Schneider firmware really has issues

[romangeek]

So I basically got my XW Pro with MPPT 100/600 and eg4 batteries working... sort of! The intent with my system is that I have a minimum of batteries but I am also grid tied. The batteries should only come into play if the grid drops out and there is not enough PV or it is night time. And this basically works BUT...

I came to realize that though the MPPT was producing 2.5 Kw, it should have been producing 5 Kw on sunny days. So after a bunch of experimentation... I found that if the MPPt is rebooted then it produces 5 Kw... but after a while (different every time) it drops to half output. If I force it to Absorb charge mode, then for a while it jumps to 5 Kw. I can also put the MPPT to two-stage charging and sett he Absorb time to a max of 8 hours... that then produces maximum output for 8 hours. BUT what about the rest of the day!!!!!
Schneider tech support is not only useless, they really test one's patience with stupid suggestions, etc.

Has anyone else run into this?

 

[400bird]

Post all your settings.
Also, try disconnecting battery communication.

 

[ElectricIslander]

It's not clear from your post about your battery state of charge but do you realize that if your batteries are fully charged and you don't have loads on - your MPPT and solar system overall won't "produce" any power at all even if the sun is shining. The purpose of the charge controller is to only provide the power into the battery or loads that is needed at the time to keep the battery and inverter DC voltage at the set level.

On the other hand, if you have full batteries and significant loads - it's the inverters job to route power from either the grid or the batteries to the loads according to your settings. Is this the problem perhaps?

 

[romangeek]

Here are my settings
I should add a note to ElectricIslander that my inverter settings are to export up to 15 amps. So between the grid feed and the house loads, I am always using the maximum available from the PV.

 

[400bird]

Post all your settings.

Here are my settings

?
Emphasis added to my initial post. There's way more settings than that especially considering the separate settings for the XW's charger and the MPPT.

What is your battery voltage when the system is not doing what you want?

 

[romangeek]

The inverter XW PRO:








More for the MPPT:

 

[400bird]

Can you post your battery graph to try and show what your concern is?

The MPPT recharge voltage is set to 51.5 volts, so it would let the battery voltage drop to 5.15 volts before trying to recharge.


In the MPPT the bulk, absorb and float are all set to 54 volts, meaning that is as high as it will even try to charge.
The inverter is only going to sell to grid when above the grid support SOC, 95%
Is 54 volts below 95%? Might be.
If the MPPT isn't reaching the point where the inverter will start selling, all the inverter is going to to is support your loads. That leaves the MPPT no option but to taper off the output.

Can you better define the issue? A graph would be best.
If you can also define what you want the system to do, that would be helpful if you want assistance.

 

[romangeek]

The goal of my solar system is as follows:

• Keep the battery charged and never use it to support the load unless the grid goes down and there is no or insufficient PV.
• At night, draw power only from the grid.
• During the day, if PV is low such as clouds… supplement the house from the grid and NOT the battery.
• Harvest full potential of the PV array to feed the house and whatever is left over, sell back to the grid.

What works:

The inverter is working as it should and does all of what is stated above. The problem is with the MPPT… as stated below.

What does not work:

• Assume for this description of the problem, that there is plenty of sun to be able to provide 5 Kw of power and there are no clouds.
• Instead of the MPPT converting all of the 5 Kw (minus inefficiency losses) to feed the inverter, it converts only about half of the power… say 2.5 Kw.
• If I force the charge mode to be Bulk, then for some minutes, the MPPT jumps to convert all 5 Kw to feed the inverter and the grid then eventually it goes back to only 2.5 Kw.
• The same can be accomplished (as in # 3) by rebooting only the MPPT.
• So the mystery is… Why does the MPPT go into this “limiting mode”?
• My system is exactly the same and as my son’s and as a mutual friend of ours. ALL 3 SYSTEMS DO THE SAME THING!

The most frustrating thing is, that Schneider tech support is absolutely no help in this. Unfortunately, by this time of many hours experimenting with my system, I know more than they do! They will not pass me to anyone who would really know the system and be able to help. I think only a YOUTUBE video about this problem would get their attention.

It appears that the MPPT is really stupid and has no knowledge of the rest of the system. For example, it has not clue as to the fact that I have a LIPO4 battery with a BMS.

Again, the hardware I have is a MPPT 100/600, an XW PRO, Insight Home controller and EG4-LL batteries.

 

[400bird]

Excellent description!
That's amazingly helpful.

In short, when connected to the grid, the battery should stay at 100% and excess PV should be sold to the grid?

The XW will power itself from the battery over night (listed as 30-40 watts from memory) but the loads should pass through, as you are seeing.

• Instead of the MPPT converting all of the 5 Kw (minus inefficiency losses) to feed the inverter, it converts only about half of the power… say 2.5 Kw.

Is it by chance 3.6kw? Because that's what you have grid support (sell) limited to at 15 amps.

You didn't show it, but check that EPC is off and turned all the way up.

• So the mystery is… Why does the MPPT go into this “limiting mode”?

Likely because you are limiting the MPPT to 54 volts and also have the XW set about the same.
So, the MPPT is hitting it's target of 54 volts and starting to taper down charge current, which means it's limiting output, just as the XW is starting to hit 54 volts and able to start exporting.

