(pdf) high output/efficiency alternator white paper

[mobile_bob]

Bruce:

no problem sir, i too would have expected no problems
and i really didn't have any problem until i exceeded about 100amps on my way to 130amps
at 28.8 vdc
thats when my rectifier bridge which is rated at 1000vac PIV and 150amps started to get very irratic

it seems that at about 75amps i could count on it working as designed "most" of the time, and as
you know "most" of the time is not good enough.

from my vague memory my thinking was i would not need fast recovery diodes until i got to many khz
and so i just assumed i got a faulty surplus unit, i ordered two of them so i tried the second and got similar
results,, that is when i called the manufacture and asked for their engineering dept.  thats when i learned
about high power bridge rectifiers for the most part are rated at up to about 100hz before they get flakey.

who knew?  i just figured a bridge rated at 1000 volts and 150amps per diode was good enough for my use
luckily it didn't damage anything, didn't seem to damage the diodes either.

upon looking further they make stud mounted unit with the 4" cable comeing out the end, single power diodes
rated at anywhere from 50-1000volts and 75-150amps and beyond that are fast recovery diodes, apparently they
are used in some sort of high frequency rectification apps, i don't now,, i think i have enough of both types
to build a nice bridge if it comes to me needing one.

so far i have figured a way around needing such a beast with some aftermarket automotive press in diodes
rated at 200volts and 75amps each they seem to be sufficient to hold up for 57.6 volts DC and ~130-150amps
collectively in the oem heat sinks

bob g

[BruceM]

I'm very happy to learn from your experience.  (So much cheaper and less aggravating!)

I never would have found out as I usually go for gross overspec (2-3x) on power components.  Cowardly but cool and covers sins in design and fabrication.

[Geno]

"Quote from Bobs white paper.
Next we will adopt an architecture from the alternators big brothers, that being separate
excitation of the field by a separate belt driven exciter. We usually have a 12 volt electric start on our engines, and as such must have an alternator or other charging source for charging the starting battery. This starting battery system will be the backbone of our excitation buss. We now have a 12volt excitation source for our 110-555jho (reapplied 12volt machine) so that if we were to turn the machine at a sufficient speed it would begin to charge. Because its cut in at 12volts was approx 800rpm, its cut in speed for 24 volt operation would now be approx 1600rpm. For reasons explained later we shall set our design speed to ~4500rpm, which is well above the requisite 1600rpm needed for cut in at 24volts. Such a basic system would work to charge a 24volt bank "if " one wanted to babysit a field control rheostat, but this is both ineffective and can be either injurious to your batteries or dangerous, so
End quote"

Some of you have been following my " Portable battery charger" thread and I'm ready to gather ideas to bring it to the next level. That being proper, 3 stage charging of a battery bank using an external 3 phase rectifier and my Flexmax60.

Goal:
To be able to charge a 12-48vdc battery bank at whatever power/voltage/hz the various components are capable of. Keep in mind I'm not looking for enormous amounts of power here either.

My question is:
Could I do what I think Bob states and use the starting battery as an excitation source with a field control rheostat to control the output voltage of the alternator? If I can do that I can then fine tune the alternator output to provide the most efficient voltage to the rectifier/Flexmax for whatever battery bank voltage I'm charging. From what I've read the flexmax is most efficient with an input voltage close to, but above battery equalize voltage. There's a table in the manual.

By the way, I'm a bit over my head here so if my idea will smoke something or lead to ridiculous inefficiency I apologize in advance.

Thanks, Geno

[mobile_bob]

does the starting battery get its charge from the engine's integral alternator?

if so you can use the starting battery as the excitation power source, and would be in keeping with the suggested method
in the white paper.

you will need to disable the internal rectifier bridge, because it has avalanche diodes that will clamp at about 30volts,
which leads to the alternator becoming a massive engine brake, lots of smoke and a burnt belt.

the next problem comes in what to use for a rectifier, it will have to be fast recovery diodes, those typical for use on 60hz
will not work for this duty because you will likely be anywhere from 400-600hz, the common diodes will go into full time conduction
and not rectify, and you end up same as the prior example, smoke and squealing belts

these are the major hurdles to overcome,

if you are content with a 24volt system, the oem rectifiers will just barely get the job done without going into avalanche provided you
use heavy enough cables so that you don't have anymore voltage drop than necessary, any drop limits the available charge level
before you trip into avalanche.

