Strange efficiency readings

Discussion about the OutBack OpticsRE system monitor

Moderators: OutBack Moderator Team, OutBack

Strange efficiency readings

Postby cflocker on Fri Dec 29, 2017 2:17 pm

Hi,
The new update to OpticsRE is amazing with the additional details, but I seem to be getting really strange numbers across multiple days. Here was an example from 12/28/17
Solar: 4.5kwh
Gen: 0kwh
To Load: 2.2kwh
To Battery: 4.69kwh
From Battery: 5.35kwh

This was with my new battery bank: 700ah of AGM batteries. It seems that there are a ton of losses if I produced 4.5kwh and only sent 2.2 to the load. So 2.3kwh are missing...?

I recently went through my system with a fine toothed comb to ensure that all shunts and settings were correct per the manual. Hoping that someone can offer some insight.

Thanks!
-Chris
Attachments
Screen Shot 2017-12-29 at 2.08.25 PM.png
cflocker
Forum Member
 
Posts: 22
Joined: Fri Jul 28, 2017 7:06 am
My RE system: Radian GS4048A
GSLC175-PV1-120/240
FlexMax 80
FN-DC
Mate3
700AH AGM batteries
3.8kw Canadian solar panels
Generac Ecogen 15kw

Re: Strange efficiency readings

Postby David LeBow on Fri Dec 29, 2017 3:46 pm

It won't answer all of your questions, but it may address some. Take a look at this post: http://www.outbackpower.com/forum/viewt ... =46&t=9465
David LeBow
Vice President, Software Engineering


Alpha Technologies Inc. / OutBack Power Technologies
3767 Alpha Way Bellingham WA 98226 USA
http://www.alpha.com
User avatar
David LeBow
OutBack Management
 
Posts: 301
Joined: Sun May 03, 2015 10:42 am
Location: Bellingham, WA

Re: Strange efficiency readings

Postby Greg T Fordan on Fri Dec 29, 2017 4:35 pm

Hello! Happy New Year! Newbie in the Forum here, but been lurking for a while. Here's my input:

1. According to FM80, the total harvest for the day of 28 Dec was 4.5 kWh
2. If we go by FNDC's accounting, as reported to MATE3, its "To Battery" shunt is showing 4.69 kWh of energy went to the battery bank. There is some discrepancy there because all the other devices (CCs and inverters report the kWh values to MATE3 in .1 kWh resolution, while FNDC reports in a higher accuracy of .01kWh.
3. According to FNDC, the total discharge for the day, "From Battery" is 5.35 kWh. You might ask why is this number higher than the "To Battery" of 4.69 kWh. Because this is the total energy extracted from the battery even when the sun was not yet shining, from midnight to the next midnight.
4. Now, on the 2.2 kWh load. Your issue was 4.5 kWh of energy was harvested according to Solar (CC), but only 2.2 kWh went to the load. Because that was the only loads demanded by your system on the day of 28 December. You had more room for maybe another 2.2 kWh free energy from the sun but did not take advantage of it. If you decide to use the grid sometime, your type of inverter is very much capable of selling. That way the unused harvest will be exported (sold) to grid instead of getting wasted.
Greg T Fordan
EngTech
Outback Power Technologies
User avatar
Greg T Fordan
OutBack Technical Service
 
Posts: 19
Joined: Fri Dec 29, 2017 2:56 pm
My RE system: Radian, FM80, FNDC, 48V batt bank

Re: Strange efficiency readings

Postby cflocker on Fri Dec 29, 2017 6:03 pm

Thanks David and Greg,
I think I understand what you are referring to from the other post and the explanation. The curious thing for me is that this is happening over multiple days. I am off-grid, so no selling or outside input.

I totally get the precision explanation for shunts and small variations, but this is ~2x which precision would not seem to account for. The other heartburn is that when I sized my battery bank, I sized it for the average load reported by the system ~70AH/day, but when I look at the data now on OpticsRE, I see nearly double that coming and going from Battery; therefore an undersized bank based on total system

I think the root of the question lies with Greg's #3. If FNDC says that it took out 5.35kWh from the battery, where did overage go if it did not go to the 2.2kWh of load?
cflocker
Forum Member
 
Posts: 22
Joined: Fri Jul 28, 2017 7:06 am
My RE system: Radian GS4048A
GSLC175-PV1-120/240
FlexMax 80
FN-DC
Mate3
700AH AGM batteries
3.8kw Canadian solar panels
Generac Ecogen 15kw

