Days Since Parms met

An area for discussing the OutBack FLEXnet DC product.

Moderators: OutBack, OutBack Moderator Team

raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

Screenshot_20210717-064255_DuckDuckGo.jpg
7/15, "Absorb" took 4 minutes
Screenshot_20210717-153409_DuckDuckGo.jpg
7/17, "Absorb" took 2 minutes

I tried to correlate to SoC when entering "Absorb", but its too narrow a slice of the status graph to make sense.

Guess I have to buy that $40 Victron USB cable and fire up a PC.
gtarolli
Forum Guru
Posts: 422
Joined: Thu Dec 21, 2017 10:39 am
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 1 FLEXmax 60
- 30 (10x3) 300w panels (9000w total)
- 6 SimpliPhi 3.4 kWh LI batteries (400ah , 20kWh)
- 1 Honda EM5000 generator + one EU3000
Location: Wainiha, Kauai, HI

Re: Days Since Parms met

Post by gtarolli »

.1v makes a lot of diff between 52 and 53v, it doesn't matter much between 53 and 56, or below 51. Almost all the charge is between 52v and 53v.

My system just takes a few more minutes to reach Return Amps, not a huge diff. Instead of 6 minutes it takes 12-16 mins.

As for float, my batteries are at float voltage for about 4-5 hrs per day, so while I understand holding them at float constantly is not good, holding them at float for 4-5 hrs, and then going down to 50% SOC and back up to 100% every day seems like it shouldn't be an issue.
JRHill
Forum Emperor
Posts: 1874
Joined: Thu Feb 17, 2011 11:12 am
My RE system: VFXR3648, FM80, 3k panels, Mate3s, FNDC, 4ea 3.8 PHIs. EU7000is AGS. X240 with big Bertha way off in the shop. Off grid.
Location: PNW
Contact:

Re: Days Since Parms met

Post by JRHill »

raysun wrote: Sat Jul 17, 2021 5:08 pm
I can't recall why I have absorb set at .2 but it's working and I'm staying there.
Absorb=56.3V is 0.1V below recommended so it likely takes another few minutes to get to the charge current "knee".

0.1V makes a lot of difference for lithium.
I remember. The CC always over shoots the target voltage so I set it back that a bit.
provo
Forum Czar
Posts: 843
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Two SimpliPhi 3.8-m-24V
(300Ah total)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC
Grid connection (no selling)
Honda EU3000is generator
Location: Sierra foothills

Re: Days Since Parms met

Post by provo »

Even when I get Charge Parms Met using 2% for Charged Return Amps (which took 8m today), the highest SOC reported on the Mate webpage is only 98% when the charge current goes negative. This has been happening ever since I started using Charge Termination with the Simpliphi's (I never used it with the Rolls.)

Is the reported SOC wrong, or am I resetting the SOC to 100% every day when it’s really only reaching 98%?
provo
Forum Czar
Posts: 843
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Two SimpliPhi 3.8-m-24V
(300Ah total)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC
Grid connection (no selling)
Honda EU3000is generator
Location: Sierra foothills

Re: Days Since Parms met

Post by provo »

JRHill wrote: Sat Jul 17, 2021 10:20 pm
The CC always over shoots the target voltage so I set it back that a bit.
I've noticed the same thing. I set the FM80 absV to 28.2V, and I see several excursions to 28.3V over the course of the absorb, some for as long as 10s.

I'm seeing this on the Mate webpage, and I haven't actually caught it on the FM80 display. The FNDC Device Map also confirms the 28.3V as the highest voltage today. I guess I should turn absorbV down to 28.1.
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

provo wrote: Sun Jul 18, 2021 1:03 pm Even when I get Charge Parms Met using 2% for Charged Return Amps (which took 8m today), the highest SOC reported on the Mate webpage is only 98% when the charge current goes negative. This has been happening ever since I started using Charge Termination with the Simpliphi's (I never used it with the Rolls.)

Is the reported SOC wrong, or am I resetting the SOC to 100% every day when it’s really only reaching 98%?
That's interesting. I usually have the opposite issue: SoC = 100% before CPM (generally by 5 - 6AH).

What is the FNDC->Battery Charging: Battery AH, and Charge Factor settings?

Perhaps the FNDC is miscalculating Peukerts (more than the normal 1.05/1.15 Peukerts Constant error between lithium and lead that's "baked in" to the firmware) due to a small voltage calibration error?

No matter what the Net Battery and SoC readings say, the 2% Charged Return Amps value is the "truth" for 100% SoC, IMO. Syncing the FNDC to 100% SoC via CPM is a good method for keeping the readings on an even keel.

