FNDC and SimpliPhi batteries

Discussion about OutBack Inverters in Off Grid Applications

Moderators: OutBack Moderator Team, OutBack

Mauricio
Forum Virtuoso
Posts: 78
Joined: Fri Apr 16, 2010 2:49 pm
My RE system: 24 LG275, 275W panels, 2 FlexWare500, four VFX3648, series/parallel Quad system, two FlexMax80, X240 transformer, MATE3, FN-DC, 6 SimpliPhi 48V 3.5kWh Lithium Ferro Phosphate (LFP) batteries, Cummins Onan 20GGMA generator (20 kW, LP Vapor fueled), OpticsRe.
Note: 12 Surrette Rolls 4Ks25P Batteries ( 4V, 1350 AmpHr), were replaced by the 6 Simpliphi PHI 3.5 on June 21/19.
Location: Todos Santos, Baja California Sur, Mexico
Contact:

FNDC and SimpliPhi batteries

Post by Mauricio » Fri Jul 12, 2019 8:25 am

On June 21/2019, I have retired my old Rolls 4KS25 and replaced them with 6 new SimpliPhi 48V 3.5 kWh, Lithium Ferro Phosphate (LFP) batteries. The transition went smoothly, and thanks to Outback/SimpliPhi integration guide, denoting all the required settings for Inverter, Charger, Charge Controller, FLEXnetDC and MATE3, the new batteries are performing very well.
My issue is with FNDC. While I think the unit is working properly and as intended by its designers, I believe that FNDC as it is now, it is not "tuned" for these type of batteries.
Compared to the Rolls, the voltage discharge cycle for the PHI 3.5 is very flat and lingers above the 50 Volts until the sudden drop at 49.5 Volts, so, SOC% from the FNDC and Battery Voltage do not agree with battery status I would have to increase the battery capacity setting 100 folds to have the two measures agreeing with each other. Not a solution.
Here is a real condition: From sunset to next sunrise, 13 hours, I consume about 12 kWh (real consumption), FNDC indicates SOC at about 79%, a bit low, but close enough considering my setting for battery capacity (Please note that FNDC was set with fully charged battery), but real battery voltage is still measuring 53.2 Volts. As per SimpliPhi technical support: ÔÇ£Anything over 53V is roughly a fully charged batteryÔÇØ.
So, after burning 12 kWh, I find myself still, with a set of fully charged battery. Free power? No, I think that since at sunset my battery voltage measured 54.5 Volts, the 1.3 Volts drop explains the consumption. Operating at such high levels of SOC, this discrepancy is not a real concern, the problem will be when, in need of more battery power (days of bad weather), recognize when the time has come to start that generator.
Comments from users that have also install these type of batteries will be greatly appreciated. Thank you, Mauricio.
Mauricio - Todos Santos, BCS
http://www.arribadelaroca.com

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Thu Aug 15, 2019 8:34 am

Welcome to the group of (hopefully happy) Simpliphi owners. I started off with 3 batteries and now have 6, so approx 20 kWh. I am very happy with them so far. I find the SOC reported by FNDC pretty accurate as it is based on amp hours, not voltage. As you state, the voltage is hard, but not impossible to correlate to SOC. If you had lead acid batteries, it is probably hard to forget about voltage and pretty much ignore it. At least until it gets low like < 50.5v and starts dropping really fast.

Here's my observations and settings. I absorb at 56v for 6 minutes and then float at 54v. I find the temp. compensation gets in the way, so I use the limited setting at 54 and 56v and then fudge the charge controller absorb and float voltages to 56.4v and 53.5v. The temp comp limit keeps the target voltages within [54,56]. Without this I find that sometimes the temp comp messes with the target voltage which isn’t necessary in the 50-90 degree F range with LiFP.

After the PV stops outputting , my voltage drops to around 53.5v with a small load, e.g. 300-500 watts. The voltage drops down to 52.5v at about 50% SOC. Remember the Simpliphi chart is under a C/2 load or 10kw load for 6 batteries! I find that if I turn on a toaster and microwave, creating a 2000 to 3000 watt load I can lower my battery voltage by .5v to .8v, so I believe their chart is fairly accurate. Regardless, the voltage drop is so minimal and load sensitive that I pretty much ignore it and focus on amp hours which is what SOC reports. It is important to have the charge factor set right for FNDC, I use 96%.

