Voltage doubling circuit question...

[windstuffnow]

   I have an alternator that I've been thinking of using on a small Lenz turbine but the cut in speed is considerably higher than I'd like.  The alternator is complete so this is an attempt to use it and save some time.

   My thoughts were, and I have no idea if it would work.... using the voltage doubling circuit in paralell with the normal rectifiers.  The alternator is wired in star so I'd be tying in to 2 of the 3 wires for the low power circuit but still leave the main leads connected to the rectifiers.  In my mind, and drawings, it would pass through the low voltage side until it reached optimum speed for the normal side.  Granted it would only deliver 1/2 the amps on the low side but thats 1/2 more than it would normally do so it would be a bonus.

   Could something like that work?  I don't see anywhere where it would back feed into the system, it's almost like using a separate rectifier... I might be missing something...  Could be a quick fix for a couple alternators I have laying around.   I'll have to find some diodes that could stand a little power.

   Also, When the speed was up and the normal rectifiers were passing current would the small circuit need to be disconnected or would the current pass mostly through the main (least resistance ) rectifiers?

   Just a passing thought that sounded like it might work...  Comments please...

.

[hvirtane]

With a friend of mine

we have used on a single

phase alt a transformer

in parallel with the leads

to the rectifier, to get

the cut in speed lower.

That seemed to work just OK.

I see no reason your idea

wouldn't work. Maybe there

would be some vibrations

if you'll use only two

of the leads to go through

the voltage doubling circuit.

- Hannu

[satchel]

would it be possible to build a step up transformer and wire it in on the ''hot'' side? i dont know much about all this,learning as i go,but i talk to freinds that do ,and sometimes i should just not talk at all.

but its a thought

[hvirtane]

Also, When the speed was up and the normal rectifiers were passing current would the small circuit need to be disconnected or would the current pass mostly through the main (least resistance ) rectifiers?

With our transformer system in parallel

we found out that there was a limit of

the current passing through the transformer.

When the speed was up the most current

went straight passing by the transformer.

I think that you should just

test the circuit and see what will happen.

- Hannu

[BigBreaker]

Look up a "Buck Converter".  You definitely do NOT want a core transformer.  A Buck converter is like a simple switching transformer and can be far more efficient than a traditional cored transformer.

I haven't really looked at them but I think you basically charge caps in parallel and dump them in series on a duty cycle.  Come to think of it, I believe buck converters are DC only, so you'd need to rectify first (unfortunately).

[windstuffnow]

  I was hoping a simple add on circuit to increase the low end would work.   I don't like the idea of using a transformer, it would simply steal power from the alternator and put it in constant load.  With the doubling circuit, it seemed simple enough, it could be run in parallel without interfering with the normal operation and would simply load the alternator earlier through the doubler.  

  I guess I'll have to experiment with it and see what happens.   I need to find some diodes that will handle a few amps or I would have tried it already... I built a small one and spun it up by hand and it didn't seem to interfere with its operation.   No rpm data, just watching the meters.  We'll see what happens... Need a couple of electronics guru's to step in and tell me I'm wasting my time or it will work but... I don't quite understand the circuit, (how it works) just that it does which inspired the idea.

.

[monte350c]

Hi Ed,

Sounds like you could use a boost converter...

http://www.discovercircuits.com/P/pwr-dctodc1.htm

Discover circuits is a pretty good resource.

Ted.

[Bruce S]

Ed;

   I am guessing the doubler is the "normal" type and the rest is going to be the standard setup for rectifying.

Here's the scoop on what we/I would see happening and what you may want to work with .

The doubler will continue to work just as if the rectifier wasn't there. The circuit is going to see the path of least resistance, this you'll need lengthen the charge timing by using larger than normal caps on the doubler side. The consensus here is that twice the norm would be best. This way you'll get a very low resistance looking circuit during the low power point. Could have some reactance issues later with high speed, but that will probably be no big deal. The higher caps will help by lengthing the charge timing. There's a math formula for figuring out what the RC timing is and i can put it up here if you're really worried, but usually not a big deal until the frequencies hit above 350Hz.

The rectifiers will not be getting thier entire amount even at the higher speeds due to the push to the doubler and it's much lower resistance.

The rectifier will however, also be able to push probably 98% of their design though and the diodes will make sure theirs no feed back ( be sure and isolate the grounding from each other as well, otherwise there will be a feedback loop).

What outputs are you looking for?

