Conect XW voltage too inaccurate for LFP charging?

twistedtree
twistedtree Solar Expert Posts: 135 ✭✭✭
I recently cut over to LFP batteries for my off-grid, XW6848 + dual MPPT60 controllers.  Also included are SCP and AGS controlling backup generator.  The LFP batteries are stand-alone with respect to the XW system, and I am trying to get the XW charge settings to work appropriately.  While doing this, I'm seeing two problems.

1) Reported voltages between the MPPTs and XW6848 are different, and the spread increased as battery voltage rises.  The two MPPTs are within about 100mv of each other, and within about 100mv of fluke measured bank voltage.  But the XW6848 is reporting about 500mv higher, and towards the end of charge as bank voltage climbs to 54+ volts, the difference seems to grow to as much as 900mv.  That's a huge inaccuracy for LFP operation.  I have checked voltages at the device terminals while under XW charge load of 115A and measured voltages are the same as bank V, so not wiring issues or losses.  This would appear to just be crappy metering in the XW?

2) The Bulk to Absorb transition doesn't appear to be happening at the programmed voltages.  I have bulk and absorb both set to 55.2v (3.45Vpc).  But the transition to Absorb happens at about 54.5v.  And that happens for both the XW and the MPPTs.  Both are stopping prematurely.  As an experiment I tried raising the XW bulk/absorb set point to 52.7V, and the transition point rose, but only by 200mv.

So far I have been using the Custom settings in both devices, and not the Lithium settings.  But for 3 stage charging with Lithium, there is no documented difference between what I'm doing now.
«1

Comments

  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    And I see Schneider no longer has customer phone support - only a web form that you fill out and submit.  They have perfected the art of hiding from/ignoring their customers.  They were pretty good at it before, but this is real genius.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    You have to fill out the form and then depending on your request you can choose phone support.

    If you have equipment that is not in support life they may just tell you that.
    Schneider gives a field direct phone number to people like me who install and sell their equipment.

    I have heard that they are working on an interactive FAQ for users with non supported equipment.

    They have investor share holders and have to be profitable. Going out of business is not acceptable to me and so I respect their decision.

    You have choices, no one makes you use their equipment.

    Good Luck with your battery setup. What make is it?
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    As a matter of interest, is temperature compensation disabled on all devices?  Does the magnitude of the anomolies vary with any correlation in battery or device temp?

    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    The MPPTs are old, but all other equipment is just over a year old.  I registered and will try calling.

    The batteries setup is 32 CALB CA180 cells wired 2P16S, plus a custom PLC-based BMS.  That part works great.  I just need the XW and MPPTs to switch into bulk/absorb/float as they are supposed to, and with some hint of voltage accuracy.
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    Estragon said:
    As a matter of interest, is temperature compensation disabled on all devices?  Does the magnitude of the anomolies vary with any correlation in battery or device temp?


    Good question, and sorry I didn't mention it.  I have temp comp set to zero.  "Interesting" observations are:

    - Both the MPPTs and XW transition to absorb prematurely, even though the XW is what's metering high, not the MPPTs.  But perhaps the MPPTs are switching based on the highest metered reading available on the xbus?  If it were using the XW meter to switch, it might explain why it switched too soon as well.

    - The XW's over metering seems to grow as the battery voltage climbs.  Maybe that's related to the base voltage, or maybe it's related to cumulative heating?

    - I'm really confused about why the XW, when forced into bulk mode, immediately switches to absorb even though it's well away from the bulk voltage.  Could it be related to the Recharge Voltage, which is set to 50.0v?  Do the batteries have to drop below that value for force bulk to work?  Recharge Voltage is described as the voltage at which a new charge cycle is automatically initiated, but maybe it does more than is not documented?



  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    I just got off the phone with Schneider after registering that I am an installer.  Phone support is now only for dealers and installers.  No phone support for end customers.  Interesting choice, and I hope it works for them.