I am curious what you have you battery capacity set to, I have mine set to 400ah. If you used say 12kwh overnight out of a total of 20, you should have been around 40% SOC with 8 kWh remaining, not 79% which is way too high, or maybe you meant 79% depleted and 21% remaining? We have a 7200 PV array and usually the batteries reach 100% and then drop to about 50% overnight. Once in a while with bad weather the SOC drops to 20% and the voltage seems reasonable per the chart (again with a small load). I only run my generator a few hours every few months. We average 12-20 kWh per day , it used to be a lot less but now we are spoiled with more panels, lots of sun, and great batteries.

Mauricio
Forum Virtuoso
Posts: 78
Joined: Fri Apr 16, 2010 2:49 pm
My RE system: 24 LG275, 275W panels, 2 FlexWare500, four VFX3648, series/parallel Quad system, two FlexMax80, X240 transformer, MATE3, FN-DC, 6 SimpliPhi 48V 3.5kWh Lithium Ferro Phosphate (LFP) batteries, Cummins Onan 20GGMA generator (20 kW, LP Vapor fueled), OpticsRe.
Note: 12 Surrette Rolls 4Ks25P Batteries ( 4V, 1350 AmpHr), were replaced by the 6 Simpliphi PHI 3.5 on June 21/19.
Location: Todos Santos, Baja California Sur, Mexico
Contact:

Re: FNDC and SimpliPhi batteries

Post by Mauricio » Fri Nov 22, 2019 3:48 pm

Helo gtarolli, My system is running well and I’m very satisfy with the Phi 3.5. Having used flooded batteries for ten years, I still cannot believe the difference, is like being connected to an imaginary grid. They, (the batteries) have been working for almost six month now and my only concern is to go and see if they are still there. I have the old inverters, so I am using the FLEXnet DC relay to control inverters ON/OFF in case of a fast discharging conditions, as detailed in Outback Application Note.
After 6 months the only thing that does not work for me, is the battery SOC. FN-DC was designed for standard batteries, and the FN-DC Battery Ah setting is based on the battery’s Ah @ C20 discharge; PHI rated capacity are at C/2 and their discharge curve is flat. I cannot correlate SOC reading from the MATE to the true SOC of the battery. The best example is this: Over an overnight 13 hrs. period, with no devices charging the batteries, my average total consumption is about 10 kWh. I start at sunset with a battery voltage of 54VDC, and MATE SOC at 100%. Just before sunrise battery voltage has dropped to 52.8 VDC with MATE SOG at 55%. Now the MATE calculations are correct according to my the FN-DC battery Ah setting, 10 kWh consumption represent about 45% discharge, but battery voltage measures 52.8 VDC, and, according to PHI battery SOC voltages, my batteries are still at 100%. (Free energy???)
I have discussed the problem with Outback support, their answer: disregard SOC.
Mauricio - Todos Santos, BCS
http://www.arribadelaroca.com

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Fri Nov 22, 2019 4:27 pm

Glad you like them, yes they truly are maintenance free! Just an occasional dusting off is all they need.

SOC works fine for me, it basically is always 1 + Net Battery Ah/Battery Bank Size. i.e. what percentage of my Ah are remaining. Note that Net Battery Ah is negative. From your example it looks like it is working. Note that the PHI battery chart is for C/2 discharge, i.e. that is the voltage when your are drawing 10.5KW !!! A load like that could drop the voltate by 1.0v That is why battery voltage is hard to use to measure SOC as there is so little change in voltage based on SOC, but a large change based on load. I think they have a C/10 curve for the new batteries (3.8) , it would be nice to have one for the older batteries as well. My experience is 52.8 is around 70% with a minimal load, but .1v makes a big difference in this range, so battery calibration can easily make a 10-20% difference in trying to figure SOC from battery voltage.

My experience is this - use the SOC reported by FNDC and don't use battery voltage , except watch it if it goes below 51v as it will drop suddenly. Also you can easily create 4kw loads with a microwave plus toaster or something like that and observe the voltage drop, but this is only a C/5 load. And make sure your charge factor is 96-98% as that is important also. I don't understand why Outback says don't use SOC, I would say don't use the PHI charts unless your current load matches the chart.

