diy solar

diy solar

Schneider xw pro 6848 not powering up with eg4 lithium battery

Agreed, my XW PRO does this as well, when trying to pre charge it before I hooked it to the lead acid battery bank.
It seem the EG4 batteries must have a longish pre-charge timer causing the XW to fail during initialization.
If the built in resistor is time based, and is staying on longer than the XW is happy about, I think we have solved the mystery "why". Now comes the "What" as in what to do about it.
 
If the built in resistor is time based, and is staying on longer than the XW is happy about, I think we have solved the mystery "why". Now comes the "What" as in what to do about it.
I never gathered if @Koldsimer tried this or not and it may not be the "recommended way" to start the XW, but you could turn the batteries on with the MAIN breaker to the invertor OPEN, give them a few minutes to stabilize, THEN CLOSE the MAIN breaker to inverter. It will be a large inrush "cap charge", but if the batteries only use the Pre-Charge resistors for a certain amount of time when turned on, it would get you past that at least.

Signature Solar any thoughts on this? Hopefully they are working directly with @Koldsimer on the issue and we will hear the verdict soon.
 
Well, success! To some degree anyway...

Schneider Tech made some recommendations that finally worked.

He read this thread and agreed that somehow, the battery was cutting current likely related to the resistor and bms settings. They recommended I hook up the batteries in parallel with no communication between them in case they were talking to each other and telling each other to shut off. He theorized that maybe the other batteries could "pick up the slack" if there was some kind of delay. Tried to power up, same thing. Tried closing dc breaker then powering up, same thing. Basically, we tried all the things that I had been doing with no success. At one point, I let the inverter click and hum until alarm lights came on with the battery, and my scp gave a warning of w48, just like I had done several times last week when trying to power up.

The Schneider tech said that we could try to power up the batteries with no communication between them with the dc breaker open. Then he had me start the midnite classics- the thought being that they may act as a dc source buffer. He said the chargers may go into a search or open mode and cut voltage but we could try. The concern that the caps would overheat or have a problem with the fet was there but he said the inverter should protect itself and turn itself off if things got too hot. He emphasized the xw needs constant voltage above 40v to effectively power up. My measurements with a dvm at the inverter posts showed the eg4's would only put out about 33 volts max on their own in any configuration.

So I closed the dc breaker, inverter clicked for half a second and powered up! Holy crap was I happy! Almost can't believe it!

I'm sure im forgetting some stuff...


This leaves me with an issue. I still can't start my system without dc input from the scc's? Never had that issue before with fla. Does that make any sense?

Many thanks to Alejandro and the EE he was working with at Schneider!
 
If it's time based, they @Koldsimer , maybe you are not leaving the battery connected to the XW (when the XW is in pulse and blink mode). That is not long enough for the battery to switch from resistor to full battery.

I've left it connected for up to about 5 seconds until i get warnings from the xw6848 W48- low voltage, fault 45 capacitor over temp, f68 transformer over temp. At that point the batteries alarm light comes on but turns off after a minute or so and the batteries indicate they are fine.
My only problem with this is, I do not know if this is good for the XW to pulse and blink with supper low voltage. Its like it gets stuck in a "cant start" mode. It does not seem like this would be good for the electronics in side.

Yeah, agreed. At least it protects itself and gives the faults and warnings.
 
I'm pleased to hear Schneider tech support is working the issue from their end. Signature Solar seems to be on it as well, based on their recent messages.

I'm pretty sure the issue is with the EG4 LifePower4 Pre-Charge feature. I'm still not clear in how it is setup and works, but it sure seems to be causing the Schneider inverters grief while starting up.

Hopefully Signature Solar gets their Schneider on the bench soon and finds a solution to the issue. I wonder if the BMS is software updatable by the end users? I sure hope it's that easy.
 
I believe it's done described here:

OH I fully understand how Pre-Charge circuits work.

What WE don't know is how EG4 LifePower4 Pre-Charge circuit works, EXACTLY. How many seconds does it stay on? When is it active? They confirmed it's timer based, but there's still more info needed to understand how this is impacting the Schneider startup.
 
