Schneider xw pro 6848 not powering up with eg4 lithium battery

[Mr-Sandman]

"Too long" a delay is only if somebody gets impatient, like inverter automatically starts operating with loads connected.

In the XW's case, there can be too long. It doesn't wait to power on and has some "bootup" current as it goes thru it's internal test, EVEN if loads aren't connected. Another words, as soon as it see's DC power, it's going to start pulling DC amperage to "boot up". This is in ADDITION to the in-rush for the capacitors all inverters have.

 

[Hedges]

Yep. tried it that way as well. Tried it with just one battery and every combination imaginable with same result every time.

Determine how fast inverter AC output comes up, after DC is provided by a battery that is able to start this inverter. Time from closing breaker to lightbulb illuminating, for instance.
Determine what DC voltage rises to in that amount of time, as inverter capacitors are precharged by the battery from SS.

It needs to have pretty much full voltage, then hard connected by BMS, prior to current being drawn.
If precharge isn't complete when BMS fully connects, there will still be a current surge instantaneously completing charge of caps.
If BMS isn't fully connected when inverter begins drawing higher current, voltage on capacitors will drop back down (linear ramp) then there will be a current surge when BMS connects.

 

[Mr-Sandman]

That's what it seems, that the pre-charge isn't finished when the inverters is trying to pull current to boot, has low voltage and doesn't finish "booting". Maybe needs less pre-charge timer OR more pre-charge current. Timer can be changed easily it seems, but that might not solve it, if it's not on long enough to charge the capacitors before locking.

 

[Mr-Sandman]

@Koldsimer one other thought that just popped in my head while explaining the "bootup" test it does to Hedges. What firmware are you running in your XW Pro? Maybe that's been covered, but I'm at the latest on mine. They certainly could change the bootup procedure in their firmware, so that it waits longer or draws less possibly. Latest is 1.11.01bn49

 

[Hedges]

In the XW's case, there can be too long. It doesn't wait to power on and has some "bootup" current as it goes thru it's internal test, EVEN if loads aren't connected. Another words, as soon as it see's DC power, it's going to start pulling DC amperage to "boot up". This is in ADDITION to the in-rush for the capacitors all inverters have.

Yes, that is why I suggested earlier timing within inverter firmware could be the problem. You want a longer delay, a programmable delay, or handshaking. Coordination of inrush limiting with downstream loads determining "power good" and turning on is a problem in product designs.

If "bootup" current can be determined, precharge resistor value (ohms) and wattage could be selected accordingly. I've suggested a coil of Romex for resistor of DIY precharge circuits, could even supply current for modest inverter output wattage.

For BMS built-in precharge, if inverter settings can't be changed it is a race. Careful adjustment might make it work.

 

[Koldsimer]

That's what it seems, that the pre-charge isn't finished when the inverters is trying to pull current to boot, has low voltage and doesn't finish "booting". Maybe needs less pre-charge timer OR more pre-charge current. Timer can be changed easily it seems, but that might not solve it, if it's not on long enough to charge the capacitors before locking.

One of the things i did in the past was to measure voltage at the terminals in my ebox when attempting to power up. As soon as I connected breaker, voltage immediately dropped from 52.8 to low 30's and just hovered there while the inverter clicked away and wouldn't start.

 

[Koldsimer]

@Koldsimer one other thought that just popped in my head while explaining the "bootup" test it does to Hedges. What firmware are you running in your XW Pro? Maybe that's been covered, but I'm at the latest on mine. They certainly could change the bootup procedure in their firmware, so that it waits longer or draws less possibly. Latest is 1.11.01bn49

That was one of the first things Schneider suggested. I'm running that exact version.

 

[Hedges]

If AC loads are disconnected, I think an external DIY precharge circuit (with optimal resistor value and size) enabled after the battery's BMS is done ought to work.
Of course, that's not what you're looking for in a ready-made battery.

One of the things i did in the past was to measure voltage at the terminals in my ebox when attempting to power up. As soon as I connected breaker, voltage immediately dropped from 52.8 to low 30's and just hovered there while the inverter clicked away and wouldn't start.

"ebox" being between the SS battery and Schneider inverter? Sounds like voltage drop through battery's precharge circuit due to current draw. Meaning current starts flowing before fully precharged and FETs on hard.

That was one of the first things Schneider suggested. I'm running that exact version.

Check time for inverter to turn on output (with other battery that works), as I suggested earlier.
Can you measure (i.e. clamp meter) current draw during boot-up? Detect time when this starts? If a moderate current draw begins instantly, any precharge circuit must supply that current while maintaining voltage.

Can you try my suggestion of putting a switch between battery and inverter. Switch open, turn on battery. Use a coil of romex as resistor to precharge, then close switch. (Wire nut joining white and black at one end of romex, use white and black at other end to precharge, so the coil doesn't act as an inductor.)

