el-guapin
Trust no One
Good observation! I better understand your issue now.
An Enphase Ensemble Installation
- Thread starter svetz
- Start date
How do you get that level of access? Maybe the account the installer granted access to the system is your account. I will try to get that to happen.
svetz
Works in theory! Practice? That's something else
While playing around with the zigbee stuff I took the dongle out and put it back... my worse signal strength dropped to -72 dB! I put it back repositioning it similar to the way it was before and it went from:
to (2 samples at various times, seems to more often be the latter):
Whew! So finicky.
to (2 samples at various times, seems to more often be the latter):
Whew! So finicky.
svetz
Works in theory! Practice? That's something else
Updated the OP with the year's numbers, the utilities average rate (including all fees and taxes) without solar for my usage would have been $0.154357/kWh, ended up saving $2733. That's a lot better than prior years, sadly mainly due to the increase in the cost of electricity.
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,723
I know exactly what you mean here. As the cost of grid power goes up, it makes our solar produced power worth more.
My Enphase/SilFab panels produce 7,951 kilowatt hours for the 2022 calendar year. My cheapest electric rate is now $0.24 per KWH.
That works out to at least $1,908 in savings off my electric bill. But thanks to the battery shifting half of that to the high rate time, The real savings is probably over $2,500, even after the losses from charging and discharging the batteries through the XW-Pro.
The DC connected BougeRV/NewPowa panels produce power that was only used at the high rate time. That system made 192 kilowatt hours in just 83 days of operation. 192 x 0.37 = $71 paid back so far. That is almost one of the panels paid off so far. A full year should be well over 4 times that production, since this was mostly winter time. I should get around double the daily energy production in late spring to early fall.
Obviously we won't get rich off the energy produced, but having it pay off and turn any profit is a good thing.
My Enphase/SilFab panels produce 7,951 kilowatt hours for the 2022 calendar year. My cheapest electric rate is now $0.24 per KWH.
That works out to at least $1,908 in savings off my electric bill. But thanks to the battery shifting half of that to the high rate time, The real savings is probably over $2,500, even after the losses from charging and discharging the batteries through the XW-Pro.
The DC connected BougeRV/NewPowa panels produce power that was only used at the high rate time. That system made 192 kilowatt hours in just 83 days of operation. 192 x 0.37 = $71 paid back so far. That is almost one of the panels paid off so far. A full year should be well over 4 times that production, since this was mostly winter time. I should get around double the daily energy production in late spring to early fall.
Obviously we won't get rich off the energy produced, but having it pay off and turn any profit is a good thing.
svetz
Works in theory! Practice? That's something else
Did you work it out or take their numbers? From their numbers my lowest power rate in 2022 was 0.07784 and the highest was 0.14867 - a whopping 200% increase!
Despite their average rate ($0.0973), when I calculated it with all the charges and such it worked out to $0.154357/kWh on average. Took a while to make the spreadsheet and figure it all out. Of course, at 60% of your rate I don't want to sound like I'm complaining ; -)
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,723
The 24 cents per kilowatt hour (they raised it 1 cent, it's now 25) is their estimate after adding in all of the fees and taxes. I have to give So Cal Edison a little credit here. They may be a total rip off, but they are at least honest about it. The one part that throws a wrench into calculating it is the minimum monthly charge. When you use so little power, it makes it look very bad.
Last billing period that ended Dec 17th I did net consume 61 KWHs for the whole month. They charged me a total of $19.36 for using that amount of energy. That was after all the NBCs, taxes, etc. That means I had to pay $0.3173 per KWH. That is average for the entire day, not breaking down the different time of use rates. Since most of my consumption is at the lower rate, the difference is likely the added NBC charges.