Re: Strange efficiency readings

Postby David LeBow on Sat Dec 30, 2017 1:53 am

As a general statement, keep in mind that the devices themselves require power to run. That power won't be shown going to a "load", per se.
David LeBow
Vice President, Software Engineering


Alpha Technologies Inc. / OutBack Power Technologies
3767 Alpha Way Bellingham WA 98226 USA
http://www.alpha.com
User avatar
David LeBow
OutBack Management
 
Posts: 301
Joined: Sun May 03, 2015 10:42 am
Location: Bellingham, WA

Re: Strange efficiency readings

Postby cflocker on Sat Dec 30, 2017 10:51 am

Thanks David,
Totally understand that the system needs power and there are static power requirements and variable power requirements to run the system and invert power.

I ran some more numbers today. From midnight to 8A this morning, the system delivered 0.4kWh to Load and consumed 1.4kWh from Battery with 0 power coming in. This means that over 8 hours the Radian, Mate, FM80 and FNDC (system) consumed 1kWh or 125W/h to deliver an average of 50W/h to AC devices.

Please check my assumptions there, but I was hoping that folks could sanity check that to see if it was similar to their experiences. I did doublecheck the Radian brochure and noticed that there is a sharp efficiency climb from 0-2kva where it starts to level out. It's reasonable that at the low load of 50W/h I'm at the very base of the curve :)

If the numbers that I calculated turn out, I'm thrilled to have the extra data from OpticsRE, just wish I would have run the battery bank calculations based on Battery Ah/day rather than Load Ah/day and I may have gone with a larger bank :) Good learning experience.
cflocker
Forum Member
 
Posts: 22
Joined: Fri Jul 28, 2017 7:06 am
My RE system: Radian GS4048A
GSLC175-PV1-120/240
FlexMax 80
FN-DC
Mate3
700AH AGM batteries
3.8kw Canadian solar panels
Generac Ecogen 15kw

Re: Strange efficiency readings

Postby David LeBow on Sat Dec 30, 2017 12:43 pm

cflocker wrote:This means that over 8 hours the Radian, Mate, FM80 and FNDC (system) consumed 1kWh or 125W/h to deliver an average of 50W/h to AC devices.


Again... this is the software geek speaking and not the power engineer, but: your load is low so the consumption to load ratio looks kind of whiffly. Maybe you need more load! ;-) This isn't an official OutBack statement, but if I had that much electronics gear hanging on my house, I somehow wouldn't be surprised that it used the equivalent of one bright light bulb. I think I read somewhere that the inverter can consume about 60W, so the number seems (again, to me, the software guy) to be in range given the other devices.
David LeBow
Vice President, Software Engineering


Alpha Technologies Inc. / OutBack Power Technologies
3767 Alpha Way Bellingham WA 98226 USA
http://www.alpha.com
User avatar
David LeBow
OutBack Management
 
Posts: 301
Joined: Sun May 03, 2015 10:42 am
Location: Bellingham, WA

Re: Strange efficiency readings

Postby EMCF on Sun Dec 31, 2017 12:54 am

@cflocker, one thing I noticed, your idle no load battery consumption is too high. If you look at hours where there is single light blue bar (from battery), the consumption is .2 kWh, that is 200 watts in a span of one hour. The idle, no-load power consumption as per website is only 34 watts. So, theoretically, in those hours when there is no load at all, the battery consumption (from battery) should only show .03 kWh. CC and FNDC's effects are negligible. Maybe it is about your battery discharging on a not so normal pace. How long have you been using them?
EMCF
Forum Expert
 
Posts: 61
Joined: Thu Jul 21, 2016 10:22 am
My RE system: FX3048, 4 200a batteries, FM60. 3 kw array

Re: Strange efficiency readings

Postby cflocker on Sun Dec 31, 2017 8:37 am

Thanks, the batteries are brand new, but I was seeing similar symptoms with the previous battery bank when I go back in history. Although this is an off grid cabin, there is always some load because of the networking gear. I have 3 Ubiquiti radios, a POE switch, weather station and Smart things running 24/7. There is also a 200w fish heater that can randomly turn on. I'll have to grab a kill-a-watt and try to measure the network stuff. Possible that the wattage is below the threshold for inverter measurements, but enough to change it from idle and that a 10-20 watt load is just enough to ramp up some electronics in the Radian.
cflocker
Forum Member
 
