Minigrid Behavior I don't understand

Dual Input, 120/240, stainless steel form factor design for unsurpassed surge capability

Minigrid Behavior I don't understand

Postby pss on Mon Feb 11, 2019 12:22 pm

I will admit that the longer I use my system, the less I understand it.
I recently replaced my FN-DC. I have been carefully measuring the voltages with meter of the battery bank. The inverter is accurate, the FN-DC is about 0.1 v off. The temperature sensor is accurate. My three charge controllers all read a different voltage of the battery bank with a variance of 0.8 volts. Why, I don't know.
This morning, my SOC was 96%. My load 2.2 kWh. Sunny outside. My PV output was 1.7 kWh. Inverter was in pass through with all load being supplied by the grid, all PV into charging the battery bank towards 100%. When looking at my CC's. two were in silent mode and only one was producing PV. I went to the Mate 3S, and manually dropped the AC input. This immediately woke up the two silent CC's and the PV output went from 1.7 to 3.4 kWh in about 20 seconds. By manually dropping the AC input, I doubled my PV output and saved grid use. I either don't understand how to correctly set voltages on the Mate 3S in Optics RE or there is some type of logic flaw in the Minigrid setting which allows the CC's to be shut down, reduces the harvesting of PV in favor of grid use and consequently costs more money from the utility. It just seems to be that as soon as PV can equal load and SOC or V is above a set parameter, the grid should be dropped regardless of "timers" somewhere.

Please help me understand this.
Also, I suggest adding separate forum tabs for each of the Radian settings such as Minigrid, Grid-Zero, etc.
Thank you.
pss
Forum Guru
 
Posts: 134
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 7820 watts in three strings, Flexmax 60 x 3, Radian 8046A, GLSC load center, Mate 3S, Hub 10.3 and 900 Amp, 48V Trojan T105-RE battery bank.

Re: Minigrid Behavior I don't understand

Postby Kurt Lundquist on Mon Feb 11, 2019 3:09 pm

To disconnect from the grid the inverter has to meet the Absorb charge target voltage and time. If the charger on the inverter is turned on then this should be a pretty easy target to hit. Buying grid power to charge up the batteries isn't that efficient though. For best performance make sure the PV Absorb charge target is at least .2 V above the inverter target. Next try reducing the inverter Absorb charge time to .1 of an hour. Keep the normal Absorb time on the charge controllers. You need enough PV to make up for whatever capacity you used last night, as well as handle all the daytime loads while Absorb charging. You may have to keep raising the connect to grid voltage until you find a sweet spot.

I don't know your loads but I would probably connect to grid after the battery is below 52 volts for more than 10 minutes. I would reduce inverter Absorb to .1hr and keep the charger off. The next day it shouldn't take long for the charge controllers to meet the drop grid targets so you run off-grid through the day. If your batteries are flooded every so often you will probably want to drop the grid to work them over at a deeper discharge. I only pay ~10c/kWh so for me it wouldn't make sense to burn battery power to save on grid costs overnight.

You will need to watch the system and tweak things until you find what works best. Winter and summer settings will probably vary unless you are near the equator.

If you are allowed to Grid-Tie/Sell I would go that route instead, even if they don't credit you for what you produce. I'm assuming you have loads on the AC input side of the inverter as well and selling power can help take care of those.

Good luck,

-Kurt
Kurt Lundquist
Renewable Energy Project Engineer
http://alphaenergy.us/
User avatar
Kurt Lundquist
Alpha Energy Engineering
 
Posts: 1185
Joined: Tue May 31, 2011 7:23 am
Location: Arlington, WA
My RE system: VFX3524, Mate3, 200Ah 24V AGM Battery Bank

Re: Minigrid Behavior I don't understand

Postby pss on Mon Feb 11, 2019 5:16 pm

Kurt, thanks for your replay.

My inverter settings are Absorb 57.2V and time 0.2 hours.
My charge controllers are Absorb 57.6V and time 0.3 hours.
My inverter charger is off.
My Minigrid setting is connect to grid at 48V and 4 minutes.