• My system is exactly the same and as my son’s and as a mutual friend of ours. ALL 3 SYSTEMS DO THE SAME THING!

It's probably because you have the exact same settings commanding the exact same actions out of all 3 systems, that really doesn't surprise me.


As a test I'd recommend:
Turn off SOC control, leave grid support at 54 volts.
Set grid support sell amps higher, something that would allow the entire array to sell to the grid, 21 amps should cover it.
Set MPPT to 55 volts for all 3 charge voltage settings (bulk, absorb, float)
These settings aren't ideal, they should not be identical, but for testing it should work.

If that works, you'll want to reassess the voltage/SOC limits so they will still function off grid.

 

[MisterSandals]

Harvest full potential of the PV array to feed the house and whatever is left over, sell back to the grid.

Your settings start sell back at 6:15PM. If your batteries are full before that (like all afternoon), the MPPT would taper solar harvest to meet loads (right?).

Why not sell back anytime your batteries are full (or over 95%)?

 

[400bird]

Good catch, I missed that one.
But, the timers are dumb, it works backwards. Starting at 6:15pm it blocks selling until 7:00am

 

[romangeek]

So the mystery is solved and I finally have all three systems working!!!

So many parameter settings on this Schneider system are:

• Either undocumented as to how they are used.
• Interact in undocumented ways (e.g. – The breaker amps value for the generator affects how many maximum sell amps you can have)
• Or are plain wrong (e.g. – the way they tell you to setup “enhanced grid support” simply does not work. I had to stumble around to figure out the settings for the inverter that finally make it work properly.
• Or are plain misleading (e.g. – the Battery Temperature Coefficient has no documentation as to how it is used)

The bottom line issue is this… The MPPT is absolutely stupid and knows nothing about LIPO4 batteries. To boot, my battery is a EG4-LL which has its own BMS. The BMS communicates with the Insight Home unit and it can display all the parameters related to the battery including the temperature, voltage, SOC, amps in or out, etc. However, the MPPT is completely blind to this. I had the MPPT battery temperature probe sitting on the battery… the MPPT “thinking” that it is charging a lead-acid battery at the temperature of 70 degrees F. I had the temperature sensor connected because the Schneider tech told me to do so and the manual reads that way. My batteries are in the basement where it is cool. They also told me to set the Default Battery Temperature to Warm! So apparently, what was happening, is that the MPPT “thought” that what it thinks is a “lead-acid” battery should be current limited when it is 70 degrees! Add to this that the MPPT is simply feeding a DC bus to which the battery and the inverter is connected… it has no idea how much is going to the inverter and how much to the battery! So it has to assume that it is only going to the battery. This is plain crazy since the XANBUS has all the information about each element of the system… the MPPT, inverter, batteries, etc. I am a computer programmer and all this to be handled in a logical manner would be trivial.

So what is the fix? Simply unplug the battery temperature sensor, set the Default Battery Temperature to “cold” and BINGO… full power from the PV. The battery has a BMS to guard the battery as to temperature, etc. In my use-case which I spelled out in the previous post, the battery is never used and stays charged at 99%. If the grid should go down while there is PV power, the battery would not be used or minimally used… when the grid would come back, the battery would be recharged again.

The hardware seems solidly built and is efficient and handles tremendous surge currents such as my 6 horse-power compressor starting without a hitch! The starting current is 94 amps! However the software is BAD! I surely hope that they will scrap it and rewrite it. I think only public pressure will get them to make changes. After all my wasted hours, now exceeding 80 hours, after frustrating hours waiting for Tech support and getting useless answers, I would have to tell others looking at a system to consider other manufacturers!

 

[ChrisG]

So before the temp sensor change, the MPPT wouldn’t even cover the inverter loads because it thought that battery temp was too high? Would figure the XANBUS would have picked up that the inverter needed the DC power or could sell back.

 

[romangeek]

Unfortunately, all the information is available on the XANBUS but the MPPT makes no use of it! It could actually get the battery temperature from the BMS that is in the battery. The main issue however, is that the MPPT always thinks it has some sort of non-lithium battery and therefore tries to handle the temperature compensation itself. That said, I can't believe that 70 F would be high enough temp to limit current to the battery. They even have available from the XANBUS, how much current the inverter is using so they could do the math... BatteryCurrent = MPPTOutputCurrent - InverterInputCurrent. So if they wanted to do the temperature compensation, they could base it on the true battery current.
Perhaps the craziest thing of all is the pathetic technical support... I spent hours on the phone with them... my son spent hours on the phone with them and they were absolutely clueless about the problem, how the system works, what talks to what, etc.

By the way, the sell to the grid issue is a secondary one. My friend who had the worst issue because his battery was actually in a warm room, he was not selling to the grid (not approved yet) but he does have large loads and most of the loads were supported by 30% solar and 70% from the grid.

I hope that this experience will be helpful to others!

 

[400bird]

Unfortunately, all the information is available on the XANBUS but the MPPT makes no use of it!

That's documented by Schneider

It could actually get the battery temperature from the BMS that is in the battery.

