Auto wye/jerry switch

[hmccoy99]

Another approach to the wind loading problem..  

Using the wye-jerry switch posted by Flux I added a current sense circuit to auto switch..

the current sense circuit....

the circuit is voltage regulated can be used with 12v or greater.

select R1 for approx .5v drop at the desired switching current

adj R2 and R3 to toggle at the desired switching current

adj R4 for the hysteresis band.

Any constructive reviews appreciated

Henry

[Ungrounded Lightning Rod]

I'd consider using a VOLTAGE sense circuit.  That way you don't have to put a sensing resistor in series and drop some power into it.

With the right filtering and hysteresis you can avoid excessive switching.

[hmccoy99]

Hi

Voltage sense will not work as well, because the load will clamp the load in wye and all the reactive power heat up/burns up your windings.

with current sense there a crossover point will all systems when delta/gerry will start

to make more current than wye that is where you set the switch point.  this circuit

uses a sensitive voltage comparator to determine that switch point. The power wasted in

the sense resistor could be selected to be no more that a couple watts. depending on the system.

Henry

[hmccoy99]

[Voltage sense will not work as well, because the load will clamp the load in wye and all the reactive power heat up/burns up your windings.]

sorry I meant to say:

Voltage sense will not work as well, because the load will clamp the voltage in wye and all the reactive power heat up/burns up your windings.

henry

[hmccoy99]

That is assuming you are using a battery for a load. a big voltage clamp. but with resistive or with a MPPT load the voltage sense would work OK. with the voltage 1.73 times the voltage in delta. you could set the switch point at that point.

Henry

[Ungrounded Lightning Rod]

Right.  (Sorry.  My mind was back on a previous posting where it was an MPPT control rather than a Jerry-Y switch that was in question.)

For a Jerry-Y I'd consider reading frequency rather than current.  Again you don't need to put a resistor in the circuit to sample the current and waste power.  (Do need to filter it, though.  But that's just a big resistor and a big cap.)

By the way:  If you want to sample current you could light-gauge sensing wires a couple feet apart on a heavy gauge output wire.  (Since you'll have the wire's resistance there anyhow, why not use it rather than adding some more?)  Just be sure both sensing taps are on the same length of wire and don't span a joint or a section that flexes.

[hmccoy99]

(For a Jerry-Y I'd consider reading frequency rather than current.  Again you don't need to put a resistor in the circuit to sample the current and waste power.  (Do need to filter it, though.  But that's just a big resistor and a big cap.)

Freq will work but not as but not as well because your have to translate the power to rpm load curve it will work but you have to know the power curve in wye and delta/jerry

to determine the switch point. not as elegant.

henry

[Flux]

From my experience frequency ( speed) is the easiest to deal with. I did get one of the others to work on my first attempt, might still have the circuit somewhere but it was not as good as frequency.

You just need to make the changeover speed adjustable. It will spend a lot of time in the wrong mode whatever you do unless you have better wind conditions than I had.

Basically you want it in star on low wind days and Jerry on high wind days but you will have to accept that wind doesn't come like that. Mine was series parallel and the change was a bit more drastic, you may not suffer so much when it is in the wrong mode.

Flux

[tecker]

The  star - Jerry that you have there is a good stress reliver and an overall good idea for a good current jump to second gear  .Just an observation on your  circuit : The relay is not the best device for this circuit probably a triac would be better as the circuit is going on and off rapidly and the points or the acuator coil is in a pickle.

   One of things your missin ( maybe not also you have an efective  voltage sense switch) here is that the voltage from jerry phase can be annded to get your low rpm cut and not stall as bad as star .  

The series connection can relieve high rpm stress by gateing the bottle neck out of the stator.

 

[hmccoy99]

tecker,

there is a hysterysis adjustment to create a dead band so the relay wouldn't be switching in and out in the circuit.

I agree with you a solid state switching device like a triac would be a good choice

with a Pulse Modulator to control the triac to smooth out the transistion from series to Parallel so as not to induce spin up stall oscillation.

henry

[hmccoy99]

you said:

[One of things your missin ( maybe not also you have an efective  voltage sense switch) here is that the voltage from jerry phase can be annded to get your low rpm cut and not stall as bad as star .  

The series connection can relieve high rpm stress by gateing the bottle neck out of the stator.]

Tecker

I am not following you here the series connection will not relieve stress at high rpm.

it will burn out your stator. because of the mismatch of stator to load voltage assuming you are using a battery.

