Apparent discrepancy between metering systems

[sathornlover]

On 7/20/2024 at 4:10 PM, 4myr said:

@sathornlover

 Have you checked the PEA smart app? There you can see your monthly bills. I don't know if they will report credit bills as well.

Thank you for suggesting the PEA Smart App. I had installed it before but didn’t realize its potential benefits. Upon reinstalling, I found that the app allows access to all past bills, including the one for July, issued on August 1. Future bills can also be sent via email if desired. There is really nothing negative to report about this app and all the relevant data is presented relatively well, but also less comprehensibly if you look at it in more detail. Another positive thing I noticed is that the service fee is 24.62 baht, just like the old meter. This solves at least one of our problems regarding digital access to invoices, so thank you once again for this very useful information, which we unfortunately didn't receive from the PEA during our last visit for whatever reason ...

 

 

Even at first glance, I noticed that we now have a completely new problem. Strangely enough, there is a considerable discrepancy between the units used 302 shown on the bill for the month of July and what our smart logger from Huawei records (208).

 

 

 

 

 

Then on closer inspection of the July bill, I also noticed the feed was off by 100 units, this time in my favor. Around 790 versus 699 units.

I also found the first bill suspicious, which was presented to us during our last visit but we weren’t allowed to photograph. My rough calculations suggested a payout of around 1000 Baht, but the bill showed only about 650 Baht. This bill combined the last days of May, when the new digital meter was installed, with June, and I didn’t want to dispute it.

 

For the July bill, I expected a payout between 500-700 Baht, but it turned out to be only 357 Baht. I was confused and reached out to my PV installer via Line last night:

 

“Dear Anan,

 

I apologize for reaching out to you again, but I’ve noticed a significant discrepancy in the last two monthly statements from PEA. The discrepancy is between the units of electricity consumption recorded on the PEA bill and the units logged by the Huawei dongle logger. The difference is so substantial that I can’t explain it, and I doubt I’ll get satisfactory answers from PEA.

 

Basically, in my view there are three possibilities:

 

1. PEA is not correctly recording the units of electricity we consume from their grid.

2. The Huawei system is not accurately recording the consumed units from the PEA grid or is misconfigured (for example, the system might prioritize feeding electricity back to the grid over meeting your household’s consumption needs first).

3. I’m making a calculation error, or PEA is using an algorithm to calculate the units that I’m not aware of.”

 

 

I received an immediate response and assurance that he would look into it today, Sunday. He not only reviewed the July bill but also the combined May/June bill, created a data sheet, and advised me to present it to the PEA during our next visit, as he found both bills suspicious as well.

 

For those unfamiliar with digital meter bills, unlike the old analog meters, the digital ones have three reading marks: P (peak), OP (off-peak), and H (holidays, i.e., weekends and public holidays). These three values add up to 302 units. The upper part of the bill strangely shows values from the old analog meter, which don’t factor into the calculation and make no sense. I hope to get a reasonable explanation for this oddity soon …

 

Since I have no need for a TOU tariff and we charge our EVs for almost zero baht during the best daytime hours with our ABB charger, all units we draw from the PEA grid are charged at the standard electricity rate.

 

 

 

 

[lom]

You have a discrepancy of 94 unit in import but you do also have the same 94 unit discrepancy in your export.

At a first glance that seems to indicate that the 94 units you import flows out directly and the inverter doesn't see any of that.

Maybe it is just a coincidence that the error is the same in both directions..

 

[sathornlover]

14 hours ago, lom said:

You have a discrepancy of 94 unit in import but you do also have the same 94 unit discrepancy in your export.

At a first glance that seems to indicate that the 94 units you import flows out directly and the inverter doesn't see any of that.

Maybe it is just a coincidence that the error is the same in both directions..

 

I don't think it's a coincidence, I really didn't notice it. I had 690 export units in mind, thank you for taking the trouble to look into it more closely 😉 and for your further conclusions. I would like to give it more than a thumbs up! I'll just wait for the bill for this month and instead discuss the problem further with the installation company...

[sathornlover]

6 hours ago, sathornlover said:

I don't think it's a coincidence, 

"Coincidentally," my ABB charger used 186.7 kWh in July, which is almost exactly the sum of the export and import discrepancy in the July electricity bill. Another „coincidence“, or are we already on the trail of the culprit?

 

[UWEB]

2 hours ago, sathornlover said:

"Coincidentally," my ABB charger used 186.7 kWh in July, which is almost exactly the sum of the export and import discrepancy in the July electricity bill. Another „coincidence“, or are we already on the trail of the culprit?

 

Where do you get this Report from?

[lom]

3 hours ago, sathornlover said:

"Coincidentally," my ABB charger used 186.7 kWh in July, which is almost exactly the sum of the export and import discrepancy in the July electricity bill. Another „coincidence“, or are we already on the trail of the culprit?

