Grid Tie and Induction Motors

Yes, I know it's nearly 3am on the east coast... but and idea struck me. So the idea is to feed small amounts of mechanical power into the power grid. Not necessarily run the meter backwards, but supplement power consumption.

I've researched grid tie inverters - which are very expensive. For those wondering, a grid tie inverter is feeds mains power back into the grid by syncing phase angle and phase (no dead shorts :) ) and applying slightly higher voltage. They are very efficient and really not within a college student experiment budget :p

So I was thinking... Rather than go from mechanical to DC to AC to grid - go from mechanical to AC to grid VIA an induction motor. As a proof of concept, use a DC motor + battery to turn an induction motor. Plugged into the grid, in theory, should apply current.

Oh, but the phase you say? How do you prevent a dead short?"
I've thought of this -- before applying mechanical power - have the grid bring the induction motor up to speed. Then try to turn faster (apply a torque) with the DC motor, for example. In theory, the amount of extra power put into the grid will be related to the slip angle of the motor - which will also control the speed of the input (so you can't go over speed by too much).

Keep in mind that this whole battery business is just a proof of concept sort of thing - I'm not talking perpetual motion or any hohaa craziness. In the end, the final mechanical input will be around 200 watts. I expect this to be very low efficiency (likely 50%ish), 100W isn't an answer to the energy issues - but it's an experiment. It's also not going to come even close to driving the meter backwards, but it should run (as supplement) my laptop + two to three 13w CFL's :D


I think the theory is feasible -- the inspiration comes from flywheel driven UPS systems. An induction motor is driven while mains power is on to keep a flywheel in motion. When the power goes out, the FW drives the motor and feeds to local grid.

I'm thinking of using a "low" rpm induction motor.... If I recall, ceiling fans are 16 pole? So that's 60Hz2*2/16=450rpm... Add ceiling fan motor to the list of things to hunt for :) Looking at the one above my head, it looks like it even has a nice bolt pattern for some sort of pulley shenanigans :D


Can someone either throw some ice water on me and slap me for being an idiot -- or let me know if I've found a boat to Valhalla.

Oh, and my apologies for dancing around the "mechanical input" details.... There's a reason for this, I promise :) In any case, insight and information is appreciated :)

sort by: active | newest | oldest
R.Blakely5 years ago
Feeding electricity back into the AC powerling is possible by driving an AC motor with a DC motor. But the speed of a DC motor is proportional to voltage that is applied to the DC motor. To work as a grid tie inverter, the DC motor would need to have a speed the same as the AC motor. This means that the DC motor would need a variable speed transmission between the DC motor and the AC motor, so that the DC motor would apply torque to the AC motor.
Another solution to the problem is to use a wound-field DC motor. This type of DC motor can vary its speed depending upon current flow in the field winding.
However, a grid tie inverter can be built more easily than by using a DC motor driving an AC motor. An amplifier can generate AC voltage if AC is fed into its input. Simply generating a slightly higher voltage than the AC line voltage would allow energy to be fed into the AC grid.
I'm very interested in this topic. I really want to do a DIY electrical generating project of some kind, but I think it's critical to be able to dump the energy into the grid. Batteries are not a good solution for people like me who live in urban areas. Like some of the comments have said, it would be nice to make small projects of 50-100 watts and be able to input that energy into the grid. My house uses around 440 W average power, so even 5% of that from a renewable energy green project would be worthwhile.

At the Maker Faire last weekend there were two presenters dealing with this issue. Here are links to them:
http://www.3ub.com/
This guy is trying to patent technology that would allow anyone, with permission from their utility, to put power into the grid.

www.kwiksolar.com/
This company offers a plug and play 200 W solar panel for around $1000. I suppose they probably use the standard inverter technology. It would take a long time for that unit to pay for itself.

