• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

3 Phase Bridge Rectifier Specification with Alternator Question

Status
Not open for further replies.

Puzzeled

New Member
I should preface this question indicating I'm a house power storage and generation system enthusiast, not an electrical engineer.

I am building a Remote Rectifier for a small case P style 12V 120A engine driven alternator that will have external voltage regulation. The reason being the alternator is located in an area of high heat and under a high constant load so wish to reduce the heat of the alternator so it doesn't experience as much output loss or burn up. The existing internal rectifier will be removed and the 3 phase AC taken off the alternators stator via heavy gauge wires (say 6g) to the Remote Rectifier. It will have a heat sink and a fan. The alternators voltage regulators power source will go via thermal switch on the Remote Rectifier's alloy case in the event the fan malfunctioned and the rectifier overheated thus causing the regulator to switch off the alternator. Alternator output should be in the order of 100A hot with this modification so not to over stress the windings.

My knowledge of bridge rectifiers in this application is only limited to knowing they are taking 3 phase AC and via 3 sets of diodes converting it to DC and their DC current capacity to exceed that of the alternator to allow some headroom. I assume any excess current capacity of the rectifier will simply mean it will operate at a cooler temperature? I have attached a picture of the bridge rectifier I'm contemplating which indicates it is a full wave module with 3 sets of diodes and rated at 1600v and 300A.

My questions are;

1. Other than the current rating of 300A is there any other rectifier specification that I need to be concerned about for this application?
2. Am I correct in assuming the 3 phase AC cable run length even at 6g will need to be limited otherwise DC current output will be reduced?

As you can see this is my first post so I would be greatly appreciative of any guidance and assistance given.

Thanking you in anticipation.
 

Attachments

Puzzeled

New Member
Thanks Ci but your words "not exactly to subject" are very understated..good looking girl in the vid though.
 

ci139

Active Member
good looking girl in the vid
:meh: aren't they all if not over stressed
it was just the 1-st youtube result that included the kW spec in it about active rectifier
you likely go near 3kW -- but this is a fast estimate
 

MikeMl

Well-Known Member
Most Helpful Member
Think of it this way: Alternator x meters from load. You have a choice. The power is transported from alternator to load either on two wires (DC, rectifiers on alternator) or three wires (3ph AC, rectifier at load). Power lost is proportional to Irms^2 * wire resistance. What are the currents in the two or three wires? What does this say about wire sizing?

Since the regulator compensates for Ohmic drops along the wires, other than overheating the wires, what do the drops matter in either method?
 

Puzzeled

New Member
Think of it this way: Alternator x meters from load. You have a choice. The power is transported from alternator to load either on two wires (DC, rectifiers on alternator) or three wires (3ph AC, rectifier at load). Power lost is proportional to Irms^2 * wire resistance. What are the currents in the two or three wires? What does this say about wire sizing?

Since the regulator compensates for Ohmic drops along the wires, other than overheating the wires, what do the drops matter in either method?
Thanks. It was not a length of AC cable from alternator to the rectifier versus DC cable (much heavier) from rectifier to the load question. There is an opportunity to locate rectifier in a better location with a longer AC cable run but no point doing that if that impacts on alternator output capacity. I don't know if it does or doesn't?
 
Last edited:

Puzzeled

New Member
For the rectifier ratings, you also require Peak Inverse Voltage values.
Thanks but I wouldn't know what peak inverse voltage was if I fell over it. My simple thinking was if this rectifier is rated at 1600v then that is way above the voltage generated by this application so there would be no problem using it?
 

MikeMl

Well-Known Member
Most Helpful Member
Thanks. It was not a length of AC cable from alternator to the rectifier versus DC cable (much heavier) from rectifier to the load question.
Nope, the DC wire only needs to be slightly bigger than the AC wire for equal losses. That is because the power is being carried by only two conductors instead of three.
There is an opportunity to locate rectifier in a better location with a longer AC cable run but no point doing that if that impacts on alternator output capacity.
By moving the rectifier out of the body of the alternator, you will gain slightly because the heat dissipated by the rectifier stack is moved out of alternator body, and heating determines the ultimate power output.
 

Puzzeled

New Member
Nope, the DC wire only needs to be slightly bigger than the AC wire for equal losses. That is because the power is being carried by only two conductors instead of three.

By moving the rectifier out of the body of the alternator, you will gain slightly because the heat dissipated by the rectifier stack is moved out of alternator body, and heating determines the ultimate power output.
AC wire size is limited to 6g with regard to space practicalities of getting the 3 phases out of the alternator case. There is no physical limitation on DC cable size and can be sized accordingly for current and minimising voltage drop which is important for battery charging. Sounds as though AC cable run length will effect current output so shorter is better.

Alternator longevity and output gain from reducing heat within alternator and at diodes by going from internal to external rectifier is what drives this project.
 

MikeMl

Well-Known Member
Most Helpful Member
The current in the DC wires is only 30% higher than in the AC wires.

37..png
 

Puzzeled

New Member
The current in the DC wires is only 30% higher than in the AC wires.

View attachment 106015
Wow Mike. Your effort towards doing the calcs and diagrams incl wave is greatly appreciated.

You have answered in detail for a comparative ludite my Question 2/2 being; "Am I correct in assuming the 3 phase AC cable run length even at 6g will need to be limited otherwise DC current output will be reduced?"