Battery temp sensors are for lead acid installs, not needed for lithium as temp compensation is unnecessary.

The main issue however, is that the MPPT always thinks it has some sort of non-lithium battery and therefore tries to handle the temperature compensation itself.

You programmed the temp compensation into the MPPT, you could have just as easily programmed in 0 for the temp compensation. I'm really surprised and sceptical this made the difference. I thought temperature compensation was for low temps.

But, I agree there could and should be much better communication between the Insight and MPPT.

 

[Hedges]

Temperature can swing both ways. I think it is just a point-slope equation that is used.
Temperature compensation should reduce voltage for high temperature lead-acid battery.
That could drop charge voltage of SCC below export voltage of inverter.
I haven't used the feature, but for Sunny Island it says if DC charger charges above inverter's setpoint, then it will export.
(Only when connected to grid, of course, so that would hold voltage too high off-grid.)

 

[romangeek]

In reply to 400bird...

• Schneider needs to update the MPPT to actually be smart in the system and communicate properly.
• Nowhere in the documentation is there an explanation that with LIPO4 batteries the sensor should not be used! And their tech support people say it should be used!
• I had tried 0 for the temp. compensation and then the MPPT stopped producing output entirely! I found this strange but figured that 0 must be considered an illegal value, so I tried low values like -5 instead of the default -108... that too did not work as it fed only about .5 Kw to the DC bus.

Like I said, it is incredible that on top of poor software and documentation, Schneider does not train their support techs. One would think that they would have a section in their training that would include those "fancy new batteries called LIPO4"!

The following is from their release notes:
Release 120-240V Firmware 02.04.00 Build 0029
Released: 06 Feb 2023
Summary of Changes
Added IEEE1547-2018 and HECO SRD 2.0 2020 grid codes
Added SoC Based Enhanced Grid Support
Added Low SoC Hysteresis
Increased the LBCO setting range to 50Vdc
Improved frequency shift response time for AC Coupling
Improved BMS comms lost operation
Improved transitions to backup power for multi-unit systems to optimize load sharing
Known Issues:
Enhanced Grid Support in voltage control mode (SOC Control Disabled) has the following limitation:

• When MPPT in float and DC coupled PV production is less than the sell Amps, the battery voltage may dip and XW pro result in sell cycling. The workaround for this would be to set Absorption time to 8 hours, so that absorption timer doesn’t expire before end of the solar day.

Enhanced Grid Support with External BMS, SOC Control Enabled was added in this release but has the following limitation:

• The XW Pro output to AC is limited by the Max Discharge current of the BMS, even if available solar from the MPPT exceeds the BMS limit. Ensure the battery maximum discharge current is greater than the maximum MPPT output if using this feature.

My comments: First, the main feature of this inverter is that it is "Hybrid" ... yet the main feature that makes it hybrid does not work properly. Second, their workaround solutions listed above do not work.
In their manual, they claim that to enable enhanced grid support one should set the "Grid support Voltage" to 64 as though that is a magic switch to turn it on - that does not work either (it deep cycles the battery continuously)! My son was on their tech support call for 2 hours and in "guiding" him to set up the system for EGS. Their "guidance" did not work at all! It was only by my experimentation that I figured out what values for the many parameters would actually make the system work. So I wonder how much testing they actually do??

So in the end I have a system that works but at an incredible cost of time!

 

[ChrisG]

But, I agree there could and should be much better communication between the Insight and MPPT.

@400bird
Is there any benefit to getting a Schneider MPPT outside of string voltage over a midnite where you have lower voltage but benefit of ground and arc fault?

 

[400bird]

In reply to 400bird...

• Schneider needs to update the MPPT to actually be smart in the system and communicate properly.

Agreed, but "communicate properly" is not how I'd word it. I think they need to finish the programming, as you found in the release notes, it's like they stopped part way through just to hit a release date. The communication is there, you can see data in the Insight.

• Nowhere in the documentation is there an explanation that with LIPO4 batteries the sensor should not be used! And their tech support people say it should be used!

See below, I found it at least twice in two different documents in about 3 minutes.

• I had tried 0 for the temp. compensation and then the MPPT stopped producing output entirely! I found this strange but figured that 0 must be considered an illegal value, so I tried low values like -5 instead of the default -108... that too did not work as it fed only about .5 Kw to the DC bus.

That is odd

Like I said, it is incredible that on top of poor software and documentation, Schneider does not train their support techs. One would think that they would have a section in their training that would include those "fancy new batteries called LIPO4"!

Part of the problem is that they don't consolidate and update the documentation for each software update.

 

[400bird]

Is there any benefit to getting a Schneider MPPT outside of string voltage over a midnite where you have lower voltage but benefit of ground and arc fault?

I went with a Midnite charge controller to get the Arc Fault Detection. To get Arc Fault Detection with a Schneider MPPT you need to purchase the separate "MPPT RS disconnect"
When Schneider finishes/fixes the grid support and BMS communication to allow the XW and MPPT to coordinate and export excess PV it may be worth it.

Also, having all your solar components on one interface is a very nice feature.

Schneider does make a 150 volt MPPT, which is what I would have gone with.