If the FNDC is off 2% low, I wouldn't really sweat it all that much. However, you may wish to join my "Hey Outback! How's about an FNDC-Li, huh?" fan club.
provo
Forum Czar
Posts: 843
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Two SimpliPhi 3.8-m-24V
(300Ah total)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC
Grid connection (no selling)
Honda EU3000is generator
Location: Sierra foothills

Re: Days Since Parms met

Post by provo »

raysun wrote: Sun Jul 18, 2021 1:20 pm
No matter what the Net Battery and SoC readings say, the 2% Charged Return Amps value is the "truth" for 100% SoC, IMO. Syncing the FNDC to 100% SoC via CPM is a good method for keeping the readings on an even keel.

If the FNDC is off 2% low, I wouldn't really sweat it all that much. However, you may wish to join my "Hey Outback! How's about an FNDC-Li, huh?" fan club.

Here's the FNDC Settings:

Screen Shot 2021-07-18 at 1.41.26 PM.png

I just wouldn't want to be "fudging" the SOC from 98% to 100% every day, and a few weeks later think I'm at 75% when I'm actually at 25%. I suppose that could never happen :shock: ....
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

I just wouldn't want to be "fudging" the SOC from 98% to 100% every day, and a few weeks later think I'm at 75% when I'm actually at 25%. I suppose that could never happen :shock: ....
You're not "fudging" anything. You just have a clock that runs slow, and needs to be reset every day.

The Charged Return Amps is the "correct time" showing 100% SoC. The FNDC gets reset to that.

My wife bought herself a very nice Rolex watch before I met her. It keeps terrible time. One Christmas, I bought her a $10 Timex that keeps perfect time, so she could reset her Rolex.

Some day, as an experiment, allow the battery to stay in PSoC for an extended period. (Winter overcast will assist in this naturally.) When the battery fully recharges, note the difference between FNDC reading and 100% SoC as indicated by CPM. It will be interesting to see if there is daily "scope creep" or if it stays near 2%.
provo
Forum Czar
Posts: 843
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Two SimpliPhi 3.8-m-24V
(300Ah total)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC
Grid connection (no selling)
Honda EU3000is generator
Location: Sierra foothills

Re: Days Since Parms met

Post by provo »

raysun wrote: Sun Jul 18, 2021 1:59 pm
You're not "fudging" anything. You just have a clock that runs slow, and needs to be reset every day.
The Charged Return Amps is the "correct time" showing 100% SoC. The FNDC gets reset to that.
I like that explanation, thanks!
raysun wrote: Sun Jul 18, 2021 1:59 pm
Some day, as an experiment, allow the battery to stay in PSoC for an extended period. (Winter overcast will assist in this naturally.) When the battery fully recharges, note the difference between FNDC reading and 100% SoC as indicated by CPM. It will be interesting to see if there is daily "scope creep" or if it stays near 2%.
Another post for the Raysun folder :grin: .... Now if I can just remember to LOOK in that folder every so often....
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

The reason I suggest the experiment, is I too have an FNDC that drifts, but apparently in the opposite direction. It shows 100% SoC while the charge phase is still in Bulk (near the end.) Its generally off by 2 - 3% by my calculations, but that doesn't seem to materially change with number of days in PSoC. I'm speculating there are compensating "errors" in charge and discharge.

I also have a Victron BMV700 in series with the FNDC, and the two are rarely more than 2% off each other, even after two or more weeks at PSoC.

Empirically, things look good, but I'm still trying to make sense of it.

In the meantime I most trust charge termination at 56.4V @ 2% C20 charge current.
sbrownian
Forum Czar
Posts: 655
Joined: Mon Nov 30, 2020 9:23 am
My RE system: GS8048 FM80 FNDC Mate3 4 port HUB
4kw solar
7kw Kubota propane gen
Off grid.
Originally installed with 1200 AH of FLA Locomotive batts.
Replaced after 9 years of service with 2 x 300 AH ReLion Lithium packs.
MUCH more efficient!!
Location: Central Washington State

Re: Days Since Parms met

Post by sbrownian »

I still wonder a bit about the FNDC "firmware" reset...
Information, you get not; if incorrect question, you ask..
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

sbrownian wrote: Sun Jul 18, 2021 4:14 pm I still wonder a bit about the FNDC "firmware" reset...
It resets the parameter values, and deletes the FNDC "learned" battery characteristic data.
sbrownian
Forum Czar
Posts: 655
Joined: Mon Nov 30, 2020 9:23 am
My RE system: GS8048 FM80 FNDC Mate3 4 port HUB
4kw solar
7kw Kubota propane gen
Off grid.
Originally installed with 1200 AH of FLA Locomotive batts.
Replaced after 9 years of service with 2 x 300 AH ReLion Lithium packs.
MUCH more efficient!!
Location: Central Washington State