My only remaining original and unanswered Question is; "Other than the current rating of 300A is there any other rectifier specification that I need to be concerned about for this application?"

schmitt has kindly advised above that; "For the rectifier ratings, you also require Peak Inverse Voltage values" As noted I have no idea about peak inverse voltages (and this particular chinese made rectifier is not specified in detail in english) so my simple thinking was if this rectifier is rated at 1600v/300A then there is ooddles of AC headroom voltage with this alternator application so there would be no problem using it? In fact alternator size could say double to 250A at 12v DC and that would still be well inside the AC specs of this rectifier. Or am I dreaming??

Thanks again.
 

MikeMl

Well-Known Member
Most Helpful Member
I repeated the sim to ask about the rectifiers.

The green trace is the PIV required for D1 (the others are the same in all plots). Note that typical automotive alternator rectifiers are rated at 100V.

The yellow trace is the current through D1. The waveform has an average value of 39.7A, and a rms value of 62.3A.

The red trace is the power dissipated by D1, assuming that this is a Silicon Ideal Diode. The average power per diode is 37.7W. A real Silicon diode will have a higher dissipation because of internal resistance. You could do better with Schottky rectifiers.

If you move the six diodes out of the shell of the alternator, this raises some serious heatsinking issues. Note that inside the alternator, the diodes are mounted inside the metal shell, but are cooled with a built-in fan. To get rid of a couple of hundred Watts if mounted externally, the rectifier stack will need a big heatsink, with either serious convection or a fan.

37a.png
 

Puzzeled

New Member
Thank you very much for that Mike. I am not expecting the diodes in that particular rectifier to be top of the line efficient. If you see my original post I'm intending to put in a large heat sink with fan and over temp protection switch in the event the fan malfunctions. Cheers.
 

shortbus=

Well-Known Member
Most Helpful Member
I'm probably missing something here, but why not just use the original diode bridge you take out of the alternator as the 'remote' bridge?
 

Puzzeled

New Member
I'm probably missing something here, but why not just use the original diode bridge you take out of the alternator as the 'remote' bridge?
Fried with the heavy workout it got plus for a few reasons don't want to reuse the same bridge size remotely.
 

Puzzeled

New Member
I'm probably missing something here, but why not just use the original diode bridge you take out of the alternator as the 'remote' bridge?
short

I should probably have also mentioned that an alternatators internal diode bridge is designed to be cooled by one or more large fans drawing air in through the back of the case over the bridge and to cool the windings and case and running at around 6000 RPM and whereby output drops of they fail when a battery is looking for max output at lower revs. Hence they are not suitable for reuse in remote rectification even in association with a large heat sink and electric fan.
 

ChrisP58

Well-Known Member
Most Helpful Member
Does anyone know what the typical frequency of the AC coming out of the alternator might be?

Just wondering about the impact of the skin effect on the wire resistance.

For instance. At 2KHz, the penetration depth is about 60 mills, so a 6 gauge wire would have a higher effective resistance in this application than expected.
 
Last edited:

tcmtech

Banned
Most Helpful Member
A typical automotive alternator is a 12 pole unit so if you know you're working RPM's you can figure out the base frequency from there. At 3600 RPM you would have a 720 Hz base frequency which if its a commercial unit rated for continuous duty would be rated at 3600 RPM Vs the typical 6000+ RPM that most automotive units are.

Now as for the rest of the system and components there are some common misconceptions regarding rectifiers three phase power and general wiring.

First of any decent rectifier is rated for average amps which means that you don't need a huge amperage rating for rectifying three phase power. To be honest the diodes in your original alternator despite it being a 120 amp rated unit were likely not rated for much over 40 - 50 amps each tops. I have never found any alternators that used huge current rated diodes for their relevant amperage ratings.
Reason being for practical purposes to get 120 amps DC out of three phase AC they average current per phase would be about 34 amps which for our application a 50 amp three phase bridge would be plenty enough.

Next is realistic overhead voltage rating which for a 12 volt systems any diode with a 50 or more volt rating is way more than you need so the 1600 volt rating is just wasting power due to it inherent higher forward voltage drop which is likely around 1 volt Vs what a common germanium type diode which was commonly used due to it low forward voltage drop of ~ .3 - .4 volts at full load hence how they get so much power through a bridge rectifier with such a small heat sink even with air cooling.

So given that, by realistic and past personal experimenting with modified alternators of all makes and ages your alternator using 6 Ga wire for each phase should be able to supply a good 100+ amps at the 12 - 15 volt range it's rated to work at a good 50+ feet away! Odds are if it was a quality commercial unit that has a fair amount of reserve capacity that could be used to compensate for a bit of voltage drop in the lines even 10 ga wire for a 50 foot run would be more than big enough. ;)
 

Puzzeled

New Member
Does anyone know what the typical frequency of the AC coming out of the alternator might be?

Just wondering about the impact of the skin effect on the wire resistance.

For instance. At 2KHz, the penetration depth is about 60 mills, so a 6 gauge wire would have a higher effective resistance in this application than expected.
Thanks Chris. I had to look this up to see what meant. The output frequency calc is as follows I find and decided by two factors:
  1. Rotor Speed
  2. Number of Poles
and where,

N = Speed of rotor in rpm, see below

f = Frequency of generated emf

P = Number of Poles ie 12 for this alternator.

N = 120f/P
or
f = PN/120

So in this case where there are two principle generating rotor speeds of say 3500 and 7000 RPM their resultant frequencies are:

(12 x 3500) /120 = 350 Hz
or
(12 x 7000) / 120 = 700 Hz

I am assuming that even at 700 Hz the AC "skin effect" of the AC conductor is negligible.
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top