Mauricio
Forum Virtuoso
Posts: 78
Joined: Fri Apr 16, 2010 2:49 pm
My RE system: 24 LG275, 275W panels, 2 FlexWare500, four VFX3648, series/parallel Quad system, two FlexMax80, X240 transformer, MATE3, FN-DC, 6 SimpliPhi 48V 3.5kWh Lithium Ferro Phosphate (LFP) batteries, Cummins Onan 20GGMA generator (20 kW, LP Vapor fueled), OpticsRe.
Note: 12 Surrette Rolls 4Ks25P Batteries ( 4V, 1350 AmpHr), were replaced by the 6 Simpliphi PHI 3.5 on June 21/19.
Location: Todos Santos, Baja California Sur, Mexico
Contact:

Re: FNDC and SimpliPhi batteries

Post by Mauricio » Sat Nov 23, 2019 9:16 am

Hello gtarolli, Doing the math, the SOC reported by FN-DC is correct. I had never disputed that. Last night for instance, just to increase consumption, my wife and I spent time in a hot tub with a 220 VAC, 2 HP pump motor running continuously, (I thought creating a load this way being better than a microwave + toaster), my total overnight burn was 13.82 kWh. SOC displayed 36%, again doing the math FN-DC, showed correctly. What I cannot wrap my head around is this: Pure battery voltage, no loads nor chargers attached, remained at 52.8 VDC, where I would have expected it to be nearing the 51 VDC. According to SimpliPhy, Voltage vs SOC chart, and as told by their technical support person, “anything above 52.4 VDC, can be considered fully charged”. Thus, the big discrepancy in my head.
There are no doubts that I removed the 13 kWh from the batteries, and knowing I cannot get things from nothing I explain this discrepancy with the fact that I start with a fully charged bank voltage of 54.4 VDC, so there is a drop in voltage that explains the consumption, but no one at OB support nor PHI support seams to zero in, care to discuss or acknowledge that as the reasons of my dilemma.
Then of course my reasoning is completely …up the creek. ](*,)
Mauricio - Todos Santos, BCS
http://www.arribadelaroca.com

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Sat Nov 23, 2019 10:05 am

I checked my history and I observe 51.8v at 36% with minimal loads, under 500w. Your reading of 52.8 seems really high, for me that is around 70%. Between 52-53v (with small loads) there's a lot of SOC so voltage calibration matters and .1v matters. 52v for me is 40% and 53v is 80% , again with small loads under 1000 watts.

Since voltage can vary .8v by varying the load and SOC is constant, it is hard to correlate the two without factoring in the load. Note charging can increase voltage by the same amount.

raysun
Forum Emperor
Posts: 1028
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower One: FX3048T, (2) FM80, MATE3s, FlexNetDC
Outback 200NC batteries (8 @ 48v)
Outback IBR3 battery enclosure
Suniva 330 watt panels (12 - 6 Strings @ 2 Series)
Hyundai 355 watt panels (6 - 3 strings @ 2 series)

Re: FNDC and SimpliPhi batteries

Post by raysun » Sat Nov 23, 2019 10:12 am

There's a simple answer to clear up your confusion - stop looking at voltage to derive state of charge. It works with uncertainty for lead acid batteries under load, and it works with even more uncertainty for lithium batteries in any state, except two: when fully charged, and when fully discharged

The claim: "Anything above 52.4V, can be considered fully charged", is just as valid as "52.4V can be considered 20% discharged @C5 rate". Both of those statements are derived directly from SimpliPhi's published specifications.

Using voltage as the determinate is the wrong way to go. With a lead acid battery, voltage and SOC only correlate under a very restricted set of conditions: battery at rest for at least 24 hours, no load, specific temperature. With Lithium ion batteries a voltage/SOC correlation is much more difficult even under those conditions. Looks like it would take more elaborate measurement gear than most folks have.

The little "head banging wall" emoji is appropriate here. Stop banging wall with head and pain goes away.

Oh, and there's a trick to getting through a wall without pain - use a doorway. So what"s the doorway here? Coulomb Counting.

No matter the technology, every battery is the same thing: a chemical bucket to hold electrons. If one knows the size of the bucket, knows how efficiently it can be filled and emptied, and starts with either a completely full or completely empty bucket, one can derive the State of Full by carefully measuring fluid in and fluid out. The fluid in this case is electrons. The measuring device is the FNDC. It's not a perfect method, estimates must be made to define the size and efficiency of the bucket, and the discharge algorithm defined by Peukert's Law is based on lead acid, not lithium, but the Coulomb Counting method is leaps and bounds more accurate than voltage for deriving State of Charge.

Looking at several posts where SimpliPhi battery users report FNDC as tracking SOC to reasonable accuracy leads one to believe this is a reasonable monitoring approach. Wired into the system and configured correctly, the FNDC seems to do a proper job.

Another consideration is how accurately the SOC needs to be tracked in order to manage the battery effectively. With lead acid, SOC is extremely significant to battery longevity and excursions 'over the line' can have material impact, as can holding the battery in a partial state of charge. It would appear lithium batteries are far more tolerant to deep discharges. Not immune to them, mind you, but more tolerant. If minimum SOC is kept within the ballpark of the desired target, Lithium claims a reasonable service life.

