maybe a new project

[mobile_bob]

because i have all the parts and a need for the finished unit i am thinking of building the following

an inverter/generator using the following components

an R175 changfa water cooled
the leece neville 48 volt alternator
a xantrex xar 3 step regulator
a set of ups batteries, quite small but adequate for use as a buffer
an exeltech mx series 4kwatt pure sine wave 48/120vac inverter

what i am thinking is setting it up to drive the alternator for max output at the full rated engine speed of 2600rpm

the engine speed will be regulated via a custom wound current solenoid that is used to bias the governor spring
the idea being if there is the full load current flowing from the battery to the inverter based on a large inverter load
the current increases in the solenoid coil and it pulls the spring and governor to full throttle, as load decreases so
does the current in the solenoid and the engine rpm drops off.

i don't expect a linear operation of the engine speed based on load, but i think i can get it close enough to certainly
pull it to full throttle at max amp draw and then reduce to some level at a lower load.

the theory is that it might be more efficient in fuel consumption to not have the engine running at full power all the time, and
not have a large battery bank to have to necessarily charge, ideally i would add the electric start option sent to me by a board
member and have the ability to provide a bit of automation to the system such as autostart.

what i am thinking is this, charging batteries has an efficiency hit of its own, batteries are not 100% efficient, so you have to recharge
by running the engine at some point anyway, so why not run the engine when a predetermined significant load is sensed? this would eliminate
the additional run times required to offset a good part of the battery inefficiency.

a smaller battery bank would likely suffice as well,

there are times of the year in my planned offgrid installation where the need for heat is minimal, and the need for electricity is also minimal
probably 6 months out of the year i only need battery power to cover minimal lighting needs overnight, a bit of tv, puter and little else
the rest of my loads can be scheduled for the most part, save for a well pump and refrigeration that come on whenever they feel the need to do
so. it is those times that the batteries even though small in capacity can provide for starting of the load, and in doing so initiate a startup
of the genset, and the fuel rack solenoid goes to full throttle for max power output, until the load has been serviced and the batteries have been
replenished to the point that the amps drop to a very low level, where the solenoid is down to fuel cutoff.

it might take some work to make the fuel rack current solenoid and get it tweaked into proper operation, but i think it is doable.

thoughts?

bob g

[Henry W]

Hello Bob,

I like the idea! I never thought of using smaller batteries for buffering loads. You got me thinking.

Henry

[rl71459]

Very interesting Bob... I like the concept. Could this approach also minimize the engine run time required by not letting the batteries voltage drop as much as a traditional arrangement?

Rob

[bschwartz]

I like the idea, BUT am concerned that run times might not be enough to get the engine to full temperature.  You could have lots of short start/run cycles.  What implications this would have for engine maintenance I'm not sure.

[mike90045]

I think a larger battery bank, using AGM batteries (longer BULK charge time, higher BULK allowed amps, less adsorb time), would be better, as engines throttle back, they get logarithmically less efficient, keeping up on the higher point of the run curve may be better.  AGM's are less prone to sulphate too, but cost more.

[Henry W]

That is easy to take care of. You can use a 15-60 min. time delay once the batteries are satified. I think a delayed shut down can be controlled in some controllers.

Henry

[mobile_bob]

i think all you guys concerns are valid and could be worked out, some with proper battery sizing to match the end use,
some with getting the governor control solenoid worked out, and the remainder on microcontroller.

probably will spend a bit more time working out the engineering concerns, such as all the sizing aspects before starting assembly.

in the end there may be some issues with run times being too short for proper warm up, and the engine might pay a penalty
in lifespan, i guess one would have to weigh that against whatever is saved in battery first cost (due to amp/hr reduction) and
fuel saved not running to cover the charging inefficiencies of those batteries (at least the goal would be to dramatically reduce
those inefficiencies)

because it uses some of the subsystem control that is incorporated in its bigger brother the 195 based trigenerator
much has already been worked out and can be brought over for use with the 175 unit.

at least that is the idea

bob g

[mobile_bob]

upon more thought, i do have a 170F changfa, which is approx 4.4 hp aircooled
perhaps it might be a better option for intermittent cycle use, it would likely heat up much
faster than the watercooled 175.

i might just go that direction, it is setting on the shop floor and not in a crate... so
ya.. that is the one to go with.

bob g

[BruceM]

Why duplicate a bigger off grid power system on a small scale?  I do see the value for a no-PV or wind situation  where meeting a smaller demand via generator efficiently is the goal.  So I'll assume that's the case.

I  agree with Mike-  AGM batteries fit this scheme well because of the greater charge efficiency, and ability to take a higher charge current without penalty.  They do have to be carefully managed (can't be overcharged or take frequent protracted equalization to make up for mismanagement) , but your specified charge controller should take care of that.   I only wish I could afford them for my 10 battery string.