The bill before that looks completely wrong. For the whole month, I used a net of -80 KWHs. That's right, I gave them more energy than I used. For that bill, they gave me a credit of $6.70 after you tack on the NBCs and taxes. So for this export month, they only paid me $0.08375 per KWH. The bulk of that difference seems to come down to the minimum charge of $9.85 and the $0.022 NBC. I kept the flow very low with the battery, but that still came to 109 KWH they charged on the NBCs. $2.40 is a small number, but when it is on such a small balance, it becomes a fair percentage. I am still not completely sure where they came up with 109 KWH for these charges, that does not work out to the amount of energy in+out. Adding the in and out together is still only 89 KWHs. So it must be something to do with the energy that went into my home, even if I later exported energy.
Here are the rates quoted by SCE.
And here is what they charged me just for the energy on the actual bill.
We are in winter, so the rates quoted are Mid Peak $0.41, Off Peak $0.28, and Super Off Peak $0.25
Looking at the actual bill, Mid peak rate is $0.27161 for delivery and $0.17254 for the energy. Added together it is $0.44415 but then I save $0.09086 on the baseline credit, bringing it down to $0.35329 before taxes and NBCs. And the quoted rate is up at $0.41 to cover off those extra costs.
The other rates are lower on both the generation and delivery. And the numbers they are still using on my NEM 2.0 plan is just the total for the whole month. If I consume 400 KWH and then Export 400 KWH in any given rate, the total here should be zero, but not on the NBC's.
Here is the NBC's for the same month bill
It all totals just over $0.03 per KWH, but where is the 109 KWHs coming from? Since I did consume power on this month, there was no added minimum charge.
They certainly do play a bit of a numbers game to make it more confusing. Even though I am on a monthly billing plan, they still total up the energy separately on my bill and give two separate running totals. I know they do this for the yearly billing since they charge all the taxes each month, but then bill the energy at the end of the year. If I owe for energy, I pay it each month on this plan. But when I have a credit, they keep adding it up until the end of the year still. I currently have an energy credit of $14.61 and thanks to the energy rebate, I still have a $33.16 credit to keep covering taxes and NBCs. With this lousy weather we are having, I am going to likely burn up all of those credits at the end of my Jan. billing cycle.
Do I need to start writing down the numbers off my power meter again to track what they are seeing? I know it has a separate running total for the energy into my house and the energy export from my house. No count ever actually goes backwards.
Last billing period that ended Dec 17th I did net consume 61 KWHs for the whole month. They charged me a total of $19.36 for using that amount of energy. That was after all the NBCs, taxes, etc. That means I had to pay $0.3173 per KWH. That is average for the entire day, not breaking down the different time of use rates. Since most of my consumption is at the lower rate, the difference is likely the added NBC charges.
The bill before that looks completely wrong. For the whole month, I used a net of -80 KWHs. That's right, I gave them more energy than I used. For that bill, they gave me a credit of $6.70 after you tack on the NBCs and taxes. So for this export month, they only paid me $0.08375 per KWH. The bulk of that difference seems to come down to the minimum charge of $9.85 and the $0.022 NBC. I kept the flow very low with the battery, but that still came to 109 KWH they charged on the NBCs. $2.40 is a small number, but when it is on such a small balance, it becomes a fair percentage. I am still not completely sure where they came up with 109 KWH for these charges, that does not work out to the amount of energy in+out. Adding the in and out together is still only 89 KWHs. So it must be something to do with the energy that went into my home, even if I later exported energy.
Here are the rates quoted by SCE.
And here is what they charged me just for the energy on the actual bill.
We are in winter, so the rates quoted are Mid Peak $0.41, Off Peak $0.28, and Super Off Peak $0.25
Looking at the actual bill, Mid peak rate is $0.27161 for delivery and $0.17254 for the energy. Added together it is $0.44415 but then I save $0.09086 on the baseline credit, bringing it down to $0.35329 before taxes and NBCs. And the quoted rate is up at $0.41 to cover off those extra costs.
The other rates are lower on both the generation and delivery. And the numbers they are still using on my NEM 2.0 plan is just the total for the whole month. If I consume 400 KWH and then Export 400 KWH in any given rate, the total here should be zero, but not on the NBC's.