Re: Days Since Parms met

Post by sbrownian »

raysun wrote: Sun Jul 18, 2021 4:34 pm
sbrownian wrote: Sun Jul 18, 2021 4:14 pm I still wonder a bit about the FNDC "firmware" reset...
It resets the parameter values, and deletes the FNDC "learned" battery characteristic data.
Did ya do that when converting to Li?
Information, you get not; if incorrect question, you ask..
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

sbrownian wrote: Sun Jul 18, 2021 4:46 pm
raysun wrote: Sun Jul 18, 2021 4:34 pm
sbrownian wrote: Sun Jul 18, 2021 4:14 pm I still wonder a bit about the FNDC "firmware" reset...
It resets the parameter values, and deletes the FNDC "learned" battery characteristic data.
Did ya do that when converting to Li?
In a manner of speaking. The conversion to Li was a conversion to 100% all-new equipment.

Well, not exactly true. I kept my IBR-3-48 battery rack (but replaced all the cabling), two Midnite Solar PV combiner boxes, and the Honda EU7000is.
provo
Forum Czar
Posts: 843
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Two SimpliPhi 3.8-m-24V
(300Ah total)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC
Grid connection (no selling)
Honda EU3000is generator
Location: Sierra foothills

Re: Days Since Parms met

Post by provo »

sbrownian wrote: Sun Jul 18, 2021 4:46 pm
raysun wrote: Sun Jul 18, 2021 4:34 pm
sbrownian wrote: Sun Jul 18, 2021 4:14 pm I still wonder a bit about the FNDC "firmware" reset...
It resets the parameter values, and deletes the FNDC "learned" battery characteristic data.
Did ya do that when converting to Li?
Yes, curious minds want to know! I considered doing it but didn't -- I wonder how long it would take the FNDC to forget the Pb-acid "learning" and get educated about Li....
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

provo wrote: Sun Jul 18, 2021 5:09 pm
sbrownian wrote: Sun Jul 18, 2021 4:46 pm
raysun wrote: Sun Jul 18, 2021 4:34 pm

It resets the parameter values, and deletes the FNDC "learned" battery characteristic data.
Did ya do that when converting to Li?
Yes, curious minds want to know! I considered doing it but didn't -- I wonder how long it would take the FNDC to forget the Pb-acid "learning" and get educated about Li....
It will be an interesting test. I'd imagine a few cycles would give some indication.
JRHill
Forum Emperor
Posts: 1874
Joined: Thu Feb 17, 2011 11:12 am
My RE system: VFXR3648, FM80, 3k panels, Mate3s, FNDC, 4ea 3.8 PHIs. EU7000is AGS. X240 with big Bertha way off in the shop. Off grid.
Location: PNW
Contact:

Re: Days Since Parms met

Post by JRHill »

raysun wrote: Sun Jul 18, 2021 8:54 pm
provo wrote: Sun Jul 18, 2021 5:09 pm
sbrownian wrote: Sun Jul 18, 2021 4:46 pm
Did ya do that when converting to Li?
Yes, curious minds want to know! I considered doing it but didn't -- I wonder how long it would take the FNDC to forget the Pb-acid "learning" and get educated about Li....
It will be an interesting test. I'd imagine a few cycles would give some indication.
If it was done 'by the book' exactly. But with the change over to lithium there were all kinds of considerations and un-learning FLA batteries and worry wart issues for the LI investment. But to the question, if you hard reset the FNDC when the Phis hit the Bulk target voltage with minimal Absorb you're are on the way. As Raysun said It's gonna take a few more cycles to get the Days Since tuned. By the time that is done the FNDC is probably as close as it will be when the batteries are full or what you determine is full for your purposes.

So here is an answer as to why I put Absorb at .2hr. If you are charging to max voltage, 56.4, the BMS is gonna cut things off when 'a' battery is full. But if you have 2, 3, 4, 6, 8+ PHIs is their internal BMS gonna cut them off at the exact same time? I would doubt it. They don't communicate amongst themselves that anyone knows of. So I went with .2hr as an experiment just in case there was a laggard. If its a laggard once it is going to be a consistent laggard? This was my reasoning for the .2 absorb. [also note I dialed back the absorb voltage to 56.3 because my CC always over shoots to 56.4 every day.]

