building a 12 voltbattery charger

[Ronmar]

Glort
 Car alts do indeed run thousands of hours and many many years in service.  But they do not do that at extreme loads such as would be applied charging large battery banks. Their typical load in vehicle use is usually far below their max rated output.  The unit you describe does sound like a good unit, it having dual fans.  Can you put up any pics of the alternator you describe so we can take a stab at positively identifying it?

Or is this the beast?

http://www.rockauto.com/catalog/moreinfo.php?pk=2219582&cc=1428791

[Tom Reed]

You guys are really getting me interested in this direction, but I need 48v. I guess some testing with the voltage regulator thomasw is working on might produce some interesting results.

[Ronmar]

Have you read Bobs white paper?  You will probably need a 24V alt, or at the very least better rectifyer diodes, but the alt can deliver 48V, even using the internal 12V regulator.  You just need to separate the regulator sense from the B+ output of the alternator. The B+ would be connected across the 48V bank, but that regulator sense would only get connected across 1 of the 12V batteries in the 48V series string. The regulator sensing across the single battery will drive the output of the alternator high enough to get it's desired regulated voltage across that single battery, which means it will be pushing 48V+ out the B+ line across the 48V battery bank.     

[Tom Reed]

I saw Bob's paper and IIRC it did not really address 48v systems. Although it was such a good read, perhaps I should read it again.

[glort]

I was rummaging up the shed today and came across a Bosch alternator off a car over 12 years old now. It was the same 2 fan design as the ones I have been using.
Although it was marked Bosch, I have a feeling it might be a locally made version of a Japanese alternator given the radical change they made in bosch alternators virtually over night and the extreme similarities.

Having a look on Evil babe, I can see a heap of alts that look exactly the same from a wide variety of makes and models of Vehicles.
I also notice that all the internal fan designs also appear to be Dual fan designs

The alt in the link you posted looks like the alts I am using. Even the control connection plug is identical. I'm familiar with them as i took about half a dozen off some battery leads I chopped out of cars the other day and was sorting and organising them today.

As for not running at high load for long periods of time, I'm not so sure about that.  If you look at the features and electrical demands my wifes car has which these alts come from, there are some significant loads there. The things also in standard form run a battery that looks like something I would fit to a heavy duty flash light. I changed the battery terminals and put a battery about 3 times the size of the original in it before I even drove the thing home. I like having some battery capacity but these cars in std form leave a lot of work to the alt.

If one were driving on a wet night with the air and blower on, the rear demister going, wipers front and rear, Electric power steering, the seat heaters, Fog and High beam lamps , stereo, and radiator fans all on which is almost a given on a wet night, ( and there have been plenty here of late!)  I reckon the alt is going to be real close to max.
I think the cars run close to 20 amps just running. There is the computer and fuel pump plus the dash and other systems for airbags, antilock brakes, electric power steer etc that all run whenever the car does.

The supposition seems to be that these alts won't run long periods at full tilt.
Why not?
Is there some factual and or logical reason why not or relative experiences people have had I'm not aware of?

I can't see any reason and I think there is a significant amount of " Bob Factor"  to skew the advantage the way of the alt surviving. Just like with the Leece units he modified, as he pointed out, they run in far cooler conditions, at the same speed and in a clean , dry environment in stationary power supply use. 
From what I can see, the only thing that may not take kindly to high constant load use is the windings and Diodes.
In the case of the twin fan internal units, each heat loaded system has it's own fresh air supply.

I think a highly significant factor is that for the most part, an alt in stationary use would be running in air temps about 1/4 of that they operate in under the bonnet of a car. Electronics like cool environments so that has to be a BIG plus in their favor.
Other than cooling, I don't see any other issues.  I'm also thinking that in standby use, an alt wouldn't be running flat out ALL the time would it?  As the thing charged the batteries they would come up and the amps should back off even iff that took some time on a large battery bank.  It could be worked out from a given size bank and depth of discharge how the charge curve and amperage from the alt should taper off.  If one were charging a bank with 2 alts  producing 160A Surely in reality they wouldn't need to pump that out for more than 2-3 hours before the batteries came up and the amps started to back off.  If they didn't, then you wouldn't be actually charging the batteries, you would just be using the alt as a straight out power supply if you were pulling all the power back off again.