    The call was less than satisfying.  The metering variation from device to device is considered normal tolerance.   I can see that for LA, but for LFP it's problematic.  The only work around is to adjust my charge voltages to compensate for the metering error in the Conext devices.  And depending on which device, each gets it's own settings, and I have to remember how to interpret each, doing offsets in my head.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    edited January 2019 #8
    I understand their stance... They have distributors and retailers (like our host NAWS) that are there to provide first level support and relate issues back to Schneider up their support structure.

    If Schneider was in end sales--They would be getting more of the retail $$$ to support a larger support crew.

    I ran into this issue decades ago with outsourcing and offshoring of manufacturing.

    I, as a design engineer, used my local suppliers and mfg representatives to get me the support I needed to design and implement their parts on my designs.

    In the meantime, my MBA VPs of Manufacturing were telling the upper management how much they could save with outsourcing to other states/countries.

    And my support from local suppliers/distributors/Mfg. Reps was starting to dry up as they were nowhere near the money flow other than for the first 5-10 prototypes.

    They were literally telling me to contact the overseas contract suppliers' vendors/mfg reps in China for support. What we got from data books and the Web was all were were going to get locally.

    Buying complex/expensive hardware from a reliable local "authorized" retailer/distributor does raise costs. But you ain't going to get very much support from somebody on EBay that sold a FOT (fell off the truck) unit.

    Good support costs money. It can make or break your product. I am sure that Schneider is not making these decisions in a vacuum.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    Setting aside their support model, what success (or failures) have people had using their products with LFP, excluding systems where the BMS meters and controls charging and the chargers are slaves?  The later model side steps any shortcomings in the Conext product metering, and might be why Schneider went down that path with their official offerings.  Being off by .5 to .75 volts strikes me as a lot of slop for an LFP charging system.  With all the challenges in designing and building a BMS, the precision of the Conext chargers is the last thing I expected to be a problem...
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Zero problems. Over 10 LFP systems on XW+. Your BMS should be able to tweak the voltage, and based on Soc you should be able compensate. I usually set the Soc at 97% for float and the BMS just deals with it.

    BMS is the key and how well it is integrated. The ones I use are designed by the battery manufacturer.

    One of the Engineers there said the most common call was what does the red LED mean on an mppt-80.....


    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    Zero problems. Over 10 LFP systems on XW+. Your BMS should be able to tweak the voltage, and based on Soc you should be able compensate. I usually set the Soc at 97% for float and the BMS just deals with it.

    BMS is the key and how well it is integrated. The ones I use are designed by the battery manufacturer.

    One of the Engineers there said the most common call was what does the red LED mean on an mppt-80.....


    I think you are describing the type of LFP system that I'm excluding in my question.  Namely systems where the BMS controls charging and directs the chargers, e.g. LG's RESU.  Not that it's a bad approach - I actually think it's the best long term solution - but the economics are still a struggle to make work, at least for my applications.  The question is around stand-alone LFP battery systems where the chargers need to be adjusted to operate within acceptable LFP ranges.  From what I've seen so far, the XW+ isn't so great for that give the voltage slop.
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Estragon said:
    As a matter of interest, is temperature compensation disabled on all devices?  Does the magnitude of the anomolies vary with any correlation in battery or device temp?


    Good question, and sorry I didn't mention it.  I have temp comp set to zero.  "Interesting" observations are:

    - Both the MPPTs and XW transition to absorb prematurely, even though the XW is what's metering high, not the MPPTs.  But perhaps the MPPTs are switching based on the highest metered reading available on the xbus?  If it were using the XW meter to switch, it might explain why it switched too soon as well.

    - The XW's over metering seems to grow as the battery voltage climbs.  Maybe that's related to the base voltage, or maybe it's related to cumulative heating?

    - I'm really confused about why the XW, when forced into bulk mode, immediately switches to absorb even though it's well away from the bulk voltage.  Could it be related to the Recharge Voltage, which is set to 50.0v?  Do the batteries have to drop below that value for force bulk to work?  Recharge Voltage is described as the voltage at which a new charge cycle is automatically initiated, but maybe it does more than is not documented?



