Autumn in Southern Oregon

[mbryner]

Ahhh, the heat of summer is finally over!   Rain for the past 2 days.  Temps not even up to 60F in the daytime.  The maple trees along the creek are starting to turn yellow.   Just finished canning 89 quarts of applesauce w/ my wife last week, and now she's working on canning a few more boxes of pears.

But that means not enough solar insolation, and time to start burning diesel for at least a few hours/day.

Yesterday, I ran the 'roid for 24+ kwh worth.   When all is said and done, about 1800 watts are going into the charge controller when charging w/ the Listeroid.  There's efficiency losses from the gen head, transformer, choke, rectifier, and I'm running an A/C fan over the radiator, so I might be in the 2500 watt range.   Respectable for a 6/1.   If I run much more watts into the Outback MX80 than that, the thermosiphon cooling will start to boil over but the engine can handle the load -- Looks like it needs a little circulator pump.   

A question was brought to mind with the talk of large battery banks and small DC alternator charging them on another thread (dudes, you were very gentle w/ the guy.  His aggressiveness and arrogance reminded me of the guy who got banned about 1 year ago).    I have 16 x Deka L-16's at 48 Volt nominal.   My only backup source for charging in wintertime is the 6/1.   Is that enough?   Am I going to damage the batteries w/ too low of charge rate?

Marcus

[LowGear]

Home made applesauce.  Wow!

I thought they left just the right amount of rope in the room.

Are you totally off grid?

Casey

[mike90045]

  I have 16 x Deka L-16's at 48 Volt nominal.   My only backup source for charging in wintertime is the 6/1.   Is that enough?   Am I going to damage the batteries w/ too low of charge rate?   

The batteries need to get some vigorous bubbling once in a while, to stir the electrolyte up to keep it from stratifying. This can sometimes happen in a long absorb cycle, or more often in the EQ cycle.  Perhaps use the 6/1 for a bulk charge in the AM on a forecast sunny day, let the solar top it off, and then while you still have some solar, start an EQ cycle for 30 minutes or so, with the 6/1 running.

That's 2, parallel banks of batteries right, about 800AH @ 48V ?

[mbryner]

@Casey:  ...and he used it.     Yep, totally off-grid.

@Mike:  Yes, 2 parallel strings.  I just never know how much or how often to equalize.   All summer long, the batt bank never got below 80%, and most days not below 85-90%.   Now that we got rainy weather, the Listeroid takes forever to get them back up to Full.    2 days ago when I came home late in the evening from work, my wife had drained the batts to 25% on the Trimetric / 60% on the Outback Mate with cooking and cleaning and canning and kids watching some TV.   Of course, that was w/ a few days worth of use and poor sunshine.    I don't know why the Trimetric and Outback Mate are so divergent in their readings.

[Lloyd]

@Casey:  ...and he used it.     Yep, totally off-grid.

@Mike:  Yes, 2 parallel strings.  I just never know how much or how often to equalize.   All summer long, the batt bank never got below 80%, and most days not below 85-90%.   Now that we got rainy weather, the Listeroid takes forever to get them back up to Full.    2 days ago when I came home late in the evening from work, my wife had drained the batts to 25% on the Trimetric / 60% on the Outback Mate with cooking and cleaning and canning and kids watching some TV.   Of course, that was w/ a few days worth of use and poor sunshine.    I don't know why the Trimetric and Outback Mate are so divergent in their readings.

Marcus,

Do you have the shunt module for the Outback? The Triemetric will show only a gustimated SOC. If you have the shunt module for the outback it will be the most accurate for the SOC, and the real amps being used and charged.

Because an equalize charge is a controlled overcharge it will add to the positive grid plate corrosion. The best way to determine the need of equalize charge is;

3.4.2. Equalizing (flooded/wet batteries ONLY)
Equalizing is an overcharge performed on flooded/wet batteries after they have been fully
charged. Trojan recommends equalizing only when batteries have low specific gravity, below
1.250 or wide ranging specific gravity, 0.030, after fully charging a battery. Gel or AGM
batteries should never be equalized.
• Confirm that the batteries are flooded/wet
• Check electrolyte level to make sure plates are covered with water before charging
• Check that all vent caps are secured properly on the battery before charging
• Set charger to equalizing mode
• The batteries will gas (bubble) during the equalization process
• Measure the specific gravity every hour. Discontinue the equalization charge when the
gravity no longer rises

As far as charging in general, using the top 20% seems to be a good regime with AGM's.

I say if you use that regime with FLA's especially L-16's let the bank draw down to 50% once a month, then recharge the bulk cycle with a current equal to at least 20% of the amp hr size of the bank. L-16's need a lot of current during the bulk cycle to tickle it's toes.

Running to long of an absorption cycle, at to high of a current/voltage on shallow cycle is basically the same thing as an overcharge. edit and won't really tickle the toe's of an L-16.