Posts: 22
Joined: Fri Jul 28, 2017 7:06 am
My RE system: Radian GS4048A
GSLC175-PV1-120/240
FlexMax 80
FN-DC
Mate3
700AH AGM batteries
3.8kw Canadian solar panels
Generac Ecogen 15kw

Re: Strange efficiency readings

Postby gtarolli on Sun Dec 31, 2017 11:55 am

I have a new Outback system (off-grid) and noticed similar issues. Below is my screenshot. Here is my take based on lots of reading and reverse engineering

The To-Load data is what comes out of the inverter. Since my screenshot shows the system data before the sun came up, all the load came from batteries. The big issue in my mind, which I raised in another thread on inverter efficiency, is why was 2.8KW drawn from the batteries to satisfy 1.9KW load. That is 50% more than the load. The GS8048 inverters have about 93% efficiency, but the curve drops off steeply below 2KW. During the night we draw about 300w (internet modem, routers, fridge, UV sterilization light, etc), which from the inverter graph looks to be well below 50% efficiency. So that might explain it. However I still see this discrepancy during the day, although a little smaller. I too undersized the battery bank based on my previous data from a smaller outback system with smaller inverter that I think was more efficient at lower loads - I think the curve dropped of at 1KW instead of 2KW.

image2.JPG


As you can also see, there seems to be some small phantom current coming from the charger shunts, about .1 amp at times, this can lead my daily To Battery total to be a little higher (~5%) than the Solar production total, I can live with that and is within reason of accuracy. But my From Battery is often 30-40% higher than the To Load even for an entire day.

Is this real? Is this inverter inefficiency? If it is real and due to inverter inefficiency, would I be better off with a smaller inverter that is more efficient at small loads, in a stacked config? so that the smaller inverter runs at night at higher efficiency?
gtarolli
Forum Whiz
 
Posts: 48
Joined: Thu Dec 21, 2017 11:39 am
Location: Wainiha, Kauai, HI
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 21 (7x3) SolarWorld 300w (6300w)
- 5 SimpliPhi 3.4 kWh LI batteries (335ah)
- 1 Honda 3000 generator + one spare

Re: Strange efficiency readings

Postby gtarolli on Sun Dec 31, 2017 1:41 pm

One other thought - is that the GS8048A is a two-leg 240v system. My house and guest house are 120v and I have attempted to split the loads across the legs, but can't control things like large loads (microwave = 2000 watts, water pump = 1300 watts) from creating an imbalance on just one leg. Is that a problem? There often will be a 2: or 3:1 ratio between the legs. In bad cases, there can be a 10:1 ratio.
gtarolli
Forum Whiz
 
Posts: 48
Joined: Thu Dec 21, 2017 11:39 am
Location: Wainiha, Kauai, HI
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 21 (7x3) SolarWorld 300w (6300w)
- 5 SimpliPhi 3.4 kWh LI batteries (335ah)
- 1 Honda 3000 generator + one spare

Re: Strange efficiency readings

Postby cflocker on Sun Dec 31, 2017 7:07 pm

Yeah, I think that we may be closing on an answer. I also wonder if a smaller inverter could offload the 24/7 network gear so that the Radian can idle.

Does anyone know the maximum wattage that can be supplied in "idle" mode?

I'm hoping that stove / microwave clocks and stuff like that are small enough that it would not push past idle. I can use some SmartThings outlets to turn off battery chargers and other stuff like that after dark so they don't have parasitic losses.
cflocker
Forum Member
 
Posts: 22
Joined: Fri Jul 28, 2017 7:06 am
My RE system: Radian GS4048A
GSLC175-PV1-120/240
FlexMax 80
FN-DC
Mate3
700AH AGM batteries
3.8kw Canadian solar panels
Generac Ecogen 15kw