    I'm a bit confused about your confusion.  By definition, absorb holds at Vabs while current drops off.  Presumably it's actually going to float?  It could well be it flips a "we've done a bulk today" bit which prevents a new cycle absent hitting Vrebulk.  I can force a new cycle on Classics, but maybe your gear needs power cycling?

    The Classics also can sync charge states such that whoever gets to Vabs first triggers everyone else to change mode also.  Wouldn't be surprised if yours does similar.

    .75V strikes me as a lot of slop even for LA, let alone lithium.

    Back in the day, some of my (better) network gear makers supported usenet groups or forums (not unlike this one) where the vendor had someone clueful lurk and sometimes help otherwise clueful but stuck folks using their gear.  The RTFM stuff like what does the blinky light mean got weeded out by more clueful users.  Actual vendor engineers got to interact with actual clueful users though too, to the benefit of both IMHO.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    I'm a bit confused about your confusion.  By definition, absorb holds at Vabs while current drops off.  Presumably it's actually going to float?  It could well be it flips a "we've done a bulk today" bit which prevents a new cycle absent hitting Vrebulk.  I can force a new cycle on Classics, but maybe your gear needs power cycling?
    I have bulk and absorb set to the same value, 55.2V in particular.  So far, all solar and gen charge cycles have switched from bulk to absorb at around 54.6V which is .6V shy of the target. I have the absorb time set very short (3 minutes), and the chargers switch to float after that time.

    So the problem is that bulk/absorb is acting like it's set to 54.6 when it's actually set to 55.2.  My best guess is that they are stopping based on the highest metered voltage being reported on the xbus, and that would be coming from the XW+
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    That makes sense, same as if one classic thinks it sees Vabs, all classics in the "follow me" group go to absorb.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    Estragon said:
    That makes sense, same as if one classic thinks it sees Vabs, all classics in the "follow me" group go to absorb.
    I always figured that only a device that was actually charging could trigger a follow-me change of state, and that an idle device would not.  So I would not expect an XW+ that is only inverting (no grid or gen input power) to be able to trigger a state change in actively charging MPPTs.  But that's just an assumption on my part, and I don't think I've ever seen any more than a one line explanation in the manuals.  Do you know if it works as I presumed, or maybe differently?
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    I don't know. 

    These devices generally have fairly simple brains, and it could be just "when [condition] do [something]", without much sanity checking.  The CCs, for example, may not be able to check and say to the inverter "hey bud, you aren't charging anyway, so thanks for the tip but we'll carry on charging anyway".  They may just hear and obey.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    edited January 2019 #17
    @twistedtree
    With regards to the voltage difference displayed, I too have noticed deviation between MPPT display and SW via SCP. Since the resolution of both is only 100 milivolts, not that accurate, I choose to monitor actual battery voltage as well as individual cell voltages via a BMS display which corresponds with my Fluke meter readings. The conductors from CC to battery, battery to inverter, will always have some voltage drop and the higher the current, the larger the drop. Depending on actual inverter load these discrepancies are greater, why I ignore them, as long as the voltage is correct at the battery that's what I consider important.  Are the device in you case all connected to a common point such as a DC distribution, or on seperate feeders? 

    As for settings what I use are,  bulk/absorption 56V, absorption 6 minutes float 54V, EQ disabled, temperature compensation 0mV/C, battery capacity 20Ah per Kw of installed capacity, default battery temperature warm, recharge 50.5V, maximum charge rate 10A per installed Kw.

    Inverter charger settings same as above except, 2 stage no float, LBCO 48V, LBCO hysteresis 2v, LBCO delay 10 seconds, high battery cut out 60V.

    Notes and observations: Voltages doubled from my actual 24V nominal values, LBCO is the highest setting available 50.2V would be more desirable. During solar charging the controller display voltage reaches the programed 28V but battery terminal voltage is 27.6V when charge current is ~36 A, voltage drop, cell voltages 3.454 +/- 18mV once in float all cells are <10mV. apart. Been operating for 5 months without issues, haven't needed generator so little testing done with regards to that. Have  contacted Schneider support regarding LBCO they have been useless, never reply despite auto reply stating will respond next business day, I'm still waiting a month later, pretty poor.