Here is the NBC's for the same month bill
It all totals just over $0.03 per KWH, but where is the 109 KWHs coming from? Since I did consume power on this month, there was no added minimum charge.
They certainly do play a bit of a numbers game to make it more confusing. Even though I am on a monthly billing plan, they still total up the energy separately on my bill and give two separate running totals. I know they do this for the yearly billing since they charge all the taxes each month, but then bill the energy at the end of the year. If I owe for energy, I pay it each month on this plan. But when I have a credit, they keep adding it up until the end of the year still. I currently have an energy credit of $14.61 and thanks to the energy rebate, I still have a $33.16 credit to keep covering taxes and NBCs. With this lousy weather we are having, I am going to likely burn up all of those credits at the end of my Jan. billing cycle.
Do I need to start writing down the numbers off my power meter again to track what they are seeing? I know it has a separate running total for the energy into my house and the energy export from my house. No count ever actually goes backwards.
svetz
Works in theory! Practice? That's something else
This morning I looked at the realtime graphs to see what was using power, 7:05 is probably me turning on some lights...
But the big notice for me is the jagged purple line. It's at 7:10-7:12 so it's probably my PC starting up, is the switching power supply making it jagged??
Then my wife got up and went into the kitchen, 7:20 is probably the coffee going on.
7:23 the coffee pot is probably going off. But the "Prod Real Power Aggr" is still going up despit the 15 amp load on the coffee pot going off.
So, what the heck is "Prod Real Power Aggr"? I get power factor and real power versus apparent power, seems like unabbreviated it means "Product Real Power Aggregate", but I could be wrong. Mainly I don't see why it didn't drop when the coffee pot went off.
But the big notice for me is the jagged purple line. It's at 7:10-7:12 so it's probably my PC starting up, is the switching power supply making it jagged??
Then my wife got up and went into the kitchen, 7:20 is probably the coffee going on.
7:23 the coffee pot is probably going off. But the "Prod Real Power Aggr" is still going up despit the 15 amp load on the coffee pot going off.
So, what the heck is "Prod Real Power Aggr"? I get power factor and real power versus apparent power, seems like unabbreviated it means "Product Real Power Aggregate", but I could be wrong. Mainly I don't see why it didn't drop when the coffee pot went off.
svetz
Works in theory! Practice? That's something else
It's probably going up as it's an "Aggregate", but it was flat at the start and if that was the case it should have been going up from the phantom power throughout the night. Past 8:00, so doesn't seem to reset hourly... also noticed that it's being measured in "watts", so it's not as big as the graph would make it look.
In case the jagged lines were from the PC I turned it off for 10 minutes starting at 8:00, didn't seem to make a difference:
Removing "Prod Real Power Aggr" from the graph and the scale resets so you can see what's going on:
Running a load of laundry so the current loads are probably mainly the electric hot water tank and washing machine.
The red lines are "Prod Cur" for the two phases and purple/blue "Cons Cur". The "Cons Cur" must be the actual amps. For a couple of points along the graph for L1 I see:
The 8:12 point is on the far right where the prod line exceeds the cur line and the 8:39 point isn't shown on the graph. I checked that timestamp later after the graph reset with new values in case it was a measure over the lifespan of the chart, but the values didn't change suggesting it's something else. The red curve did fall back as power consumption dropped.
The "Prod Real Power Aggr" dipped at 8:42:19 with a corresponding dip in amps. 530W/240V ~= 2.2 amps, so it must be some aggregate over time trying to give an idea of actual power consumption rather than snapshots of currents at given times.
In case the jagged lines were from the PC I turned it off for 10 minutes starting at 8:00, didn't seem to make a difference:
Removing "Prod Real Power Aggr" from the graph and the scale resets so you can see what's going on:
Running a load of laundry so the current loads are probably mainly the electric hot water tank and washing machine.