OK, beat me up for my reasoning. I get a full load of lithium every day.
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

The genius of the SimiPhi type battery packaging is all battery blocks are in parallel. (And the reason they are such stickers about equal impedance feed cables.) Every block has the opportunity to accept an equal portion of the charge current, and contribute an equal portion of the discharge current.

Given the physics of lithium, and its marked difference from lead-acid, its far easier for each cell to be near balance with its neighbors - in a monobloc, and across a battery (with monoblocks in parallel).

Its true each BMS is managing 16 series connected cells (in a 48V monobloc, 8 series connected in a 24V monobloc) autonomously, but that does not impact the other BMS doing the same thing.

When a cell reaches "full charge" (really full migration of ions to the anode) it simply refuses to accept any more charge current. The BMS may look at all the cells, and decide to "bleed off" some charge from the full cell so the others in the string can continue accepting charge current. This is called "cell balancing". Every cell in every monobloc in the battery is handled the same way.

Eventually, the charge current declines to 2% of C20, and every cell in every monoblock in the battery is considered fully charged.

One can continue to "put the pressure on" the battery by holding the chosen "Absorb" voltage (called Diffusion Voltage in lithium terms, except in certain northern California neighborhoods, where for some reason, they call it Shirley) for as long as one desires. Eventually all that is being created is heat. Good thing we're using LiFePo4.

The upshot of all this is: picking the factory specified "Absorb" Voltage, and holding it for the factory specified Absorb Time should result in Charged Return Amps reaching the 2% target, and the battery will at 100% SoC.

When the above condition is met, the FNDC should be (re)set to 100%, no matter what it thinks.
JRHill
Forum Emperor
Posts: 1874
Joined: Thu Feb 17, 2011 11:12 am
My RE system: VFXR3648, FM80, 3k panels, Mate3s, FNDC, 4ea 3.8 PHIs. EU7000is AGS. X240 with big Bertha way off in the shop. Off grid.
Location: PNW
Contact:

Re: Days Since Parms met

Post by JRHill »

raysun wrote: Mon Jul 19, 2021 10:15 amThe BMS may look at all the cells, and decide to "bleed off" some charge from the full cell so the others in the string can continue accepting charge current. This is called "cell balancing". Every cell in every monobloc in the battery is handled the same way.
Seriously I am just trying to understand the big picture. Heck, FLA batteries are probably worse (in time) between cells than LI, IMHO. FLA overcomes that with time and voltage if correctly done, neither of which the LI wants once its full.

So here's the deal: All this is still new. Just staying within a depth of discharge is interesting for cycle life, just like it is for lead batteries. But there has to be more when it comes to lithium. And of course, its gonna vary by Mfr just like lead batteries.
raysun
Forum Emperor
Posts: 5623
Joined: Tue Jul 26, 2016 5:57 am
My RE system: Flexpower Two: (2) FXR3048A-01, (2) FM80, MATE3s, FlexNetDC
SimpliPhi 48-3.8 (6 @ 48v)
Outback IBR3 battery enclosure
REC Alpha 440W panels - 2 arrays: each of 4 strings of 2 in series
Honda EU7000is gas fuel generator

Re: Days Since Parms met

Post by raysun »

JRHill wrote: Mon Jul 19, 2021 10:49 am
raysun wrote: Mon Jul 19, 2021 10:15 amThe BMS may look at all the cells, and decide to "bleed off" some charge from the full cell so the others in the string can continue accepting charge current. This is called "cell balancing". Every cell in every monobloc in the battery is handled the same way.
Seriously I am just trying to understand the big picture. Heck, FLA batteries are probably worse (in time) between cells than LI, IMHO. FLA overcomes that with time and voltage if correctly done, neither of which the LI wants once its full.

So here's the deal: All this is still new. Just staying within a depth of discharge is interesting for cycle life, just like it is for lead batteries. But there has to be more when it comes to lithium. And of course, its gonna vary by Mfr just like lead batteries.
Yes to what you say, but don't make it more complicated than it really is. At the cell level, Li is ridiculously simple. At the physics level, even moreso.

Researchers have been studying lead acid battery chemistry for nearly 200 years and still haven't mastered all its complexity. If they had, lead acid batteries would cost $50 and last 50 years.

The simple truth of the current lithium ion battery cell is: don't boil the electrolyte, don't try charging if you've frozen the electrolyte, don't charge the battery in a way that causes sharp spikes to form on the anode, don't exceed the cell voltage maximum and minimum, don't exceed the cell maximum current capability.
Post Reply