    Edit, found document where settings were sourced, see attached pdf.

    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    mcgivor said:
    @twistedtree
    With regards to the voltage difference displayed, I too have noticed deviation between MPPT display and SW via SCP. Since the resolution of both is only 100 milivolts, not that accurate, I choose to monitor actual battery voltage as well as individual cell voltages via a BMS display which corresponds with my Fluke meter readings. The conductors from CC to battery, battery to inverter, will always have some voltage drop and the higher the current, the larger the drop. Depending on actual inverter load these discrepancies are greater, why I ignore them, as long as the voltage is correct at the battery that's what I consider important.  Are the device in you case all connected to a common point such as a DC distribution, or on seperate feeders? 

    As for settings what I use are,  bulk/absorption 56V, absorption 6 minutes float 54V, EQ disabled, temperature compensation 0mV/C, battery capacity 20Ah per Kw of installed capacity, default battery temperature warm, recharge 50.5V, maximum charge rate 10A per installed Kw.

    Inverter charger settings same as above except, 2 stage no float, LBCO 48V, LBCO hysteresis 2v, LBCO delay 10 seconds, high battery cut out 60V.

    Notes and observations: Voltages doubled from my actual 24V nominal values, LBCO is the highest setting available 50.2V would be more desirable. During solar charging the controller display voltage reaches the programed 28V but battery terminal voltage is 27.6V when charge current is ~36 A, voltage drop, cell voltages 3.454 +/- 18mV once in float all cells are <10mV. apart. Been operating for 5 months without issues, haven't needed generator so little testing done with regards to that. Have  contacted Schneider support regarding LBCO they have been useless, never reply despite auto reply stating will respond next business day, I'm still waiting a month later, pretty poor.

    Edit, found document where settings were sourced, see attached pdf.

    Thanks, that's very helpful.  To answer some of your questions....

    My BMS monitors per-cell voltage, and of course bank voltage.  It's accurate to within 5mv using a NIST Fluke as a baseline.  But I think like yours, it is stand-alone with respect to the Conext system, and requires proper charger programming in the XW and MPPT.  My BMS will throw up warnings if cell voltage start to get too high or too low, and will disconnect the bank if things really get out of hand.  But so far not even a warning, which is good.

    I have checked voltages at the batteries and inverter terminals while charging at ~115A (~C/3), and the differences are what you would expect in a properly working system.  Certainly noting even close to the discrepancy reported by the XW metering.  I have also shot the whole power path with both a thermal gun, and a FLIR imager to be sure there aren't any hot spots or other build issues.  All looks excellent, so I am really confident there aren't any wiring or connection issues.

    Actual wiring is via a Conext wiring/breaker box, so 100% per Schneider guidance.

    My voltage settings are very similar, though not exactly the same.  Bulk/Absorb 55.2, absorb 3 min, 53.6 float, no EQ, 100% charge rate (<= .4C), 0mv/c temp comp, 50.0 recharge.  Note that my AGS is set to trigger at 50.4 for 15 min, and 49 for 30 seconds.  So that really controls the bottom end and recharge point.  But looking at this, I think my recharge point should be above the AGS start point, not below.  That said, I can't see how that would relate to the issue I see, which it that charging advances to absorb/float before the programmed bulk voltage is reached.  I am running 3-stage on teh XW and MPPT, but it's kind of moot on the XW since the AGS is set to shut down the generator when float is reached.

    Another thing I see as we are discussing this is that my HBCO is set to 58 and your have 60.  It makes me wonder if the XW will only charge to withing some min distance to the HBCO.  It's just speculation, but I might change it to see what happens.





  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    The settings you have are close, there shouldn't be much difference as once the individual cell voltage nears 3.4xxV things happen pretty rapidly, the battery is in the mid to high 90% range, no point going higher. The setpoints provided are a guideline which I used, not an absolute setting, had to experiment with something and they seem to work. The HBCO value actually seems high, in my opinion, however as the charging sources are limited, as well as the BMS,  to a lower threshold, it would be a last resort should something go wrong.

