Conect XW voltage too inaccurate for LFP charging?

twistedtree

Here are a bunch of screen shots showing settings, and reported voltages from various devices.  Screens are the XW settings.  The MPPT settings are the same, except Bulk/Absorb are 55.2V

Below is during a gen/XW charge cycle, and you can see the large spread in reported V between the XW and the two MPPTs.  Actual battery V is closest to MPPT 0.  Over the course of this charge, the spread in V between the XW and MPPTs has grown from about .3V to .7V.  Current has remained constant the whole time at ~115A which is about 0.3C

Below is during a solar charge cycle.  Here you can see a much smaller spread between the XW and MPPTs.  In general, the MPPT readings are pretty stable, and close enough to actual battery V to be workable.

The other observation is that the SCP's "system" info, and the Combox "System Performance" info is all based off of MPPT 0's voltage.  I don't know why that device was picked vs the others, and whether it might change to a difference device if I restarted everything.  I've seen nothing that states the selection criteria.  If the metering were consistent, it wouldn't matter, but in this case it throws things off in a number of ways.

twistedtree

Dave Angelini said:

Just because a voltage anomaly does not vary with current, does not mean that there is not a wiring defect. You should triple check any older equipment by removing the wires and inspecting. Sometimes just removing the connection will remove a "diode effect" defect.

What's the "diode effect" in this context?

Also keep in mind that the voltage at the XW terminals was confirmed to match actual battery voltage, which further exonerates the power path leading up to it.  And the XW is metering high, not low.  Power path losses are of course just that - losses, never gains.

Even if there is a lot of variation in XW metering, it's to be expected that some units will meter quite well where other's will be off by more.  I probably got one of the "off-by-more" units.  And as I said right at the beginning, this just isn't that important with LA.  Heck, I ran one XW for 10 years, and this XW+ for over a year without noticing or caring.  It's only when I swapped in an LFP bank and started paying close attention to charge voltages and correct bulk/absorb/float transitions that all this became apparent.  And hence the thread title about accuracy with respect to LFP.  And with an LFP setup like you are using where the battery controls everything, XW and MPPT metering is again moot because it's not used for charging.

Dave Angelini

First of all you are wrong about my set-up and the ones I use for clients having the battery control everything.

 In my set-up LG is no longer being controlled by Schneider. In my clients, the only thing controlled by Schneider is LBCO. The BMS in both cases charges to to a Soc based on voltage which is the same for bulk, absorb and float. I run both systems here and tested for the other battery manufacturer. The point is the BMS was designed for the battery by the company that warranties the battery.

I think you need to check wiring again. I have never seen anything like your screen shots. NEVER !  Many dozens of systems.

I assume you have cascading off ? Also assume the voltmeter you used was backed up by another decent meter. One of the reason I use calibrated voltmeters is part of the cal is making sure the input impedance/circuit has not been damaged. This gives non-repeatable readings.

I can fix your metering error but you would have to become a client. You know how to contact me. 

Who did you buy this XW+ from please?

twistedtree

My meter is NIST calibrated in the past year, and agrees (within reason) with other meters that I have.

Predictable answer from you.  "If you were my client this wouldn't have happened".   Or "become my client and I will fix it".  It's not going to happen, if for no other reason than that's how every conversation winds up.  I'm surprised such solicitation is tolerated on this forum.

I'm an installer with a wholesale account with a local distributor.  In my experience, installers, resellers, and distributors know very little about the products they sell beyond what's in the manuals, and I can (and do) read them myself.  I don't think I have ever had a question that anyone in the distribution channel was able to answer to any level of satisfaction, and that's pretty uniform across industries, product lines, etc.  There is lot's of BS tossed around by people who pretty quickly reveal that they really don't know what they are talking about.  It's like people saying that a high voltage reading is due to cable losses.  This is why direct manufacturer support is so important, and even then they often don't know what they are talking about.  Lots of companies see support as an overhead expense that needs to be reduced in any way possible.  It's actually the easiest way to distinguish yourself from others.  I would argue that a dollar spend on support is way more valuable than a dollar spent on marketing.  But that's the CEO in me, and not what I'm doing now.

There are a number of companies, Schneider included, who strive to place their channel our front to shield themselves from us pesky, needy customers.  It seems the last thing they want is customer contact.  Some products need very little support, like Schneider relays, electrical boxes, etc.  But throw in a complex power system, typically installed and serviced by someone with at best a high school diploma and likely zero electrical circuit background beyond basic wiring hookup,  and a lot of things will fly right over their heads.  Distributors are in the business of stocking and moving boxes, and typically know even less.

I was amazed, but actually not, to see your response to McGiver blaming his problems on everyone except Schneider.  It's him, his crappy dealer, and the crappy distributors in his region.  But wait a minute, that's Schenider's distribution channel and the front they have presented to that region.  They own that channel and it's performance, not McGiver.  If customers are not successfully using a company's products, there is nobody to blame except the company, and nobody who will suffer except the company.  And that's for any distribution model.  If a company wants to hide from their customers behind a distribution channel, then it darn well better be a spectacularly competent channel.  Some are, but not many.  And many companies (Cummins is a good example) are pulling the distribution channel back in house, all to control quality.  Cat is another example of a company that has a small number of very large dealers with very large territories who have been in business for a very long time and have deep knowledge.  That works for Cat.  Parker, in my opinion, slipped on the same banana peel as Schneider when they acquired several water maker companies, having no idea how much more support would be required compared to inanimate hoses and fitting.  They are still struggling.

In all frankness, my experience with Schneider over a number of interactions is that they have managed to be the worst of all worlds.  They hide from their customers behind an incompetent channel, are ridiculously slow to respond, if they ever respond at all, will go completely dark on you for unknown reasons, and have incompetent techs when and if you do get a hold of them.  I think most of the issues I have contacted them about involve one or more things that should have been escalated to engineering, and at some point resulted in product fixes.  But they have all been brushed off or disappeared into the black hole.  To me, they are a giant Fail.  I have already moved all my power system business to other vendors, and am moving all my automation business too.  Other companies do this sooo much better, there is just no reason to put up with it.

Estragon

FWIW, and being the sort who doesn't mind tinkering some, I think Dave's notion of isolating the inverter to test makes sense.  If it still can't reasonably accurately read voltage, it would rule a lot out.

On the business side of it, I can understand where Schneider is going, and maybe why.  I might do it differently, but it's their choice to make.

MarkP

Schneider makes a battery monitoring system. I assume it communicates with all the other equipment. If there is any sense in the world it would be the "official" battery current and voltage. Anybody know if that is true?

Dave Angelini

The battery monitor would definitely be the current. I think you should ask Schneider about where the voltage reading is derived when using their monitor. I put that on my list as it is an interesting thought. I am going up there soon but you should ask also. The XW+ drives the xanbus power source for some devices as do the mppt-80's. The 60's do not source xanbus power.