I don't think it is related and I have a problem understanding the export discrepancy, that's why I suggested that the same units might be a coincidence.

The inverters metered import can be wrong if there is a consumer connected between the PEA smart meter and the inverters external current transformers which measures power flow and its direction. The inverter will not see the power going to that consumer because it is on the wrong side, before the current transformers

I can't understand though how you can export more than what the inverter think it has exported.

Is your system a single phase or three phase system? Where/how is your ABB EV charger connected?

[sathornlover]

perhaps a moderator reading this could separate this discussion thread because it has nothing to do with the actual topic and it was not my intention to steer it in that direction.
Sorry about that!
 

The report is a statistic of my three-phase ABB charger, which also supports load balancing, for our electric car.

 

Two years ago I ordered a three-phase electricity meter for this. Then a year later a three-phase solar system was installed. The original house installation, which was still running on the old one-phase electricity meter, was also integrated, so that everything now runs on just one electricity meter and finally two months ago the switch from analog to digital electricity meter.

[Crossy]

43 minutes ago, sathornlover said:

perhaps a moderator reading this could separate this discussion thread because it has nothing to do with the actual topic and it was not my intention to steer it in that direction.
Sorry about that!
 

The report is a statistic of my three-phase ABB charger, which also supports load balancing, for our electric car.

 

Two years ago I ordered a three-phase electricity meter for this. Then a year later a three-phase solar system was installed. The original house installation, which was still running on the old one-phase electricity meter, was also integrated, so that everything now runs on just one electricity meter and finally two months ago the switch from analog to digital electricity meter.

 

Done.

 

If you want to adjust the topic title please let me know.

[lom]

1 hour ago, sathornlover said:

Two years ago I ordered a three-phase electricity meter for this. Then a year later a three-phase solar system was installed. The original house installation, which was still running on the old one-phase electricity meter, was also integrated, so that everything now runs on just one electricity meter and finally two months ago the switch from analog to digital electricity meter.

I would check your current transformers, it could be that one is faulty or disconnected which would make the inverter only measure power on 2 phases.  That is the only thing that  came to mind which will cause a discrepancy in both directions.

 

On 8/4/2024 at 8:28 PM, sathornlover said:

The upper part of the bill strangely shows values from the old analog meter, which don’t factor into the calculation and make no sense.

I think those are reactive power which they just measure but don't charge you for. I do also have a digital meter but not an export one and therefore get a bill looking a bit different from yours but it shows a monthly slightly increasing value for reactive power.

All digital meters display this, analogue meters can't measure it.

[lom]

2 hours ago, lom said:

That is the only thing that  came to mind which will cause a discrepancy in both directions.

 

but I still don't like that the units in error are the same in both directions. There is something fishy going on..

[sathornlover]

21 hours ago, lom said:

I would check your current transformers, it could be that one is faulty or disconnected which would make the inverter only measure power on 2 phases.  That is the only thing that  came to mind which will cause a discrepancy in both directions.

I have to add to my three-phase ABB charger that I am not at all sure how it was actually wired, because two years ago the first charger installed was a no-name single-phase charger and this was then integrated into the solar system installation as already described. The single-phase charger was then simply replaced six months ago by the three-phase ABB charger. I have now asked the installation company to send a technician to me next month after the August electricity bill, as this is when the annual inspection should have been carried out anyway.

 

Quote

 

I think those are reactive power which they just measure but don't charge you for. I do also have a digital meter but not an export one and therefore get a bill looking a bit different from yours but it shows a monthly slightly increasing value for reactive power.

All digital meters display this, analogue meters can't measure it.

 

 

For this helpful and comprehensible answer alone, which even my installation company still owes me, I pity again that I can only give it a single thumbs up.

 

Quote

but I still don't like that the units in error are the same in both directions. There is something fishy going on..

 

Full agreement but at least a clear symptom to work on now 🙂

 

 

[lom]

Is your ABB charger able to feed back from the cars battery to the grid (V2G, vehicle to grid) ?

If it is then I think I know what is wrong.

[sathornlover]

The ABB charger TAC-W22-G5-R-C-0 was a "free" gift from BYD and to my astonishment is one of the better wallboxes, but the v2g/v2h capability is missing from this device, neither in the descriptions nor on the net can I find any reference to it.

 

[lom]

Ok, so then it isn't your ABB charger that is responsible for the additional export so it must be your inverter exporting more than itself is aware of.

I'd say that is very unusual, close to  impossible..  I'm looking forward to hear the result in end of August.

[UWEB]

9 hours ago, lom said:

Ok, so then it isn't your ABB charger that is responsible for the additional export so it must be your inverter exporting more than itself is aware of.

