I've been thinking about ways to maximize efficiency in large buildings, and one thing I thought of was to place some sort of turbine in the exhuast vents or smokestacks of factories, power plants, and large office buildings. Is there a reason that this hasn't been done? We're expending energy to move the air, so why don't we try to recover at least some of it? It seems like it's basically free energy.

  • $\begingroup$ According to my late neighbor who worked in the business, lighting systems in most large retail stores are cooled with the water, if possible, then used to assist in heating as needed based on the environment. This tidbit was also from the days of incandescent lighting so may not be as applicable today. $\endgroup$
    – jko
    Commented Oct 5, 2020 at 14:31
  • $\begingroup$ Then the fans will have to work harder to turn the turbine? $\endgroup$ Commented Nov 3, 2020 at 18:43

3 Answers 3


So, if I understand correctly you are proposing to use a turbine in the exhaust vents in order to reclaim the energy from the flowing air.

However, in most cases the air is flowing due to forced flow, if you use a turbine to "suck" energy out the air then it tends to slow down. So then you would probably need to use bigger fans to push the air. So its a chicken and egg situation.

The main problem is that, as you increase the air speed, air flow tends to exhibit greater losses, so by the time you get from the fan to the turbine you would have lost more energy (compared to slower speeds). So IMHO, its going to be very difficult to harvest anything meaningful.

There are of course some exceptions. If for example you could create air flow without expending electrical or mechanical energy (see for example Trompe wall) and if you could dimension it in such a way that you would have an excess of air flow, then it would be possible to harvest "energy" for free.

Also, if you are talking about reclaiming other types of energy (not just kinetic energy of the air), as @SolarMike understood, then yes, it is more possible. However, more often that not it is not done. Actually, thinking it, and implementing it efficiently is a characteristic of a great (not just good) engineer.

  • $\begingroup$ But in Solar Mike's case, the heat recovery was designed in to the building design. So it is possible, but to be effective it must be a component of the design. $\endgroup$ Commented Oct 5, 2020 at 16:51
  • $\begingroup$ I upvoted SolarMike's answer precisely because I agree with it. My point was that most engineers (especially in small scale installations), don't bother with unification of systems. A great engineer considers the opportunity and implements the most efficient solution. $\endgroup$
    – NMech
    Commented Oct 5, 2020 at 16:54
  • $\begingroup$ I come at it from ship design, where heat recovery from stacks for a diesel driven ship could power ships auxiliaries, but there is a problem synchronizing to ships diesel generators, so it is not done in many ships. It has to be a conscious decision from designers to be effective. $\endgroup$ Commented Oct 5, 2020 at 16:58
  • $\begingroup$ When you come across it being implemented successfully doesn't it fill you with awe and admiration for the "craftmanship" of the head designer/engineer? $\endgroup$
    – NMech
    Commented Oct 5, 2020 at 17:01
  • $\begingroup$ Yes. And it has more to do with better tools, coupled with a better focus on energy efficiency. At times, all you have to do is make small decisions to always do it. $\endgroup$ Commented Oct 5, 2020 at 17:10

Energy recovery works best when it is a key component of the design. Solar Mikes example. From planning, design, operation and maintenance. Lose one of the four and it will fail.

But as an addon, it has limited success. NMech's issue with blocking air flow, which requires increasing airflow. You cannot get something for nothing.

You could add heat exchangers to smokestacks of factories and power plants to recover a portion of the heat going up the stacks. Ultimately, the viability of this approach depends on the amount of heat going up the stacks. Can heat recovery be justified via cost, energy efficiency, space, etc.

50% of the fuel burnt for ship propulsion is heat and up to 11% (5.6% of total) of this loss can be recovered by economizers and heat exchanger from gases going up the stacks. Here, the scale is sufficient to make energy recovery from stacks viable. 5.6% of 69.72MW is 3.8MW, which is more than enough to power most 2 stroke diesel driven cargo ships.

From MAN Waste Heat Recovery System (WHRS) for Reduction of Fuel Consumption, Emissions and EEDI

enter image description here

Heat can be used to generate steam for heating or power turbines and generate all of the electrical power for the ship, but the power system must be designed to allow operation of steam turbine generator and diesel generators. This is do-able but adds a level of complexity to the power system.

This is the extreme, not directly related to HVAC and ships have generators so recovered heat can be used, but it illustrates the need for energy recovery to be a component of the facility design and not an add-on. A holistic design approach.

  • $\begingroup$ Motor exhaust is hotter than from a boiler, large boiler exhaust will be around 140°C. This solution cannot be transplanted to a domestic heating system. $\endgroup$
    – mart
    Commented Oct 6, 2020 at 13:01
  • $\begingroup$ Oh definitely not, temp is 300C, but air exchangers in modern homes is viable. $\endgroup$ Commented Oct 6, 2020 at 13:21

Yes, this is possible.

Where I work, there are several very large fridges and freezers. The energy removed keeping them cool is pushed into a water tank that is the pre-heated water feed to the boiler, so the harder the freezers have to work means the water temperature gets raised before the boiler (fueled by gas) gets to raise the water temperature to the set value.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.