Early Numbers

plongson · January 2012

No, not on the Iowa caucus (for us Yankees). Ya know we just got the new home powered up and i'm still in a steep learning curve, but getting there.

I reset the AH meter, MAX and MIN volts on the Magnum remote the other morning at 0900 to see what our 24 hr AH consumption would be without running the genset. What I got was 113AH in about 24 hrs. recorded, with my wife doing a couple loads of wash and running the dishwasher once, evening lights and some TV time. I'm calling that 13.5 kWh, correct?

Upon rising that first morning of the test, the genset kicked on for "VDC RUN" and ran for 1 hour. It did that two days in a row and I'm thinking I like the morning bulk up on the batteries.

What's curious is, I have the AGS VOLTS set to start at 47.2 vdc and off at 58.6. The lowest recorded volts was 47.8 on the remote. Is that normal? Is there a voltage discrepancy from what is displayed vs what is in control? It's not a big deal because it seems like it is working, but it got me thinking.

What do you think of these set points, too conservative or about right?

The only other AGS parameter I have set is the State of Charge (SOC) and it is at 70% to 85%. That parameter never started the genset.

I also finished automating the water system with the Allen Bradley Pico PLC (which performs other duties), a float switch and 1" ASCO solenoid. It refills the cistern about every three days from the well.

Hope I didn't bore you too much, but I'm tickled so far with the performance and wanted to share...

Paul

KeithWHare · January 2012

Re: Early Numbers

Paul,

I don't know if the Magnum amp hour meter is measuring the DC input or the AC output.

113 Amp Hours at 120 volts would be 13.5 Kilo watt hours.

113 Amp Hours at 48 volts would be 5.4 kilo watt hours.

Keith

icarus · January 2012

Re: Early Numbers

I might be a bit of a contrarian, and this might be a bit off topic but I will comment anyway,,,

If I need to run the genny to do some charging I tend to do it at night. In the net/net I have found this to be a bit more efficient. Since the bulk of my loads occur at night, by running the genny at night, it reduces the net draw on the batteries, leading to small net need to charge. For example, if I reduce the load on the batteries by X ah by running the genny at night, that is fewer AH I have to recharge the next day.

As it is, I seldom have to run the genny at all, but when I do, I charge every thing including computer batteries, tool batteries, pump any additional water I can (top up the hot tub etc) so that I off load as much load off the battery as I can.

Just a thought, good luck with your learning curve.

Tony

Cariboocoot · January 2012

Re: Early Numbers

I'd suggest looking in to why the AGS is operating at all, as 47.2 Volts is less than 50% SOC (more or less) on a 48 Volt system. Just my opinion, but the batteries shouldn't be pulled down this low. (Although that is what the gen start is for; to give them a charge if the loads take the batteries down and the panels are unable to charge them up). The SOC parameter should be starting the gen before this low Voltage point, so it sounds like something is not right there. Perhaps the V-set point overrides the SOC-set point?

Most of the time any two Voltage meters will not agree as to readings at any given point in a system. Even the same meter will vary +/- its tolerance and due to other factors. Two different meters reading from two different places will likely never agree. I doubt there's any calibration function of the remote unit to bring it in line.

Check the battery Voltage at the gen start with no load on them and with a load on them. This should give you two different numbers. If you set the gen to start at a no load Voltage of 48 it should be fine (the reading will be lower with a load on, the difference being the Voltage drop under load).

plongson · January 2012

Re: Early Numbers

Hi Tony, I see your point and I believe you are on target as usual. This is what I'm getting from this...

To be fully automated, the genset will always run in the AM because of the preceding night load. Tougher on batteries, because of the added loading and heavy recharge in the morning.

Evening charge is not automated but is easier on batteries for the same inverse reasons.

So it's mostly a be "Kind to the batteries" philosophy. I don't have enough PV or batteries to never run the genset at our current electrical usage. This was known from the beginning because we already owned the big Kubota genset and the additional PV and batteries would have added probably $15k to the house. If genset runs 1 hour a day, that's 1/2 gallon of (red dyed) diesel at $3.50 us, the payoff would come in something like 23 years at current prices, but you see what I'm getting at.

I have no problem running the genset a little bit once a day, I just want to run it at the BEST time of day. It looks like from what you are saying it should be in the PM. What parameters should one use to determine the run time in the evening, or should I just wing it at about an hour...

As always I appreciate your knowledgeable input and great feedback!!

Paul

icarus · January 2012

Re: Early Numbers

Is you gen set auto run prescribed to be AM or is it prescribe as to how low the batteries are? One could certainly change the run time based on clock time, changing it based on battery condition is another matter. If, for example you gen set auto starts in the morning because the AH are low, if you were to run the genny for the same amount of time in the evening, running house loads at the same time (I am assuming that your genny is big enough to do both) thenfore the same run time, you are in essence getting more bang for you 1/2 gallon of diesel.