While vehicles may not run the constant high end loads ( But I'm pretty sure lots of them do!)  standby use may require,  they also don't have to cope with the high under bonnet temperatures and heat isn't friendly to most mechanical / electronic components.
I spoke to my father last night and raised the subject of Alternator longevity and his comment that he didn't sell them in his wrecking yard.  He said the main times he sold them now was for accident repair where they can be impacted and have a mounting on the casing snapped. He said that virtually every 10 yo car he gets in still has the original alt on it and he hasn't seen one burned out or failed in the last 10+ years.

He said years ago alternators were a good seller but now they hardly move. He asked if I hadn't seen the pile of them on the floor under the shelf where he keeps them? I said I did. He said that was what he had accumulated again from the about 40 he put in the scrap metal bin 6 months ago because he was tripping over the things all day.  He said starter motors sold OK but power steering pumps of the old mechanical type barely sold either except for accident replacement. He said manual gearboxes were on a waiting list but autos were another thing he got no call for and ended up dumping every so often.
His thoughts were that if the alts survive as well as they do in cars with the temps they get to under the bonnet, particularly here in the OZ country summer, they ought to last forever when they were running off an engine in a shed that would be extreme and rare to get to half the temp they would see in an engine bay.

Far from scientific evidence but I'd be interested to hear of any experiences that contradict this from people who have had them in backup situations and experienced failures.

The other thing I'm thinking is how long do people expect them to last in the stationary application and what qualifies as being reliable or having an unacceptably short life?  How long do other machines of the same capacity with full loads placed on them last for?
I could set up a test running an alt over loaded for a day but is that enough or would it have to be 3 years before the things were proven as being OK?
I tend to think If I couldn't cook the thing when overloaded for 6 hours or so, the chances of ever doing it are reduced at a minimum.

As the alts I get are 80A, I could easily drive one off an engine and put some inverters on it with 1500W worth of load. I could also run the things direct without a battery which I also hear is a no no but I have done it loads of times for long runs and still haven't popped one. Only thing I have done is collapsed the field so had to re boot them off a battery to get them to generate again but a quick flashing of the sense wire is all that takes.

Would anyone be interested in a day test and if I did it, would it prove anything or would the testing requirement to prove the acceptable longevity of these units be a lot longer?
I don't know what sort of standards and expectations there are with off grid power as far as what people expect and deem to be acceptable service times. If there was some sort of number than we would have something to work with as to the units being viable or not. 

I'm going away for a couple of weeks but If I think of it, I'll grab another alt or 3 when I'm back in the yard and dissect one and take pics of the thing as to how it's set up. From what I can see with the ones I have without pulling them apart, they are pretty standard, just have an extra fan between the windings and the diodes.

The other thing I see with these alts is if they can do a decent runtime ( whatever that is) even if you did burn them out and had to change them, they are cheap enough and  once you have them set up, would take minutes, if that to swap out.  There is also the ability to drive several of the things at a time to reduce the load on any one unit.

I have seen youtube vids where the young deaf people ( or soon to be!) with 1000W stereos run like 6 of them at once. Some of those Irragi's etc are 300+ amps. I think that has to be a wank though because if you were driving 6 alts at 250A ea, you would need to be pushing about 50+ HP through that one serpentine belt and if it took it at all, I can't see it taking that much HP for long.  I also wonder what the response of the vehicle would be like?  They must feel like there is a caravan  with jammed brakes behind the things!


For a critical application, a savvy operator could set up  4 of the alts and just drive 2 at a time. If there was a failure, you could just move the drive belt to the other pair which with a serpentine belt and the spring loaded adjusters, would be a job that took seconds.

Anyway, I'm always keen to learn so if anyone has any experience or data that shows why these things may not work long term ( and can define that so we have a measure) I'd be interested to learn what I may be over looking atm.

[mobile_bob]

Glort

i like your thinking, and why not do some testing to see what readily available alternatives
might be put to use.

something you might consider, (as i have also) is the use of a temp limit switch, such is is typical on furnaces and clothes driers. they come in varying temp ranges and might be quite useful as protection devices for an alternator that is being called on for heavy duty charging.

my thinking is to epoxy the temp switch to the stator core on the outside of the case, maybe a 180deg F or thereabouts would be appropriate as the temperature down inside the stator core and windings would likely be significantly higher than 180 F

use the switch inline with the field so that when the stator temp exceeded 180F it would disconnect the field and let the alternator cool down then switch it back online again.

maybe one could do some testing and determine that 190 or 200F might be ok too, i don't know
other than 180 in my testing would be good for a 555jho in a 24 or 48volt charging system as the temperatures at the stator stack don't exceed 180F unless i am asking for more than 100amps charging current, which equates to between 2800 and 5600 watts depending on whether it is driving a 24 or 48volt load.

