The mechanical relay could easily be replaced by a solidstate one, in fact for AC voltage control they are much more available and cheaper, compared to DC. check this part# from McMaster Carr, 7456K14 , its on page 816 in the 109 catalog. whatever circuit you had in mind could be prototyped with a mechanical relay, or three relay's in progresive stages. then once you got that how you like it just switch over to solid state. it will use 3-30 vac or dc to actuate. The load side will handle 25amps@300vac. I'm not quite sure how you going to determine the "neutral"(white). but you should switch "hot" lead(black). I guess with a WYE connection instead of DELTA. Also it would be neat to know if you can parallel the three(hook together) and get your "hot" lead and tap the center of the "wye" connection to get "neutral". but mostly I am curious about the hot leads, all three phases should be "hot" most likely (watch for sparks) its okay to connect them.
Another point thats interesting is rectifing some dc for the control circuitry. I would suggest some form of regulator circuit, since depending on wind speed voltage is going to vary. and your LM3915 is not going to like this. But on the otherhand you may be able to ditch the regulator, and the LM3915, by using zenier diodes hooked in to the dc rectified part, and the voltage surges, from more wind will be enough to actuate the solidstate relay and the zeiner together. pop a led in each relay circuit and you have some indicator of whats happening.
when I say "switch" the hot lead I am refering to the switching leads of the relay or load circuit.
just tossing some ideas around
-JW[ Parent ]
The zeiner diodes are definitly worth a try for what you are doing. I think Xeroid brought these up on the board before. I believe he was using one to transfer a dump load from the charging circuit, once the battery was full. I think he was working with a pure DC circuit.
If you are trying to make DC power, I think its best to use a Delta wired stator and the six rectifier diode arrangement. But in this case you are intending to make and use AC power. This is why I brought up a WYE connection for the stator. this will accomodate a neutral leg. But its probably not as efficient as a Delta connection. And the delta connection will not give you a neutal. I am pretty sure STAR and WYE are the same type of connection, but not absolutly sure.
Now, with a WYE connected stator you should have a total of 4 leads, from the three phase arrangement of the coil connection. three of these leads should be hot and the center tap should be neutral. At this point you are going to have to do some bench testing with your windmill generator. Isolate the four leads comming from the connection. note the three that are suposed to be hot, and the one, thats suppose to be neutral. spin up the generator to a speed you feel comfortable with. then try to short any two, of the three hot leads, and dont mess with the neutral. check to see if there is a negative effect when shorting the hot leads together. If there is, or the geny loads up or slows down this may not be that good of an idea. However... if you short any or all the hot leads to the neutral the geny should slow down from load. this is triky business, observe all safety practice while trying this. In the event that shorting the hot leads together does not cause any problems, then this is very do-able.
On to the hybred AC/DC circuit. assuming you can hook the hot leads together plus the neutral, you should only have two leads. this is your main AC output circuit. since you are using this output for simple heating and water could be involed, WIRE IN A GFCI, or ground fault interupter. then take the neutral and connect it straight to all your loads. all you have to do from this point is hook up the hot lead and complete the circuits to your loads. use your relays (single position single throw should be all you need) to make the hot connections to your loads.
The DC part. before your GFCI you have your AC output, connect a simple run of the mill bridge rectifier here. now you have some DC to power your ziener diodes and relay coil circuit. the zieners are wired in series with the relay coil, the ziener is placed in the circuit backwards so it will not conduct, until its activation voltage is reached, then it will conduct in both directions. and thats pretty much it. I think some serious playing will be in order by the time you reach this stage. remember ziener doides have all kinds of activation voltages many to choose from. current should not be a major selection factor since you are only activating a small load "ie" relay coil, or solidstate relay. the solid state relay may be obtained for less then ten dollars from http://allelectronics.com . but Im pretty sure your ziener rating will be very different from mechanical to solidstate relay coil activation.
the GFCI is where you will get your main ground that can be connected to the water tank. I would love to make a schmatic for this, but the computer I do that with is having some problems. but more importantly is the question if you can connect the three "hots" from the WYE connection.
-JW [ Parent ]
I was hopeing you would specify your coil wiring arrangment. Sounds fine to me the way it is, seems you already have a two wire output. This is very good news, you should not have to allocate any coils for DC voltage production. Use them all to generate AC.
The start up and run features that you are aiming for will not be a problem. Upon further thought, you will be able to run AC to the house, and do all the DC buisness at the point/place of use or load.
My suggestion is this, at your tower install the GFCI. Take your two wire AC output run it to a small box at the bottom of the tower, get a GFCI type outlet mount it in there and hook the output from the genny into the GFCI outlet, then ground the GFCI to the tower. now you have a grounded outlet at the bottom of your tower. use a grounded (three prong) extention cord to get the AC to the load, house, whatever. I would set up a reciever box with a pig tail, cut off the male end of an extention cord to make a pigtail, make sure it has a ground pin. take the three wires and run them into a box of some sort to keep things neat, use a terminal block in there to recieve the three wires.