None of this explains the behavior of seeing two of the charge controllers go into silent mode. My guess is that under a load of almost any kind, it is very hard to get a 48V battery bank up to 57.2 or 57.6 volts. At most, my 900 amp hour bank gets anywhere from 2000 watts to 4000 watts above load to charge. Today, after I dropped the AC input, the PV generated load plus 1700 watts.

Today, my SOC hit 100 while V = 56.0.

So maybe I still need to reduce the absorb voltages on both inverter and CC's or CC's alone?

Thanks.
pss
Forum Guru
 
Posts: 134
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 7820 watts in three strings, Flexmax 60 x 3, Radian 8046A, GLSC load center, Mate 3S, Hub 10.3 and 900 Amp, 48V Trojan T105-RE battery bank.

Re: Minigrid Behavior I don't understand

Postby Mike Curran on Mon Feb 11, 2019 5:39 pm

Maybe this will help:

My GVFX's can't use mini-grid mode and even if i had newer Radians or FXR's, my PV resource isn't large enough to use it. However, my system works to use PV to support my loads as long as a) charge parameters have been met for the day, and b) my PV watts exceed what's required to maintain my battery at the sell setpoint, which is roughly the same as my float setpoint. As soon as there's a cloud, my inverters go into pass-through mode until the cloud passes. You obviously don't have a problem with part (b).

Contrary to Kurt's advice, my inverters' and charge controllers' absorb times are set high at 3 hours. For me it's reaching the FNDC ""charged time" that allows me to stop passing AC thru my inverters and sell to the grid each day, and that's set for 5 minutes. You don't say what your FNDC charged time is set for but that's what I'd be looking at. Take a look at my voltage for a sunny, productive day when I met the FNDC charge parameters (absorb voltage and charged time):
Screenshot_20190211-192954.jpg
You can see during the hours between noon and 1 pm when I had max PV production (and after I'd met charge parameters), "from grid" fell to zero. Then later when sunlight tapered off, I started taking power from the grid again. So I think the key for you is making sure your system meets charge parameters each day, which is part of what Kurt is suggesting.

Again, hope this helps.
http://www.tigoenergy.com/site.php?95b2dca2-ca6c
Mike Curran
Forum Czar
 
Posts: 962
Joined: Sat Aug 28, 2004 6:28 pm
Location: Chagrin Falls, Ohio
My RE system: Outback:
- 8 (2x4) Evergreen 180's into one MX60 (2007)
- 6 (2×3) ALEO S79-300's into one FM80 (2017)
- 2 grid-tied GVFX3524's classic stacked for 120/240VAC (2007)
- 12 Surrette/Rolls 2V x 1766Ah (2007)
- Hub10.3, Mate3, FNDC, RTS, OpticsRE. Tigo Energy ES maximizers on each PV module.

Westinghouse Solar: (2012)
- 30 (2x15) 235W panels with Enphase M215 microinverters, grid-tied.

Re: Minigrid Behavior I don't understand

Postby pss on Mon Feb 11, 2019 6:20 pm

Mike, thanks for your input. My FN-DC charge time is set for 1 minute.

What I am trying to reason out is that I do not want to have PV available from 7AM now and draw from the grid until say 11 AM or Noon while my batteries reach a magical "100%" as the FN-DC calculates. This is because what happened today illustrates that at 9:10 AM I manually disconnected from the grid by using the AC Input hotkey and the drop AC key. My V was between 55.7 - 55.9 and SOC was 95-97%. Two CC's silent, one charging (can't remember if bulk or float). AT 11 AM I hit the 100% SOC while V = 55.0 - 56.0.

By manually dropping the AC Input, I was able to re-activate all of my CC's into Bulk, dramatically increase my PV output (load + batt chg instead of 1/3 CC's for batt chg only) and still supply my loads and complete my charge cycle.
In SoCal, our utility is hitting me for $0.47 /Kwh. Ouch. That is why this is so important. An extra 5-10 kWh/day times 30 days is $70 - 140 per month. It is important that the programming logic of the Mate 3S and FN-DC and Radian inverter somehow take into account that PV is free, grid is not and getting my battery bank to 55V is okay with me instead of 56.7V if I save $140/month. And the rate is $0.55 in Summertime.
pss
Forum Guru
 
Posts: 134
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 7820 watts in three strings, Flexmax 60 x 3, Radian 8046A, GLSC load center, Mate 3S, Hub 10.3 and 900 Amp, 48V Trojan T105-RE battery bank.