those limit switches are easily found in electric driers, and i suspect gas fired ones too, and are also common in electric and gas furnaces... clothes driers use two such units one is a low limit and the other is a high limit , the latter is a failsafe should things get really out of control.

i think the low limit is typically about 140-150F and the high limit maybe 180-190F but i don't recall for sure and various manufactures might well use differing degree's depending on location and how far from the heat source etc.   

probably could find them easy at an appliance repair/parts shop, in the scrap or bought new from wwgrainger or similar.

most of the aftermarket controller/regulators offer optional temp sensors for both the alternator stator and the battery negative post, and they use the temp info to protect both the alternator and to determine the battery temp so that they can calculate the appropriate temp compensation along with provide for protection from overheating the battery bank.

if they (oem) realize the need for temp monitoring, we ought to as well?  maybe not monitoring but for protection?

bob g

[SteveU.]

Hey guys
M.Bob I actually did do this back in the mid 80's on those totally enclosed 24 volt Prestolite Cat alternators  that would always come back totally internally fried and melted down. Some sailboat marine units I would have like to do this on also. But  no room  to put it inside. And external I figured the cutomer/user would just remove and throw away a visiebly added on temp control away. I did start putting the glue on heat buttons with warratee disclaimers like they do on remanufactured engine cyclinder heads.
I actually had a personal watercooled VW that I could on hot summer nights driving along see the piggyback regulator turing the alt off to cool down then back on.
The new digital factory regulators might be electronic safer to tie into the regulator turn on or computor contol line to prevent rotor coil on and off spiking the whole system. The weaker systems with brush wearout hopping and skipping circut spiking almost aways wipes out the digital regulators.

Hey glort my wife just drove off to work in her 2007 Hyundai Tuson. She drives year around with her headlights "on for safety". Year around with her seat heater on high for her back problems. Both the windshield/wiper/mirrors heater AND the rear window electric grid heaters on. "To warm up the car." Sigh. Her poor low down on the V engine buried agaist the fire wall 80 amp alternator NEVERS see less than a 60 amp contious load on her Mitsu derivative Korean alternator. Only by refusng anymore to replace the factory halogen driving light bulbs every six months have I gotton her to reduce her electrical consuption. It was another "more lights" switch to her.
The manufactures had to up grade the insulations, cooling and continuos usage duty cycles on all of their vehicles early 90's and on or face the wrath of the consumer market place. Power limitations was no longer accepable. Short servive life's would get you Internet broadcast brand smeared.
Factory auto units are as good or better heat wise, duty cycle wise as the very best of the 80's Lestec/Pynix Gold hotamped units. The better ND and on machined housings gear reduction starter factory units (without the plastics planetaries inside) better than anything that could be direct drive made up.

Ha! Ha! The world turns. Why after 16 years I left the electrical unit rebuild field and went other directions.
My NipponDenso alt on the Chrysler 1998 minvan is 230,000 miles+ and still on the original long pac brushes and bearings.
Nothing actual factiry automotive 60's, 70's could do that. And damn few truck, bus and industrial units from that era either.

Regards
Steve Unruh

[BruceM]

I like Bob's themostat idea but just wanted to make one caution:

Make SURE that the thermostat switch being used is rated for DC (automotive), if used for switching the DC field.  The snap disc type switches in AC appliances are NOT rated for DC, and will fail quickly if no immediately.  If using an AC thermostat, you can protect with a solid state switch, Mosfet power transistor, or low coil current relay.

DC requires specially designed switches and contacts to have any durability.  The arc is not interrupted by zero crossings like AC, so while switching ON DC always works, switching OFF DC of any meaningful current can cause arcing and melt down.  For example, I tested 5 different heavy duty (AC 25A-35A) wall switches for switching my homestead's 120VDC to incandescent lights. With a 4 amp DC load, every single switch was totally destroyed (arcing, melting) on the very first "OFF".  Fortunately lamp switches today are still using the original Edison DC designs, and have no trouble with DC, and I was able to find some wall switches that were 150V, 5A rated.

[mobile_bob]

i think BruceM makes a valid point about protecting the switch against DC issues
i might suggest one of the little bosch style automotive relays, they are cheap enough, very durable and last nearly forever.

i wonder how long the temp switch might work however because the DC voltage is usually under 10volts and just  few amps in most alternators?

it might be that the switch would live a reasonably long time without a relay, however it is unlikely that its lifespan would be improved without the of the relay. 

bob g

[BruceM]

I don't think it's worth the bother to life test a cheap thermostat on DC, as it's just too time consuming.

