3 phase rectifers

[(unknown)]

I was wondering why it is that no one (that I've found) makes 3 phase rectifiers for "wye" connections.

Say for example I had a 208/120 3 phase alternator and I wanted to get 120VDC, I would like to think that I could get a 3 phase rectifier with a connection for the neutral wire, rather than have to use 3 single phase rectifiers.

Is there something I'm missing here?

[Flux]

To use single phase rectifiers you have to open the star point. If you can do this, why not connect the alternator in delta and use a 3 phase bridge.

If for some reason you can't open the star point, you can't use 3 single phase rectifiers unless each one supplies a separate load, you can't common the dc outputs.

In this case you could use a half wave rectifier. Make the neutral the dc negative and either use the positive half of a 3 phase bridge as the positive or use 3 individual diodes with the anodes to the 3 phases and common the cathodes for dc positive.  Dont expect to get full load rating continuously without checking for excessive temperature rise in the alternator winding, you are not using the windings at best efficiency.

Flux

[commanda]

When you say "single phase rectifier", I presume you mean a bridge rectifier. A 3 phase rectifier uses 6 diodes, and doesn't care whether your source is connected in star or delta.

You can, of course, use 3 bridge rectifiers as a 3 phase rectifier; just connect the 2 ac pins on each bridge together. A lot of people use 1 1/2 bridge rectifiers to get a 3 phase rectifier.

The short answer is; just ignore the neutral.

Amanda

[JW]

"Is there something im missing here?"

Yes, and it is infact, the neutral. If a stator is connected in delta 3phase then 6-diodes or 3 rectifier junctions are needed, a bridge rectifier is basically 4- diodes wired in parallel, that a 2- diode recifier configuration could handle easly, except with the bridge, a wheatstone configuration is utilized, allowing 4- cheaper diodes to be used in the place of 2 extra-speical ones .  

 Anyhow your not going to be able get away with converting an AC stator branch(one of three) that will infact produce 120VAC per-phase, since RMS will deterime a VDC voltage "less than that". This is due to rectification of an AC current RMS too- VDC.

JW

[JW]

I was just pondering the switchable STAR/WYE to DELTA ideas, that have been disscused here on the board preiviosly. Seems to me, that one would need 6- SCR's to do this, considering the neutral, could 7 SCR's be needed? is anyone pondered this, with SCR's? nevertheless youd still need the rectifier circuitry after the SCR's to convert AC to DC. Just seems to me to make a star/delta swithchable stator circuit youd experience heat losses thru the SCR's, perhaps using mechanical relays a star/delta swichable ciruit would be more efficient?

JW

[Flux]

With scrs and triacs having a forward volt drop of over 2v this is only a viable option on high voltage machines. At 12v it would be hopeless.

Flux

[commanda]

A star-delta switch using 3 relay contacts is very simple & easy to do. You can also make it automatic, switching at some pre-set rpm. Need a tacho circuit to give a dc control voltage proportional to rpm. Hugh Piggott has a basic circit on scoraigwind. For a circuit with all the bells & whistles, see my diary.

Amanda

[(unknown)]

Thanks for the comments. Some of them I need to think about more / ask a few more questions before I understand.

First a few clarifications.

I don't know if I can open the star point or not, I suspect not.

I do want to "common the outputs" as was asked/said in one comment.

By single phase rectifier, I do mean a bridge or full wave rectifier.

I understand how RMS will reduce the DC voltage, but for 3 phase conversion I believe the reduction is minimal?

So let me restate, given 208/120 3 phase AC, what's the best way to get

120VDC +/- 5% ?

[Flux]

Unless you can open the star point you are stuck with the half wave circuit using neutral and 3 diodes. What voltage you get will depend on the type of load.

Flux

[(unknown)]

"... you are stuck ..."

Well that's unfortunate.

I would like to understand better what the issue is that prevents being able to use full wave.  Can you point me to a reference?

Also, could using an isolation transformer on each leg overcome the difficulty?

[Flux]

You can use full wave as Commanda suggested by ignoring the neutral but unless you can find a way to reduce the alternator voltage you will get too high a voltage.

You can indeed solve the problem with transformers if you can stand the expense. In that case the simplest and most efficient way would be to use a 3 phase auto transformer  or 3 single phase ones to reduce the line voltage from 208v to 120.

My thought is that if you have the neutral it should be possible to open it without a great deal of problem and connect the winding in delta.

Flux

[(unknown)]

"My thought is that if you have the neutral it should be possible to open it without a great deal of problem and connect the winding in delta."

Sorry for being so dense, but I don't see what that gets me.  Opening up a 208/120 star just gives me 208 delta doesn't it?

[Flux]

No. The phase voltage of the alternator is 120v. with the thing star connected the line voltage is 120 x root 3(1.73). this is where you get your 208v.

If you want 120v the normal idea is to connect the windings in delta and then you get the lower 120v.  If you just open the star point and don't do anything else it becomes an open circuit, it doesn't become a delta without re connecting.

Normally if marked with dual voltage the winding has 6 leads brought out to make the star/delta change a user option. If this is not the case it must mean 208v line 3 phase and a single phase option of 120 between each line to neutral. To get 120 3 phase you need to re connect.

