Best Practice for 48v Altrenator?

[thomasonw]

I had a chance to review postings in Microcogen forum looking to see if there was a kind of 'best way to get a 48v alternator' agreement, but I am unable to find that.  I see posts around maybe using the 555, and questions its efficiency at 48v vs. 24v.  I see post of folks suggesting two 555's in series to get 48v.  Some about large frame Leece Nevilles with external Diode packs, and perhaps a re-config of the stator.  As well as opportunity buys of manufactured configured 48v alternators. 

So here is my question:  Given all that is known here.  If one is looking to install a 48v capable Alternator, say in the 50-100A range, what would be considered the 'best practice'.  Best practice being  a balance between cost, efficiency, and availability.  One that will not break the bank, rely on being at the right-spot at the right-time, and does not require too much magic.  (ala, likely external Diodes will be needed?   But if one needs to hand-wind the stator, well, that seems over-the-top).

And for those who have deployed 48v alternators, what did you do and how is it working?

Thanks,
-al-

[Tom Reed]

I concur with your assessment and eagerly await the outcome. Also don't forget rewiring an ST5 to put all the windings in parallel to get a rectified 60vdc output. Any candidate platform will need the ability to reach around 62 volts in order to be able to equalize a battery bank.

[BruceM]

+1 for Tom's suggestion.  It would be hard to beat the ST head wired for a nominal 60 V, and your existing regulator design with little modification could do charge/voltage/current regulation.  Not compact or lightweight, though.

[mobile_bob]

+1 for the suggestion Tom made

so far i have not had the time to further explore the 48volt thing
what i know so far is this

all of the oem 48volt units that i am currently aware of do not come close to the efficiency of the st genhead

there might be a couple of high end units (costing greater than maybe a grand each) that might approach the st efficiency levels, i am not sure though.

the idea was kicked around some time back about splitting the st windings and using switchgear
(relays) one could produce a series/parallel setup so that the st head could be dual purpose, charging 48volt nominal or switching back to 120/240vac operation... in my thinking it might be the best of both worlds.

the only downside i can see is one of higher ripple generating 48volt nominal from a single phase head, although maybe that can be tamed down a bit with enough filtration.

fwiw
bob g

[thomasonw]

<SNIP>
the only downside i can see is one of higher ripple generating 48volt nominal from a single phase head, although maybe that can be tamed down a bit with enough filtration.

fwiw
bob g

That would be a mighty big inductor!  Maybe I should have hung onto those two old-iron inverters I put into the give-n-take shed.   

Has anyone on the list actually deployed an ST head in this way?  And learning's / experiences?   And guess the same question for non-ST head deployments, has anyone built and is using a 48v direct DC charger?

-al-

[BruceM]

The filter choke would be manageable size if a toroid core is used.  For my single phase 240ac to 120vdc charger I used 2 surplus toroidal transformers (L-C-L), with a 10K uF cap using the 120vac primary winding for the choke windings and my ripple is about 25 milivolts.  The best bargain I"ve seen for big power toroid cores are the autotransformer (single winding) step up/down toroid transformers being sold on ebay.  I think they have 3KW and up in size. 

Too bad you couldn't get a three phase 60V ST head...

[Tom Reed]

Well ..... if a ST could be modified to make all the windings in parallel, perhaps an STC could also. I have to admit that 3PH is a bit of a mystery to me. A member of the lister board has successfully completed this modification of an ST5 and fed the resulting power into an Outback MF60. I don't remember what he used smooth out the ripple though.

[thomasonw]

OK, going to take another run at this - aside from the ST approach, which might be a bit costly and heavy for some applications, looking for suggestions on 'alternators' to use for 42/48v DC generators.

A quick note:  In addition to the work Bob did looking at the 555 alternator in 24v applicaitons, here is a rather interesting paper which in part gives some insight into the advantages of using a '12v' alternator in higher voltage systems:  www.researchgate.net/profile/Nesimi_Ertugrul/publication/237464668_EXTRACTING_MORE_POWER_FROM_THE_LUNDELL_CAR_ALTERNATOR/links/0c960532a1b4458951000000.pdf

In short:  we get better efficiency but need to spin it faster (higher cut-in RPMs, just as Bob noted).  The attached paper is looking at one way to overcome this higher cut-in w/o doing the traditional approach of rewinding (which lowers peak efficiency) by the application of a boost switching power supply.  None of which is needed for a DC generator; we can just take advantage of the stock '12v' alternator - spin it fast and get an efficient 48v charging source.

But which alternator to target? 

• Which ones do not have 'avalanche' diode packs - clipping things around 29v?
• Is there an easy way to determine?
• use the 555?
• How about a more modern alternator?   

Suggestions / thoughts? 
Has anyone actually made a 42/48v DC charger using a '12v alternator'?

-al-

[BruceM]

Some folks are using the Midnight classic controller to buck convert down to nominal 48V battery bank, and then you can start with any 120VAC generator head that you can down regulate just a bit with your AVR. 

[mobile_bob]

i too would be interested in seeing more hands on testing of alternators for this purpose

fwiw,  these are the reasons i like the 555 alternator

1.  economies of scale allows for new units to be found at sub 200 bucks without core charges
2. it is easy to access the brushes to bypass the oem regulator
3. the oem (new) avalanche rectifiers which limit output to about 29vdc can easily and readily be replaced with the earlier non avalanche rectifiers with piv rating of 200vdc at 50amps each, which makes 48vdc nominal no problem
4. there is at least one source for custom wound stators, which i have not tried yet to see if a stator could be wound to get the 48vdc nominal at higher efficiencies.
5. the unit is built well, using front and rear ball brgs
6. the unit is large frame with a low cut in rpm

to my knowledge, which admittedly is limited to alternators commonly available up to about 7 years ago, there is just nothing else that really fits the list of requirements that i would at least have to have met before i invested much money or time to experiment with.

the issues i have with running the 555 at 48vdc nominal relate to having to run at ~6500rpm to get to the needed output levels.  this is still within the rpm rating of the design (oem states 8krpm max speed) however the losses are higher at the required rpm.  i have tried to reconnect the stator from delta to star, this of course lowers the rpm needed to get there, however the increase in resistance of the reconnected stator increases losses which seem to offset the gains made from lowering the rpm.   i think it might be worth the effort to have a custom stator wound with a few more turns, maybe 5 in hand to get the resistance down to nearly as low as the stock 12volt stator, maybe where the rpm requirement could be down to around 5-5500rpm and do some testing... we might get there doing that.

failing that, i have resigned myself to running twin 555 series connected to get the 48vdc nominal charging ability.

if anyone comes up with another large frame alternator that is reasonable in first cost to work with that shows promise i would surely like to hear about it too!

bob g

[thomasonw]

<SNIP>

failing that, i have resigned myself to running twin 555 series connected to get the 48vdc nominal charging ability.

<SNIP>

bob g

Bob,  are you actually running a 2-alt in series setup?

-al-