Tarek Yag

Thank you @monsam and @pintopf for your contribution. You went way off-topic here in this thread, please open a new thread to continue your discussion. I'll contact moderartors asking to help move your off-topic posts to your new thread. Kindly,

I'd highly recommend you to disconnect your BMS immediately and use manual charging settings instead for a while, just to make sure as soon as possible that your inverter and battery cells are all good! At least, this way you can have more time to diagnose and discuss/negotiate with your seller. And this way, I think you might rule out the earlier incidents you had with your inverter. I'm sorry, it's not personal at all. it's just for the sake of the community's security. Though, I'm still planning to test it personally, and will let you know for sure if fans are slower than with version 60.95 Wings, huh? I totally understand since I've seen examples of their marketing! Unfortunately, it's common here for reseller companies to be out of the field!!!

Does it happen only when Grid input is available? Did you check if Frequency output is correct (50/60 Hz) on display and/or using an external ammeter? Do you use an AVR (voltage regulator) before/after your inverter? Thanks for your feedback in your first point!

I will try to check more components around the DC/DC MOSFETs as you suggested. We're planning to do the repair job just like last week, on Friday. I just purchased all planned replacement parts, and we're good to go now! I'm hoping to get better repair outcome this time, and it won't hurt if we need one more repair session! I like the journey so far since it's not my main study field! Most probably not! We already checked the soldering condition of every component on the inverter last session.

There should not be any other files/folders on the root directory of the flash drive other than the update file itself. In your case just make sure that the filename is correct (refer to Coulomb's details). It is known that it might cause a lot of issues when performing a firmware reflash for any kind of equipment. The file readers in these basic equipment are not full fledged file managers, they're so tiny and too simple to handle file management.

Just close your browser, then clear the cache, then visit again. This usually happens due to corrupt downloads of the site's JavaScript files.

Not the buck diode, it is the small separate one indeed (all photos attached already). Yeah, it won't hurt, and I will replace it after your statement 😄 New discovery in my newest reply here, which seems I wrote at the same time you were writing yours. Four pieces only from three different batches is way too much for me! I will check this out now, thank you for the hint! Bad news to me, and I won't figure this out until the very end of the repair! Fingers crossed with all the planned replacements this time it could fix the issue somehow someway.

Yeah, it helped, thank you! Actually, after a few photo reviews, I noticed that my 4 INV Bridge IGBTs were replaced with pieces from 3 different batches!!!!!! I don't know how a sane man could do this! So, even the professional who did that repair, didn't do an acceptable job from my perspective. I'm planning now to replace them all again with brand new four pieces of model YGW60N65 and will check today if I can find a higher Amperage specification of this component. I'm tired of this mess! But at least, this way there will remain no components not re-replaced by me and my close friend at our own standards. @Coulomb

@monsam, thank you for all the details you provided, WOW! Super interesting to me!!! Can't wait to know the end explanation for this! This is most probably the case. As the reflash process takes around 10 minutes, if your Windows Power options (Turn off screen timeout, or Sleep timeout) are less than the period of time needed to reflash, your laptop will interrupt the reflash process when it goes idle and beyond these timeouts. Hence, as soon as you turn on the screen again, you'll get that exact error on the reflash tool. It's known and happens to me if I forget to disable the timeouts before I reflash if I'm to leave my laptop unattended. Nothing to worry about here, you're good to go! I very much agree with your theory, and I'm really impressed of how you analyze things! No wonder, Chinese products really need special logic to understand and explain it 😄 @esmail-kassir, any relatable context here with your 24 V one-time issue? Good luck! This makes a lot of sense, no hurry for the photo then. Decent logic! It's obvious you've read everything in this thread! 😄 Best wishes!

Here's an update that I wanted to post last Friday, right when I've had my PV & Battery (DC/DC) MOSFETS replacement job done. Replaced MOSFETs had different values each, with two of them having significantly different values than the rest. I, and my professional friend, tested the new MOSFETs before installation, and to our expectation, they all had almost the exact same values as manufacturer specifications. Good indicator, huh? Bad news is that we've found two open resistors (drivers) behind a single DC/DC MOSFET, which I guess explains all the differences I have in my inverter's PV readings when grid is not present. Am I correct? We made sure again that all four INV IGBTs were replaced properly on the previous repair which was conducted by another professional, as I stated before. The only -perhaps- negative thing we noticed is that one IGBT seems to be from a different batch, however it has the exact same model number. Additionally, all these IGBTs have almost the exact same values. Is that okay? One last component I don't feel okay about, and I think it was tampered with on the first repair / mess, is component D82 which is located at the bus soft start circuit. When I checked my inverter the very first time I noticed a very bad soldering job and a mess on the board around it. So, I'm planning to replace this at the same time with the MOSFET drivers (resistors) I need to replace next. This time we re-soldered it quickly till we replace it the next and hopefully last repair. Any thoughts? @Coulomb Would you please confirm that I'll be good with replacing all DC/DC drivers (resistors) with their original same values (SMD 47R0 and 4702)? And please answer my questions at the end of each of my previous paragraphs. PV & BAT MOSFETs Repair.zip

Please open a separate thread for your request, or post in another thread related to Axpert King firmware.