I did some checking and some do have DC ratings, though at pretty low amps, but only in special models with special contacts and then at less than 1/10th the AC amps.  The problem is that the snap disc is an inherently poor design for breaking a DC arc.  Look down the rating list to see the special DC model.

http://www.wolfautomation.com/products/25554/12-disc-thermostat-automatic-reset-open-on-rise-ncbrselco-oa-series

This one can only handle 1 amp at 30VDC, and that will not change much at 12V.  And as usual pushing parts to the max spec is always a bad idea.

Many I've seen have no DC rating and usually that means they WILL fry on DC quickly.  The contacts must be DC rated or they just won't last on DC unless it's only the current needed for a relay coil.  A half amp of relay coil current would be too much for many AC only rated snap discs, I'd rather use a mosfet or solid state DC relay.

Check your parts spec sheet or don't use AC parts for DC except as a control signal.



 

[Ronmar]

Well it would be a pretty simple matter to put a small AC coil relay in the circuit, but if you are generating strictly DC for batts, you would have to grow some AC somewhere...  I more like the temp sense as a function of the controller.  I think it would be a little more consistent in operation, as well as allow you to monitor the running temp of the alternator in operation in your particular installation...

[mobile_bob]

well the switch ought to handle the dc coil requirements of a bosch style cube relay, their coils require about a half amp iirc, maybe less?

the contacts in the relay can certainly handle the field current requirements over hundreds of thousands of cycles i suspect.

the relay would be really easy to integrate into the design

bob g

[SteveU.]

Ha! Ha! Fellows here we go again. Creating a compex eleborate solution to a non-problem.
Those totally enclosed Prestolite Cat equipment alternators I refered to we often covered with mud and dust having to charge up a from dead a dual 4D or 8D battery  bank. Of course they would output overheat and cook themselves to death. These same equipments when stuck dead battery outback were often rechaged back up good enough to crank over and start with a man packable chainsaw engined driven little cased "A" series 24 volt Motorola air cooled alternator portable charger. They would have to run balls out for hours to do this. I NEVER saw one of these lottle air cooled alts ever burnt up. Brushes worn out sure from eating field dirt, dust. and volcanic ash.
You are not going to cook out one of these late model dual fan alternators. They are using the best grade of epoxy insulations. Stuffing them down low crammmed into a hot underhood engine envitoment they HAD to upgrade the fan circulation and on boarg electrinics packages also. Mitsu units used to rectifier die from cold wet driving slush  and snow throw ups. Stupid engine Engineers could not just swap the sealed power steering pumps from the top of engine down with the alt postions like American positioning. No. The Japaneese alt guys had to upgrade the alt units to servive in this horrible dusty dirty thermal shocked area. Enclosed sealed brushes/slipring areas was manditory then also.

My Cat alt modification were for a proprietary market advantage. Any time you want to make somthing survive in a collaping coil circut you best use selected condensers (capacitors) to absorb and dampen out the up to 350 volt spikes. This is old well proven tech. This was easy to set up with an oscilloscope and a rosebud torch unit heating on the motor driven test bench onto  a pair of discharged 8D's. I used heavily white themal paste installed off the shelf DIXON DC high amp thermal resetting circut breakers. That was it. Just an in-circut thermal breaker with a paralleled to ground selected automotive grade condenser. Cycled it on and off for 100 cycles. This was to be unseen discrete for these now old HD analogue voltage regulated systems. No more overheated melted down comebacks on that product line. Wearouts from then on.
Today now I'd do it differently. With all digetal controls, and with now high amp/high watt rotor coils -  IF THERE IS A PROVEN NEED!! skip completly around the field coil circut wattage/voltage issues and go for contolling the always now present external override engine contoller regulator input line. Very, very low watt circut with no kick back. Shunt this line to ground to kill the signal. The regulator and computors are well protected now for shorts to ground on control lines.
Are your creations??

Honstly if it can't be proven to be broken then don't fix it. That is the basis of good personal use YOU PAYING FOR IT IT; KISS developing. Break it first to prove it needs a solution. If you are not willing to use-it, break-it, then  up-grade it (or just back off abusing it) then you are in the wrong game and should be working for pay with an OEM or a University.

Just be sure fellows you are failing-safe and breaking-safe out there. Getting coil/condenser discharge jolted is a very pee yourself expereince at the least. Probably stop a pacemaker.