Re: Minigrid Behavior I don't understand

Postby Mike Curran on Mon Feb 11, 2019 8:01 pm

Wow, those are some rates!

Well, re-reading Kurt's advice, I'd say he's exactly on point. Sounds like there's no way to get around waiting for your system to meet charge parameters. You'll have to experiment with your settings as he suggests. "it shouldn't take long for the charge controllers to meet the drop grid targets so you run off-grid through the day". Gonna take some trial and error with your settings to get to that point, though. Good luck.
http://www.tigoenergy.com/site.php?95b2dca2-ca6c
Mike Curran
Forum Czar
 
Posts: 962
Joined: Sat Aug 28, 2004 6:28 pm
Location: Chagrin Falls, Ohio
My RE system: Outback:
- 8 (2x4) Evergreen 180's into one MX60 (2007)
- 6 (2×3) ALEO S79-300's into one FM80 (2017)
- 2 grid-tied GVFX3524's classic stacked for 120/240VAC (2007)
- 12 Surrette/Rolls 2V x 1766Ah (2007)
- Hub10.3, Mate3, FNDC, RTS, OpticsRE. Tigo Energy ES maximizers on each PV module.

Westinghouse Solar: (2012)
- 30 (2x15) 235W panels with Enphase M215 microinverters, grid-tied.

Re: Minigrid Behavior I don't understand

Postby ralphday224 on Tue Feb 12, 2019 6:18 am

Have you investigated HBX? I use that mode. The inverter absorb voltage is same as float (53v), absorb time 1min, float time 24hrs. At night the inverter holds at float voltage, drops charger as per manual information...drops grid when sun and/or wind drives the voltage up past HBX setpoint. If, on a day like today with lots of wind, I "drop" grid and shut off the breaker to the inverter.

I'm doing this because my 15 year old batteries don't like cycling anymore, at least deep cycling. It takes days and lots of amps to bring some of the cells back to full SOC if they've discharged/cycled for a couple of days.

Ralph
ralphday224
Forum Whiz
 
Posts: 44
Joined: Sat Mar 07, 2015 4:25 pm

Re: Minigrid Behavior I don't understand

Postby mikedowdell on Tue Feb 12, 2019 9:53 am

Pss, I'm in SoCal as well and also use MiniGrid for the very exact same reasons you do. Thankfully, with SDGE we got off the Tier'd rate plan where we were getting dinged at 54 cents plus overuse penalties and onto the Time of Use plan where only 4pm-9pm daily is the 54 cent rate.... and that cut our bill in half by itself. I'm using MiniGrid as I think you expect where after about 20 min each morning im full PV for all loads until end of day, then use the battery bank to get me past the 4pm-9pm higher charge window before returning back to Utility power. From your settings, it seems like you have everything set correctly, however you didn't mention if your Mate's Load transfer or HBX features are turned on. I had similar problems when they were on, but disabling them has corrected those issues.
mikedowdell
Forum Whiz
 
Posts: 36
Joined: Wed Sep 12, 2018 10:04 am
Location: San Diego, CA
My RE system: Outback FXR2524A-01, FlexMax FM80, Mate3s, Power Hub 4
6x Solar World 250w panels on a 2 axis solar tracker
4x 6v 220ah Trojan Batteries
Hybrid Off Grid with AC as backup

Re: Minigrid Behavior I don't understand

Postby Kurt Lundquist on Tue Feb 12, 2019 10:35 am

pss wrote:My inverter settings are Absorb 57.2V and time 0.2 hours.
My charge controllers are Absorb 57.6V and time 0.3 hours.
My inverter charger is off.
My Minigrid setting is connect to grid at 48V and 4 minutes.


.3hrs Absorb on the Charge controllers is not enough to take care of your battery bank. I would bump it up to 4 hours unless you have a specific recommendation from the battery manufacturer. If the batteries are flooded your Absorb voltage is probably too low as well. Trojan should have advice for the proper charge settings. Either way once the inverter sees the batteries over the Absorb target for 12 minutes it should disconnect from the grid. The charge controllers will continue to target the Absorb settings so if your load isn't too much you should be able to meet Absorb depending on how much sun you have this time of year.