pnunes689 years ago
This idea is similar to something I was thinking about for a while. I wondered if it would be feasible to use one those datacenter quality UPS units to feed power back into the grid. This idea assumes your utility allows for both "Time-of-Use" metering (peak/off-peak rates) and "Net Metering" (utility pays for excess at same pay it charges). The idea is to to allow the UPS to charge a battery bank during off-peak hours. Then, when the peak rates begin, switch the UPS to dump its power back to the grid at the higher rates. Depending on the capacity of the batteries, you may be able to have a net income by selling the utility's power back to them. At the very least, you may be able to significantly reduce the bill. The questions in my mind is the following: 1. Efficiency - there will be power losses associated with the charging, inverter, battery storage. 2. Synchronization - inverter needs to be in-phase with utility. Most UPS units automatically synch to a signal, but may need a special circuit. 3. Policy - my utility allows Net metering for alternative energy sources like solar, but they may not like buying back their own power. 4. Equipment - although high quality UPS units are available cheap on places like Ebay, the batteries may need frequent replacing due to the daily deep cycling. Maybe marine type batteries can handle the stress without breaking the budget. 5. Switching - special circuit needed to allow UPS to safely feed power back to the grid at the right time. Any comments?
> 1. Efficiency - there will be power losses associated with the charging, inverter, battery storage.
. Yes. A lot. Probably more than enough to make the project a bust, unless the price differential is pretty steep.
. The first hit on "inverter efficiency" (http://www.solar-electric.com/solar_inverters/inverters_for_solar_electric.htm) says "as they have the best efficiency with batteries - you will get about a 5-10% loss. With some older inverters, such as the Xantrex SW series, that can sell back excess power to the grid overall losses can be as high as 50%." So there's a minimum of 5% (bet it ain't cheap) just in the DC-AC inverter.
I agree that efficiency would probably be the deal breaker. In my case the spread between peak and off-peak is about 18 cents/kwh. NY metro area ConEd gets 9 cents/off peak and 27 cents/peak. Even if the efficiency was high enough to cover the spread, I suspect battery replacement would wipe out what was left. I wouldn't mind setting up a test rig to get the numbers. However, I don't have a design with which I am confident won't fail.
LinuxH4x0r9 years ago
Or, because it is less than you total power consumption, you could just use it to power a dc generator and use that to power a small inverter (400W ~$15) or directly power things (12v car phone charger, laptop, charge batteries, etc). Has anyone ever thaut of powering a refrigerator compressor directly with a wing turbine and making a wind powered AC system? Good luck, and please reply.
trebuchet03 (author)  LinuxH4x0r9 years ago
you could just use it to power a dc generator and use that to power a small inverter (400W ~$15) or directly power things (12v car phone charger, laptop, charge batteries, etc).

That's totally possible, but that meets the demand of another market ;) I'm looking for seamless integration. A second line sort of system means your devices stop working once you stop generating power or when your battery dies. The psychological benefit for that is you see the fruits of your effort immediatly - versus when you put a small amount of power into the grid, everything is literally hidden in the wall :P
I meant to use them side by side. When the sun sets or the wind stops unplug everything and plug it in to the coal burning, pollution causing goodness we call the wall. Any thought about the AC idea- I'll post a new forum about it.
NachoMahma9 years ago
. I don't think you can just hook up and start feeding power into the grid without permission from the utility. . I'm no EE, but your basic scheme sounds good. I imagine the util will want pretty tight control as far as staying in synch. . I doubt it will be very practical, but, if everybody had a small generator, even a few Watts each would help. It could be driven by just about anything - wind turbines, PVs, exercise bikes, hyperactive kids - any source of free energy. Use batteries to collect energy at night (guess that leaves out PVs) and feed it into the grid during high-demand periods. . I doubt if the util would be able to pay much for anything you produce. The bookkeeping to keep track of a few WH here and there would be a big burden. . Very interesting idea. I hope someone that knows what they're talking about chimes in and tells you how easy (and cheap) it is to do.
trebuchet03 (author)  NachoMahma9 years ago
Hyperactive kids :p Yes :D

I saw your first post - about checking with the utility company... And yes, that does have to be done..... I might not necessarily tell them the mode of generation at first, but methinks if I use a UL listed motor - it should meet some sort of compliance :D

As for book keeping, if the power company lets you feed back into the grid - they may be obligated to buy the power back from you at the rate they sell it for (up to your consumption). If you have a rotary type meter (not a digital one - the type with a spinning disk), typically you can feedback without any special equipment. Otherwise, the power company has to install a second meter... Depending on local laws, they may be required to do so for free.

But that assumes your power actually leaves your home - in my case, that shouldn't happen. While making power, it should go straight to whatever devices are currently plugged in (and the meter will just turn slower as less power is needed from the grid).

I hope someone that knows what they're talking about chimes in and tells you how easy (and cheap) it is to do.

Me too :) I'm hoping the answer is.. "Yes, that will work." and "Yes, it will self regulate."