Great and inexpensive monitoring system and a problem to solve.

[scrambler]

I have an SMA inverter that specifically requires a certain firmware upgrade to be commissioned before TIGO optimizer commissioning so it would recognize TIGO optimizers properly. I have no idea why this specific order of commissioning is required, but SMA is very specific about it being required.

SMA acquired Tigo a while back, and my understanding is that the requirement you mention is so that their reporting can be integrated, not the functionality per say.

[Mike90250]

if your grid voltage rises from lots of solar pumping power into the grid, that can cause an inverter to either "throttle back" or shutdown for 5 min and restart when voltage is more normal

[oregon_phil]

SMA acquired a minority stake in Tigo. SMA was going the Tigo optimizer route, but now they seem to be going Tigo rapid shutdown with SMA Shadefix MPPT algorithms. I noticed SMA America has removed all Tigo devices from their website except for the JMS-F rapid shutdown unit.

I have an older -40 model that does not have the integrated Sunspec communication with the Tigo/SMA JMS-F rapid shutdown. My installer specifically told me that I needed a certain firmware version in order for the SMA inverter to operate correctly with the Tigo TS4-RO Optimizers.

Last year I had my year old SMA inverter replaced due to peculiarities. I spent over six months documenting each fault by day, by event, by screen shot, by etc.

One thing that happened a couple of months into my SMA inverter ownership is an auto firmware update. As we all know, if anything happens during a firmware update (like a power glitch) then all bets are off. The inverter may or may not work properly. My inverter happened to reboot on its own a random odd times.

Has Darfon pushed firmware updates to you without you knowing it? maybe they pushed a firmware update at 5 months. They should have a log.

If you are going to call Darfon, maybe they can tell you how their MPPT works in detail. I know my SMA optitrack used to look for a global MPPT point every 6 minutes, which is different than the local MPPT point. I could see power rhythms that had a 6 minutes intervals as shadows moved across my panels. That was before I had Tigo's installed. Now the power seems to be smoother over the time of day. I'm guessing SMA changed the optitrack MPPT algorithm once it figured out Tigo optimizers were in the mix.

[scrambler]

if your grid voltage rises from lots of solar pumping power into the grid, that can cause an inverter to either "throttle back" or shutdown for 5 min and restart when voltage is more normal

Wouldn't that also affect the Micro inverters?
What Voltage are we speaking of? Given these events last quite long, I can check the voltage while they happen.

[scrambler]

Has Darfon pushed firmware updates to you without you knowing it? maybe they pushed a firmware update at 5 months. They should have a log.

We did have firmware updates, and the fact the symptom got less severe after another one a few month later, makes me think it is a strong possibility.
The inverter also blew its battery board due to a flaw in the inverter a few weeks before it started and that was replaced, so there could be some leftover damage...

[Ampster]

Wouldn't that also affect the Micro inverters?
What Voltage are we speaking of? Given these events last quite long, I can check the voltage while they happen.

It is possible and worth seeing if there is a correlation. Also frequency changes can accompany changes in voltage. The typical scenerio is that lower voltage may be associated with lower frequency and higher voltages with higher frequency. My guess is the standard that determines the devices response is a range and that may explain different responses of devices to the same event. The standard may give you the answer about what voltage or frequency change light make a difference. I would start with UL 1741SA but there are others as well.

[oregon_phil]

I understand you did have firmware updates that you knew about. Were there any pushed to you that you didn't know about? Sometimes companies think some firmware changes are house cleaning, but in the wild these changes might not be so minor.

I'm with you, if the battery board blew and now I have troubles, I would be pushing to get a replacement inverter. How can Darfon prove your inverter problems are distinct from the battery board failure? Only by replacing the entire inverter. Your Darfon inverter is different and is not a factory version anymore. Darfon should want the inverter back in their factory so they can do all the testing they want.

Your manual has voltage ride throughs to determine when to trip offline. 13 secs 264 to 288VAC; 0.166sec above 288VAC.

I did take a closer look at one of your suspect graphs. If you look at it closely, there are times when the power output is correct (including one longer ~15 minute stretch); and a bunch of places that are wrong in between. Possibilities: Darfon and Tigo's have an interaction that needs to be addressed (Darfon firmware change); The Darfon is damaged from earlier battery board failure.

I am not a EE and don't understand the inner workings of the Tigos and how that might interfere with inverter MPPT.

Maybe an electrical type can explain all this stuff to us.

Darfon-modified.jpg

[scrambler]

It is possible and worth seeing if there is a correlation. Also frequency changes can accompany changes in voltage. The typical scenerio is that lower voltage may be associated with lower frequency and higher voltages with higher frequency. My guess is the standard that determines the devices response is a range and that may explain different responses of devices to the same event. The standard may give you the answer about what voltage or frequency change light make a difference. I would start with UL 1741SA but there are others as well.