Sorry I can't easily produce a connection diagram for the change from star to delta but search the board others have posted it in the past.

Hope this makes some sense.

Flux

[(unknown)]

"Hope this makes some sense."

It makes sense. Thanks.

One last question, is this same kind of reconnection applicable to a 3 phase transformer as well?

For example could one reconnect the output of a 480 - 208Y/120 3 phase transformer to get 480 - 120???

[Flux]

In theory yes, it all depends on the construction and how easy it is to get at the star point. If it is not too big it is generally fairly easy but larger strip wound ones may be rather hard going especially if brazed or tig weleded together.

Flux

[tecker]

 This is a stranger view point the 120 will be a single phase . Each leg of the 208 to the neutral is 120 single phase . pick a leg put a bridge from neutral to the line and wala 120 dc.

[(unknown)]

"pick a leg put a bridge from neutral to the line and wala 120 dc"

I neglected to state the requirement for full power conversion at maximum efficiency.

[(unknown)]

In an effort to better understand the issues discussed in this thread, I started poking around my shop and found this transformer whose nameplate I photographed and attached below. The primary voltage says 480 Delta but the wiring diagram clearly shows a Wye connection.  So is it wye or delta?

[Flux]

I shall have to let someone familiar with North American rating plates answer this one. It makes no sense at all to me. It states that primary is delta but shows star.

It shows a star secondary but if X0 is a neutral, why is it part way up the winding.

If the bottom drawings are just representations they are at best confusing and if they are phasors and the angle is really 90 deg then it must be some sort of Scott connection. I am not surprised that you are confused.

Flux

[JW]

Gnarl,

 What the hell is that a 6KVA Variac? you must work for NASA. Ummm, just make 165vac on your stator 'per-phase' rectify that, and whalaaaaa, you've got 120vdc at the proper load, per-phase. Parallel all that thru the dc circuit and whaabing there you go probably wouldnt even need a transformer if the set up was engineer'd for the average wind speed in the area. Qurious to what the diameter of the prop would be, on such a machine. By the way, a half wave rectifier, is a single diode, 2- diodes wired in series with three connection point's makes a basic fullwave rectifier, the center-point-inbetween the 2- diodes, is the ac input, and the cathode and anode of the remaing diode terminals defines the +/- terminals of the dc circuit. This can just as easily be done, with 4- diodes, aswell as 2-, and is commonly called a fullwave bridge rectifier, actually, the way that the ac terminals on a standard bridge rectifier block can be paralleled, lends itself to this. Not to mention the advantages of increased heat-sink efficiency.

 But seriously, star is probably better for a low cutin speed.

JW

[ghurd]

How about triacs or scr's used for controlling mosfets to carry the current?

G-

[LEXX]

I gotta agree with ya there Flux, I've set up a few square D transformers and they have never had such confusing diagrams.  Two were even the same specs but they were rated star they showed star and guess what... they were star, that one I wouldn't really make a guess at.

LEXX

[Nando]

Amanda:

The rectifiers exist, it is the same for a DELTA or STAR but one needs to look the arrangement.

You want to connect Neutral and the 3 phases, the only solution would be half way rectification and not to generate the negative voltage. (or the reverse if you want a negative voltage output)

To get 3 phases STAR full wave rectification for 120 Volts, there are two paths:

Conversion to DELTA and have the 3 phases 6 diodes or to separate the connection to neutral- common-( STAR) and connect 3 single phase full wave rectifiers -- which is the same thing that a DELTA connection .

Regards

Nando

[dinges]

G'day

I think the original poster is looking for a bridge rectifier, but for 3-phase instead of single phase, so he can use 1 'brick rectifier' instead of 3. I've been looking for these too, and they exist. They have 5 (instead of the usual 4 for single-phase) connections, 3 ~ ACs in, 2= DC out (+ and -). I want them because it'll make the wiring easier, neater and mechanically simpler & more stable/durable. However, the only thing is the price: Farnell (www.farnell.com) wants about 20E (25$) for one of them (15amps, if I'm not mistaken). That's a bit too much, especially when you have plenty of diodes & bridge rectifiers in your junk boxes. If anyone knows another place where they're available for more reasonable prices, I'm interested.

It surprises me that no one else uses these 3-phase bridge rectifiers; especially for the less-electronically inclined ('electrically challenged'?), it would make wiring it up much easier: 3 wires (star or delta, doesn't matter which you choose, you'll have three wires out) to rectifier, 2 wires (plus and min) go to battery/ charge regulator. The downside is that they're harder to find than single-phase versions, but they do exist!

BTW, it's quite easy to build a switchover from star to delta; after an evening of thinking, I've come up with the solution (probably a re-invention of the wheel), needing just one one switch (with 3 decks). Building it is easy, once you get the wiring diagram....

Hope this has been of help,

Peter.

[ghurd]

The problem is mostly the high power these machine are making.

Like the 800 watts at 12 volts is around 65 amps.

The $50-$100 3-ph are 50 to 100 amps, and 240 to 1000 or more volts.

G-