That said, the reason that many of my clients have gone to PV is so they don't have to run a genny in the evening so they don't have to hear it. It doesn't seem to matter as much in the daytime to most folks. The great thing about the small Hondas is that they are so quiet, it really doesn't impact lifestyle much.

As with anything, good load calcs are important,, and to the point, when those loads occur. It used to be that nearly 100% of my load was at night, because all we used were a few lights and the radio. As is wont to do,, the loads have grown with time,, and the availability of 24/7 AC power. We now keep the modem lit all day, run lap tops, sat radio, ipods, internet radio etc. Additionally we now pump water with the PV instead of the gas powered gravity pump. I try to shift as many of my loads to times of good sun as I can. For example, I kill the pump after the sun goes down (unless I need more water) since there is usually enough water to get through till morning (doing little things like filling the coffee pot in the late afternoon for the morning, filling a pitcher of drinking water etc), then I pump water once the sun comes up. Same with the lap top batteries. While the modem may be on all evening, I will draw the laptop batteries down, and recharge them in the morning. By these simple load shifting regimens I am able to reduce the daily discharge of the battery to ~10% over night. Of course this strategy means that fewer amps are available to charge, but fewer (significantly fewer) AH are required to re charge the bank. Not counting the extra ~20% required to put AH in vs AH out, just in the matter of the pump, I can run the pump during the day for the few minutes I need with no draw on the bank at all,, giving me a +20% increase in overall efficiency.

Just some thoughts,

Tony

Cariboocoot · January 2012

Re: Early Numbers

In figuring out what is the best time to run you have to think about a few things regarding your system and how it is used.

1). Gensets are not efficient for light charging. They are best for bulk charging, preferably when running additional loads (making the most use of the generator's capacity).
2). Batteries do not like to sit discharged, even for a few hours. This usually means have them fully charged before you go to bed because loads are generally lightest when everyone is asleep.

This is why Tony suggested running the generator at night; finish up the day's work without eating into the power stored in batteries, leaving them charged 'til morning.

But if you have a condition where there's inevitable significant overnight use (furnace blower or such) and there may be some delay before panels start providing charge (due to installation restrictions) or the panels are "a bit on the lean side" you may want to run the gen in the morning to handle initial loads (electric coffee maker perhaps? :grr ) and get the bulk charging done. In that case you need to balance the generator shut-off Voltage with when the panels can take over the charge, and that may need to be seasonally adjusted.

And of course you need to be able to have the gen start when the panels aren't capable of supplying any practical amount of charging due to inclement weather.

This is one reason why I don't like auto gen start; lack of flexibility. But if the system must be functional unattended there's no choice but to rely on the automation.

P.S.: One reason why I like Outback systems; a large amount of programmable variables including in controlling the AGS. It can have mandatory start/stop times, forced "quiet times", day of the week programming, and Voltage set points all working together. It's not easy to figure out, but the functions are there.

plongson · January 2012

Re: Early Numbers

Cariboocoot wrote: »

I'd suggest looking in to why the AGS is operating at all, as 47.2 Volts is less than 50% SOC (more or less) on a 48 Volt system. Just my opinion, but the batteries shouldn't be pulled down this low. (Although that is what the gen start is for; to give them a charge if the loads take the batteries down and the panels are unable to charge them up). The SOC parameter should be starting the gen before this low Voltage point, so it sounds like something is not right there. Perhaps the V-set point overrides the SOC-set point?

Most of the time any two Voltage meters will not agree as to readings at any given point in a system. Even the same meter will vary +/- its tolerance and due to other factors. Two different meters reading from two different places will likely never agree. I doubt there's any calibration function of the remote unit to bring it in line.

Check the battery Voltage at the gen start with no load on them and with a load on them. This should give you two different numbers. If you set the gen to start at a no load Voltage of 48 it should be fine (the reading will be lower with a load on, the difference being the Voltage drop under load).

Hi 'Coot,
The whole AGS thing confuses me anyway. I've set it to start on AMPS and never got it to start at all. I called Magnum and they were confounded as well. I'm wondering if it works best with just ONE parameter on and the rest off, like if it has too many "IF's and "AND's" in the logic before a "THEN" can be obtained. Maybe I'll turn off the VDC RUN and try just SOC.

If I stay with a voltage setpoint what is a good number? 75% is about 49.6vdc

Checking the SOC when the VDC was at +/- 47.xx it was at 83%.

Curious indeed.

Paul

Cariboocoot · January 2012

Re: Early Numbers

Paul;

I've never set up a Magnum system, but you may be right about the "pick one criterion for gen start" limitation.