    Can you please provide a link or pdf for the BMS you're using, it just interests me to know what's out there, may I ask if you if previously  used LA and made the switch to LFP, or just took the plunge? What are  your options regarding the performance, in comparison to LA, if applicable. There aren't many who use a DIY system in this theater, dialog is extremely useful because I believe it's a hidden treasure many overlooked primarily out of fear of the unknown.

    Personally I did have LA before, the primary reason for the switch was temperature, the efficiency of LFP which  greatly reduces battery temperature, the enemy of LA, in a tropical environment there is a distinct advantage, perhaps not so in cold climates, where there are other challenges in keeping the bank warm.

    For what it's worth, I have 2 Fluke 179's and Fluke 20, all measure different voltage +/- 10 mV, sometimes consistent, there are tollerance variations even with the best tools.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭


    This system has been running on LA for 20 years, 10 years on the Schneider gear, though all of the Schneider gear except the MPPTs were replaced about a year ago after a DSL power spike in a conduit shared with the xbus took out everything else.  Thank you Fairpoint for not grounding the network interface!

    I just cut over to the LFP bank about a week ago, and so far I love it.  I'm disappointed in the Conext gear, but can probably make it work acceptably.  For me the big advantages over LA are:

    1) Much more efficient, which means less heat and less power loss.

    2) No endless absorb cycle, which is a giant pain for any sort of generator powered recharge.

    3) No consequences to operating at a partial state of charge.  This tends to force oversizing of charge sources for LA

    4) Longer life.  Nominal life of 2000-3000 cycles, and that's pushing the cells through the full rated SOC range.  If you are lucky you can get 3000-4000 cycles out of Surrettes, but they are costly, and gastly inefficient because you need to run them at high charge voltages and long absorb in a solar setup.  Mine are still running, but close enough to EOL that it seemed a good time to switch to LFP under non-emergency conditions.

    There are lots of people doing DIY on boats (and I gather RVs).  For some, the BMS is the owner with a meter and a switch.  Others have bought engineered systems of one extend or another.  CruisersForum.com has some good discussions, and a number of very smart people.  Also some posers, so proceed with caution.

    My BMS does monitoring, alarming, and last-resort battery protection via disconnect.  It does not control chargers as currently implemented, though it has signals to enable and disable charging if desired.  It's largely unnecessary if your chargers are programmable with reasonable accuracy, and that seems to be the Conext issue at hand.  My BMS also does not include any active cell balancing.  That has been demonstrated to be unnecessary as long as you operate away from the extremes of the SOC range.  My per-cell alarming will give me plenty of heads up if I have a cell that's drifting high or low.  So far all are tracking within 10mv or so.

  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    OK, just went through another generator XW+charge cycle, and think I see what's happening.

    The SCP and Combox are reporting battery voltage using the metered value from MPPT 0 in all the system level displays and graphs.  The SCP top level screen shows the battery voltage as the MPPT 0 value.  Same fro the Combox System Performance overview, and in the Battery Bank Daily Summary graph.  Why it has picked the MPPT 0 is not clear, but it happens to be pretty accurate.  Now if the metering in these boxes was worth a damn, it wouldn't matter, but that brings us to the next issue.

    The metering in these products, or at least in my XW+ sucks.  There's just no other way to describe it.  MPPT 0 & 1 meter within 100mv of each other, and MPPT 0 is withint about 50mv of teh battery bank actual voltage.  But the XW+ is metering 600-700mv to high.  And the error grows, seemingly as the unit heats up.  Over the course of the day the XW was reading 200-300mv high, and over teh course of charging it has grown to 700mv.  Schneider told me metering is accurate to within 2%, which would be a 1v variation.  Did I say that sucks?  Completely unacceptable for LFP.