The best way to determine the length of the absorption cycle is to draw the bank down to 50%, disconnect any loads and start bulk charge, watch the bank like a hawk and measure the time it takes reach the absorption cycle, then set your absorption timer to run about 110% of that time.

This is where having a charge source large enough comes seriously into play, first it tickles the toes of the bats without overcharging. Second when the bank is in bulk and absorption cycle you need enough current, to maintain the cycles plu any additional loads that come from the normal system use.

If you don't have a large enough charge source, I would break that bank in 2 for the once a month 50% discharge then run the regime.

Lloyd

[mbryner]

Do you have the shunt module for the Outback?

Yes, I have the FlexnetDC module and 3 shunts: 1 for input from the solar/ST-roid MX80 charge controller, 1 for output to the inverters, and 1 for total I/O to batts.   Wait, I have to double check that later...

disconnect any loads and start bulk charge

The hard part is disconnecting the loads for equalization, or running longer calibration tests, etc.   How do you tell the family "the power is going down now for hours, go light some candles".   That's what the system is for. 

[Lloyd]

@Mike:  Yes, 2 parallel strings.  All summer long, the batt bank never got below 80%, and most days not below 85-90%.  

One note when running L-16s in this type of cycle. Using only the proverbial top 20% of the bank, you're actually using the bottom of the L-16 plates to provide that 20%.

Now if you have short to medium bursts of amp loads that exceed the Pukert rate of the bank, then your depleting the lead anatomy at the bottom of the plates. Under these loads a large bank using only 20% day in and day out will stratify even worse, and eventually will sulfate. As the large Pukert Discharges will use the acid at the bottom without being able to replenish. Water being heaver, is what cause the stratification, that's why you have to tickle the toes once in a while.

lloyd

[Lloyd]

Do you have the shunt module for the Outback?

Yes, I have the FlexnetDC module and 3 shunts: 1 for input from the solar/ST-roid MX80 charge controller, 1 for output to the inverters, and 1 for total I/O to batts.   Wait, I have to double check that later...

disconnect any loads and start bulk charge

The hard part is disconnecting the loads for equalization, or running longer calibration tests, etc.   How do you tell the family "the power is going down now for hours, go light some candles".   That's what the system is for.  

Your outback will be accurate and this is the poof in the pudding that the Triemetric is only a guesstmate.

That's why I suggested breaking the bank, you can use half the bank to supply the loads, and draw it to 50%, then switch banks, and properly charge the previous bank. During summer that's only a once a month event. In winter if you don't have a large enough charge source then it may be more often.

lloyd

[Lloyd]

Do you have the shunt module for the Outback?

Yes, I have the FlexnetDC module and 3 shunts: 1 for input from the solar/ST-roid MX80 charge controller, 1 for output to the inverters, and 1 for total I/O to batts.   Wait, I have to double check that later...

disconnect any loads and start bulk charge

The hard part is disconnecting the loads for equalization, or running longer calibration tests, etc.   How do you tell the family "the power is going down now for hours, go light some candles".   That's what the system is for. 

The calibration test only needs to be ran once in the beginning to establish a base line, make sure you also establish a baseline for the SPG of the cells. The as the battery bank ages or shows signs of becoming sulfated you need to run another calibration. Maybe two or three times over the life of the bank.

Taking SPG at the end of the full charge cycle should be a once a month ritual, keep a log and this will advise when the bank needs an equalize. A properly built and charged bank should only need to EQ'd once a year.

It's tough to be off-grid...but when that's you're only choice, you can choose to manage a bank, or just buy a bank more often. Everything in life seems to be a trade off.

lloyd

[mbryner]

OK, thanks Lloyd.   What do I do if we don't have the baseline SPG?   Just equalize at the recommended SPG, no?   I'm bad, I haven't been taking SPG measurements (ever).   I do check the water levels periodically and never run too low, and I equalized once for a few hours this summer.

you can choose to manage a bank, or just buy a bank more often.

Of course you are correct....

[Lloyd]

OK, thanks Lloyd.   What do I do if we don't have the baseline SPG?   Just equalize at the recommended SPG, no?   I'm bad, I haven't been taking SPG measurements (ever).   I do check the water levels periodically and never run too low, and I equalized once for a few hours this summer.

you can choose to manage a bank, or just buy a bank more often.

Of course you are correct....

If you didn't establish a base line when you commissioned the bank, do it at your next opportunity, sooner than later.

A large FLA bank is a problem that most Solar Dealers never want to address. You'll also note that most dealer are promoting grid-tie over bat banks.

In your installation AGM's may have been a better choice, because they don't stratify, as badly, that's why the Telco's use them, bc they follow the 20% discharge/charge regime, with a charge source large enough.

AGM's actually benefit from a larger charge source then FLA's.

In your case, your fighting a double edged sword, to small of charge source. So what everyone of the Solar dealers I see propose, is regular EQ"s. The trade off is if you don't EQ  often the bank dies early from sulfation, and if you do it dies early from grid plat corrosion.

The only remedy is being able to be able to meet the charge requirements, in you case you need to find a way to boost your charge capabilities. Or an efficient way to break the bank for the proper charge.