OH I fully understand how Pre-Charge circuits work.

What WE don't know is how EG4 LifePower4 Pre-Charge circuit works, EXACTLY. How many seconds does it stay on? When is it active? They confirmed it's timer based, but there's still more info needed to understand how this is impacting the Schneider startup.

ok, my bad.
 
OH I fully understand how Pre-Charge circuits work.

What WE don't know is how EG4 LifePower4 Pre-Charge circuit works, EXACTLY. How many seconds does it stay on? When is it active? They confirmed it's timer based, but there's still more info needed to understand how this is impacting the Schneider startup.

If the battery pack has an inverter cap charge surge system it must act on timing. It initially triggers like a a true short circuit has occurred opening up the BMS's main series MOSFET's but connecting a separate single MOSFET, or mechanical relay in series with power resistor. If short is maintained longer than they think appropriate, indicated by voltage drop across power resistor, it assumes it is an actual load short circuit and the power resistor connection also opens and BMS remains shutdown.

If the voltage drop across the series power resistor fades to a low value, in the time they allow, it assumes it was a temporary inverter cap charge surge current then it turned the main BMS MOSFET's back on. All this should happen is less than a few milliseconds.

If you heard clicking than the power resistor likely has a mechanical relay instead of a MOSFET pair for switching resistor.

Problem with this method is if an inverter immediately goes into active inverter operation the current may not drop off enough to allow the system to recognize there is not a permanent short so it does not re-activate main MOSFET's, opens path on surge resistor, and BMS remains shutdown.

If their power resistor has too much resistance and/or inverter caps are large in value, the cap charging current may last longer than their allowed time interval before voltage drop across power resistor reduces to a level acceptable to them.

This is a tough problem prone design when they need to be compatible with all possible inverters that may be connected to their battery pack and they also likely must modify the timing delay depending how many battery packs are connected in parallel. Would be interesting to see how a system with multple inverters in parallel react.

How it recovers from this type of shutdown is another question. Probably just tries again after some time interval.. They cannot continue to leave power resistor connected to sense current.
 
Last edited:
This is a tough problem prone design when they need to be compatible with all possible inverters that may be connected to their battery pack and they also likely must modify the timing delay depending how many battery packs are connected in parallel.

This is why I'm of the opinion that a circuit like this has to be part of the inverter, not the battery. As it is implemented right now, it won't work properly in the event that the you turn off the inverter and the caps discharge, and then you turn the inverter back on. The pre-charge circuit only works at the time you turn the battery on. If the inverters implement a kind of 'soft-start', this wouldn't be an issue.
 
So to be clear:
4x 48v LIFEPOWER4 packs are being powered on, then the DC breaker for the Schneider 6848 is being engaged after the 4 packs have powered on?
Yes, I have tried that. Multiple times.

Guys, please read this thread in its entirety. It may take say, 30 minutes but it will let you see the steps I have taken to try to get these to work.

You need to understand how frustrating it is to get asked something like this when it's clear that I have tried that probably half a dozen times.

I'm glad to see you guys are now acknowledging that there is a potential issue with the battery and it's compatibility with some inverters.

Im curious, when your tech team suggested I get a resistor and try to pre-charge the caps, were you aware then that the batteries had the pre-charge resistor feature built in? Seems there have been some changes to the eg4 lifepower batteries that you guys were not aware of.
 
Im curious, when your tech team suggested I get a resistor and try to pre-charge the caps, were you aware then that the batteries had the pre-charge resistor feature built in? Seems there have been some changes to the eg4 lifepower batteries that you guys were not aware of.
If you did the manual precharge your problem would have gone away,
 
And if you did it your problem would have gone away,
Well, I ordered the resistor and it came in last night after I got my batteries going. I was going to try it.

I really appreciate all your ideas and input here. They have helped a lot.

Sorry but i dont have a 250' roll of romex and if I did jumping terminals in my battery to start my plug and play batteries is not something I should have to do.