Hope this helps

Bruce S

[johnlm]

Ed,

Without sitting down and doing some serious AC circuit analyis  I offer the following:  How much effect the doubler presents to the standard rectifier route kind of depends on the amount of current you get through each path.  I tend to agree that the doubler path will look like a higher impedance because looking into both circuits from the alternators point of view on the doubler side it sees the effective dynamic impedance of the rectifiers in series with the caps; and into the normal bridge side it just sees the diodes (assuming the outputs of each section are working into the same load - the battery).  It will unbalance your 3 phase output somewhat with the windings which have the doubler connected to them being more heavily loaded than the standalone winding. I have found this to not be too serious of an issue if within reason on a WYE connected alternator but would probably cause more problems on a Delta connection.  The impedance of the doubler side will decrease (cap impedance will decrease) as the frequency of the alternator increases - which is the opposite of what would be preferred, as it would be nice to have it drop out of the path and the bridge section take over as the frequency (and output voltage of the alternator) increased.  Probably the redeming factor in this is that my experiments with (thrown together) voltage doublers finds them to not be terribly effecient at transferring power through them; but then I was only using cheap electrolytic caps which likely had a reasonably high ESR and may not have been sized for optimum performance.  Getting a couple of amps through a voltage doubler, at the lower frequency range (20-60 HZ) I suspect you will be seeing, may be difficult without careful selection of the caps used.

Johnlm

[K3CZ]

If we are talking about a car/truck alternator with separate field control, I have a suggestion.  A number of years ago you could buy a kit that would apply full battery voltage to the alternator at a flip of a switch, while also disconnecting the vehicle battery.  Then, when the engine was cranked up to 2500 rpm or so, the alternator is spinning at app. 6000 rpm, and putting out 90vdc or so.  The rpm's are not exceeding its normal usage, but the DC voltage was good enuf to operate a universal motor of 1/2 hp or so (leaf blower) at near full power with no bad side effects, except fuel consumption on the prime mover (!).  It works; I used it many times for an hour or so at a stretch.  Flip the switch, and you are back on the road.

This also idicates the normal alternator, which will put out full charging volts at about 2500 rpm, can be speeded up a bit (50%?) to produce double its voltage output with no other modifications or accessories.  If you are trying to obtain this kind of output at wind speed, it obviously is impractical.  Good luck.

                                             K3CZ    Van,   PE

[windstuffnow]

  Here is a diagram of what I was thinking...

  After drawing it out, and I'm not sure its right, it looks like it will actually back feed into one or the other phase.  

  The pos(+) and neg (-) would be paralleled together feeding the battery.  

  Bruce, I'm not sure what you meant by separating the grounds... since both pos and neg would feed the same battery bank.  Also, I have no idea how to calculate the capacitor size.   I experimented with two 1 amp diodes and 2 1800mf 80V caps ( no idea what they came out of ).

  As usual, what I thought might be a simple fix to re-use an alternator might turn into more trouble than it's worth...

.

[Bruce S]

Here's a hacked up addon. Sorry, paint + lots of coffee= crappy redraw.

There's still the simple solution just a few extra parts needed. perhaps just one extra diode on the doubler's neg side, otherwise your going to feed the higher voltage right up the lower powered circuit's wire, which will/could back feed into the cap and ....not good.

The RC network can be calculated by the resistance of the wire being used in the circuit and the mf of the caps. the voltage you have listed should be twice of what ever the highest possible voltage output will be.

Assuming your wire's resistance is 1.3ohms and the 1800mf comes out to a frequency of about 70Hz (hand calcs) a 3600mf would bring the frequency down to about 40.

What does your wire read?

Hope this helps

Bruce S

[hvirtane]

I didn't mean that you should

use a transformer, but exactly

the circuit as you suggested.

I just wanted to tell that

at least with a transformer

it would work. I see no reason,

why it wouldn't work with that

circuit the same way.

Bruce is right by pointing out

that you should isolate

the grounds by diodes.

I have no idea about the values

of the capacitors needed.

- Hannu

[johnlm]

Ed,

Try this site for some info on calculating the voltage drop from a multiplier at given currents and cap values.  Also has a schematic for a 3phase voltage doubler which would be pretty easy to just connect in parallel with the standard 6 diode 3 phase bridge.  It would not unbalance your windings and the impedance of the doubler would just limit the current being supplied by it after the normal path started conducting.

http://home.earthlink.net/~jimlux/hv/cw1.htm

Johnlm

[Bruce S]

John;

   Gotta say that 3-phase is a beauty, much simpler that using the old single unit and trying to balance the outputs. The CW type is what we use to use for making soild-state ion gennerators. Still have one and it's been working since the day I built it.

Thank's for the site post.

Bruce S

 

[windstuffnow]

  Wow Johnlm... that looks promising.   It will take a few reads for it all to sink in.

  Thanks John and Bruce for all your help ! Thanks everyone for you helpful contribution, I'll see if I can actually make one work... It looks simple enough but my hands and fingers were designed to fit nuts n' bolts.   When it comes to electronics my co-ordination is like assembling parts with my feet.