The red lines are "Prod Cur" for the two phases and purple/blue "Cons Cur". The "Cons Cur" must be the actual amps. For a couple of points along the graph for L1 I see:
| Prod Cur (A) | Cons Cur (A) | |
|---|---|---|
| 8:06:07 | 2.59 | 7.35 |
| 8:06:09 | 2.44 | 12.82 |
| 8:06:27 | 2.53 | 7.85 |
| 8:12:33 | 4.23 | 3.73 |
| 8:39:30 | 3.27 | 2.94 |
The 8:12 point is on the far right where the prod line exceeds the cur line and the 8:39 point isn't shown on the graph. I checked that timestamp later after the graph reset with new values in case it was a measure over the lifespan of the chart, but the values didn't change suggesting it's something else. The red curve did fall back as power consumption dropped.
The "Prod Real Power Aggr" dipped at 8:42:19 with a corresponding dip in amps. 530W/240V ~= 2.2 amps, so it must be some aggregate over time trying to give an idea of actual power consumption rather than snapshots of currents at given times.
| Prod Real Power Aggr | Prod Real Power Aggr removed to increase scale of other lines |
|---|---|
 |  |
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,723
Only a guess here...
Is that jagged purple line the output of the Enphase microinverters as they ramp up making power as the sun rises?
My guess is the jaggedness is the MPPT searching for the maximum power. The current from my iQ7's certainly wiggles a bit, as does my DC charge controller. But I have no way of looking at it at that resolution.
Is that jagged purple line the output of the Enphase microinverters as they ramp up making power as the sun rises?
My guess is the jaggedness is the MPPT searching for the maximum power. The current from my iQ7's certainly wiggles a bit, as does my DC charge controller. But I have no way of looking at it at that resolution.
svetz
Works in theory! Practice? That's something else
The purple line (e.g., "Prod Real Power Aggr") seems to be fairly close to Σ (current x voltage ÷ PowerFactor ). It's a lot easier to see when the lines aren't moving around (the line is no longer as jagged, although that might be scale). There's possibly time averaging going on or the datapoints aren't all collected at the same instance as the numbers are close, but not exact.
Just noticed there's another line available I haven't seen before, "PoC Real Power Aggregate", it can be negative! PoC? Point of Contact? Guessing it must be exporting power to the grid when negative and importing when positive. So, at 8:27 a cloud impacted production so I was about net 0, at 9:30:30 some load came on temporarily that cause me to import power.
Thanks for the feedback! That was driving me nuts! Now it all seems to be making sense.
Last edited:
svetz
Works in theory! Practice? That's something else
| Hmmm, another mystery! In the post above I speculated at 9:30:30 there must have been a large internal power draw to make the system import power. I eliminated the watt curves to increase the amp scale and saw the opposite of what I was expecting as shown to the right. Something shut off, reducing the load. (The red line is the solar production, the purple is the household consumption, the cyan line is the power factor). Does that mean for those 30s the system stopped exporting power because of some sort of instability when something in the house turned off, destabilizing the power?? I checked the event logs and saw a "Grid Instability: Set" & "No Grid Profile: Clear" on microinverter 4219, but that was only a single inverter. From the PoC line in the post above it looks like a complete drop of exporting power which would have had to happen from the IQ System Controller (aka Enpower). No events around that timeframe in either the Envoy or the System Dashboard. |  |
Just did a refresh and see it happened again at 10:03. I'm starting to think it's just an error in their math as when PoC > 0, I still see my system exporting power on the live statistics. Running the drier to get some of the blips.
Using Gimp to overlay the two scales it looks like below. At 10:56 when L2 is stable and L1 drops 10 amps the line should drop but instead it raises. At 10:57:09 when L1 goes up 8 amps and L2 goes down 9 amps, then PoC Real Power Aggr becomes > 0 while L1 > L2.
I'll report it and see what they say.