Since a lithium battery should not be held at 100% SOC, and isn't useful at 0% SOC, keeping it in the 100%>SOC>[Pick your desired minimum]% partial state of charge seems a much healthier for the battery and an easier task than the SOC maintenance for lead acid.
Last edited by raysun on Sat Nov 23, 2019 1:15 pm, edited 4 times in total.

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Sat Nov 23, 2019 10:42 am

Very well said raysun! and I agree. As I said you need to use both voltage and net load to compute SOC, which is not easy, and even then you are better off simply using net amps I believe. I will reiterate that the Charge Factor is important as that derates the amps that go into the battery for these computations. So with age, you might need to decrease that a percent or two. I have mine set at 96% to be a bit conservative, they recommend 98%.

While I normally don't ever approach 80% DOD, at times I do go down to 80% and sometimes 90% (a few times a year maybe). Does anyone know what this does to battery life. I assume it has little effect as the vast majority of discharges are around 60%, which I assume should be better than 80% so I hope they offset each other. But I am not a chemist....

Mauricio
Forum Virtuoso
Posts: 78
Joined: Fri Apr 16, 2010 2:49 pm
My RE system: 24 LG275, 275W panels, 2 FlexWare500, four VFX3648, series/parallel Quad system, two FlexMax80, X240 transformer, MATE3, FN-DC, 6 SimpliPhi 48V 3.5kWh Lithium Ferro Phosphate (LFP) batteries, Cummins Onan 20GGMA generator (20 kW, LP Vapor fueled), OpticsRe.
Note: 12 Surrette Rolls 4Ks25P Batteries ( 4V, 1350 AmpHr), were replaced by the 6 Simpliphi PHI 3.5 on June 21/19.
Location: Todos Santos, Baja California Sur, Mexico
Contact:

Re: FNDC and SimpliPhi batteries

Post by Mauricio » Sat Nov 23, 2019 1:51 pm

Raysun, first of all let me thank you for taking the time to address my concerns, I am very appreciative.
You are right, using voltages to determine SOC, except in those rare occasions you mentioned is the wrong way to go, and, as a lead acid battery user for over 10 years, I had never done that. End Amps was a good measure on the fly, with Specific Gravity being the only reliable measure. But even then, cells SG was recorded with cells Voltages, linking the two together. With Lithium that is not possible, leaving voltage as the only parameter. I’m not the one saying that and my confusion possibly arises from OB application note that states: “The SOC Voltages from SimpliPhi listed in table 3 can also be a useful alternative to the FN-DC SOC % readings”. Note that when discussing discharge rate, the discharge rate for SimpliPhi we find on “Available Capacity vs Current” chart, is flat, showing no significant loss from C/20 all the way to 1C. Your argument of the C5 discharge rate (you probably meant 5C) is correct, but we are no way near those conditions. I totally understand the bucket analogy you mentioned and do not dispute it, but that’s the reason of my ](*,) . Nevertheless, I will take your advice and stop doing it, unfortunately the pain will remain for a while longer.
Mauricio - Todos Santos, BCS
http://www.arribadelaroca.com

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Wed Dec 04, 2019 11:06 am

I emailed Simpliphi tech support regarding the new recommendations for Outback integration for the 3.8 model battery. These settings are quite a bit different from the older 3.4/3.5 batteries. I was wondering if these are newer/better and should be used for all models, or if they were just specific for the 3.8 model, and so I asked this:
I just saw your new 3.8 specs and the charge params are way different than my batteries, e.g. the voltages are lower and you recommend a longer absorb time and a monthly EQ charge. Is this ONLY for the 3.8 batteries, or have you changed your recommendations for 3.4/3.5 batteries (perhaps with different voltages etc).
I received this reply. As a result , I am lowering my absorb voltage a little and lengthening the time, so allow for a slower, gentler absorb cycle.
Thank you for emailing and asking. That is an excellent question. The new settings are actually not only for the new 3.8 PHI batteries but are recommended for all our previous generation 24v and 48v batteries, however it is not necessary to follow the new setpoints if you have a previous generation and are following the previous parameters.

There is no difference between the 3.8 BMS and previous models. The only change is in total capacity due to different cell form factor internally. Thus both sets of parameters can work for both batteries. I would recommend trying the new setpoints. I’ve attached these in the latest Outback Integration Guide. If you observe changes that you don’t like in your daily use, please let us know what you didn’t like and why. You can then make adjustments to the previous setpoints if you would like.