that is the short answer, if we have more info on your end use maybe we can determine the best approach to get you there.

bob g

[mobile_bob]

i am familiar with the niehoff alternators, at least the large coach units that are 24volt nominal and ~300amps
bloody expensive units that are not for the casual diyer in my opinion.  but if you can afford one it should be one
hell of a unit.

as for remote mounted rectifier, if you can remote mount the oem unit then you will be ok, if not
perhaps you can get a set of replacement diodes for the unit and remote mount them in a heat sink
with fans?

if that is not an option you will need fast recovery diodes, because most of the really nice heavy duty units
are rated for 50/60 hz operation and will lock up into full time conduction and not rectify, don't ask me how i know
this, suffice it to say it cost me a very expensive balmar controller.

what battery bank voltage are you wanting to work with? and
how large a bank in amp/hrs, and of course
will it be flooded lead acid, agm or some other technology?

btw, thanks for the kind words, speaking for myself "i am no guru" but i too am looking for the holy grail of dc battery charging w/cogen
and i think i am a lot closer than some of the commercial suppliers in that regard, probably because i have less concern for profit margins.

as for the aura-gen
i am only familiar with the product from their literature, and am very familiar with the design
it like everything else under the sun is nothing new, it is an axial machine, likely very expensive
and probably does what they claim, however
it looks to be targeted toward military use and as such has a lifecycle measured in perhaps hundreds of hours
as is the case with most military and aviation stuff, where weight and size take a front seat to longevity.

of the two units, the neihoff is the bullet proof one that will likely run forever with very low maintenance.

in my opinion of course, ymmv

bob g

[Geno]

Yes, the little diesel has an internal charger for the battery.

I remember reading the previous posts about the diodes going into avalanche and causing problems.

There are a bunch of new listings for 3 phase rectifiers on ebay. I don't know what frequency they can handle though.

My potential uses are
1)   (Little engine) At camp to recharge 12 volt batteries but I can probably get away with the stock alternator there if I'm careful.
2)   (Little engine) At home to recharge a 48 volt bank with a remote rectifier if the Lister is down.
3)   I may get another 110-555jho and hook it up to the Lister with a remote rectifier to charge the 48 volt bank. This would also give me a more powerful, CHP ready prime mover and another option if the st head is down.

I guess I'm just exploring the possibilities and I just can't stop tinkering with my stuff.

Thanks, Geno

[mike90045]

I think that the "Stack Rectifiers" will yield disappointing results.  Simple radio shack floor sweeping diodes will likely out perform them.
Modern schottky diodes are the way to go, with proper voltage and current ratings.

[BruceM]

I can't say much about the "stack" rectifiers as I haven't seen any specs or prices.  Custom rectifiers would not be something I'd be ordering for my own use due to cost. 

Schottky diodes are great, but you want plenty of headroom on the voltage rating.  Schottky diode forward voltage (loss and heat) is reduced with a lower maximum voltage rating, so it's tempting to cut it close, but if you do, you'll have to thoroughly evaluate the transient (spikes from switching) situation and perhaps add some suppression circuitry...all it takes is one over voltage spike and poof, your Schottky diode is now converted to a piece of wire (shorted).

I've seen one solar PV charge controller that uses a (too) low voltage Schottky diode to save cost on heat sinking. Just the inductance of the leads running to the solar panel is enough to generate spikes when switched off to fry the Schottkys. Within a few days of installation, every unit's Schottky diode fails.  In this controller, that means a much higher night-time drain than there should be, both from circuitry that didn't get shut down at the charge controller and because of PV night-time losses. 

The amount of commercial gear with poor or "fragile" design is pretty depressing.

