4 ms·
This impeller design is clever, but it seems that a DC generator and a resistive load would be an easier way to brake the windmill and generate heat. The first
by jdblair 8y ago
This impeller design is clever, but it seems that a DC generator and a resistive load would be an easier way to brake the windmill and generate heat. The first advantage is off-the-shelf parts. Electric generators and resistive loading are well known technologies. You can locate the resistive load (literally, an electric heating element) close to where you need the heat and there will be very little mechanical wear.
- bluGill 8y agoYou are missing something important: they are storing enough heat to last for a couple weeks of no wind (or a cold snap). Enough DC batteries to heat a house that long are expensive and dangerous (fire hazard). A large water tank (10 tons of water or more!) is a lot safer to have around the house. The water tank also means the ability to locate your load anywhere isn't important: either way you have to design your system around the tank. Also, DC is less efficient as the losses at your generator is heat that isn't captured. Generators are typically 80-90% efficient. This is pedantic though and doesn't matter (if this was the only concern DC is wroth the losses because you can divert it to lights or something else that is also useful)
- jdblair 8y agoI didn't spell everything out. I never suggested you store the electricity in batteries. The electric heater would be used to heat some mass, like a tank of water or bricks. I look at this impeller and I see a custom mechanical design, with seals that are prone to failure. You replace that with an off-the-shelf DC motor and a resister bank in a tank of water, and you can put the tank of water in your basement or wherever you already have the heater for your hot-water heating system.
- apendleton 8y agoThey also discuss an "eddy current retarder" design, which uses a magnetic mechanism that sounds vaguely like an electric generator but generates heat directly, as an alternative to the water break/impeller design. Sounds like maybe you get the best of both worlds, in terms of a smaller package but avoiding the motion -> electricity -> heat conversion losses due to the extra step.
- xyzzyz 8y ago> You are missing something important: they are storing enough heat to last for a couple weeks of no wind (or a cold snap). Enough DC batteries to heat a house that long are expensive and dangerous (fire hazard). A large water tank (10 tons of water or more!) is a lot safer to have around the house. Why would you use batteries? That’s silly, you’d simply heat water in both scenarios.
- teambob 8y ago10 tons of boiling hot water? What could possible go wrong? https://en.wikipedia.org/wiki/Great_Molasses_Flood https://en.wikipedia.org/wiki/Great_Molasses_Flood
- severak_cz 8y agothis is not boiling water AFAIK, it's more like warm water
- fourthark 8y agoIt's ten cubic meters, not as crazy as it sounds. Yes it would flood your basement if it burst, and you probably wouldn't want to be directly under it, but not going to drown you. Water is heavy!
- mNovak 8y agoI think heat storage for domestic use is a red herring--the point isn't to replace all forms of gas/electric heating, but rather to reduce the usage. So, all the turbine heat is essentially consumed immediately, and the conventional methods kick in to fill the gaps. The easiest way to accomplish this is feed your lukewarm turbine water into the water heater intake. Every joule added to the input is a joule the gas heater doesn't have to deliver.