A lot of this stuff is over my head- I still don't know why I would need to buy a resistor and rig up a setup that i have to guess as to how long to hold it and what size resistor etc.. Isn't that why they build this into the battery that i was sold as plug and play?
 
There are a lot of BMS's out there that will shut down for overcurrent when attempting to charge the inverter input caps.

I will give EG4 battery packs credit for attempting to design a solution for this common problem but it is difficult when you have to deal with a myriad of different inverters and system setups with multiple parallel inverters,

You cannot really solve the problem in the inverter without putting a high current series switch in inverter which would add a lot of expense and degrade reliability.

For BMS's that don't have the problem they just allow 100 usecs of extreme overload before tripping off. That is enough time to charge the inverter caps.
 
Last edited:
I think my EG4's are still sitting at Signal Solar even though I have had a tracking number for days now. If that's the case they won't be here by this weekend unless I go get them. Frustrating.

As soon as I get them and can do some testing myself, I will hopefully be able to determine what's going on better. The thread is long and it's hard to understand every troubleshooting step @Koldsimer has tried, which he has tried a LOT of things to try to make these batteries work with his inverter.

I'm very anxious to hear back from Signature Solar on this also....
 
There are a lot of BMS's out there that will shut down for overcurrent when attempting to charge the inverter input caps.

I will give EG4 battery packs credit for attempting to design a solution this common problem but it is difficult when you have to deal with a myriad of different inverters and system setups with multiple parallel inverters,

You cannot really solve the problem in the inverter without putting a high current series switch in inverter which would add a lot of expense and degrade reliability.

For BMS's that don't have the problem they just allow 100 usecs of extreme overload before tripping off. That is enough time to charge the inverter caps.
As Signature said in this thread and as they told me when I bought the batteries and when I picked them up, they have plenty of their batteries working with Schneider equipment. No problem at all.

So i guess they changed something in this new lifepower battery that IS causing them to have problems with Schneider. HMMM seems like they designed a flaw into the new packs. Sorry, i'm not going to give them any credit for that. I am glad they are looking into it and I hope they offer a solution.

And you still didn't answer my question....

Why would Signature Solar recommend I need to purchase a resistor and switch to pre-wire into my system to charge my caps if they are selling a battery pack with those items inside for that exact reason?
 
You cannot really solve the problem in the inverter without putting a high current series switch in inverter which would add a lot of expense and degrade reliability.

All you would need to do is add the pre-charge circuit that's in the battery, to the inverter. It's a simple contactor/relay, just like described in this document. Contactor open by default, resistor coupled. Voltage sense on the capacitors, and when charged, switch off resistor and turn on inverter. When the inverter turns off, reset the contactor, put resistor on the input.
The complexity doesn't change whether it's in the BMS or the inverter, but it'll work much better if it's in the inverter.
 
All you would need to do is add the pre-charge circuit that's in the battery, to the inverter.
That would not be a good solution for a number of reasons. Where do you get the power to precharge the caps for off grid setups, and what consequences arise from having inverter DC input terminals with charged large capacity capacitors.
 
That would not be a good solution for a number of reasons. Where do you get the power to precharge the caps for off grid setups, and what consequences arise from having inverter DC input terminals with charged large capacity capacitors.

You pre-charge from the battery of course. The resistor is in series with input until the caps are charged through it. This would be automatic through the NC side of the relay. This could be a fully hardware-only solution: once the voltage ramps up, the relay coil energizes and switches the relay, disconnecting the NC contact (and disconnecting the resistor) and powering now through the NO contact. When the inverter is off and the capacitors discharge over time, the NO contact opens again and the NC contact closes, putting the resistor back in series with the input.

Those DC inputs right now already have large charged capacitors after precharge with the BMS.

Edit: let me clarify that last sentence: if you have an inverter running, then turn it off, the caps stay charged for a long time. The DC input terminals therefor are already coupled to large charged capacitors.
 
Last edited:
Back
Top