48V is too low of a discharge unless your daytime loads are light. Under load 48V is probably ~30% depth of discharge. On your 43kWh battery bank 30% is ~13kWh. If we assume your battery charge cycle is ~80% efficient it will take ~16kWh to recharge the 13kWh that you used. If we assume your sun hours is 4 hours this time of year (I get about 1 sun hour in Wetstern Washington) and the PV array harvest is 75% of the PV STC rating you have 5.9kW PV x 4hrs = 23.6kWh. These are generous PV harvest numbers assuming you have perfect tilt and azimuth. In this scenario, you have ~8kWh of load that you can operate during the day while still taking care of the battery bank. This is an overly simplistic assessment of the situation but I hope it helps to outline the considerations of PV, to battery, to load size when using Mini-Grid.

pss wrote:Today, my SOC hit 100 while V = 56.0.


It sounds like the FN-DC is not calibrated to the battery bank right now. This can happen when the batteries are not fully recharged day over day. After getting the charge settings from Trojan you will probably need to run a charge on the battery bank from the grid. For this charge you should temporarily increase the inverter Absorb time to whatever is required by the battery manufacturer. After it is done you can set the inverter back to the .2 Absorb time.

As I said before it will take some trial and error for you to find the settings that work best. Please let us know how it goes.

Thanks,

-Kurt
Kurt Lundquist
Renewable Energy Project Engineer
http://alphaenergy.us/
User avatar
Kurt Lundquist
Alpha Energy Engineering
 
Posts: 1185
Joined: Tue May 31, 2011 7:23 am
Location: Arlington, WA
My RE system: VFX3524, Mate3, 200Ah 24V AGM Battery Bank

Re: Minigrid Behavior I don't understand

Postby pss on Tue Feb 12, 2019 12:10 pm

Kurt:

Thanks for the reply. This morning, I again watched as the battery bank went to 97% SOC, two charge controllers went to "silent" mode and a single controller was in float mode at 1600 watts. V = 56.4 and stayed there. Power kept coming from the grid for the load. All parameters were met for FN-DC and inverter charging and still no switch to PV only.
So I manually dropped the AC input at 10:30 AM pacific and immediately all three CC's went into Bulk mode and PV increased from 1600 to 3900 with load about 2500 watts. So then, with SOC at 97%, I recycled the FN-DC by unplugging the ethernet and then plugging back in and resetting SOC to 100%.

After reading your comments, here's where I now stand. For the three CC's, I changed bulk to 59V and float to 54V per Trojan values. I increased the absorb time to 2.5 hours. I set the inverter charger levels to just below the CC voltages (58.4 and 53.2 and absorb 0.2 hours).

I will run a full charge cycle tonight using grid up to 59.2 and absorb for about 3 hours and then reset. I do get bubbling of my batteries when voltages are near 56-57 and up, so I know I get a good electrolyte stir.

I am in So Cal, so on a sunny day like yesterday and today, with peak temps in the 63-69 range, my output is 24-27 Kwh. In June/July, my output peaks at about 44 kWh. The long days are nice, but ambient panel temps can reach 145 degrees here.
pss
Forum Guru
 
Posts: 134
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 7820 watts in three strings, Flexmax 60 x 3, Radian 8046A, GLSC load center, Mate 3S, Hub 10.3 and 900 Amp, 48V Trojan T105-RE battery bank.

Re: Minigrid Behavior I don't understand

Postby Kurt Lundquist on Tue Feb 12, 2019 12:29 pm

pss wrote:This morning, I again watched as the battery bank went to 97% SOC, two charge controllers went to "silent" mode and a single controller was in float mode at 1600 watts. V = 56.4 and stayed there.


If it was in Float then it went through Absorb. FN-DC charge termination may be cutting your charge off before your inverter parameters are met so I suggest you disable that. Make sure to update the FN-DC charge parameters too. Next check the battery voltage calibration on all the CC's vs the Inverter. You want them to all read the same battery voltage as close as possible. Ideally, you take a battery voltage reading when the system is at rest. Early in the morning before the PV is producing while the system is connected to the grid would be best. Measure right at the battery bank then use the Mate3 to calibrate all devices to that voltage. The .6 V difference between the inverter and charge controllers should help.

