Making them work

Post Reply
JRHill
Forum Emperor
Posts: 1866
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:

Making them work

Post by JRHill »

We've had a few cloudy days. The batteries didn't get much from solar. So last night I stayed up later and later and finally said screw it and went to bed at 11pm. Darn it, they just wouldn't go lower. The PHIs were at 52.0 then.

This morning I wake up at 5am to emails and flashing lights and TG there's not a Claxton hooked up. Down to 20% SOC and 51.2. 51.0 is 50% SOC per the factory docs (no load). The average load is a few hundred watts early morning. I tried to go back to bed. Nope. What craziness.

More later but this is how I'm 'recovering':
Charge8-1-21.jpg
61 amps. Cool.
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

AGS didn't work this AM. 2hr start was set at 51.6 and it passed right by the set point. No errors, nothing until hitting Warning and Critical SOC.

After a coffee or two I realized that while I have verified the start/stop from the Mate3s/VFXR upgrade, this is the first actual AGS 2 minute start I've done. It didn't work so good. The genset is fine. It works 'perfect'.

All of a sudden I got embarrassed telling people how easy the AGS is to set up and use. I've been using it for a long time with a VFX and with all the settings set the same its not working so good. Here is what happened:

- I had to manually fire the generator from the Mate3 - it didn't fire from the AGS settings. It warmed up and connected and I switched to Auto at 51.2 and it charged full out. It went to 51.6 and disconnected with batteries full. I was dumbfounded.

- I didn't know what to do so I checked the connections. The 30amp twist lock was in but not locked. I wiggled it and locked it. I changed the inverter absorb voltage from 53.5 to 56.0. I hopped up and down on right then left. I got another coffee.

Apparently I'm not so good at AGS.

[BTW, I haven't used it for months. Connections? The silly little automotive relay? I don't know - hopefully I can find it]
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

Things are really wonky. Hit 100% SOC at 54.7vdc. On the downhill side I hit 20% SOC at 51.3 the day before.

I've been charging from the FM all summer taking it to Days Since Parameters Met, reset, and continuing forward. Mostly every day it resets. But oh, no, this is not true SOC. I didn't know until I challenged the system - top to bottom. The top was good, the bottom was way off. In other words it wasn't linear with the PHI chart of resting SOC. My battery voltage said 50% but I was at 20% from the PHI chart (resting voltage).

I thought I was good hitting the parameters. But on the down hill side my Simpliphis batteries still have a huge amount left IF I WAS JUST GOING OFF SOC. On the uphill side you hit 100% at 55?

Raysun has gone through a bunch of recharging stuff.

Soon I'll get to the AGS stuff again.
raysun
Forum Emperor
Posts: 5608
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: Making them work

Post by raysun »

Is the FNDC - Battery Charging - Battery AH set to the battery capacity?

When syncing to 100% SoC during charging, I have found the FNDC reports 100% SoC before the battery reaches the Phi specified Charged Parameters (Charge Current = 2% of C20 AH capacity.) I scarcely pay attention to battery voltage, as during a long, slow charge session, the battery lingers on the 54.n V plateau for a very long time, and 100% SoC can be reached (according to net AH replaced) well before the programmed "Absorb" voltage rise to 56.4 V.

Think about Phi's other "Long Absorb" charging profile. I believe its 54.6V for 2H. That voltage occurs naturally during charge, and if the charge current is "just right", will linger there, even though still in Bulk.

The only way I know for disambiguation of 100% SoC is to drive the charging cycle to 56.4V termination at 2%.

I wasn't aware Phi had a Resting Voltage - SoC equivalence chart. Where did you find it?

The published Voltage - SoC under load charts are a joke, IMO. Who discharges their battery consistently at C5, or even C10? Not me, that's for sure.

I find the FNDC, properly calibrated and parameter values correct, is far more effective at tracking SoC than reading voltage is.

Its interesting that Phi used to thump the "read voltage to determine SoC is The Way" drum pretty hard in the past, and now are quietly saying "voltage is difficult to relate to SoC, and current counting is The Way".

In any event, Romulus and Remus, my OB and Victron coulomb counters, contine to faithfully perform their pas de deux without stepping on each other's toes. It must mean, they are both wrong an equal amount across all battery SoC readings, or they are both right (enough to use for the purpose.)

On to AGS, (which, again, I don't use) gross failures of the system to read its trigger points not withstanding, perhaps voltage isn't the clear "line in the sand" it should be. At least with lead acid, voltage declined more or less in step with SoC. With lithium, no such luck. I've run some tests with a LiFePo4 battery (not in my Phi battery bank), and could, at a low rate of discharge, cause the battery to go from 100% to 30% SoC with less than a 0.2V drop in terminal voltage.

