Variable capacitor

[bandmiller2]

Is there any such critter.?Still thinking about induction generators.I'am a pilgrim when it comes to electrial circuts, could larger caps be used with a reostat in series to adjust voltage or would it take multiple sets of caps and a means to switch them.??Thanks Frank C.

[mobile_bob]

yes there are variable capacitors, however they are typically rated in picofarads
(billionths of a farad) and used in tuning circuits of things like radio's

motor start capacitors like the ones used to provide excitation of an induction generator are rated in microfarads (millionths of a farad).

i don't think you will find a microfarad rated variable capacitor.

you could do a switched network of fixed capacitors which would give you stepped
capacitance, or

you could add more capacitance than what you need, and taper the voltage back with a variable inductor, more inductance added to the circuit would lower the voltage.

variable inductors large enough to do the job are available, or once were. they were used in some large welders.

one could in theory use one with a bias lever and counterweights, weighted to provide the correct voltage, it would require another coil to provide counter control
in that as the voltage begins to rise the lever tips causing an increase in inductance, and conversely as the voltage dips the lever tips the opposing way and reduced inductance which increases  voltage.

too complex? probably unless you can find an old variable inductor from a welder than used one, and could adapt it to do what you want.

bob g

[BruceM]

4-6 opto-isolated solid state relays, coupled with motor run caps, controlled with a Picaxe or Arduinio controller chip would do the trick. (Preferably the switch on zero crossing SS relay types.)  You could do the RMS voltage calculation via software or use an RMS chip (about $10 more).

Not a project for the electrically challenged, though.

[mike90045]

4-6 opto-isolated solid state relays, coupled with motor run caps, controlled with a Picaxe or Arduinio controller chip would do the trick. (Preferably the switch on zero crossing SS relay types.)  You could do the RMS voltage calculation via software or use an RMS chip (about $10 more).

Not a project for the electrically challenged, though. 

Will zero crossing SS relays work with large inductive/capacitive circuits - it all depends if it senses voltage or current, 'cause they all get all out of phase.

[LowGear]

Posted by: mike90045

Insert Quote
Quote from: BruceM on Today at 05:00:23 pm
4-6 opto-isolated solid state relays, coupled with motor run caps, controlled with a Picaxe or Arduinio controller chip would do the trick. (Preferably the switch on zero crossing SS relay types.)  You could do the RMS voltage calculation via software or use an RMS chip (about $10 more).

Not a project for the electrically challenged, though.

Will zero crossing SS relays work with large inductive/capacitive circuits - it all depends if it senses voltage or current, 'cause they all get all out of phase.

That certainly makes the choice a lot easier for me to make - How about you, bandmiller2? 

Me, I'm still looking for the unit that's working successfully so I can clone it.

Casey

[BruceM]

Good question Mike, I'd have to study the specs for a specific model of SS relay to see how they do the zero cross detection.  Switching caps in and out on an inductive circuit does cause some serious spikes, so you'd want to pick units that are rated for higher voltages, and then add some spike suppression on the output.  It's not a trivial project. (Nothing is, it seems.)

[bandmiller2]

Yea Lowgear I think I'll just settle for close enough on the voltage. Thanks for the answers guys. Frank C.

[mobile_bob]

from a practical standpoint this is how some folks do it

they arrive at the capacitance needed for the load being supplied, via trial and error
and a little luck...

then they attach the capacitors to the load, and place a contactor/relay between that load and the generator, and use a switch to control the contactor/relay coil.

contactors are better than relays because of the contacts being dual break, so they are better at handling inductive loads.

this way whenever a specific load is brought on line, it comes with the needed capacitance to stabilize the voltage with that load online.

it ain't perfect, because you can have various loads, coming on at any number of combinations that might be additive in capacitance during starts or runs, and have the voltage rise higher than one would like.  for simple systems it works, and for more complex systems it can work if you control what loads come on and keep them separated if possible, or at least determine which combinations are dangerous and not go there.

to do it right, one would likely want to go with Bruce's method, it is dynamic instead of static, so the controller can be programmed to handle whatever loads or whatever combinations of loads and apply the needed capacitance to keep the voltage within
acceptable limits.  the controller could sample the line voltage and add or remove capacitance as needed, and do it in fairly fine resolution via the use of a variety of capacitor values. for example a 5mfd, 7mfd, 10mfd, 15mfd, and a 20mfd could provide everything from 5 to 57mfd of capacitance, in a variety of combinations.

5, 7, 10, 12, 15, 17, 20, 22, ...  you get the idea. 

five contactors to switch the capacitors in various combinations controlled by the programming of the microcontroller and you have a very neat solution.

the contactors are relatively cheap, very robust, easily repaired of rebuilt (most of them anyway) and available all day long on ebay.

there is the electronic solution via hardware, now all you need is some empirical data and know how to program the micro to do the switching.

bob g

[bandmiller2]

What I'am thinking of is a standby situation in a power outage.A large three phase motor would be driven with a tractor PTO so power is in good supply.I could put the whole load on and adjust the caps to it,then when needed just run the whole load its adjusted to.Should something else come up could drop some loads to add new,keeping the system more or less balanced.Sound like a fair plan.?? Frank C.

[mobile_bob]

i don't think your plan will work, for the following reasons

1. the generator  is likely to drop out and stop generating if there is insufficient
capacitance when a large load comes online... or

2. you run the very real risk of serious voltage drops when the load comes on, until you can switch in the needed extra capacitance... this can cause damage to certain loads that might not start properly... or

3. when a large inductive load drops offline the voltage will spike, which in turn runs the risk of damaging other loads connected, until you have time to switch out the needed capacitance.

i don't think you can react anywhere quick enough, when things go wrong they do so in a hurry, which usually freaks out the operator and slows his response further,, or worse he makes the wrong decision under stress and adds capacitance when he should be removing it... which makes matters worse in a hurry and adds more stress to the operator, let alone the connected loads.

personally and in my opinion only, i think you would be better served in the long run (if not the short run) to just buck up and find a genhead big enough to do the job to start with... time you buy all the caps and work out all the related control circuitry, you likely will be within a stones throw of buying a genhead that you can simply spin and use as is.

on the other hand if you have almost all resistive loads, then go ahead and try the induction generator, also if you have only a couple of inductive loads that you can manage or schedule their operation then it might be worth a try. If however you have many varied loads that come  on when they like, and think it likely that over time you will be adding or removing some additional loads, then maybe the induction generator is not for you?

bob g

[Tom]

Just buy an ST gen head or STC head if you want 3 phase and you're good to go.

[dieselgman]

2nd vote for that solution... short and sweet!

dieselgman

[millman56]

My  adventure with a 4Kw induction motor as a generator cost me £400.00 for a new PCB on a packing machine I was testing, more than enough to buy a 4Kw alternator.

[bandmiller2]

Advice taken,I'll stick with the Lister and ST-10 I have already.If I need more power I'll wait til it comes back on.Read Bill Rogers book this afternoon,  mostly theory,his induction gen.was hooked to the grid for regulation.My time would be better spent geting my haying equip in order. Frank C.

[millman56]

A year ago I did the repair and reinstalation on a 100 plus year old water turbine, this is now driving an 18 Kw 3Ph induction motor hooked up to the grid and producing 15 Kw continuously, this works perfectly.                   As Mobile Bob explained very well, capacitor excitation is fraught with problems on mixed loads.

Mark.