Lastly, you can make adjustments anywhere in the range from the previous settings to the latest settings in case you can hone in on a nice “sweet spot” setting that works best for you – so long as it is within the ranges of the old settings and the new settings.

Excellent question, and thanks for reaching out to ask.

Best regards,

Brandon

sodamo
Forum Czar
Posts: 603
Joined: Mon Apr 03, 2006 1:55 am
My RE system: Totally off grid - Hamakua side, Big Island, Hi
36 Trina 280 & 16 Phono 250 in 48 volt array (fixed) 14080w
Radian 8048/4048
4 FM 80 charge controllers
FNDC w/3 shunts
Mate 3s
OpticsRE - MMKL - Ninole Hi
12 SimpliPhi 3.8
14Kw Kohler 14RESA

2FX3048T
Mate 3s
10Kw MEP 803a

Honda EU7000i

Davis VantagePro2 Wx Station
On line at: http://www.weatherlink.com/user/sodamo/
Location: Ninole, Hi

Re: FNDC and SimpliPhi batteries

Post by sodamo » Wed Dec 04, 2019 4:26 pm

Very interesting response from SimpliPhi that either settings for either battery or find a sweet spot in between.

Any chance you asked about AGS settings?
David
Please visit http://vacation.ninolehawaii.com

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Wed Dec 04, 2019 5:00 pm

No I did not ask about AGS setting. I haven't programmed the AGS yet (I don't have a 240v auto-start genny yet), but if I did, I think I have enough info. Basically, the LIFP batteries do not need a complete charge cycle, as I understand it. So you are free to just add some amps to them as you please, assuming that this doesn't occur every day, maybe 5-7 days in a row at most? Therefore, I would just determine at what point in SOC I would want to start and stop the generator. From observations, I know how to translate that to voltage based on my normal loads etc. And I would use both SOC and voltage settings for the start/stop. I would do something like 20-25% for starting and 40-45% for stopping. From my experience that will get me through the bad day, and hopefully PV will give a good charge tomorrow.

I would take their battery charging param suggestions with a little caution. It makes perfect sense that you can sink amps into batteres at a lower voltage, it will just take longer. This seems gentler on the battery, so that is good IMO. However at some point, you are risking not getting to 100% for a long time, e.g. an hour or more, in which case you might run out of good sun, e.g. afternoon showers or haze. So my goal is to strike a balance, e.g. right now I am in absorb mode for 6 minutes and already at about 97% when it starts. While I can wait 15-30 minutes to absorb the final few percent, maybe 30-60 minutes with intermittent sun, I don't want to wait an hour or more. Due to intermittent sun/clouds, an hour of absortion at a lower voltage might take 2+ hours which in the winter might not complete. As you lower the absorb voltage, absorbtion starts at a lower SOC %. You might only be at 90% at the start of absorbtion, and that could make a 5% difference in your SOC at the end of the day if the sun disappears shortly thereafter because PV production is throttled back at the lower voltage. The difference between 54.8v and 56v is pretty large. I am slowly going to lower my absorb voltage and increase the time and see how things go. So long as I start absorbtion above about 95% SOC I think I can accept the worst case of ending up at say 96% if the sun disappears (as it often does here). But if you start at a really low voltage and say 85% SOC I would NOT want to end the day at 90% if I could have reached 100% by absorbing to a higher voltage. So it will take some experimentation to get to a good balance. I definitely have no problem sacrificing a few percent of SOC (on some days) in exchange for longer battery life. But when "few" becomes "a lot", that is where I draw the line :-)

raysun
Forum Emperor
Posts: 1028
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower One: FX3048T, (2) FM80, MATE3s, FlexNetDC
Outback 200NC batteries (8 @ 48v)
Outback IBR3 battery enclosure
Suniva 330 watt panels (12 - 6 Strings @ 2 Series)
Hyundai 355 watt panels (6 - 3 strings @ 2 series)

Re: FNDC and SimpliPhi batteries

Post by raysun » Wed Dec 04, 2019 9:33 pm

Regarding lithium, I think we need to get the lead out of our brains, I hear it's not good for us. ;)

We all are conditioned by the physics and chemistry of lead acid batteries, and are expecting to cater to their charging requirements (Bulk, Absorb, Float) to reach the 100% SOC holy grail at the end of each charge cycle.

Lithium couldn't care less about that. In fact, religiously charging them to 100% SOC, especially if holding them at 100% SOC in a float application, will shorten their service life.