-Kurt
Kurt Lundquist
Renewable Energy Project Engineer
http://alphaenergy.us/
User avatar
Kurt Lundquist
Alpha Energy Engineering
 
Posts: 1185
Joined: Tue May 31, 2011 7:23 am
Location: Arlington, WA
My RE system: VFX3524, Mate3, 200Ah 24V AGM Battery Bank

Re: Minigrid Behavior I don't understand

Postby pss on Tue Feb 12, 2019 2:08 pm

Kurt:

I disabled the FN-DC auto charge termination control on the Mate 3S.
So far today, all three CC's are in float mode, power output is 400 to 1800 watts above load (we are getting some high clouds), SOC staying at 100% and V m= 53.
Two days ago I calibrated the CC's and inverter to the actual measured battery voltage while all load was being supplied by the grid.
I also used a battery tester on several random cells and all electrolyte measures were in the green.
We are expecting some rain next two days, so I will report back when I have the next sunny morning.
pss
Forum Guru
 
Posts: 134
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 7820 watts in three strings, Flexmax 60 x 3, Radian 8046A, GLSC load center, Mate 3S, Hub 10.3 and 900 Amp, 48V Trojan T105-RE battery bank.

Re: Minigrid Behavior I don't understand

Postby Kurt Lundquist on Tue Feb 12, 2019 2:33 pm

I suggest you force Bulk on the system with the new settings before the sun goes down today. Use the Charger button on the Mate3 to force the Bulk charge.

-Kurt
Kurt Lundquist
Renewable Energy Project Engineer
http://alphaenergy.us/
User avatar
Kurt Lundquist
Alpha Energy Engineering
 
Posts: 1185
Joined: Tue May 31, 2011 7:23 am
Location: Arlington, WA
My RE system: VFX3524, Mate3, 200Ah 24V AGM Battery Bank

Re: Minigrid Behavior I don't understand

Postby pss on Mon Feb 18, 2019 12:25 pm

This is a follow-up on my system settings. I can now say I am happy with the Minigrid performance after some tweaks. PV generation is starting around 7 AM now. This AM, V = 50.5 and SOC 90% in backup mode. Switch to Minigrid at 6:30 AM. Hourly load is pretty steady at 2-2.3 kWh. By 9AM battery voltage peaked at 59.3 and SOC at 95%. Grid support dropped shortly after 9 AM and then running totally on PV. Shortly after 10 AM SOC = 100% and battery V = 59.8. As the sun goes down, the Inverter switches to Backup mode. Power is supplied from the batteries until a voltage of about 48.4 is reached (the SOC is about 87% at that time). I think I have room for more battery discharge capacity at the end of the day by setting the voltage even lower than the 48.4. I just want to be able to fully recharge the next day and avoid having to use the grid to charge batteries due to local energy costs (0.44/kWh).

One charger in absorb, the other two still in bulk. FN-DC charged V = 58 and charge time = 1 minute. Not sure why the two stay in Bulk when the other went to absorb?

So I am keeping an eye on my battery voltages, but for Trojan T105-RE bulk charge is 59.3 and my CC's are set for 59. Float is 54V and that's what my CC's are set for. Battery V measured at the terminals is 7.45 which is perfect 100% charged value for 15 degrees C.

Any other tweaks or suggestions appreciated.
pss
Forum Guru
 
Posts: 134
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 7820 watts in three strings, Flexmax 60 x 3, Radian 8046A, GLSC load center, Mate 3S, Hub 10.3 and 900 Amp, 48V Trojan T105-RE battery bank.

Re: Minigrid Behavior I don't understand

Postby Kurt Lundquist on Tue Feb 19, 2019 9:03 am

I'm glad it is working better now. Thanks for the follow-up!

-Kurt
Kurt Lundquist
Renewable Energy Project Engineer
http://alphaenergy.us/
User avatar
Kurt Lundquist
Alpha Energy Engineering
 
Posts: 1185
Joined: Tue May 31, 2011 7:23 am
Location: Arlington, WA
My RE system: VFX3524, Mate3, 200Ah 24V AGM Battery Bank


Return to RADIAN Series

Who is online

Users browsing this forum: No registered users and 1 guest