 

[Quattrohead]

Hedges you are over thinking this
The evidence 100% says the pre-charge delay is wayyyy too long for this particular inverter in this situation. I don't know why SS are getting a different result, no AC load perhaps ?
AFAIK the capacitors are right on the battery input on all inverters I have ever seen.
I bet you a 6 pack the delay needs to be 100-1000ms.
The battery and inverter are chasing each other around in a non booting loop.

 

[Solar Guy Richard]

I understand. But as you can see in @Koldsimer video his is not. Additional a Discover Lithium battery and his old Lead Acid batteries power it up just fine, so there is something in HIS system that the EG4's are not working with right.

Totally agreed. I'm simply saying that if it was isolated to just this setting specifically, we could duplicate it.

 

[Solar Guy Richard]

Hedges you are over thinking this
The evidence 100% says the pre-charge delay is wayyyy too long for this particular inverter in this situation. I don't know why SS are getting a different result, no AC load perhaps ?
AFAIK the capacitors are right on the battery input on all inverters I have ever seen.
I bet you a 6 pack the delay needs to be 100-1000ms.
The battery and inverter are chasing each other around in a non booting loop.

We did have a load on the inverter when we tried 1, 2, and 3 batteries. It wasn't a huge load, but it was a load nonetheless.

 

[Koldsimer]

We did have a load on the inverter when we tried 1, 2, and 3 batteries. It wasn't a huge load, but it was a load nonetheless.

By default, the inverter is setup to turn on in invert mode and immediately start working. I have tried putting it into standby before shutting down to minimize the initial startup load and restarting that way with same results.

 

[Koldsimer]

The battery and inverter are chasing each other around in a non booting loop.

That's exactly what it seems like is happening.

 

[Koldsimer]

If AC loads are disconnected, I think an external DIY precharge circuit (with optimal resistor value and size) enabled after the battery's BMS is done ought to work.
Of course, that's not what you're looking for in a ready-made battery.



"ebox" being between the SS battery and Schneider inverter? Sounds like voltage drop through battery's precharge circuit due to current draw. Meaning current starts flowing before fully precharged and FETs on hard.

It's a midnite power center designed for this inverter iirc.

Check time for inverter to turn on output (with other battery that works), as I suggested earlier.
Can you measure (i.e. clamp meter) current draw during boot-up? Detect time when this starts? If a moderate current draw begins instantly, any precharge circuit must supply that current while maintaining voltage.

Can you try my suggestion of putting a switch between battery and inverter. Switch open, turn on battery. Use a coil of romex as resistor to precharge, then close switch. (Wire nut joining white and black at one end of romex, use white and black at other end to precharge, so the coil doesn't act as an inductor.)

With the fla bank, the inverter turns on pretty much instantly. I know that's not much for a technical answer, but it seems as soon as breaker is closed, the unit begins operating. At least it turns the display on and the fan begins spinning instantly. It does take a few seconds to power ac loads but i'm guessing that's normal as power is being qualified.

I don't have a clamp meter or a roll of romex. I've asked a few buddies and nobody has a roll of wire so i can try this.

 

[Hedges]

Hedges you are over thinking this
The evidence 100% says the pre-charge delay is wayyyy too long for this particular inverter in this situation. I don't know why SS are getting a different result, no AC load perhaps ?
AFAIK the capacitors are right on the battery input on all inverters I have ever seen.
I bet you a 6 pack the delay needs to be 100-1000ms.
The battery and inverter are chasing each other around in a non booting loop.

So resistor value has to be low enough to get the job done in a short time. I think the 50 ohm resistor mentioned earlier could require seconds. Don't think we know the resistor value (or current control if by FET) in SS's BMS.

I think I confirmed by measurement that capacitors of my Sunny Island are after its circuit breaker. Another guy with one swore capacitors were connected to battery terminals before the breaker, and said SMA told him the switch only enabled operation (to me, that sounds like European 8.0 model with membrane keypad, not present US model with big honkin' 2-pole breaker.)

It's a midnite power center designed for this inverter iirc.


With the fla bank, the inverter turns on pretty much instantly. I know that's not much for a technical answer, but it seems as soon as breaker is closed, the unit begins operating. At least it turns the display on and the fan begins spinning instantly. It does take a few seconds to power ac loads but i'm guessing that's normal as power is being qualified.

I don't have a clamp meter or a roll of romex. I've asked a few buddies and nobody has a roll of wire so i can try this.

That voltage drop at ebox means current is being drawn, and a resistor is present between ebox and battery. Therefore, I suggest powering up battery first, with switch isolating either ebox or inverter. Then apply a low-resistance high wattage precharge resistor (e.g. length of Romex), then close switch.


"A few seconds" helps. Prior to that, it is drawing enough current for internal electronics and fan. Precharge resistor needs to be low enough to supply whatever that current is.