Charging a lead acid battery requires the charge reverse a chemical reaction, forcing lead sulfate and lead oxide to shed their salt bonds and return the lead to a 'pure' state, while forcing the reformation of sulfuric acid. Lithium batteries have no such chemical reaction equivalent, and charging is an act of pushing lithium ions across a membrane and back into solution. Externally, charging just looks like charging. Internally, the reactions are fundamentally different.

I expect SimpliPhi has a variant on the "top balancing" battery management scheme. Every cell in a lithium battery will have a slightly different capacity. As the cells are charged, some will reach full SOC before others. Holding the cells at 100% SOC, or worse, overcharging them, will do damage to them. To avoid this, and squeeze as many electrons into the battery as possible, some charge will be bled off full cells and distributed to not-as-full cells. I bet there's a lot of this going on during the "Absorb" phase.

Many lithium batteries don't bother with top balancing. When any cell reaches full (as measured by voltage limits), charging stops. The battery does not get every possible electron soaked into it, but it doesn't matter. In fact the lower SOC cells will be less stressed.

Keeping a lithium battery at less than 100% SOC and more than 0% SOC, in a perpetual partial state of charge, is better for it than attempting to flog it into 100% SOC at every cycle.

I do believe the SimpliPhi tech's comments imply this. I'd expect the "sweet spot" reference is to finding a state of charge that meets ones consumption needs between charge cycles.

SandyP
Forum Guru
Posts: 322
Joined: Tue Jun 21, 2011 4:44 am
My RE system: 8 x 190W 24V Suntech panels (4 strings of 2)
4 x 325W 24V Suntech panels (2 strings of 2) added Sept2018
Mate2
Outback FM60 MPPT (max output lowered to 55amps)
12 x 2V Hoppecke GEL 660 Ah C100 - 24V System
Outback VFX3024 Inverter/Charger
Victron BMV-602s
Honda 5.5kW Genset
Location: Victoria, Australia

Re: FNDC and SimpliPhi batteries

Post by SandyP » Thu Dec 05, 2019 2:26 am

raysun wrote:
Wed Dec 04, 2019 9:33 pm
...Lithium couldn't care less about that. In fact, religiously charging them to 100% SOC, especially if holding them at 100% SOC in a float application, will shorten their service life.......
.... I'd expect the "sweet spot" reference is to finding a state of charge that meets ones consumption needs between charge cycles.....
May not be relevant, but my Toshiba Notebook has a "Extend Lifecycle" battery setting option which limits battery charging to only 80% SoC (for those that use a notebook mostly when connected to AC power).
As you say - just need to find the balance between use and charge!

raysun
Forum Emperor
Posts: 1028
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower One: FX3048T, (2) FM80, MATE3s, FlexNetDC
Outback 200NC batteries (8 @ 48v)
Outback IBR3 battery enclosure
Suniva 330 watt panels (12 - 6 Strings @ 2 Series)
Hyundai 355 watt panels (6 - 3 strings @ 2 series)

Re: FNDC and SimpliPhi batteries

Post by raysun » Thu Dec 05, 2019 10:23 am

That's exactly the point.

sodamo
Forum Czar
Posts: 603
Joined: Mon Apr 03, 2006 1:55 am
My RE system: Totally off grid - Hamakua side, Big Island, Hi
36 Trina 280 & 16 Phono 250 in 48 volt array (fixed) 14080w
Radian 8048/4048
4 FM 80 charge controllers
FNDC w/3 shunts
Mate 3s
OpticsRE - MMKL - Ninole Hi
12 SimpliPhi 3.8
14Kw Kohler 14RESA

2FX3048T
Mate 3s
10Kw MEP 803a

Honda EU7000i

Davis VantagePro2 Wx Station
On line at: http://www.weatherlink.com/user/sodamo/
Location: Ninole, Hi

Re: FNDC and SimpliPhi batteries

Post by sodamo » Thu Dec 05, 2019 10:52 am

I too think we need to adjust our thinking, get the lead out :grin: but on two different areas.

1. Just as we learned not all lead acid batteries are the same and require different protocols, what of lithium? Most often we read “lithium” but there are a number of different chemistries. Do we know if there are different protocols? SandyP made a valid point for a particular battery, but does that apply to LFP? I think we should be somewhat cautious about blanket statements. Regarding SimpliPhi LFP, I haven’t found the avoid 100% language. Can someone link a quote? I think we need to be more mindful and qualify our statements according to the chemistry unless it is a known fact to apply across the universe.

2. For the most part our old lead acid batteries were electrons in, electrons out. I think most lithium chemistries use a BMS between our electrons and their battery. Do they all work exactly the same or proprietary? I’m guessing the latter, but don’t know. Do we need a better understanding of the BMS role. Does my Apple laptop or iphone have same BMS as my SimpliPhis? Where do I find out?