    Now, despite the SCP and Combox locking onto MPPT 0 for battery voltage, the XW uses it's own metered value.  And since it meters .7v high, it changes charge states .7V sooner than the value displayed on the SCP, Combox, and in the Combox graphs.  Sure enough, the XW switched to absorb when bulk was hit according to its meter, which is 700mv too soon.

    The Combox is kind of a joke too.  Right now it's showing incoming generator power as 7889W and 7765VA.  But wait a minute, real power has to always be less than or equal to VA, right?  And it reports L-L voltage and for current it reports the sum of L1-N and L2-N.  What the heck is that?  My generator with a max of 60A is putting out 68A at about half load?



  • mcnutt13579
    mcnutt13579 Registered Users Posts: 111 ✭✭✭
    I was just working with a 6048 yesterday that was also reading higher watts than VA.  Strange duck these XW systems.  Not popular around here.
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    @twistedtree
    Thanks for the link to a very interesting blog, quite an endeavor to to actually make a BMS with a PLC, so this is what you are using?
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    mcgivor said:
    @twistedtree
    Thanks for the link to a very interesting blog, quite an endeavor to to actually make a BMS with a PLC, so this is what you are using?
    Yes, that's what I'm using.  So far it works great.

    I heard back from Schneider.  They say the metered difference shown by the XW is because of current induced voltage drop and is normal.  This is in reference to a 7A battery current, and .3V reported voltage difference screen shot that I sent them.  However the Fluke measured voltage at the inverter terminals shows no voltage drop, 7A is a tiny current, and best of all, a current induced voltage would cause the XW to read low, not read high.  So a completely ignorant and/or BS response.

    McGiver, I take it your XW metering is reasonably accurate?  All I can conclude at this point is that there is massive variation in metering in Conext products, and as such they are unsuitable for use with LFP for XW and MPPT controlled charging.  Very disappointing, to say the least. 
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    edited January 2019 #25
    Actually I have the SW Conext series, personally I don't have any use for this company, their support is atrocious, their product is good, but I believe it's not actually theirs, just an orphan they adopted but don't know how it mind functions, in a corporate mindset. How  I wish I went with Morningstar, for their controllers at least, the support is second to none, excluding others whom  I've had no dealings with, Midnight, Outlook and SMC to mention a few. However it is what I have, I can make it work, but should it fail I will definitely look elsewhere, not going to invest in more equipment that has no support, unless you're a preferred customer/deale.r  Bang! Just got an email from Morningstar regarding LFP batteries and their devotion to include  the format, now that's support .
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    edited January 2019 #26
    How can they support you when you do not have a way to install firmware.
    Why is it the people you bought it from are not there for your really old firmware?
    You turned down my offer for sending you a combox or config tool to upgrade your old firmware.
    You seem to have handcuffs on and the key is in your pocket. Shheesh!
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭
    How can they support you when you do not have a way to install firmware.
    Why is it the people you bought it from are not there for your really old firmware?
    You turned down my offer for sending you a combox or config tool to upgrade your old firmware.
    You seem to have handcuffs on and the key is in your pocket. Shheesh!
    Not sure who or what you are referring to.  My firmware is up to date, and I have a combox.  Maybe your comment is in reference to another thread?
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    No it was for Mcgivor in the previous post.

    As a general comment here, if one buys from an online store that is the cheapest price on the internet and the store has very little experience, support is just not going to be there. It really is the source of the equipment that takes on the support. If you have a warranty claim then yes Schneider will take care of that and does.

    My opinion only but when Schneider bought xantrex they were asked by many of us to reinstate the Xantrex user forum to help users who are DIY. Outback still does this and I use them often for that reason. Both Xantrex and Schneider decided that they were really after the professional installation market and could not spend resources on DIY. It is their business ;)

    If Schneider says the voltage is within tolerance and you can't offset the numbers to charge correctly, then you have a problem. I do not understand why you can't offset it. There is the solar voltage charging and the XW charging voltages, why do they have to agree?