BruceM

[mobile_bob]

your project presents some real hurdles in my opinion

getting to the high 60% in efficiency with the alternator is a testament to the quality of the neihoff product
however working at 12vdc nominal is not without serious compromises especially with large distances and high currents.

i understand now your desire to remote mount the rectifier, but i would pass on the selenium rectifiers for a couple reasons
one is efficiency, the other is the toxic smoke that is released should you have an overload or short. only recently have i found
that the selenium smoke is toxic, i remember well that acrid odor as a kid experimenting with those things,, maybe that is why
i am so strange today?? 

here is a possibility for you that might just do what you need with higher overall efficiency

check with balmar, and see if their mc612 will handle the field of your alternator, it should because it can handle the field of two
large frame alternators easy enough.

remove the rectifier from the case of the alternator,

attach three transformers from some salvaged ups systems like the apc units,
they are ~120/24 volt units, set them up as delta connected first, then maybe change to Y connected
to see which will get the result your after.

attach the regulator sense line to the battery bank,

rectify the stepped down voltage of the transformer pack,  and feed the rectified current to the battery
bank

your going to have to work with monitoring the field current, so place a 10amp fuse in the field wire
from the regulator, so if it over currents it will pop the fuse and not your regulator.

the goal is to spin the alternator up to a speed where it is making something over 60volts or so
let the transformers drop the voltage to the needed 12volt nominal for rectification and battery charging.

from what i can determine from the manufactures of transformers, 60 hz units can be used up to 400hz
with little penalty in efficiency, and up to as high as 1khz before things start to fall off significantly.
3 manufactures have told me that 500-600hz is ok with most 60hz transformers.

such a system if it were to work, would be pretty cool in that you could run your power from the alternator
on maybe 8 or 10 gauge wire, step it down to be rectified and charge at 12volt nominal, and the sense wire
being connected to the battery bank will tell the balmar what is needed to attain the charge level you program it for.

the balmar really doesn't care about how high the stator output voltage is, or that there are transformers involved
it just wants to see two things, a nice 12volt supply, and a 12volt nominal sense from the battery.

my bet is such as system would get you over the 70% barrier and quite possibly into the mid 70's in efficiency.

this scheme is nothing new, delco as well as leece neville both built piggyback transformer rectifier packs so that
a 12volt alternator could charge both 12 and 24 volt simultaneously

all we are doing is altering the voltage parameters, the underlying theory of operation is the same.

having said all this, it might be that a smaller alternator could be used and get you up to 250amps at 12vdc nominal
and do so at a lower cost, more common alternator, and perhaps as good or better efficiency.

[BruceM]

That looks like a nice solution, Bob; finding some suitable surplus transformers would make it sweet. 

12V sure is a bugger,  don't know why more folks aren't thinking 120VDC. 

[mobile_bob]

there are a number of options for press in diodes and Todd will know what to use
they are much cheaper, fast recovery, and avalanche will be fine and actually desirable for your
application.

i have some i have used with success that are 200piv and 50amps each, they cost about 5 bucks each
you ought to be able to parallel up a set to get you where you want to be.

if i were to spec a stud mount i would go for at least 200piv, maybe 100amps, fast recovery and avalanche at 30volts
or so for your application, the avalanche will catch spikes presumably and protect you delicate electronics.

its going to take some explaning to Todd, as my bet is he has not had to deal with out of the box applications such as being
described here.

if he has concerns about the ability to make such a system work, ask him to look at a A001-4417jb leece neville
it uses 12volt excitation and produces 48 volt output, so it is proof of concept should the question arise.

bob g

[mobile_bob]

your needing to stick with 12volts pretty much limits you to the neihoff, i think

given the distances involved,

how many amp/hrs is the bank again? i guess i need to go back and reread what your bank is comprised
of.

bear in mind that flooded lead acid batteries are at their peak charging efficiency being charged from 50-80% SOC
and then let the solar pick them up from there if possible.

your system requirements provide some interesting challenges that i haven't considered for a very long time, namely
the use of both a 12volt system and the troublesome long distances between the alternator and the battery.

depending on how big a hurry you are, i have one of the 4417jb's which i can use easily on my system to try out the transformer
pack and remote rectifier approach,  i also have a stack of transformers to pick from that would be nicely suited for the project.

if my back will allow me i will set it up for a trial run this weekend and see what the results are, this is an approach i have been wanting to try
out back when i was thinking of a 12volt system such as yours,

coming across the huge stack of 48volt exeltech inverters moved my focus up from 12 and 24, and am now committed to 48.