Re: Strange efficiency readings

Postby cflocker on Sun Jan 07, 2018 9:43 am

So, I did some work on my electrical loads to try and keep them offline in the evening. The networking equipment is still running off the Radian 24/7 but most of the others get switched off by SmartThings. Last night per OpticsRE the DC battery consumption dropped to ~100W/h vs the 200-250W/h where it was. Big improvement and the battery bank is happier in the morning. I've got a 48V to 120V 300W inverter on the way, should be ready to install mid-week. With that I'm hoping to take most of the remaining 24/7 loads off the Radian and see if I can get to go into Idle more frequently. I would have loved to try straight DC-DC, but that ads a ton of complications with all of the devices. I'm going to tie the small inverter to the FNDC shunt so that it still captures the load and hopefully the new setup will result in another efficiency gain. Fingers crossed.
cflocker
Forum Member
 
Posts: 22
Joined: Fri Jul 28, 2017 7:06 am
My RE system: Radian GS4048A
GSLC175-PV1-120/240
FlexMax 80
FN-DC
Mate3
700AH AGM batteries
3.8kw Canadian solar panels
Generac Ecogen 15kw

Re: Strange efficiency readings

Postby gtarolli on Sun Jan 07, 2018 10:13 am

Does anyone know if the small load of the idle Radian and other components show up in the load data?

Are you going to put all your small 24/7 loads on a dedicated circuit driven by the new inverter? Or are you going to run it in parallel with the Radian?
gtarolli
Forum Whiz
 
Posts: 48
Joined: Thu Dec 21, 2017 11:39 am
Location: Wainiha, Kauai, HI
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 21 (7x3) SolarWorld 300w (6300w)
- 5 SimpliPhi 3.4 kWh LI batteries (335ah)
- 1 Honda 3000 generator + one spare

Re: Strange efficiency readings

Postby gtarolli on Sun Jan 07, 2018 10:46 am

I think I have half my problem understood. On a typical day, my "From Battery" power is about 3kW or 60 amps higher than the To Load power. The FromBattery data does not seem to be truncated, the ToLoad (and Inverter) data is truncated to .1kw or about 2.1 amps. I think I remember reading somewhere (WattPro?) that this is truncated , not rounded. So if that is true, the average error should be about 1 amp per hour or 24 amps per day. The maximum error would be twice that or 48 amps. That could well explain msot the issue. At night I often draw about 4 amps while it says 6 was taken from the batteries - I could be closer to the maximum truncation error at night and morning?
gtarolli
Forum Whiz
 
Posts: 48
Joined: Thu Dec 21, 2017 11:39 am
Location: Wainiha, Kauai, HI
My RE system: Outback: off-grid
- 1 GS8048A, Mate3, FlexNetDC, Hub10.3
- 2 FLEXmax 80
- 21 (7x3) SolarWorld 300w (6300w)
- 5 SimpliPhi 3.4 kWh LI batteries (335ah)
- 1 Honda 3000 generator + one spare

Re: Strange efficiency readings

Postby cflocker on Sun Jan 07, 2018 11:21 am

gtarolli,
There seems to be a number of things at work.
1. FNDC is more accurate than the inverter, so at small loads you'll want to use the FNDC (From Battery, To Battery) numbers because at small AC loads (Inverter: To Load) it may get lost in the precision of the measurements.
2. Any inverter has losses. At 0-? watts the Radian should idle at 34W, at some threshold the inverter will start waking up to supply loads. The energy required to run the inverter electronics can be larger than the AC load for small devices. Peak efficiency occurs at ~2kw and it is really good up there, but for light loading, you'll see lots of energy not going to the load. This is true with all inverter systems, so Outback is certainly not alone. I work with a lot of UPS units and this is always a battle for IT folks and engineers to load the units correctly so that they can perform well.

Most likely you are seeing similar effects to what I was seeing. Small *whiffy :) loads won't get reported well from the inverter, but the total energy will get reported by the FNDC. Use the From Battery to get a true picture of what you are really consuming.

The idea with the small inverter is to use it solely for running dedicated 24/7 networking loads on their own plug. It would NEVER have any AC interaction with the Radian and would only share the DC bus with a fuse. I don't want to try any advanced AC Coupling because this will just be a cheap small unit and not rated to interact. If I can offload 24/7 loads then the hope was for more Idle time on the big inverter and therefore better total efficiency.
cflocker
Forum Member
 
Posts: 22
Joined: Fri Jul 28, 2017 7:06 am
My RE system: Radian GS4048A
GSLC175-PV1-120/240
FlexMax 80
FN-DC
Mate3
700AH AGM batteries
3.8kw Canadian solar panels
Generac Ecogen 15kw


Return to OpticsRE

Who is online

Users browsing this forum: No registered users and 3 guests