Seems to me voltage "is more what you'd call 'guidelines' than actual rules" when it comes to SoC.
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

Raysun, thx for the input. I'm still really pissed about what just happened let alone the AGS stuff.

Everything is fine. Except I really need to know the AGS works. I have to travel soon. To use the millennial thing - OMG - if it doesn't work.
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

raysun wrote: Sun Aug 01, 2021 1:30 pm 1) Battery AH set to the battery capacity?

2) When syncing to 100% SoC during charging, I have found the FNDC reports 100% SoC before the battery reaches the Phi specified Charged Parameters (Charge Current = 2% of C20 AH capacity.) I scarcely pay attention to battery voltage, as during a long, slow charge session, the battery lingers on the 54.n V plateau for a very long time, and 100% SoC can be reached (according to net AH replaced) well before the programmed "Absorb" voltage rise to 56.4 V.

Think about Phi's other "Long Absorb" charging profile. I believe its 54.6V for 2H. That voltage occurs naturally during charge, and if the charge current is "just right", will linger there, even though still in Bulk.
1) Yeah, its all correct.
2) I hadn't thought about this before - that a slow charge on a cloudy day at a particular voltage could be seen as an absorb or the dreaded extended absorb. I'm sure your wisdom was imparted previously buy I glossed over it. I will head back into it. TY.

So here is what happened for the last few days with overcast and limited panel input:
3dayProfile.png
Very little came in from the panels in the above graph. It was kind of a wintertime exercise. To preface this I have been meeting Days Since mostly for quite some time. I thought the SOC was reasonably accurate. Oh, no, not and esp not linear. I have not to date drawn the batteries this deep so this was definately a learning experience. The thing that was most apparent was the battery voltage vs SOC in the SimpliPhi chart. As the battery voltage declined I finally took action. I was at 51.3v with a supposed SOC of 20%. I should have a long way to go to 50.2/20% from the factory chart.

So I turned the AGS to Auto. Nothing - it should've strted. I don't want to confuse things with this part but it didn't work well. Finally I got it to run the cycle and there is no resolved as to what changed. Lazy circuits :-)?

Anyway I finally got this:
3dayProfile.png
And the genset shut down per the parameters.
Attachments
Charge8-1-21.jpg
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

Oh, and here was another oddity: I've powered the system down many times in the past. But this time when I powered up the FM80 it defaulted to 12vdc. That's weird. I don't remember having to go into the FM80 push buttons to verify system voltage during a power up. As I recall it read the battery voltage at start up, the panels were flipped on, and all the settings were retained.

Somebody got an annoying message over this - I couldn't find anything that spoke to startup voltage.
raysun
Forum Emperor
Posts: 5608
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: Making them work

Post by raysun »

JRHill wrote: Mon Aug 02, 2021 10:03 am Oh, and here was another oddity: I've powered the system down many times in the past. But this time when I powered up the FM80 it defaulted to 12vdc. That's weird. I don't remember having to go into the FM80 push buttons to verify system voltage during a power up. As I recall it read the battery voltage at start up, the panels were flipped on, and all the settings were retained.

Somebody got an annoying message over this - I couldn't find anything that spoke to startup voltage.
That is odd. The only time voltage is explicitly prompted for is during initial setup. However, it looks like a good practice to check it during any configuration changes/restart operations.
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

raysun wrote: Mon Aug 02, 2021 10:22 amThat is odd. The only time voltage is explicitly prompted for is during initial setup. However, it looks like a good practice to check it during any configuration changes/restart operations.
The good 'ol FM80, at startup (with the battery breaker on first), defaulted to 12v. Then this AM at 60vdc. All the parameters were screwed up with 12vdc abs/float then 60vdc abs/float. I've never seen that. I did the button push on the FM80 for 48vdc. "Are you sure?" Yup. After that everything worked and it was charging from panels. Connected and charging. My gosh no year is different from another in that things change. TG our systems don't have a forced updates. It's tough as it is.

Google will fix this.
raysun
Forum Emperor
Posts: 5608
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: Making them work

Post by raysun »

JRHill wrote: Mon Aug 02, 2021 11:10 am
raysun wrote: Mon Aug 02, 2021 10:22 amThat is odd. The only time voltage is explicitly prompted for is during initial setup. However, it looks like a good practice to check it during any configuration changes/restart operations.
The good 'ol FM80, at startup (with the battery breaker on first), defaulted to 12v. Then this AM at 60vdc. All the parameters were screwed up with 12vdc abs/float then 60vdc abs/float. I've never seen that. I did the button push on the FM80 for 48vdc. "Are you sure?" Yup. After that everything worked and it was charging from panels. Connected and charging. My gosh no year is different from another in that things change. TG our systems don't have a forced updates. It's tough as it is.

