2 Bridge Rectifiers or 3 ?

[MWatt]

Two year reader, 1st time poster. I noticed in the 10 footer that there are two different methods to rectify AC to DC.  One using bridge rectifiers and the other using stud diodes. My question is.. why don't you use 3 bridge rectifiers, one for each phase, instead of two? Won't three disipate the heat better? Since I'm getting close to final assembly I need to find this out.

 Thanks in advance, Kendall

[kurt]

one bridge per phase works fine just hook the phase to both ~ posts of the bridge. danb has a drawing of just that floating around someware if you are inclined to look for it.....

[KEG]

MWatt

Try this ! it worked for me .

kevin

[KEG]

MWatt

Try this ! it worked for me .http://otherpower.com/rectifier.html

kevin

[harrie]

I think it depends on the output of your turbine, and what voltage your battery bank is. If your in 12 volt, you may need as high as 8 or 9 35 amp bridges on a large heat sink, if your output is above 60 amps.

One of my first turbines, was in 12  volt, and saw over 100 amps at times. I had 9 35 amp full wave bridges, and they would still burn out. I than went to three large stud diodes, and this took care of the problem.

[harrie]

I think it depends on the output of your turbine, and what voltage your battery bank is. If your in 12 volt, you may need as high as 8 or 9 35 amp bridges on a large heat sink, if your output is above 60 amps.

One of my first turbines, was in 12  volt, and saw over 100 amps at times. I had 9 35 amp full wave bridges, and they would still burn out. I than went to six large stud diodes three in cathod, and three in anode with heat sink and this took care of the problem.

[ADMIN]

You really only need one half-wave bridge rectifier per turbine output phase. But full wave bridges are so cheap and readily available, it's often cost effective to just parallel both sides of each full wave bridge into a 1/2 wave for each phase. You can get by with only 2 full wave bridges -- use both sides of one bridge for 2 phases, and one side of the second bridge for the 3rd. The 4th side is unused. BUT, normally you'll need more current capacity, hence using 3 full wave bridges and paralleling their sides.

If you can get a 3-phase rectifier cheap, that makes for the easiest wiring. We often cruise Ebay for them, as they are terribly expensive at Digikey for something with enough current handling capacity.

ADMIN

[ADMIN]

Oh, and, you CAN make your own 3-phase rectifier out of individual diodes. But this often gets more expensive (and is certainly more work) that just buying the full wave bridges int he first place. ADMIN

[finnsawyer]

One might be able to find auto alternators that have ended their usable life but still have good diode assemblies.  I haven't tried reusing one but it shouldn't be impossible.  Anyway 60 amps at 120 volt reverse breakdown voltage doesn't look too bad to me.

[Devo]

Sometimes I run my 8 footer wired in star through 3 rectifiers & it seems to run fine 1 output lead to each rectifier & 1 open.

If one was to split each lead into 2 to occupy both leads does this allow higher amperage to be run?

Devin

[ghurd]

Yes, but I still wouldn't go past the rating.

A nice side effect is it will make a little less heat under most conditions.

G-

[SparWeb]

Here's a diagram of my system, as it was last month.  In the middle are shown the connections to the bridge rectifiers.  Each phase is split into two leads and goes through a terminal on two of the bridges.  It adds up to doubling the current rating, or, from the opposite point of view, increasing the reliability withing the same current rating.  And, yes, heat will be dissipated better with this scheme, although in that sense you're comparing apples and oranges.

[Opera House]

All diodes have slightly different forward voltages.  When you parallel diodes in a bridge, add 6 inches of smaller gauge lead or more to the diodes before connecting the leads together.  This added resistance will hep the currents through the parallel diodes to balance and increase the parallel capacity.

[BigBreaker]

That would be a reason to parallel diodes that are not physically (thermally) connected.  The extra current will increase their temperature which will increase the forward voltage which will decrease the relative current.  You can also use the temperature of the different diodes to roughly match forward voltages.  Better yet, measure the current with a shunt.

[Ungrounded Lightning Rod]

Even with equalizing wire resistances you'll get better current balance using the same bridge for both feeds from one phase rather than splitting each phase between two bridges as you have drawn it.

 - The diodes in the same bridge are more likely to be from the same lot and thus better balanced.

 - The diodes in the same bridge are on a common heatsink, which tends to keep them close to the same temperature, not just reducing imbalance but also possibly preventing thermal-runaway imbalance.

[Opera House]

Higher current will increase the forward voltage, higher temperature will decrease the forward voltage.  Resistive balancing is always a good idea.

[SparWeb]

ULR, that's an interesting point of view.  I never gave it much thought.  One diode could turn "on" before the other, which places a higher load on that one for an instant until the second one turns "on".

The changing load on each phase might also be making the AC current waveform even more messy than it already is...  I've got something new to think about now!

Thanks.