I'd say that is very unusual, close to  impossible..  I'm looking forward to hear the result in end of August.

I have had a Huawei Inverter as well. Even set to no export I had appr. 4 kwh ghost Export every day. You could not see it on the old disk Meter, but I have got a digital ones installed by PEA. After this my monthly Bill increased by 25%/

[lom]

47 minutes ago, UWEB said:

I have had a Huawei Inverter as well. Even set to no export I had appr. 4 kwh ghost Export every day. You could not see it on the old disk Meter, but I have got a digital ones installed by PEA. After this my monthly Bill increased by 25%/

Something must be wrong in your description, you don't get charged for true or ghost export.

Export is separately metered by PEA and you either get paid for it if you have an export contract with them or you will get nothing without a contract. It never affects your imported units.

 

Edit: Where do you see those 4 units of daily export?

[sathornlover]

On 8/7/2024 at 6:22 PM, lom said:

Ok, so then it isn't your ABB charger that is responsible for the additional export so it must be your inverter exporting more than itself is aware of.

I'd say that is very unusual, close to  impossible..  I'm looking forward to hear the result in end of August.

 

As was to be expected, there is also an equally large discrepancy in August,
 this time of ~139 units between the two meter systems: import 424 units vs. 284.6 units and export 1025 units vs 886 units.
The charger for the EVs shows ~261 units for this period, which this time does not quite match the export/import discrepancy values
when these are added together. 278/261
In the month of July: 188/187

 

 I won't be in Thailand for the next whole month, so only my wife will be charging her car, which she only uses for the commute to work - that's an average of ~100kWh, so the discrepancy between the export/import values should be smaller, if the assumption that this problem is related to the charger is correct.

 

An electrician from the installation company is coming by on Saturday ...

 

[sathornlover]

Unfortunately, I have not made any progress on this matter to date. Visits from PEA and the installation company have not produced any useful findings either. The first electrician who checked the system was convinced that the whole problem had nothing to do with the EV charger. The second electrician, who was actually the main person responsible for installing the whole system and only had time in December, was much more open to my assumption that it must have something to do with the charger, but couldn't find anything else either and was of the opinion that I should be happy with the current situation and the abundant power generation ... The visit from PEA unfortunately took place in my absence with the same negative result and the attempt not to use the charger during my almost 2 month absence failed because my wife was not willing to charge her Ev outside.

 

Today I received a message from the head of the installation company in which he apologized profusely for being so late and told me that they had thought about it again and had now come to the conclusion that it might actually have something to do with the charger.

 

He sent me two screen shots showing the entire course of the day from February 1, 25 of phases a, b and c. It is noticeable that phase a is in the negative range during the time when the charger is in operation. From this it is assumed, and I quote “I think EV charger may mis-install. From the plot, it's power of phase a, b, c character. It seem like one of them used heavily compare with other phase.”

 

To be honest, I myself am not in a position to draw any further conclusions from this. In the hope that I can get some helpful suggestions here, I am uploading the 2 screenshots here again.

 

 

 

[sathornlover]

As I've already emphasised, I'm not in a position to interpret the diagram from a professional point of view, but the conspicuousness of phase A correlates with the charger when it's in operation. Not only that, it even seems to support my suspicion that one phase is only drawing current from the grid and is being looped past the load, in this case the car battery, and inverter and passed straight back into the grid instead! This would also explain the strange coincidence that the sum of the discrepancy between imported and exported current units is relatively equal. As I am unfortunately not a real expert in this field, I am not yet sure whether this extended thesis would even be tenable from an electrophysical point of view....

[lom]

What I can understand from the charts is that the inverter sees that the  EV charger only uses 1 phase . Are maybe the other 2 phases connected on the wrong side of the current transformers that measures import/export?

[sathornlover]

2 hours ago, lom said:

What I can understand from the charts is that the inverter sees that the  EV charger only uses 1 phase . Are maybe the other 2 phases connected on the wrong side of the current transformers that measures import/export?

That is correct. The ABB charger is actually operated in single-phase mode, although it can also be operated in 3-phase mode. Does this show the green phase A curve in the diagram?

[lom]

1 hour ago, sathornlover said:

That is correct. The ABB charger is actually operated in single-phase mode, although it can also be operated in 3-phase mode. Does this show the green phase A curve in the diagram?

To me it looks like everything below the 0 line is import and everything above it is export. You see the yellow phase has almost only daytime usage (8am shower?) and only a little of it, the blue phase used around the clock and the green phase mainly being used for EV charging (and something else? between 9am-10am). Nothing in the graphs explains the metering discrepancy.