Google will fix this.
Hmmm... I wonder what's confusing the FM80?
jainsw
Forum Guru
Posts: 156
Joined: Mon Dec 28, 2009 10:54 am
My RE system: My setup is a 48v system and is comprised of the following components:
12-ES190 Evergreen solar panels-pole mount non-auto tracking, 4 220 Volt GE panels fixed mount
12-Hanwha Q Peak 300watt panels
1-Outback FM60 PV Charge controller 60a MPPT, 12-60 vdc. 60a output
1-Outback FM80 PV CC, 80a MPPT
1-Outback VFX3648 3.6kw, 120vac output
1-Mate
1-Outback Hub
1-Outback FNDC
24-,Trojan L-16RE-2V lead acid 1110ah C20 batteries

Re: Making them work

Post by jainsw »

JRHill wrote: Sun Aug 01, 2021 11:28 am AGS didn't work this AM. 2hr start was set at 51.6 and it passed right by the set point. No errors, nothing until hitting Warning and Critical SOC.

After a coffee or two I realized that while I have verified the start/stop from the Mate3s/VFXR upgrade, this is the first actual AGS 2 minute start I've done. It didn't work so good. The genset is fine. It works 'perfect'.

All of a sudden I got embarrassed telling people how easy the AGS is to set up and use. I've been using it for a long time with a VFX and with all the settings set the same its not working so good. Here is what happened:

- I had to manually fire the generator from the Mate3 - it didn't fire from the AGS settings. It warmed up and connected and I switched to Auto at 51.2 and it charged full out. It went to 51.6 and disconnected with batteries full. I was dumbfounded.

- I didn't know what to do so I checked the connections. The 30amp twist lock was in but not locked. I wiggled it and locked it. I changed the inverter absorb voltage from 53.5 to 56.0. I hopped up and down on right then left. I got another coffee.

Apparently I'm not so good at AGS.

[BTW, I haven't used it for months. Connections? The silly little automotive relay? I don't know - hopefully I can find it]
This is a bit disconcerting as I am about to switch over to SimpliPhi batteries with my existing OB 3648VFX inverter. Did you ever get the AGS to work properly with the reported voltage on the SimpliPhi batteries?
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

jainsw wrote: Mon Sep 13, 2021 6:09 am
JRHill wrote: Sun Aug 01, 2021 11:28 am [BTW, I haven't used it for months. Connections? The silly little automotive relay? I don't know - hopefully I can find it]
This is a bit disconcerting as I am about to switch over to SimpliPhi batteries with my existing OB 3648VFX inverter. Did you ever get the AGS to work properly with the reported voltage on the SimpliPhi batteries?
Yes. All is working fine now. It was something with the programming in the Mate3, the inverter, the FM or a combination of the three. With taking the system down and restarting it something was awry. But when I changed the VFXR and to charge the batteries to 56 at .1hr absorb and the FM to 56.4 the AGS worked perfectly from there on. It certainly had nothing to do with the PHIs.
JRHill
Forum Emperor
Posts: 1866
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: Making them work

Post by JRHill »

Oops, unintentional loads. Yesterday I did a BUNCH of watering of the garden and ground cover around the house while the DW was filling tanks. The sun was waning so I plugged the well into the house system so I wouldn't pull the well batteries too low as they're on their last leg. When I was done I left a hose tricking on some shrubs. Really, I had every intent to go out and shut the hose off - it didn't happen after supper. This morning I walk by the fridge and it was open an inch. It was like that all night. It reminded me of the recent snack commercial with the gal saying "Not me, not me!" These were not big loads, just consistently drawing .5 to .6+kWh since sundown previous.

Gotta love those PHIs. System low was 52.1vdc and 53% SOC (or as close as I can determine it to reality). If I was still using lead I'm sure of a shutdown situation. But to be honest if they were new L16s they'd had probably lasted too. But I still had a LONG way to go with the PHIs at the low point this morning. I like these PHIs more and more. And with our normally low load and moderate charge situation they are 'perfect.'

One thing though. In the last week I've pulled the PHIs down more than all along through the summer. Once to 51.9 and again this morning. I haven't changed any parameters except for lowering the AGS start point. But its been 5 days since Days Since Parameters Met has been achieved. Prior to this week the parameters were met every day. The floating averages in the FNDC are different now and it seems that's going to be more the case as the days get shorter. Sigh. Winter is coming.
Post Reply