but i still want to know how this scheme would work out, and am willing to give it a go to find out.

at some point i am going to publish part 2 of this white paper, and it may as well include the transformer pack system.

reportedly the efficiency of such a system can be quite good, it would be interesting if it is in fact a reality to actually be able to
generate, transfer, stepdown and rectify and still show an improvement in efficiency,, most would be satisfied with a modest hit
in efficiency just to be able to cover the distance without using double 4/0 cables.

fun stuff huh?

let me know how Todd reacts to this scheme, i haven't had the pleasure of dealing with him much over the years, although i have
a couple of times, the last time when i was trying to find a balmar distributor in the seattle area, that was several years ago.

bob g

[mobile_bob]

its not a matter of compensation, but rather one of time and a limited physical capability that seems to be
getting worse rather than better,

i started physical therapy last week, the result of which left me almost unable to walk, then did it again monday
and have not worked since!  damn that little PT although small and cute, is about to kill me!

i would be more than happy to help where i can, and have a few parts to use for testing if you need them.

the biggest issue i see with your proposed system is it is an out of the box problem and 99.9% of the alternator shop
guys go crazy when confronted with odd projects. those sorts of projects take more time, time being money, and bitching
from upper management.

if you have a good personal relationship with Todd, you have a huge leg up in my opinion.  keep in mind though that when
it gets too complicated he might lose patience with the project in a hurry, and understandably so.

you might offer him a bottle of his favorite solvent, or a coupon for a steak dinner for him and his wife, that might get you in
good and have his attention. i am thinking you are best to lay out the problem, the possible solutions, and ask his input somewhere
outside his normal place of business. somewhere that doesn't have a phone or some counter guy pestering him while he is trying to
get his head around your project.

none of this is rocket science, but it does take a bit of uninterrupted thought to grasp the concepts.

best to have it all laid out, so that he can see you have thought it through and have considered all the angles, and better
have explanations why standard approaches aren't appropriate for your application.

fwiw
bob g

[mobile_bob]

the old handyman book while very interesting, thank god we have progressed to where we have
something better for rectification than that described in the book.

although every mcgyver type should probably have this at his disposal 

it is also interesting how in the early part of the last century diy'ers thought nothing of building and winding their
own transformers for whatever need was at hand.

everything from small lighting transformers to arc welders, they just jumped right in there and went for it.

filling the need was the order of the day, and efficiency was probably a distant second place, followed not so closely
with safety concerns.

this battery charger autotransformer is a classic example, no line isolation with an autotransformer.

doesn't look terribly safe to me, but if faced with no other means of charging a battery??? who knows???

desperate times call for desperate measures!



not sure i would want something like that on board a boat with me!

bob g

[mobile_bob]

a bit more thought on your project

another option that might be worth a look

perhaps a parallel set of delco bridge rectifiers, some of the large high capacity delco's
did just that to increase the capability of the alternator while using common off the shelf
rectifier bridges from their smaller alternators.

i would have to do some digging, but i am almost positive there are some high output delcos
that used either 2 if not 3 bridges in parallel operation.

they are cheap, readily available, and easy to work with, and have all the attributes needed
in that they have the fast recovery, avalanche and heat sink all in a prepackaged unit for 3 phase operation
at much higher frequency than you will be running at.

that would be a simple solution, to that problem

i mentioned the use of recycled final drive transformers from old ups units, such as the apc brand
the 1400xl is rated at 1440 kva, 1050 watts per unit, and my bet is the transformer is not the weak link.

i bet they can operate at the 1440kva all day long, the secondaries are conveniently wound on the outer section
of the core as needed for this application, and are very heavy gage rectangular copper. they should handle upwards
of 100amps each on the secondaries, connected in delta i think you might get your 250amps at 14.4volts charging.

probably be prudent to put the transformers and rectifiers in a fan cooled and vented enclosure, but i think you could
put together a test unit for very low cost.

the fan could be thermostatically controlled to keep down losses, or simply have it run only when the alternator is charging
would be easy to do.

hmmm, more to think about

bob g