.

[Ungrounded Lightning Rod]

Will not work.

The output voltage will be swinging around between the full genny voltage and twice the genny voltage

That doubler circuit is designed for a setup where the wire feeding the point where the caps join is the local neutral.

(There are other approaches that will do the job but I'm too tired right now to work them out.  You do get the current from all three phases out of them.)

[windstuffnow]

  Another question about the 3 phase circuit shown on the above website...  They state that it is a diagram of a 3 phas multiplier circuit.  This circuit shows more than one doubling stage correct?  So to maintain the doubling I would only need one stage right?   3 caps and 6 diodes?  Also they show the ground comming from the neutral wire of the star connection, can this be added externally - that is, withing the circuit itself since the neutral is burried in plastic.  I do have one with 4 leads comming out but the others are burried.

.

[finnsawyer]

Not true.  There is no reason that the source voltage can not float.  It's too bad he didn't label the connections.  But when the top diode conducts, the bottom diode is cut off, and the top of the top capacitor becomes positive.  At the next half cycle the bottom diode conducts with the top diode cut off making the top of the bottom capacitor positive.  The capacitor voltages add.  You add can the ground in the output as shown by Bruce.  This does unbalance the alternator, though.

[johnlm]

Ed,

I downloaded a copy of that schematic and will have to study it with some waveforms in the appropriate phase relationships to see whats really going on.  It appears as if there  is more than 1 multiplier stage the way it is laid out but remember in a single phase full wave doubler there are 2 diodes and 2 caps in a doubler and the number of diodes and caps (other than d7,d8,&d9 which I think are used as blocking diodes) is 6 diodes and 6 caps in this circuit - 2 ea per phase.  I will have to think about the ground need of your neutral point.  I suspect it is needed to properly reference the multiplier but then you have the issue again of 2 different grounds - the ground connected to the neutral of the WYE for the multiplier, and the ground created on the negative side of the 3 phase bridge of the mainline path.  I will have to do some checking to determine if its ok to tie these separate ground points (negative polaritys) together.  I think it should be ok as in the old mechanical voltage regulators (before they made solid state regulators) had this node running through a relay coil to ground.  One could test this pretty easily by playing with a 3 phase alternator and connecting the WYE neutral node to the - connection of the 3 phase bridge.  You may just have to unbury your neutral from the plastic - if I understood correctly that it was buried.

John

[windstuffnow]

  Hi John,

    After looking at it and comparing it to the full wave doubler it looks like it's only a single doubler... don't know for sure though.  I'll be digging around looking for some parts to put one together next week.  Most of the alternators that I have laying around don't have the neutral exposed.   I do have one with the neutral comming out of the plastic ( 4 wire ) that I will experiment with until I figure out something on the others.  

Thanks!  Have a wonderful Christmas  !

.

[boB]

(sorry about the thin thin line width in the schematic picture.  There are

two capacitors C1 and C2 that may not quite show up correctly)

The basic operation of the capacitive voltage doubler works by AC coupling the

input and then clamping the "middle" point to ground with D1 so that the peak

voltage is twice as high as it would normally be.  The input is just shifted up.

Then that point is rectified and filtered by D2 and C2 to "OUT".

After this, another diode can clamp another AC coupled input to the OUT

voltage...  On and on if needed...  

One drawback of these capacitive multipliers are that the AC coupling

caps have to pass all that ripple current and they can get hot and can

severely shortened their lives.  Lower ESR caps live longer.

They're normally best suited for small currents under just a few Amps at most.

I just replaced a doubler cap in one of my Alesis audio reverbs.

Below is a simulation of that little circuit above.

Green trace is AC input centered at ground, Blue trace is the AC coupled and clamped middle point, Red is the rectified output.

Happy Christmas everybody.

boB

K7IQ

[johnlm]

Ed,

You can't connect the 3ph Wye neutral point to the negative output from the 3ph rectifiers.  The potential difference between those points is 1/2 of the DC output voltage of the 3 ph bridge.  After I tried it it makes sense that it should be.

So you are back to two separate grounds (or negative points) running at different potentials.  Maybe the ideas of coming up with some kind of simple boost DC-Dc convertor to just boost up your mainline output when it is low is the way to go.

John

[Bruce S]

Ed;

 This has been bugging me for days. I finally figured out what it was.

Here's a reload of your circuit with just a few changes, that I'm told will work.

According to all that I've shown and tested using electronic software; this should be the best route to go.

It should also give you a full use of both full wave rectifiers, along with a better balance of the outputs.

The only part will be to make sure and know the resistance of each leg of the doubler will be so the caps can be sized to stay away from loading frequencies such as 60 HZ or one of its sub-harmonics.

Sorry for the large size pic.

Bruce S