REgards
Steve Unruh

[mobile_bob]

i don't know if heating would be a problem for the little frame alternators or not, however i don't see how it hurts to provide a layer of safety or failsafe.

all we need is a clamping diode across the field, then you can excite it and cut it off all you want without fear of spikes on the DC buss?

most regulators have this clamping diode on the field output from the regulator built in, that way the regulator can pwm and not have spikes coming back or being induced in the stator?  at least that is my understanding

no if we disconnect the field or interrupt it with a temp switch then we also remove the built in clamping diode leaving no protection from spikes.

i would also think that most all quality inverters are going to have some buffering on their inputs so that if there were some spikes on the dc buss it wouldn't affect the inverter.

so that leaves the batteries, i suspect no more often than a high temp disconnect happens there would be no adverse affect on the buss to cause issues with the batteries?

my only point in using a temp switch is a failsafe not a design parameter to do active control of the alternator... just be there to disconnect it should things get really hot.

just thinking out loud

if i were to use a temp switch, i would do as Steve suggests and use it to disconnect the power to my balmar controller rather than disconnect the field between the balmar and the alternator field terminals.

because the balmar uses a temp sensor on the field core as an option, if the temp goes over iirc 220 deg F it cuts the field to half power so that the alternator can cool down... that is likely more than adequate for my needs

however if i were to build a small engine driven alternator to use as a backup charger for heavy charging of a large bank and was not using a balmar or similar controller with a temp sensor option, i sure as hell would use some sort of temp switch on my unit.

that way i can walk away from it and not have to worry about it going supercritical/thermonuclear meltdown on me.   

i don't know about the dual fan small case in question here, but i do know these newer delco' with the welded in stator leads will damn sure melt into a puddle on the ground should you try to fast charge with them into a  relatively dead or undercharged battery... every one of those have warnings to fully charge the battery before you start the vehicle to protect your warranty.

i cannot tell you how many of those things i have replaced over the years due to city delivery, stop/start/park 30-45min with lights flashing and engine off, then restart and to the floor indy style to the next stop... only to repeat over and over again in some cases 75 to100 times a day...

they go up in flames!

had there been a way to easily add a temp switch on those applications, i would have damn sure used them.

as it was i had all ours built with 13.8volt regulators rather than the factory 14.8volt units, this helped a great deal although it did not fully take care of the problem.

so what i am driving at is this, i am no fan of small case alternators for our purposes, however i realize the economics involved and the want to use them, so all i am saying is if you do, maybe get a laser temp gun and check the unit while it is under heavy charge conditions and see just how hot it is getting... if i see temps up over about 225-240F i am either going to add more air circulation or a temp switch or both.

bob g

bob g

[mobile_bob]

having given this topic another "thunk"

over the years i have moved over to the 555jho for reasons i have written several thousands of words in various places in support of...

in making that move, i got away from messing with the small case alternators, many of which are horribly over hyped by unscrupulous sellers making all sorts of claims.

it might well be that some of the newer offerings are better than the old 10si (i suspect they are much better) however trying to get 100amps or more out of such a small package on a more or less continuous basis, hour upon hour just would seem problematic to me.

even the 555jho which is rated at 160amps (cold rated) should never be subjected to much more than maybe 135amps (hot rating) on a continuous duty, and even then my recommendation has always been to limit the output to 120amps max for purposes of longevity.

these units have proven in testing to run thermally stable at 130amps and 28.8volts, however i have concluded for my use to limit them to 100amps for 24volt nominal service and perhaps 90amps for 48volt service... even though the units can surely do much more than this my thinking is why stress the unit if it is not necessary.

so using this logic perhaps a dual fan bosch or mitsu/japanese/denso whatever that is rated at 100 amps, in my opinion should be derated to perhaps 75-80amps max  for heavy charging over long hours.... surely the units probably can do significantly more output however doing so is not likely to extend the lifespan of the unit.

something i am thinking of building currently is a dual alternator unit powered by one of the harbor freight gas engines (205cc iirc ~7hp) using two of the 4800series leece nevilles and a balmar controller... the idea being to use a simple switching control scheme so that one alternator runs for a while, then it is shut down and the other runs for a while, alternating back and forth... and also have the ability to do some parallel operation as well as some series operation...  more or less a learning tool to do some research on the capabilities of such a system and the 4800 series at higher than native 12volt operation.

i think i might be able to drive the snot out of one of the alternators and as the temps get high switch out to the cold one on the fly...

just thinking

bob g