Henry

[hmccoy99]

Flux,

That is interesting it seems that a direct reading sense would be more precise the power load current curve where the parallel/series intersect would be the best switch point.  below that point series will make more power above that point parallel would make more power.

henry

[tecker]

I'll draw it and hook up ex bike for a demo .

[hmccoy99]

Flux,

the freq approach the rpm would give you a problem it seems to me. that the unit would

start in low winds in parallel when you would want it in series speed up to some rpm

then switch to series?  that seem backwards.

if you start in series then switch at some predetermined rpm to parallel the turbine would spin up and then when the winds slow down to your set point then switch to series

well below the opptimal point.  

I don't see how it would work correctly?

[Flux]

Take your star /jerry case.

It would start in star , the condition you need for low winds. At a certain speed it will change to jerry, the connection you need for high winds. Ideally you change at the point where star is becoming inefficient and you would do better with Jerry. Probably at this point your prop will be on the slow side as you are getting bogged down in star. When you change to jerry the prop will get away so it is a stable change and the increased speed will hold it.

When the wind drops you need to change back when you are just above cut in speed for jerry and when it changes to star you will see a speed drop and again it will be stable. There is no need for any hysteresis to prevent it constantly changing mode..

The thing that will most likely determine the switching speed is the change back to star. You don't want it too late or you will spend time idling in Jerry when it ought to have been in star.

On low wind days it will probably stay in star. On high wind days it will most likely spend the majority of the time in jerry but I have never seen wind conditions where it doesn't change back occasionally. Perhaps there sites where this is true.

On many days it will be doing a lot of changing. This will be true however you do the change over. If you try to devise some clever way to get it truly at the optimum point it will spend its life changing most of the time and unless you can devise an electronic switching scheme it will hammer a relay to bits in very short time.

It is crude but does significantly improve things over direct connection. There are infinitely better ways to do the job so I didn't use it except for one machine and that was a little 5ft thing and also a 3ft machine that I did for a boat that needed simplicity for the owner, he could wreck most things and an electronic solution would have been difficult as it got disconnected and all manner of things.

Give it a try, try speed and try your idea and come back and tell us how you got on. you will never sort it pondering theory here, there is always a hidden snag in practice with anything associated with wind power.

Flux

[tecker]

Well it looks like today will not be and off day so I'm out the door till later .I 'll post the rest and mod the bike for a jerry run . Here's the star set I use most of the time it allows me to wind up to a 14 + steady  charge with a Delco into my 1500 ah bank . Standard star won't come close to unless I stand on it . Give a look and and I'll post the rest tonight.

[hmccoy99]

I would build it but I live on a boat with limited equipment, parts, tools, etc..  I was hoping someone one this board would prototype it..

henry

[hmccoy99]

Hi Tecker,

That a interesting variation of star with filters. do you get a net power gain? (more

amps with the as peddle power input)

henry

[hmccoy99]

opps

i meant to say:

That a interesting variation of star with filters. do you get a net power gain? (more

amps with the same peddle power input)

henry

[tecker]

My main goal is a workout and to augment the solar kind of a kick start in the morning and a gravity boost after the usage at night (also a guard against the cold of night ).

 The open circuit voltage off the delco goes from 25 volts to 31 volts  and the two caps hold that voltage well with that configuration . I wanted to smooth out the surge of peddling and this also works the rise in cut in comes after 10 or so cranks and i can hold this with steady action . it starts to float at 13 + with to or 3 cranks with is a kinda float warm up . I am using a pwm and a 40 ah sla for the rotor and am able to kick out a nice pulse to the over the charging which helps also . this EX bike is not quite ready for prime time but well on its way . I came home with a n other alternator and will rig up the star jerry  from the star node.

[hmccoy99]

Hi Tecker,

Would you be interested in building the auto switch over circuit. I have a triac switch

design that will work with the current sense circuit.

I am not set up to do any building live on a boat... but I am a good designer.. I would

like to work with someone to prototype a good flexible design..

henry

[tecker]

No prob Here's a link to Lt spice . Just do a screen capture ,  Copy that file to Paint and open with imaging or like to save as jpeg .Or simply print the ltspice file to PDF.

 http://www.linear.com/designtools/software/

 

[hmccoy99]

Hi tecker,

here is the jpg of the triac control circuit.

henry

[hmccoy99]

Hi tecker,

here is the jpg of the triac control circuit.

and the sense circuit

henry

henry

[hmccoy99]

Hi Treker,

I posted the diagram in jpg format 1200x800 just save the pic then open it in a

picture viewer.  if you have any questions feel free email or call

email: hmccoy99@hotmail.com

phone: 321 960 8628

henry