I queried SimpliPhi re AGS settings, no reply yet. Will post response. FYI, the SimpliPhi website has loads of info.
David
Please visit http://vacation.ninolehawaii.com

raysun
Forum Emperor
Posts: 1028
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower One: FX3048T, (2) FM80, MATE3s, FlexNetDC
Outback 200NC batteries (8 @ 48v)
Outback IBR3 battery enclosure
Suniva 330 watt panels (12 - 6 Strings @ 2 Series)
Hyundai 355 watt panels (6 - 3 strings @ 2 series)

Re: FNDC and SimpliPhi batteries

Post by raysun » Thu Dec 05, 2019 12:02 pm

Yes, the Lead Acid battery structures: FLA, VRLA, Gel all require a bit of different handlng.

Different lithium chemistries do too.

Hopefully, folks are settling on LiFePO4 as home batteries for the inherent safety factor. The other chemistries are far more 'twitchy'.

The more unstable lithium chemistries demand a BMS to keep charge and temperature under control.

LiFePO4, strictly speaking, does not need a BMS, but it's highly recommended. To squeeze the most into and out of the battery balancing is needed. To keep the cells from self-destructing, voltage and current control are wise options.

Charging a lithium battery is actually simpler than LA. On the other side of the coin, lithium is far less tolerant of 'out of bounds' conditions.

sodamo
Forum Czar
Posts: 603
Joined: Mon Apr 03, 2006 1:55 am
My RE system: Totally off grid - Hamakua side, Big Island, Hi
36 Trina 280 & 16 Phono 250 in 48 volt array (fixed) 14080w
Radian 8048/4048
4 FM 80 charge controllers
FNDC w/3 shunts
Mate 3s
OpticsRE - MMKL - Ninole Hi
12 SimpliPhi 3.8
14Kw Kohler 14RESA

2FX3048T
Mate 3s
10Kw MEP 803a

Honda EU7000i

Davis VantagePro2 Wx Station
On line at: http://www.weatherlink.com/user/sodamo/
Location: Ninole, Hi

Re: FNDC and SimpliPhi batteries

Post by sodamo » Thu Dec 05, 2019 12:42 pm

Needing a BMS vs having one already built in, as I believe is the case with SimpliPhi.
David
Please visit http://vacation.ninolehawaii.com

sodamo
Forum Czar
Posts: 603
Joined: Mon Apr 03, 2006 1:55 am
My RE system: Totally off grid - Hamakua side, Big Island, Hi
36 Trina 280 & 16 Phono 250 in 48 volt array (fixed) 14080w
Radian 8048/4048
4 FM 80 charge controllers
FNDC w/3 shunts
Mate 3s
OpticsRE - MMKL - Ninole Hi
12 SimpliPhi 3.8
14Kw Kohler 14RESA

2FX3048T
Mate 3s
10Kw MEP 803a

Honda EU7000i

Davis VantagePro2 Wx Station
On line at: http://www.weatherlink.com/user/sodamo/
Location: Ninole, Hi

Re: FNDC and SimpliPhi batteries

Post by sodamo » Thu Dec 05, 2019 1:06 pm

I received answer to my AGS query to SimpliPhi. As he mentions SOC and charging parameters posting here as well.

Hi David,

Thank you for emailing for assistance. For AGS programming we advise the same charging parameters as in our integration guide. We don’t supply AGS specific programming setpoints, but in general you should have your generator kick in at 50.4V with nothing less than a 130 second delay. It sounds like you do have the correct voltage and time requirements programmed in.

As a last tip, we always recommend following the battery voltage as a guide of SOC. If your equipment gives the option to perform operations based on soc or voltage, please always select voltage. The range for our 48V batteries that you want to stay between is 50.2V (20% capacity) to 54V (fully charged).

Thank you,
Brandon
David
Please visit http://vacation.ninolehawaii.com

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Thu Dec 05, 2019 1:22 pm

I am not 100% at ease with his reply, e.g. 50.2v is 20% SOC with a C/2 load. Who typically has a C/2 load? That would be 20kw for your system! With a minimal load 50.2 could represent < 10% SOC, then a big load kicks in and your voltage drops by like a rock and yuck! 20% SOC is probably well over 51v with a smaller, more reasonable load. Their chart indicates about 51.2 for C/5 which is still a pretty big load, C/10 might be around 51.5. I would consider this when programming the AGS. I find SOC pretty reliable and would use it as the primary trigger, with battery voltage as a backup. For my batteries , I would probably use 51 to 51.2 as the voltage trigger.

raysun
Forum Emperor
Posts: 1028
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower One: FX3048T, (2) FM80, MATE3s, FlexNetDC
Outback 200NC batteries (8 @ 48v)
Outback IBR3 battery enclosure
Suniva 330 watt panels (12 - 6 Strings @ 2 Series)
Hyundai 355 watt panels (6 - 3 strings @ 2 series)

Re: FNDC and SimpliPhi batteries

Post by raysun » Thu Dec 05, 2019 3:44 pm

Huh. Peukert raises his ugly head. You'd think taking lithium, the dude would chillax.

