At the end of last month’s column (Retrofit variable speed pump considerations, Part 1), I promised that we would venture into the world of sizing retrofit circulators for light commercial projects using variable speed pumps. While there are no hard and fast rules for sizing these new variable speed circulators, it is a good idea to at least perform a new heat loss calculation and survey the existing distribution before specifying the replacement pump. Remember that the circulator that is there right now is already significantly larger than necessary.
In the course of specifying a circulator for a closed-loop system during the initial design process, there are typically two methods used in determining the size of the circulator. There is the true measured and calculated method and the partially measured and approximated method. Generally speaking, due to typical time constraints, the latter method is more commonly used, and it results in significant oversizing of the circulators.
Let’s look at both methods to show you the difference in sizing, and the resultant difference in theoretical versus real time energy demand requirements.
The true measured/calculated method requires that a distribution system be completely designed and engineered as it pertains to main and branch flows, piping methodology and material. An engineering technician will have to take actual measurements of the developed length of mains and branches, including all items that may contribute to the hydraulic resistance occurring within the system. This includes all valves, fittings, heat emitters, pipes, circuit setters and basically any component within the flow of the hydronic fluid. In larger complexes, this can be extremely labor intensive, but it will result in a much closer estimation of real time flow demand requirements.
The peak flow requirements are based upon a heat loss calculation, and it is based on a temperature differential value chosen by the system designer. In the North American market, that Delta T is typically 20°F. In the European market, it is not uncommon to see a 40°C differential, which is significantly greater than what we North Americans typically use. I have personally only seen an actual 20°F Δ T on an active space heating system once in my 30-plus years of hanging out in boiler rooms. The majority of the time, the actual Δ T is less than 10°F.
This indicates one of two, or maybe both scenarios. Either the thermal energy load is half of what we thought it should be, or the flow rate is twice what it should be, or both. Even in situations where I personally sized the system and equipment, I have seen this short Δ T scenario. I think it is telling me that the real time energy loads are not what they are theoretically supposed to be. So everything, from the heat source to the heat emitters and everything in between is much bigger than it need be, resulting in significant energy waste during less than peak demand conditions.
Obviously, a minor degree of “what if” needs to be taken into consideration when doing the measured/calculated method of pump sizing, because there is no guarantee that the field installers are going to be able to follow the designer’s intent exactly. As I have shown, though, real time demand is less than it theoretically should be, so there is some wiggle room in the equation.
In the partially measured/approximated method, the design technician will measure the lineal footage of the mains and branch piping, but will estimate the actual number of fittings, valves and other components. This is typically done by adding 50% to the calculated piping pressure drop requirements. To this, he will add the highest pressure drop of the heat emitters that are being used. Flow is based on the North American engineering standard of a 20°F Δ T.
The designer then decides if the job is critical enough to require more than one circulator for redundancy, and he selects a circulator capable of maintaining the required flow at the maximum pressure drop calculated. Even if the designer decides to put some redundancy into the system, he chooses a single circulator to run at any one point in time. Should the first circulator fail to operate, the second, or back up circulator is brought on-line. The only problem with this scenario is that the electrician who is wiring up the system wasn’t told that the pumps were to be rotated and alternated. Consequently, when the outside air drops below a certain point, both circulators are started, thereby providing twice as much pumping power as is required at design condition. All of the above scenarios are situations that I have seen in the field, more times than I care to recount.
Tune in next month as we look at the alternative method of pump sizing on new projects. Until then, Happy Paralleled Circulator Hydronicing!
Mark Eatherton is a Denver-based hydronics contractor. He can be reached via e-mail at [email protected] or by phone at 303/778-7772.