[lom]

The yellow phase has the highest export it is almost unused for consumption. The blue phase follows it but slightly below because it carries almost all of your consumption. The green phase do also mainly export like the yellow phase and the curves are overlaid on each others except for when you use the green phase for EV charging. The EV charger takes so much power that the inverter generated power is not enough so you had to import from the grid. You can, by measuring the distance between yellow and green see that you charge around 5KW to the car in the beginning of the charge cycle and somewhere around 3KW in the end..

 

Edit: are these graphs from a not so sunny day or do you have a small solar system?

[lom]

2 hours ago, sathornlover said:

The ABB charger is actually operated in single-phase mode, although it can also be operated in 3-phase mode.

Have it connected in 3-phase mode so that you can use your own produced electricity from all 3 phases instead of buying electricity from PEA!! You are selling blue and yellow phase power to PEA at a lower price than what you pay for the import from them on the green phase.

[lom]

47 minutes ago, lom said:

Edit: are these graphs from a not so sunny day or do you have a small solar system?

My brainfart.. Ylou have a system that can at least produce 3x2.5KW by looking at the graphs maybe more if the graphs were from a not so sunny day, and that is enough for charging an EV with 5KW as you do.

Use your own produce before importing, export only your surplus!

 

[sathornlover]

50 minutes ago, lom said:

Have it connected in 3-phase mode so that you can use your own produced electricity from all 3 phases instead of buying electricity from PEA!! You are selling blue and yellow phase power to PEA at a lower price than what you pay for the import from them on the green phase.

But now I'm very surprised. The third diagram below shows that only at 3:00 in the afternoon until 4:00 in the afternoon electricity was consumed again partly from the power grid, between 8:00 in the morning and 3:00 in the afternoon the electricity was generated exclusively by the sun. dark green indicates that the electricity generated by the sun was consumed and yellow by the power grid. Light green is the electricity that is exported. It was a normal sunny day for this month and the system produced 47kWh on this day, which is normal for this time of year. However, it is often over 50kWh. When the sun goes down, I can use the app to manually set the battery to charge less, so that I usually use 90-100% of the electricity generated by the PV system.
 

 

the 7 kWh from the grid is the average load that I have even when I am not using the charger

[sathornlover]

2 hours ago, lom said:

Have it connected in 3-phase mode so that you can use your own produced electricity from all 3 phases instead of buying electricity from PEA!! You are selling blue and yellow phase power to PEA at a lower price than what you pay for the import from them on the green phase.

I think you're right. It's definitely worth trying to connect the charger in three-phase mode and checking all the wiring, at least as far as the fuse box. I don't need the 11 KW charging speed. But maybe I'll be lucky and the problem will be solved or at least I can rule out this source of error, However, I am not yet sure who I will be hiring to do this ….

[lom]

Do you know how many KWh you charged your EV with on 1st of February?
My rough calculation based on your graphs is that you used around 4 hours of average 1.5KW from PV and 4 hours of average 2.5KW from PEA or in other words a total of 16KWh. 
Is that a sensible value?

[sathornlover]

17 minutes ago, lom said:

Do you know how many KWh you charged your EV with on 1st of February?
My rough calculation based on your graphs is that you used around 4 hours of average 1.5KW from PV and 4 hours of average 2.5KW from PEA or in other words a total of 16KWh. 
Is that a sensible value?

About 21kWh 

„Est. solar energy 0 kWh (0 %)“ ?????

 

The device does not seem to actually detect the source of the energy although it claims to, cause is not configured with the grid meter, probably not compatible too …

[lom]

Ok, your graphs had no time markings so I guestimated the time and missed half an hour and I did not count the short morning charge.  so 21KWh it is. 18Kwh during the noon-afternoon pass of 4.5 hours is 4KW per hour which is more than your PV could deliver that day on a single phase right?  So import is needed.

 

Your blue phase import between 00-08 and 16-24 is 16 hours of an estimated average of 0.4KW ie 6.4KWh, which is not far from your stated 7KWh or the actual 7.69KWh 1st of February.

Your blue phase power usage between 08-16 is 8 hours of an estimated average of 0.5KW or around 4KWh.

(I have guestimated average visually by looking at the distance from the yellow phase.)

 

So 21 + 7.7 + 4 = 32.7 which matches well the 31.8 that was metered.

 

Now, in order to charge the EV with 4KW you use an average of 1.5KW from PV and an average of 2.5KW from grid import (unless you have batteries which you are not allowed to have when exporting to PEA) so 2.5KW for 4.5 hours is 11.3KWh import.

Add to that the nightly import of 7KWh  for a total of 18KWh import but your inverter says 7.69.... 🤥

 

I blame the discrepancy you have from the way your  inverter measures power,  it seems to me that it counts the import correctly during night when there is no PV (ie no export) but fails in the day when there is PV and export.

It can't get it right when there is export and import at the same time?