Or just expect shorter lifetime from the bank.

Sorry for the bad news, but I hope it helps you find some balance, that you can live with.

lloyd

[Ronmar]

When all is said and done, about 1800 watts are going into the charge controller when charging w/ the Listeroid.

If I am reading this right, 1800W at say a 54V charge voltage is around 33A into the batteries?  Is that about right?  That is a big drop from the 2500W out of the generator head.  You would probably benefit from a direct controlled DC alt running your charge current into the batteries directly...

If I run much more watts into the Outback MX80 than that, the thermosiphon cooling will start to boil over but the engine can handle the load -- Looks like it needs a little circulator pump...

Do you have a link to a pic of your current engine cooling system?  maybe we can make some suggestions...  Aren't you putting the cooling system heat into your storage tank?  

If I was offgrid, I think i would want 2 gensets...  The good thing about 2 sets, besides having a spare, you could run alts on both and combine their output to meet those occasional high current charge requirements.  Do you have a tractor with a PTO?  If so, you could also make up a generator head for the tractor to provide for occasional surge requirements...

[mbryner]

Hi Ron,

1.  Yes, in fact I set up current limiting to 35A through the Outback charge controller so the engine won't boil over.   The 2500W number is just a guesstimate -- I'd have to put a clamp around the cables coming out of the ST, and I don't have that kind of current measuring device.   Where do I get a 48 V DC alternator that doesn't break the bank?   Or how to mod a decent 24 V to do 48 V?   Then I could forget my complicated ST head --> isolation transformer --> rectifier --> charge controller setup.   But having the ST head was my backup in case the inverters fail.   Then again, I do have the pull start generator that could easily be hotwired in.

2. Here's a pic of the engine in the powerhouse (old before sheetrock, etc.).   The red heater hose goes to an old radiator mounted to the outside wall, w/ the regular white bathroom vent fan blowing across it (hooked up to ST output).   Nope, it's not going into the heat storage tank.   I wish, but it's about 150 ft to the house where the tank is.



3.  Yes, I have a pull start jobsite generator for emergencies.   And I do have a 45 hp tractor & I could buy a PTO gen.

[Tom Reed]

Marcus,
It looks like you are close on the C/20 charge rate Lloyd recommended. That would be 40 amps and you are at 35. With the assist from solar, even when cloudy, you've got to be there, depending on what you're draw is.

I'm running a similar 6/1 & ST5 to charge my 1055 ah 48vdc battery bank. Charging through the Outback inverters I can consistently put 45 amps into the bank, 50 with the output from the panels. It seems like your system with the direct charge is a bit less efficient than what I'm running. I to am looking for the best way to direct DC charge the batts. To bad there is no Balmer like controller that would control the field on a ST-5 with each of the 4 poles wired through a bridge rectifier. It seems like that would be the most efficient way to go.

Just thinking out loud here, I wonder if a voltage divider on the input of a Balmer controller would work to trick it into outputting 48 volts. IIRC Bob discussed tapping the bank in the middle for a 24v input, but my concern here is unbalancing the bank due to the load from the controller.

Lloyd,
Dude, I guess I'll have to confess to arrogantly assuming I knew how to maintain a battery bank. I'll now make a humble bow to your hard earned knowledge. I'm in the 10-20% daily discharge range on my bank too. And have been doing a monthly EQ as recommended in the battery (Hawker PV-1) manual. In fact the manual says to equalize if a cell has a difference of .020 specific gravity or at least once a month. Do your annual EQ suggestion seems to conflict with the manual. Comments?

Under my charge regime I bulk to 59.2 and absorb for 3 hours. The batts do gurgle during absorption and need water about every 2 months (1 gal per batt) The EQ is to 61v for 3 hours. Comments?

[Ronmar]

Marcus
   I looked over your website and video, do you have any pics of the outside of the powerhouse showing the radiator so I can get my brain around your whole thermosiphon flow?  Can you recall what vehicle the radiator was out of?  Also where is your thermostat located and what model/type are you using? 

I have 2 relatively minor suggestions from the pic and the ones on your lister 1 webpage, but with thermosiphon, every little bit adds up to total flow volume.  First off, avoid horizontal like the plague.  Vertical adds to the flow energy, and horizontal detracts from the flow

1.  On top of that 90 degree street elbow comming out of the head, I would re-orient the "T" so the flow is straight up thru the main passage of the "T" with the sensor screwed into the branch leg of the "T" This will take an unnecessary 90 degree turn out of your flow path.  I would orient the whole affair so it points as directly as possible at the point on the wall where the upper hose heads for the top of the radiator.

2.   I would loose the section of corrigated radiator hose at the bottom.  Corrigation = turbulence.  From a flow resistance/drag perspective, that 18 or so inches of pipe seems like 3+' to the water flowing thru it.  A good part of it's length is also horizontal.  I would run a single smooth piece of hose as steeply as possible between radiator and return port at the cylinder base