristics are non-linear. For example, doubling the media depth does not make the filter twice as efficient, nor does doubling the flow rate. Under certain conditions, the latter might even decrease the efficiency. The actual relationships of these variables are quite complex, difficult to measure, and probably impossible to calculate.

The flow rate of interest is the velocity of the water across the internal surface area of the media. Realize that this velocity depends upon the choice of media as well as the physical size (flow area) of the biofilter. A denser media will result in a higher internal velocity for the same amount of water per unit time. Realize that even the highest velocities we are considering are too low to 'wash' the active bacteria off the media. However, if any air bubbles are mixed in with the water flowing through the media, bacteria can be dislodged at even fairly low flow rates.

As the velocity of the water through the media increases, more ammonia is being presented to the bacteria per unit time but they have less time to react with it. As the velocity decreases, less ammonia is being presented per unit time but more time exists to react with it. Again, all of these relationships are non-linear. There is some optimum flow rate for a given ammonia load and filter configuration. From the overall effectiveness aspect, the higher flow rate will pass the pond water through the filter more often which can reduce the residual ammonia level in the pond water if the filter removes enough of the ammonia each pass. Another characteristic of a biofilter with a given configuration is that as the ammonia load increases (for a fixed flow rate), a point is reached where approximately a fixed amount of ammonia passing through the media is converted each pass. This is believed to be due to the limitations of the maximum size of the bacterial colonies which is probably primarily related to the internal surface area of the media. I call this point saturation and the efficiency drops as the ammonia load increases beyond this point.

As the ammonia load becomes quite small, the filter efficiency also decreases. This is believed to be related to the size of the channels between the surfaces of the media. As the load decreases, the probability of a molecule of ammonia being within a required capture distance of the bacteria also decreases. At some point between saturation and zero ammonia load, the biofilter reaches maximum efficiency for a given flow rate. Fortunately, most of the non-linearities are not a problem. Although the 'efficiency' of the filter drops off as the ammonia load decreases, our filters are usually designed so that they operate toward the bottom end of the scale with a typical operating efficiency of 5 to 10% and a maximum efficiency of 20 to 30%. As the ammonia load increases slightly, the efficiency also slightly increases resulting in very little change in the effectiveness of the system. Where problems occur is when the filter is operating at or near saturation (i.e. too small a filter) and an increase in load can cause a decrease in the efficiency resulting in a significant degradation to the system effectiveness.

To determine the 'ideal' flow rate for a filter, ammonia concentrations in the pond must be measured at various flow rates allowing the pond to come back into equilibrium (24-48 hours) between flow rate changes. Special equipment is required, as these are very low levels under normal conditions. This is tedious to do and this 'ideal' flow rate further changes as fish are added, feeding rates, and/or temperatures are modified. I have found empirically that a media velocity of around one foot per minute is appropriate for most traditional filters (not pressurized filters). This is not a very firm number as there are a lot of other variables that come into play. At least combining this with a pond turnover rate (through the filter) of around an hour seems to give a pretty good design starting point for the normally constructed traditional up-flow, down-flow, or horizontal flow filter configurations. Increasing the pond turnover rate without increasing (or even decreasing) the filter flow rate (i.e. building a bigger filter) would be better. Rather than excessively exceeding this velocity, and to conform to spousal design criteria, I would suggest having a separate pump bypassing the filter and supplying any desired water features in the pond. This pump can be shut off when the water features are not desired but the filter pumps just keep doing their thing 24 hours a day.

Norm Meck

Filter Flow Through Rate, by Chris Neaves

Thanks for stopping by!

Last modified on Tuesday February 15, 2000