Last edited:
fromport
Solar Addict
Roughly same time period, also SCE:
I put 414 kWh more into the grid than I pulled from the grid. (home is totally electric, no gas to heat). Got charged $19.49 so I got charged -$0.05 per kWh
I am on discontinued plan D-A:
Weekday Winter Rates
Off-Peak: 41 cents from 8 a.m. to 2 p.m., and 8 p.m. to 10 p.m.
Super Off-Peak: 20 cents from 10 p.m. to 8 a.m.
On-Peak: 52 cents from 2 p.m. to 8 p.m.
Weekend Winter Rates
Weekend Off-Peak: 41 cents per kWh from 8 a.m. to 10 p.m.
Super Off-Peak: 20 cents from 10 p.m. to 8 a.m.
My NBC:
These are my tracked charges:
All those credits went to zero at the beginning of the new year *POOF* gone.
The generation part:
With my true-up in may, I will probably get a +-$500 check for having a surplus of 14MWh over a year.
Minus the SCE fixed costs every month, I will still have a plus amount of $$ in the pocket, but not worth it as an ROI.
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,723
That totally sucks that they just wipe out the credits at the end of the year. What month is your #1 and then #12? My January 17th bill says I am in month 4. So that means my billing year begins in the middle of August. My system is too small to end up netting negative in a full year.
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,723
When I poke in to my system and look at the two legs independently, I do see some odd activity. But when I thought about what is going on, it does make sense.
The Enphase inverters and my XW-Pro both only push power at 240 volts across the two hot legs. I have mine programmed to try and zero the export, with just a small bias towards exporting power. When the load in the house is unbalanced, that causes one leg to be exporting while the other leg is importing. The net sum of both legs does end up being about zero, or in my case, exporting about 20 watts.
I have the power meters measuring all the power my house is using out of the main panel, and I command the XW to export that much power. If I have 800 watts on L1, but only 150 watts on L2, the inverter pushes out 970 watts across both legs. That ends up being 485 watts going to both legs. So L1 is still consuming 315 watts, and L2 is exporting 335 watts.
The line transformer from the utility is acting as our auto-transformer to create the neutral for us. If they ever program the power meter to read the two legs independently, then we would have a bigger problem.
I see this exact thing when I monitor in real-time the enphase inverters using ctmeters application. In my case, the outback skybox is trying to net out my house consumption.
fromport
Solar Addict
My dec22<->jan23 bill was my "settlement bill" for the SCE part.
This is also probably the only bill of the year my solar does not provide enough for the month:
Drove 784 miles with the EV in that time period, heated the home electric, ran electric clothes dryer, induction cooking.
All for $34.64
Not Bad [TM]
On the right you can see my transport cost surplus part was reset to 0 *POOF*
Generation I used up $60.51 from my credit:
After this month it can only get better again ;-)
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,723
That still would not work, and could make it worse. The XW-Pro inverter in my system does have a huge transformer, and yes, it does have the neutral center tap, but while on grid, it disconnects the center tap. The main reason for this is because of any voltage imbalance between the 2 legs. The utility transformer is a center tapped winding. If there was even a tiny difference between the L1 and L2 turns ratios between the two transformers, there would be a massive current fighting between them. And even if they did exactly match, then my XW's transformer would also be taxed with trying to voltage balance the grid for anyone on the same utility transformer.
The only fix for this would be to have individual 120 volt inverters that independently zero the grid on each leg. And depending on the loads, that might mean one leg is inverting to help drive loads while the other leg is in charging mode to prevent grid export.
Some of the cheaper "Low Frequency" inverters try to get around the issue by not using the grid side neutral at all. This way the loads after the inverter get their neutral only from the transformer in the inverter. This has a few limitations and issue as well though. Even while in pass through mode, your neutral current is still all coming from the transformer. That should be fine as long as the loads are still limited to what the inverter was meant to supply. In the case of the XW-Pro inverter that I have, it can pass through 60 amps, but the inverter is only rated to 28 amps. But I see an even bigger issue with not using the grid neutral. You can't have the inverter output neutral ground bonded. If the transformer input is coming from the grid, and you try to bond the neutral of the inverter transformer, you are again trying to tie the two neutrals together, but this time through the grounding conductors. Any imbalance will cause a huge ground current. And not bonding the neutral out could see several volts of potential between the neutral and true ground.