Now, wire out to your loads and include your relays in series. its best to worry about these circuts one at a time.(meaning output or control) I would start with your ground, first hook that to your water tank. then feed all the relays (solidstate or mechanical) with what you designate as your hot. the neutral can go strait to your heaters. by the way as you are hooking up your GFCI "it" will determine for you, which hot and neutral, depending on the direction your mill spins. in other words when you hook up the GFCI it will "pop" automatically when the mill spins if you need to reverse the wires(of the two AC output) but it may not pop either way, in any case the outlet will have a wider and slimer pin or spade, the wider one is hot. now lets say you gought this far, and your mill is connected to everything. the loads should be all wired up at this point. the whole AC part should be done, hooked to the relays and loads. lets say you used three solid state relays. and each of these completes the circuit to each of the three heaters. since control is not energized the relays should be open and no load on the mill, no power to the heaters.
take a 9 volt battery and energize the "control" side of the solid state relay. It should turn on, and energize the heater. I was thinking about an indicator for this. you could use a small neon "bulb" conected in parallel with the heater its self. these neon bulbs are commonly used in aqaurium heaters, its the strange type of light bulb that has two strait bars in side, usually they turn orange or yellow when they light up. cheap or old style AC contiuity testers use this form of illumination also. so if the heater is seeing power the neon bulb will light. you can get these neon bulbs at radio shack for like a buck. neon bulbs use raw AC power to work. then check the other relays with the 9v battery and make sure the AC circuit works as it should. at this point you should have manual control.
DC, ok go back to the reciever box at the house, wire in a bridge rectifer. now you should have the rectifer block wired in parallel with the AC, and DC comming out of the rectifier. At this point your DC output should be a bit out of control. and will vary greatly from 5vdc to 80vdc, depending on wind speed. use a resistor to cut down the voltage, again some playing is in order here. you will probably be best off using 1 watt resistors for this, not the very small 1/4 watt type you see in circuit boards. but then agian you should be able to get away with alot, since your solidstate relay control takes practically no current. but it is the impeadance or ohm's that will drop the voltage. wattage is its current rating, ohms is what actually drops voltage. at maximium mill rpm you must not make more than 25 or 30vdc, this includes furling. if you make more than 32vdc you can damage your solidstate relay control circuit. remember you can still activate the solid state relay at 3.5v .
Zieners, you should have one main resistor to drop the dc comming out of the rectifier, this will put you in a range of maximium potental for your control circuit. then you will need an additional resistor for each relay with a different ohm value. its going to take some time to work out all these values, but you will be able to use the same value of the zeiner for all relays. Id pick a ziener with a 5v break down voltage, and use it for all three relays. then switch resistor values on each relay to get the different heater activation rpm's. wiring is very easy at this point, negative on the rectifier can go to all the relays directly. positive gets a main resistor(you may not need this) and then goes into three branches one for each relay, use the resistor and the zeiner in series with the final connection for each relay. remenber if the ziener is in place correctly, it keeps the circuit open until it goes over 5v(in my example)the the diode conducts in both directions.
-JW
[ Parent ]
Im gathering your worried about the output freq of your genny, with respect to the load control side of the solidstate relay. the best thing to do is just try it out and see what happens. I think you should be ok. these AC controling relays use an SCR to switch the load. so if your output is even at 4hz it should switch on and off. If you try to switch DC with one of these you will find it will never disengauge. however for AC this is not a problem since it goes +/-. as long as its ac at all it will shut off like it should, regardless if its over 60hz or under it. Interesting you found a minimum spec for AC frequency requirement on the relay. This could be for the control side if it does not have DC capability like it should from 3 to 32vdc. -JW[ Parent ]
I want to thank you for your kind words and suggestions. I hope this project works out for you as best as possible. And I must tell you using hot water as energy storage is a really cool idea. The way I figure it, as long as your water tank is grounded to that GFI (or ground fault circiut interupter)the worst case scenario is reseting the GFI. other than that as far as designing the circuit goes, I feel the sky is the limit. I look forward to seeing a future post here about your progress, please include some pictures.
when you mentioned that about a freq reqiurement on the load side of a AC solidstate relay, its makes sence that AC power has freq. but 1hz should be sufficient. my understanding of these AC switching solidstate relays, is that they use an SCR, silicon controlled rectifier. And you cant use an SCR to switch DC because the gate never de-energises hence it wont un-latch. A MOSFET on the otherhand frys when you try to switch AC with it. MOSFET's are used for DC solidstate swithing relays, and SCR's are used for AC solidstate switching relays. I could be wrong/ perhaps for applications where conditioned AC loads are used a special type of AC solidstate relay is used that will automatically disconnect if the ac quality isnt up to snuf, the relay kicks out as a protective device, maintaining quality of power. Stay away from this type of device, your needs have variable freq output, besides were just talking about a simple resitive load here, not some hi-tech cpu-app. read up on SCR's and you'll see what I mean. Hell Its very possible you could use an SCR if you heat sink it correctly, but then you have to build control circuitry that will turn the scr on and off with DC, thats what makes a DC controlled AC load switching solidstate relay so nice, besides you cant use the zieners without DC control on the solid state relay..
I see your having fun keep at it, cant wait till you get to the water tank and radiator part.