    Most Gen users offset all the time to get max solar and minimum fuel usage when both charging types are running. I agree that I do not see more than a couple 100mV max difference over the voltages.
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • twistedtree
    twistedtree Solar Expert Posts: 135 ✭✭✭

    If Schneider says the voltage is within tolerance and you can't offset the numbers to charge correctly, then you have a problem. I do not understand why you can't offset it. There is the solar voltage charging and the XW charging voltages, why do they have to agree?
    The settings for the MPPTs and XW can (and are at this point in my system) be different as long as the actual resulting battery voltage is correct when charging.

    Problem one was figuring out which voltages each device uses, and that it's different from what the SCP and Combox report.  That's not documented anywhere, and Schneider hasn't been able to explain it.  I finally figured it out through experimentation.

    Problem two is that the metering error in each device isn't constant.  It doesn't vary with current, so I remain confident that I don't have wiring loss issues.  But it does vary over the course of a charge cycle, both for an MPPT cycle and for an XW cycle.  It's probably related to internal device temp since that's what would change over a charge cycle.  But it doesn't matter why it's varying, the key is that it's not a constant.  So different combinations of solar charging and XW charging can result in different required offsets, and it's not possible to program that into the system.

    I currently have a set of offset bulk/absorb voltages, one set for MPPTs and a different set for the XW.  Individually they seem to work OK, but I'll need to see a bunch more cycles to convince myself it's stable and repeatable.

    And then when I look at any of the Combox reports, I need to mentally offset all the graphs.  It's just very sloppy product implementation in many ways.  It does generate power day in and day out, I gotta give it that.  But the metering and especially the combox reporting is a total joke.  It's largely inconsequential for LA, but much more problematic for LFP.
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    How can they support you when you do not have a way to install firmware.
    Why is it the people you bought it from are not there for your really old firmware?
    You turned down my offer for sending you a combox or config tool to upgrade your old firmware.
    You seem to have handcuffs on and the key is in your pocket. Shheesh!


    The gear purchased was from an authorized  distributor recommend by Schneider, not from an online store, the distributor refers me to local support, local support refers me to the distributor, after contacting support in Canada they recommend contacting support in Thailand who bounce me back to the distributor who refuse to answer emails.

    The equipment should have been sold with firmware that is up to date, as such it's their responsibility as a distributor to correct the issue, or pressure local support, failing that, it is in my opinion that head office should intervene as the parent. Your offer to assist is greatly appreciated, there was no refusal of the offer, just thought I would wait for a response from support in Canada as promised, I believe it's the companies responsibility to resolve problems, which it appears they are unwilling to do, they too do not reply as they claim they will, or perhaps I'm being impatient, it's only been a month, took three months last time.

    Schneider's choice to alienate DIY users to focus on "professional installers " should then exclude sales to all other than registered  installers, or at very least have a disclaimer stating as an independent installer all support is void. While they're at it  removing or the fluffy "we're committed to support" videos, or amend them to include unless you're not an installer. Eventually reputation will exclude Schneider from consideration at a customer level, their market share will diminish to a point where even the installers get no support, but will be left holding the ball.

    As mentioned before the equipment is great, sure I could buy a ComBox but really have no use for it, other than installing firmware, it works as is, it's shortcomings can be compensated for by external control already in use so no big deal, when it comes to replacement though, I'll look elsewhere.

    Apologies to @twistedtree for any derailment.

    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Start your own company mcgivor  and you can do whatever you want. :)

    It is a ridiculous statement that the firmware should be up to date on a new product that was going thru updates. You should have bought the more mature XW in that case, especially in a developing country with very few people there to support what they sold you. How much money should Schneider spend on someone like you or in a country with that bad of a dealer network?

    twistedtree If you post your settings and there are quite a few I will take a peak. What you describe does not sound right. There are at least 3 things that could cause metering errors. I assume this means voltage?

    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.

    You can see in my previous screen shot how little error there is in the mppts. The -60 mppt is a beta from 2006.

    Beautiful day here today. Where ever you guys are be sure and go outside and enjoy it. Electronics can be over rated  ;)


    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net