VFDs Spend a little Save a LOT

So one of the big things I have been pushing for in my job has been VFDs on every motor >2 horsepower. VFDs can be incredibly useful tools. First of all, without considering energy implications, they give you finite control of the motor. Just being able to modulate capacity is huge, especially if you have a control system that can handle the input.

Energy savings, however, are really where its at. Most of our motors in the data center are connected to a fan or a pump. For the most part, fans and pumps are variable torque loads and (with the exception of some oddball positive displacement pumps) vary their consumption with the cube of the speed, per the fan laws.

In layman’s terms, if you slow a fan down by 20% (from 100% to 80%) you will only use about half of your original energy to turn the motor (80%*80%*80%=51.2%). So a 10 kW load will draw 5 kW at 80% speed. Even better, lets say you have 2 pieces of equipment (for redundancy in case one fails). Instead of having 1 of them off and the other on at 100%, you can run them both at 50% and achieve the same capacity (hydronic pumping systems have slightly different responses due to changes in other impacts, but the pump law will give you a reliable approximation).

With 2 motors at 50%, you are moving the same amount of air or water. Each motor, however, only uses (50%*50%*50%=12.5%) of its original power. Our 10 kW motor from before is now consuming 1.25 kW, and since we have 2 of them, we consume 2.5 kW of energy. That’s a quarter of energy to accomplish the same task. To put that in financial terms, that 7.5 kW of savings is worth between $3000 and $12,000 annually in energy (depending on how much you pay for your electrons). And what does it cost to make that happen? Well 2 15 horsepower VFDs will probably run you about $1000, and add another 500$ for labor and if you need a controller, another $500 should cover it. Sometimes, you may want to change out the motor, but in reality most motors are acceptable (per the MG-1 Nema Standard, class F or better insulation and a Service Factor of 1.15 or higher are acceptable, and it is recommended to de-rate the service factor to 1.00, limit the speed reduction to 2:1, use a low switching frequency, and keep your cable from the VFD to the motor short, but not shorter than about 24″). Remember, the guys telling you your existing motor isn’t good enough stand to more than double their bid if they include a motor replacement (a good, premium efficiency motor VFD rated with internal grounding could run 2500$ plus installation from your local shop). The VFD may shorten the motors life, so depending on the redundancy present and the criticality of the environment, peace of mind may be worth doubling the project cost. Even doubled, the paybacks are still fantastic.

And remember, if the fan is in a piece of cooling equipment, and it uses less power, it’s also generating less heat which adds a multiplier effect to your savings. So if the “mechanical” PUE is 1.5, and your saving 10 kW in motors, the net impact at your meter will be 15 kW (10*1.5). Say what!

This will be a frequent topic here, and as well will EC Plug Fans. Frequently used in data center “CRAC” and “CRAH” equipment, this equipment can do even better than VFDs, savings an additional 15-35% of your fan energy while moving the air in a more evenly distributed pattern, important for data centers looking to balance their air flow.




4 thoughts on “VFDs Spend a little Save a LOT”

  1. Sometimes people ask how I know running 4 fans each at 75% is more efficient than running 3 fans at 100% and leave one idle on standby. Here’s a general formula for calculating the saving:
    % saving in energy = 1- (n/m)^2
    So running 2 fan at 50% will save 75% over running one at 100%. Running 4 fan at 75% will save 44% over running 3 fan at 100% etc.


      1. That multiplier effect is often overlooked. Reduce a CRAHs motor heat from 8-4 kW, and you also increase that units capacity as its capacity is rated as the sum of what it can remove on its various circuits less what it generates. With EC fans, the air distribution is better across the coil, less heat is generated, and both these factors can improve the cooling output of the unit while at the same time reducing the energy consumption of the equipment. And by reducing the horsepower load, for any giving cooling load we reject less heat into the chill water (or refrigerant) which means less mechanical cooling work.


    1. Actually the cube law more appropriately applies to the fans in our air handlers. If the system has a lot of components which resist flow then the behavior will more closely resemble the function of the cube law (where head is the calculated factor not volume).
      Systems like cooling tower fans follow the cube law very closely, Systems like long ductwork runs from a fan power box or hydronic systems with a long pipe run, multiple bends, reducing valves, check valves, etc, will more closely resemble the square law. Our utility company use (n1/n2)^2.3 for their approximation so they can apply one thumbrule to everything. Jack of all trades, master of none I guess.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s