There’s been a lot of talk lately about magnetically coupled adjustable speed drives (MC-ASD) with suggestions that they offer a lower operating cost form of speed control than AC drives, especially for pumps, fans and blowers. I’d like to take this opportunity to discuss this concept and, hopefully, convince you that speed control using AC drives still offers the lowest operating costs.
What is an MC-ASD?
An MC-ASD is similar to the old eddy current drive and coupling. However, it differs from that in one fundamental way. Instead of using an electronic controller to energize electromagnets in the coupling, it uses permanent magnets in the magnet rotor assembly which is fitted to the load (for example, the fan), and a copper conductor assembly fitted to the motor shaft.
As the motor rotates, the relative motion between the magnets and the copper plate creates a magnetic field that transmits torque from the motor side to the load side. An actuator is used to physically change the air gap between the two assemblies, which changes the magnetic field strength and therefore the coupling force between them and therefore the amount of torque transmitted between the motor and the load. By doing that, it controls the slip between the motor and the load and therefore the speed of the load. The motor always runs at full speed but the coupling ‘slips’, acting like a clutch, to change the speed of the load.
Why control fans/airflow with AC drives
A comparison of the power consumed/energy savings achievable for the four principle methods of flow control.
At 80% flow rate (typical for many installations for much of the time):
This graph shows a comparison of the typical power consumed versus speed/flow by a centrifugal pump or fan in a fixed system. Although traditional eddy current couplings are better than using a damper, because they physically change the speed of the fan to control the flow like an AC drive does, they are still not as good as using an AC drive as the solution is less efficient. And, although an MC-ASD might be a bit better than a traditional eddy current coupling, it’s still not going to be as good as an AC drive; the increasing slip between the two parts of the MC-ASD as the speed of the fan is reduced creates an increasingly inefficient situation and energy is lost in dissipated heat; the total power consumed will be higher with an MC-ASD at all loads and speeds/flow rates and therefore electrical energy operating costs will be higher.
You may have seen information that an MC-ASD is 10–15% more efficient than a traditional eddy current drive, but you have to be careful how you interpret this. If a traditional eddy current drive solution was, for example, 85% efficient, that means that an MC-ASD is still only around 93–97% efficient.
An AC drive is typically 97–98% efficient across a wide speed and load range (for example, for 50–100% load and 50–100% speed). From this simple efficiency comparison, you can expect the operating costs with an AC drive will be lower.
An energy efficiency calculation
The following example is a little simplistic, but it should help to explain why an MC-ASD is a less efficient way of controlling the speed and therefore airflow of a fan than an AC drive.
Let’s assume a direct drive fan is installed in a fixed system and is driven by a 4 pole, 400 V, 50 Hz, 190 A FLC, 1490 rpm, IE2, 110 kW motor with 95% efficiency. At 1490 rpm, full speed/full flow, let’s suppose the fan absorbs 100 kW. Using an MC-ASD selected with, for example, 2% slip at full motor speed and load, it wouldn’t actually be possible to run the fan at its full speed but instead, the maximum speed of the fan is only 98%.
Assume the fan is installed in a fixed system where the affinity laws are true, and no matter how the fan’s speed is changed (whether using an AC drive or an MC-ASD) the fan’s efficiency doesn’t change between 100% and 80% speed. Therefore, at 98% speed the approximate power absorbed by the fan = 0.98 x 0.98 x 0.98 x 100 kW = 94 kW.
A while ago, Drives and Controls published an article titled ‘Magnetic coupling is cheaper than a VSD’. The article stated that “the output torque of the coupling is always equal to the input torque” when using an MC-ASD. Therefore, the torque on the motor shaft is the same as the torque applied to the fan. However, due to the relationship between power, torque and speed, (with power being proportional to torque x speed) and because the motor is operating at full speed (1490 rpm) the power output from the motor is 102% (= 1/0.98) higher than the power absorbed by the fan (namely 94 x 102% = 96 kW). Taking the motor efficiency (95%) into account, the electrical power to run the fan = 96/0.95 = 101 kW.
Using a 110 kW AC drive, which would typically have an efficiency of approximately 98% at 50 Hz with the fan absorbing 100 kW (90% load), to operate the fan at the same speed the AC drive would simply set the output frequency to 49 Hz (= 98% of 50 Hz) (and the efficiency would be the same as at 50 Hz). The power absorbed by the fan would be the same = 94kW. The electrical input power to the AC drive would be approximately = 94 kW / (efficiency of the motor x efficiency of the AC drive) = 94 / (0.95 x 0.98) = 101 kW (namely the same as with the MC-ASD).
Now consider operating the fan at 80% speed/flow. At 80% speed the approximate power absorbed by the fan = 0.8 x 0.8 x 0.8 x 100 kW = 51 kW.
Remember, with an MC-ASD the torque on the motor shaft is the same as the torque applied to the fan. And, due to the relationship between power, torque and speed and the fact that even though the fan is operating at 80% speed, the motor is still operating at 100% speed, the power output from the motor is 125% (=1/0.8) higher than the power absorbed by the fan (namely 51 x 125% = 64 kW). Assuming the motor efficiency at this load is the same as at full load (95%), the electrical power to run the fan = 64/0.95 = 67 kW.
Using an AC drive, to operate the fan at the same speed, the AC drive would simply set the output frequency to 40 Hz (= 80% of 50 Hz) (and the AC-drive efficiency would typically be approximately 97% at this speed/load point). The power absorbed by the fan would be the same = 51 kW. The electrical input power to the AC drive would be approximately 51 kW/(efficiency of motor x efficiency of the AC drive) = 51/(0.95 x 0.97) = 55 kW (therefore significantly less than with the MC-ASD).
As I said, the above is just a simplistic analysis. There have been many independent studies done comparing the efficiency/power consumption of different variable speed control solutions. They all seem to come up with the same conclusions:
- an AC drive results in lower power consumed at all speeds tested
- the total efficiency (drive + motor) is clearly higher with an AC drive at all fan shaft powers, and significantly higher across a wide fan shaft power range
- the life cycle costs calculated over a 20-year operating period are significantly lower using an AC drive
The graphs below are indicative of the results published in many of these independent studies.
And, just as a side issue related to the UK market, I’m quite sure MC-ASD technology is not included on the UK government’s energy technology list and therefore users cannot take advantage of the Enhanced Capital Allowance (ECA) tax scheme. I know that doesn’t mean the MC-ASD is definitely not an energy-efficient product, but it does suggest that, at this moment in time, the UK government does not consider it a product suitable for inclusion on the Energy Technology List.
Please use the comments box below to let us know if you have experience using MC-ASD technology and what conclusions you have drawn with regard to OPEX savings. It would be good to hear from people ‘on the frontline’. In the meantime, you can read more about Danfoss Drives products here.
Author: Andrew Cooper, Global Director, Heavy Industry, Danfoss Drives