Any wire will do it. I wanted to calculate IR drop such that with about 50V shorted out by resistance of wire, it wouldn't exceed maximum current spec of BMS. If a single strand of wire, fold in half, then coil (or not) as desired. Even skinny phone wire could do it, but you'd need to close switch before inverter starts actually delivering AC. I know that 12 awg could supply a LOT of current, just gets hot after a while. Fine for multiple seconds so long as IR drop doesn't reduce voltage too low at inverter.

 

[Mr-Sandman]

The problem with everything your suggesting Hedges, the pre-charge circuit that "we think" is causing the low voltage is IN the batteries and hooking external cables and resistors up doesn't bypass that. We don't know what triggers the built in Pre-charge timer. @signaturesolarrichard could you tell us what triggers the pre-charge circuit timer?

A good easy thing to verify is what that the "CHG_UT_Delay" setting indeed = 4000ms in his batteries, to start with. Knowing the resistor size used in the EG4 LifePower4 batteries would then be useful and that would tell us at least how much power the circuit can provide while active.

 

[Horsefly]

I've not been keeping up with all this, but just read the most recent few pages. My observations:

• @Signature Solar seems to have finally re-engaged. That's good, but...
• @Koldsimer has posted a video demonstrating his system not starting up. I don't think I've seen any acknowledgement from @Signature Solar / @signaturesolarrichard that they have watched that video. If I were cynical I would say that they showed via their video that it works, so if it doesn't for @Koldsimer it must be his problem, not theirs. Sorry if that sounds judgmental.
• It would be pretty easy reading only the last half of this thread to think that @Koldsimer was the only one with this problem, but I know I've seen others with the exact same issue. Perhaps those people need to either speak up here or at least be in a list that could be pointed out to Signature Solar.
• Finally, the thread seems to have shifted back into many of us trying to help Koldsimer by workarounds or more debugging. Since Signature Solar isn't actively engaged in the thread, all these posts are probably muddying the waters and they might scan it and drop back out again.

From a distance, it doesn't seem like this is getting fixed, or even that Signature Solar believes any longer that there is something for them to fix.

 

[Solar Guy Richard]

I've not been keeping up with all this, but just read the most recent few pages. My observations:

• @Signature Solar seems to have finally re-engaged. That's good, but...
• @Koldsimer has posted a video demonstrating his system not starting up. I don't think I've seen any acknowledgement from @Signature Solar / @signaturesolarrichard that they have watched that video. If I were cynical I would say that they showed via their video that it works, so if it doesn't for @Koldsimer it must be his problem, not theirs. Sorry if that sounds judgmental.
• It would be pretty easy reading only the last half of this thread to think that @Koldsimer was the only one with this problem, but I know I've seen others with the exact same issue. Perhaps those people need to either speak up here or at least be in a list that could be pointed out to Signature Solar.
• Finally, the thread seems to have shifted back into many of us trying to help Koldsimer by workarounds or more debugging. Since Signature Solar isn't actively engaged in the thread, all these posts are probably muddying the waters and they might scan it and drop back out again.

From a distance, it doesn't seem like this is getting fixed, or even that Signature Solar believes any longer that there is something for them to fix.

1: Unfortunately the forum was not updating me that this thread had continued. I've now added it to my watched list manually so I can keep updated.
2: We are 100% communicating with Koldsimer via PM. We have 100% acknowledged his video, the fact that there is an issue with his system, and we are trying to figure out what the issue is. We 100% believe that there is NOT an issue with Schneider inverters and our batteries in general, as we have 3 different model schneiders (that were purchased and delivered this month specifically to test the issues outlined here) that all operate normally with our batteries.
3: I've repeatedly and specifically asked ANYONE who has this issue to message me or post on this thread. I have so far not received any other PM's from users with this issue.
4: I have actively posted every day and frequently, 100% showing the testing we are doing, and being completely transparent in how we are trying to figure out the issue and fix it.

Seriously, tell me what more you would like to see me do and I'll do it. I SPECIFICALLY ask in several posts for people to point out things I can do to further troubleshoot this issue, short of hiring a contractor to inspect his system specifically and troubleshoot onsite. We are a distributor for a DIY community - we would love to help each user 1 on 1 for hours to troubleshoot the array of problems people run into when they DIY a solar install. There are just too many different models, manufacturers, and system set ups to even come close to being able to accomplish something of that magnitude - and yet, here I am, working with my team (very diligently I might add) to try and help this user. I literally posted just an hour ago on this thread. Again I ask, please help me understand what you would like to see from myself and our company that I've not already done, or am actively doing.

 

[GVSolar]

FWIW, here's a thought.
I noticed this change/feature in the most recent Schneider firmware upgrade notes for my SW4024:
"improved Load Shave startup to better manage inrush currents".
Do you have Load Shave enabled/disabled? Could that be affecting your inrush current issue?

 

[Solar Guy Richard]

@Koldsimer What is your total load? Can you go over exactly what you are trying to pull? These batteries have a max amp of 200a each - is it possible your initial inrush current is exceeding the capacity of your 3 batteries?