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Thu Dec 05, 2019 3:51 pm

OK, I had to Google that .... does anyone know how the Ah ratings for Simpliphi batteries varies with discharge rate? Is it small enough to ignore?

raysun
Forum Emperor
Posts: 1028
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower One: FX3048T, (2) FM80, MATE3s, FlexNetDC
Outback 200NC batteries (8 @ 48v)
Outback IBR3 battery enclosure
Suniva 330 watt panels (12 - 6 Strings @ 2 Series)
Hyundai 355 watt panels (6 - 3 strings @ 2 series)

Re: FNDC and SimpliPhi batteries

Post by raysun » Thu Dec 05, 2019 6:01 pm

It's the two-step: Total capacity as a function of discharge rate; and terminal voltage as a function of discharge rate AND state of charge.

Inquiring minds want to know, and as SimpliPhi is earning a role as go-to storage for Outback system owners, it would serve both us and SimpliPhi to share that information. The easier it is to adapt this new battery, the more folks who will shell out the $$$$$$$.

They have it, no doubt. Somebody may think it proprietary data, but early adopters will directly benefit by not having to learn most of it from their own 'trial and error'.

A blessing and a vexation, IMO, is the flat voltage curve, which makes everything look and act hunkey-dorey, until like Thelma and Louise, one drives off the cliff at one end or the other.

From personal experience, I appreciate battery replacement warranties, but I never really want to exercise one.

Maybe in a spirit of hands-across-the-technology-waters somebody can talk to somebody and get us a fuller picture, or some more concise metrics at least.

gtarolli
Forum Guru
Posts: 290
Joined: Thu Dec 21, 2017 11:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 24 (8x3) 300w panels (7200w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda 3000 generator + one spare
Location: Wainiha, Kauai, HI

Re: FNDC and SimpliPhi batteries

Post by gtarolli » Thu Dec 05, 2019 6:42 pm

After reading some on Peukert's law, I am probably more confused than before. I think it says the total Amps you can remove from a battery are basically independent of the discharge rate, its just that voltage varies with the load, so you have to stop discharging earlier with large loads. What you do before that low voltage doesn't really matter, its just the load at the end that determines when you want to stop. So I'm confused about the lifetime of the battery at say 80% or 90% DOD. Since the voltage varies with load, how can you you use voltage to start a generator to stay above 80% DOD? Or maybe DOD doesn't really matter - it's voltage that determines the battery's life expectancy? Maybe 10% SOC isn't bad so long as the load is minimal and the voltage is above the minimum recommended? But then as soon as you turn on the microwave, you drive over the cliff and before your genny starts, your voltage goes down another volt.

So when they say start a genny at 50.4v with nothing less than 130 sec. delay , what load are they assuming, or doesn't it matter? It seems me that to be safe you want to assume a minimal load, so that if a large load kicks in you have enough headroom to allow for something like a 1v drop before the gen. starts. In other words, the real voltage you are trying to stay above is around 49.4v? Which is maybe why they have a 130 sec delay - so if a well pump starts up for a minute or two you don't start your gen., until the normal load drops the voltage down under 50.4? OK, I think i have totally confused myself now. I hate this feeling....

raysun
Forum Emperor
Posts: 1028
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower One: FX3048T, (2) FM80, MATE3s, FlexNetDC
Outback 200NC batteries (8 @ 48v)
Outback IBR3 battery enclosure
Suniva 330 watt panels (12 - 6 Strings @ 2 Series)
Hyundai 355 watt panels (6 - 3 strings @ 2 series)

Re: FNDC and SimpliPhi batteries

Post by raysun » Thu Dec 05, 2019 7:02 pm

You know that blessing and vexation thing? You just described it in the flesh.

I don't have a lithium battery, but I've been looking at it closely for the last four years. I'm almost to the point where I think I'm going to oversize it (it's only $$$$$$$ + $$$$) and use SOC, not venturing into the top 10% and bottom 30%. The 60% in the middle becomes my battery capacity.

Post Reply