[mobile_bob]

"Why duplicate a bigger off grid power system on a small scale?  "

well i got a theory on that one 

my bet is the 175 while rated at a higher BSFC in gr/kw/hr than the 195, it will still be better than the 195
at what would be light loads for the 195.

how much difference? i don't know yet,
but plan to find out.

bob g

[rcavictim]

Bob,

You really need another project?   

Actually I like your idea of using the smaller air cooled 170.  As your say it will warm up faster.  If you aren't trying to capture engine heat then the simplicity and reliability factor of not having a water cooling system is also worthwhile.

[mobile_bob]

i am still fence sitting on whether to use the aircooled or go with the watercooled engine

one the had the air cooled is as you say very simple and less to be concerned with, but

it seems just wrong to allow heat to escape without any attempt to harvest it.

so now i am thinking that because i will have a thermal sink for the 195 in place anyway, why not tie
into the same sink? this way the engine will be generally warmer than ambient air which would make for
an easier and cleaner cold weather start, and probably be easier on the engine as well.

the more i think about it the more a micro controller would benefit the project, the ability to have control
over the various aspects is obviously very good, but also having an easy way of timing other aspects would
likely improve both performance and the lifespan of the engine.

for instance if after dark and up to about an hour before waking, the house loads which would be predominately
lighting, some tv, puter and a fridge were carried by the batteries only (the micro would lock out a startup of the enigine
during this time) then at an hour before wakeup the engine is allowed a startup to recharge batteries which would likely
take a good part of an hour, during which time enough hot water could be generated to take a morning shower or
do the dishwashing, maybe both. also other intermittent loads such as the wellpump could also be locked out overnight,
then allowed a startup before wakeup to top things up.

the remainder of the day the micro could allow the engine to startup based on load demand so that the batteries don't take
much of a hit.

being able to alter the programming on the fly in real time would allow one to finely tune the operation and get to a balance
between efficiency and what would seem prudent for engine longevity vs battery life.

and of course having a manual override switch for those times when all else fails or some out of the ordinary operation is needed, such
as when other folks come over for a holiday, under which conditions perhaps the engine runs all day cycling up and down in rpm to provide
for load as needed.

having read about Jens method of tieing both the exhause heat and coolant heat into a single loop, this has made me consider
that it might be the way to go with this project for a water cooled engine. that way the exhaust heat aids in rapid warmup of the engine
and if the coolant system is relatively small (comparatively speaking) rapid warmup could be attained.

removal of the hopper and a custom built replacement that houses a small flatplate exchanger one could then keep the cooling system down
to about a gallon of coolant, and use the flatplate exch. and a small electric pump to regulate engine temperature, by means of heat removal
with the pump controlled by a temp switch/relay circuit.

with a 7lb cap i can assure higher temp operation, with a recovery tank it can expand and then return on cooldown properly
and with the use of a non poisonous coolant coupled to a pressurized domestic water supply if there were a leak the higher pressure
side would likely force fresh water into the engine side, rather than visa versa, and even if not the use of non poisonous should keep down
the risk of killing anyone.

doing more thinking on this project,

the 195 trigen took about 3 years to develop to the testing stage, and will likely take the rest of this year to complete at the rate i get
things done,, so maybe the new project will be done by 2020?

bob g

[Henry W]

Making hot water is good enough of a reason to use a water cooled engine.

Henry

[Ronmar]

Making hot water is good enough of a reason to use a water cooled engine.

Henry

I'll second that!  Engineers fight for fractions of a percent gain in efficiency.  Letting 33% of your potential energy just drift away with the breeze is not very practical when it is so easy to catch it... 

[mobile_bob]

in testing i fight for every percentage point i can find, not quite to fractions yet, but...

actually when i started thinking, the sketching, more thinking, tons of reading, more sketching
wore out a few cheap calculators, and finally at the rather blunt suggestion of a very good friend
that i value the input from very much...

"bob foregodsakes, quit figuring and get out in the shop and build it!!!"

i did just that, then came more questions, followed by testing

i would have never guessed it possible to capture in testing loads as low as a 60watt light bulb
and be able to quantify that load in specific fuel consumption.

thinking back that is where the insanity started, once you are able to measure reliably down to that level
you can then make minute changes and see the effect. along the way things were learned that flew in the face
of popular belief,, and that was really surprising to me.

personally i would never have guessed at what the possibilities are

it used to be 8kwatt/hr produced per gallon of pump diesel was about the limit, now it is universally thought of
as the limit being ~10kwatt/hr per gallon of pump diesel.

my belief now if that the 10 kw/hr barrier can be broken by a significant margin, so...

Ronmar, you are right on with your statement, wherein engineers fight for fractions of a percent in efficiency
much like found pennies they add up.

We should not  think of any increase in efficiency as being marginal, unless it costs us mucho bucks to capture it
if it is there for the taking, we should probably grab it.

getting a healthy bite out of the waste heat goes a long long way toward improvement of overall efficiency.

bob g