If we are speaking about Voltage or frequency getting above or below the inverter threshold, the inverter has error codes for that, and it is not throwing any errors during the event, so that should not be the issue.
It also has error code for derating caused by over temperature.

[scrambler]

I understand you did have firmware updates that you knew about. Were there any pushed to you that you didn't know about? Sometimes companies think some firmware changes are house cleaning, but in the wild these changes might not be so minor.

I'm with you, if the battery board blew and now I have troubles, I would be pushing to get a replacement inverter. How can Darfon prove your inverter problems are distinct from the battery board failure? Only by replacing the entire inverter. Your Darfon inverter is different and is not a factory version anymore. Darfon should want the inverter back in their factory so they can do all the testing they want.

Your manual has voltage ride throughs to determine when to trip offline. 13 secs 264 to 288VAC; 0.166sec above 288VAC.

I did take a closer look at one of your suspect graphs. If you look at it closely, there are times when the power output is correct (including one longer ~15 minute stretch); and a bunch of places that are wrong in between. Possibilities: Darfon and Tigo's have an interaction that needs to be addressed (Darfon firmware change); The Darfon is damaged from earlier battery board failure.

I am not a EE and don't understand the inner workings of the Tigos and how that might interfere with inverter MPPT.

Maybe an electrical type can explain all this stuff to us.

* No there is no firmware push without my knowledge, I keep an eye on it, and the inverter has to shutdown and reboot for it.

* I have not provided the full report to the integrator yet, so Darfon is likely not aware of it yet, so no need to jump the gun, I am just getting some feedback before I go to them with the latest info I am gathering

* I am not sure what your point was about the Graph. The Graph is there to show the major production deviation of the Darfon while the Micro inverters keep producing normally, showing that what the Darfon is doing is not caused by a change in solar radiation, as all panels (18 connected to Darfon and 2 connected to Micro Inverters have the same exact exposure.

* I have checked with both Darfon and Tigo and they both confirmed there was no interaction problem between them and they are compatible. Certification only applies to the Rapid Shutdown system and although they have not formerly Certified that aspect, it is known to work properly.

I am continuing monitoring through this week to have at least 10 days of data on the new monitoring system, then I will submit all findings and we will take it form there

Thanks all for the contributions.

[oregon_phil]

My point with the graph: In what appears to be random data, there are periods of time when the Darfon is making the correct power (like that flat spot from ~11:47 to 12:57). On either side of that flat spot is lower power and a much noisier signal. What changed?

[scrambler]

That is the point of this whole thing

Something strange is happening with the Darfon that I cant explain with anything else than failure or FW bug so far

On some days (not all), when solar production is sustained, the Darfon will suddenly drop down the production significantly and fluctuates down there, with occasional ramp up to normal.
There are no external circumstances I can find so far to justify it other than malfunction as the Micro Inverters keep producing normally.

[oregon_phil]

I had originally asked about your inverter model to see if data could be extracted real time. Due to the short timeline, you could display your Darfon inverter real time variables on your PC and use your phone/camera to record the data. Maybe DC input values and AC output values. The power output discrepancies seem to happen every day so I don't think it would be a waste of time.

I believe your assumptions on Darfon strangeness from failure or FW bug seem to be on track. With 10 days of data they should agree with you. You have gone above and beyond most what most solar customers would do. Most solar customers probably wouldn't even know they had a problem. Just because Darfon has never seen the problem before doesn't mean that yours is the only case out there. You just happen to look at the data.

[DanS26]

If you could graph voltage over the same period you may find a correlation with the production curve anomalies.

[scrambler]

I had originally asked about your inverter model to see if data could be extracted real time. Due to the short timeline, you could display your Darfon inverter real time variables on your PC and use your phone/camera to record the data. Maybe DC input values and AC output values. The power output discrepancies seem to happen every day so I don't think it would be a waste of time.

I don't have access to real time data of the inverter, I don't even have access to minute data. All I have is the Tigo Minute data of the DC drawn from the panels, and the AC output data using my new monitoring system.

This crashing event DOES NOT happen every day. On cloudy days where the production fluctuates all the time, so far the DArfon production tracks the Micro inverter perfectly.
On some sunny days, the Darfon also produces perfectly and its production curve tracks the one of the Micro inverters.
But on some sunny days, it will have these crashing period.

I am yet to establish a percentage of the days when the problem happens. I cant watch the system all the time as I have other aspiration for my time
But now that I have installed the Monitoring of the Micro Inverters output, I will be able to go back and check all the days where the problem happens, and I will be able to find out if these are a few exceptions or a large percentage of the sunny days.

[scrambler]

If you could graph voltage over the same period you may find a correlation with the production curve anomalies.

Although the monitoring I installed measures Voltage, I cant download that data yet, it is a feature planned for the Future. I will check it manually on the next event though,

But the Integrator has access to the full inverter data monitoring, so that will be his job if they agree the problem is real and requires investigation (again, I have not forwarded my new dat yet...)