As long as the utility is only measuring the total sum of the L1 and L2 power, we are just fine having them balance our split phase at their transformer. The grid feed should be sized for a worst case imbalance when there was no solar. Admittedly, the neutral wire does seem a little smaller than the two hot legs. I can't read the gauge on them though. But it is possible that my 100 amp panel could have had a full 100 amps on one leg and zero amps on the other. That would mean 100 amps on the neutral as well. Lets say my solar is cranking it's full current limit of 16 amps and my XW-Pro is also pushing it's maximum 28 amps. So I have 44 amps at 240 volts being pushed into the grid. That is 10,560 watts. If I then had the full wattage being loaded all on one leg, what happens? In this worst possible case, the main L1 and L2 out of the house will see a balanced 44 amps, with one leg importing and the other leg exporting, and the neutral will see 88 amps back to the utility transformer. This will certainly not be an ideal situation. But it won't hurt anything. And there will still be no current draw or export on the primary of the utility transformer. The power meter should still zero out as it sees 5,280 watts going in on one leg and -5,280 watts going out on the other leg. Sum is 0 watts.
Obviously, we do not want to do this. As the load in my home increases, the balance always get's better. I have my larger 120 volts loads fairly well split between the 2 phases. The only issue is that I can't promise when certain loads will run alone. At light loading, I do end up with up to 1,500 watts of imbalance from time to time. Right now it is 2.5 amps of draw L1 and 4 amps of export on L2, so I must have about 1.5 amps imbalance back in the main panel. And looking at my power meters, I see 237 watts on one leg and just 53 watts on the other. Throw in my 20 watts of export and some power factor, and it's about right.
The sun is just starting to come up and make some power at the Enphase panels, so the XW is only pulling 60 watts from the battery bank right now. And the DC panels are pushing 50 watts into the battery bank, so net right now is only 10 wats out of the battery while I am exporting 20 watts to the grid. It would be much longer and it will begin charging up again. And as I typed that, I see the battery watts go positive before 8 am.
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 20,827
If you only had 240V load, no neutral connection, you would draw equally from L1 and L2, no imbalance to be penalized for, never export on one and import on the other. If load is imbalanced but still importing not exporting, still no penalty. If you export on one but import on the other, that is the situation to consider.
So your XW disconnects to avoid rebalancing the grid. It is a 240V only source, and your loads are 120/240V, could be imbalanced.
If you connected a 120/240V auto-transformer for balancing, it would partially rebalance an imbalanced load, but could also try to rebalance the grid. Something you could be charged for "If they ever program the power meter to read the two legs independently"
If you inserted a 240/480 to 120/240 isolation transformer between grid and XW, both 240V windings in parallel and connected to only L1 & L2, the secondary would give you 120/240V split-phase with derived neutral. Even if you put 10kW load on a single 120V phase and backfed 10kW into the opposite phase, the power would be equalized on the secondary (autotransformer) side with zero imbalance on primary side.
Like this, except call center tap 0V neutral, grounded:
It is an expense and efficiency loss we'd rather avoid, but if utility started crediting export on one leg at wholesale while charging for import on the other retail, it is what we would do.
Might be a special case for utility to invest in just for PV customers, because requires separate inductive pickup for each line. If meter routes L1 and L2 through same sensor in opposite directions, simpler design. I think they already ignore voltage imbalance, don't connect to neutral. Don't know if L1 and L2 current are sensed together or separate. For 3-phase, needs to be separate.
Hmm, if L1 current is in phase with voltage and L2 is out of phase, does meter read correctly (while ignoring neutral)?