What are you using to determine SOC of the battery bank? All the battery monitors I know of have to be programmed with data at the start to reference as "fully charged" and if you get one parameter wrong they are inaccurate from then on out.

After salvaging a few systems, I've come to the conclusion that the safest no load battery low Voltage limit is the nominal rating. Under load they will go below this, but it's usually a sure-fire way to be certain SOC doesn't drop below 50% (at least not for any significant period of time). In your case that would mean the gen would be running more, which is why I'm a bit concerned about what is pulling the battery Voltage down. A no-load battery Voltage of 47 corresponds to a SOC of about 30%, and that's no good.

plongson · January 2012

Re: Early Numbers

'Coot,
This is what I was going by to set the SOC. From the magnum handbook...

02 BM: SOC – This display is the best way to monitor the actual state of
the battery. The read only display shows the State Of Charge (SOC) for the
connected battery bank. The SOC represents the condition of the battery as
a percentage of the available capacity left in the battery. The range is 0% to
100%, where 100% represents a fully charged battery and 0% means the
battery is completely discharged.
When the Sense Module is ﬁ rst connected, the batteries will need to be fully
charged (i.e., SOC = 100%) at least once to establish a SOC reference point.
While this reference point is being calculated, the display will show “Think’n”,
to indicate that the SOC reference point is being calculated. After the batteries are fully charged, the display will change from “Think’n” to “100%” and
begin to provide accurate SOC percentage values.
Info: “Think’n” can take up to several days depending on the charge
source and its ability to completely recharge the batteries (e.g., a
small PV array may not be large enough to replace all the amp-hrs
in a reasonable amount of time).
If the Sense Module is disconnected from power, this display resets to “Think’n”
and the batteries will require another full charge before this display begins
providing SOC percentage information.
Info: Try to limit battery discharging to 50% of capacity (keep
battery above 50% SOC). If batteries are allowed to be continually
discharged below 50% SOC, their effective service life will be considerably shorter. This 50% rule has been determined to be the best
compromise between available energy and the maximum number of
discharge cycles a battery can provide

Cariboocoot · January 2012

Re: Early Numbers

This function would be similar to MidNite's battery gauge: http://www.solar-electric.com/mnbcm.html
Whereas it is probably more accurate than Voltage alone, it is not as accurate as a battery monitor such as the Trimetric or Victron. I wouldn't count on it, especially as there appears to be a significant discrepancy between what it reads and what the Voltage-based SOC would be. Again, Voltage alone is not that accurate (especially under load) but we would expect these two numbers to be closer than what looks like a 50% difference in SOC between them!

The part where it says it may take several days to get an accurate 100% SOC reading depending on the charge source may well be part of the problem. I'd ignore it for now and rely on standard Voltage readings, especially for full charge. With no temp compensation, 57.6 would be Absorb set point. For the batteries to be truly fully charged they'll need to remain at that Voltage for a while. How long is going to be a question of battery bank size and how deeply they were discharged. Frankly you can't beat charging from panels with a controller. Maybe if you disconnected the gen and just recharged via solar for awhile the Magnum's self-programming SOC indicator would come up with more accurate readings. Once those are established you could program the gen to start based on SOC and it might work.

plongson · January 2012

Re: Early Numbers

Thanks 'Coot,
None of this has been too easy, why start now...LOL

I just don't want to trash these batteries, and dang if the whole system seems to working against me some times. Sure would be nice if these systems could be plug and play.

Cariboocoot · January 2012

Re: Early Numbers

It's an unfortunate fact that they aren't. Everything is stacked against it!
We start with generalities and rules-of-thumb and then refine the calculations but in the end ...
Every single install is site-specific and therefor has to be unique. Despite best efforts at planning you never really know how it will work until it's in use. Sometimes they work better than planned, but usually the loads are higher the insolation lower and the efficiency not up to snuff. If you expect that to begin with and err on the side of caution you may do okay.

See? That's one o' them "generalities" I was talking about.

plongson · January 2012

Re: Early Numbers

'Coot,
Ya just keep me thinking...LOL saying "you cant beat a controller to charge batteries"

When I look at the Midnite 250 after a daily scenario like we've been talking about, first thing in the morning it goes to bulk MPPT but later in the day it goes to float MPPT. I can't remember the voltage on the controller right now but it would seem the batteries are getting the necessary daily charge, and this is in the dead of winter. If these things will fly now, summer should be a breeze.

I'm going to change my charge regiment to the evening for a while and watch it. I'm also going to bump up the gen start voltage parameter to 49.6 (75%).

I'd like to get in a (good) routine with these things and let them do their thing. I have lots of other things to do other than hover over batteries like an old hen. But I need to get it right!

Early Numbers

Comments

Categories