@monsam, welcome to the community! Sorry for my late response as I was sick the last couple of days. Would you please provide us with a photo of your inverter's sticker? You can blur or wipe out the last 6 digits of the serial number for privacy purposes, if you would like to. Another thing, did you buy your inverter brand new? Or conducted any repairs? I recommend upgrading your display firmware, too! I'm interested too! Whether you used a Flash Disk and updated using OTG directly, or used a computer and a USB cable, would you please describe what hardware/cables did you use exactly in each update attempt? If you'd go with quieter fans, consider testing 60.95 in the future, as 60.10 doesn't have quiet fans logic, but not before solving your grid bypass mode issue! Just to make things clear, so you're saying that you're measuring the AC output with an external ammeter and it still gives you that there's an output of around 9 Amps? Did you try wiring an electrical extension lead/strip with no loads connected to it? I mean Instead of wiring your actual home loads on the AC output of the inverter.

I'll be more than happy to assist in the rewrite after a while 😁 I just had my inverter's PV and Battery MOSFETS replacment job done today. I have very important updates that I'll post in a couple of hours in my respective thread, and ask for more assistance! After I finish the whole repair process, I'll begin with the calibration project by taking the promised readings first, gotta need your rewrite then! 😁

Thanks for joining in! As far as I'm aware of, there's still only one hardware version of the VM IV display. Hence, only one major firmware version for each VM IV type (Twin and Non-Twin). As for your VM IV Twin display, the only major firmware version we're aware of is (39.xx). Though, there's one minor update that we found out recently, which is 39.27, you can find it in the following post: https://powerforum.co.za/topic/29708-collection-of-dsp-axpert-firmware-for-vm-iii-twin-and-vm-iv-twin/?do=findComment&comment=210925 Regarding whether or not it's possible to figure out which MCU version to use based on the DSP version, I'm not aware of any clues that determines which goes with which, especially after my recent findings for the VM III series.

Unfortunately, it seems like our inverters do not make as accurate measurements as our BMSs. But, I still believe it's not something that we really need to worry about, it might be due to cheaper components than other higher value Axperts. I guess, if @Coulomb by the end of analyzing this phenomenon with us, says that what we need to worry about only is calibrating the voltage reading to the BMS with no current flowing, and it's safe otherwise. In this case, I think we'll be fine with these VM models! I'm more excited now to take my readings as soon as possible, hoping to have the needed free time soon! It would be very helpful if you could provide us with more new readings for different operation scenarios from your inverter so that we can compare. @esmail-kassir too? I think this is not a bug, but due to the display design which refreshes data. It's obvious that the display in VM IV models doesn't show real time data. Thanks for joining the discussion!

Since around a couple of months I'm certain that my inverter's voltage measurement needs calibration, as I always have +0.1 V difference than the battery's BMS readings, most obvious in the middle of the day when the battery is topped off and current is 0 A. Anyway, I'm certain that the BMS readings is the accurate one as I've taken so many measurements at the battery's external terminals with my accurate multimeter in so many times and scenarios, previously. However, I'm used to observing very much higher voltage difference between my inverter and BMS (depending on the displays). The higher the charging current the higher the voltage difference I observe. I recall that it can reach up to a +1 V when charging on grid with 80 A. I'll try to take measurements on both terminals soon. @esmail-kassir would you please tell us the difference you observed on your devices when you did your experiments? I think you would relate a lot to this from what I understood earlier. All too well (like Taylor Swift says), and checked several times some time ago. Will do for sure, as I'm very much interested in finding the answer. I once tried applying your calibration instructions but didn't really understand what I was doing. Will do another and better try after taking my multimeter measurements, and might even ask you for some help.

Following up on this, I want to discuss this because I had similar doubts when I first transitioned from Tubular Lead Acid batteries to LiFePO4 with BMS a few months ago. Now, I do believe that this difference is totally normal due to the fact that a BMS measures the voltage at its own terminals, and the inverter at its own terminals, which means that they're not measuring at the same electrical point (hence a voltage difference), which explains a big voltage difference when charging an empty battery using a high current. @Coulomb would you please confirm to us that this is a normal behavior, at least for the VMs? I need to rest assured that's there's nothing wrong with this.

Yeah, me too. So, this is not a possibility for you and me. I just wanted to make sure again that you have the exact same case as me. WallK has his inverter connected to the BMS, unlike us.

Hmmm, could it be the CV timer in the BMS stopping the charging process? Solar charging is usually slower than AC charging, which might very well exceed the absorption timeout before getting the battery fully charged at first try @WallK @Coulomb @bratpit ??? If this is the case, then the simplest solution would be increasing CV timeout, or decreasing Float timeout, or both to reasonable values! Thank you @esmail-kassir for pointing this out!

If you have your inverter connected to your battery's BMS, this looks like the works of the BMS charging timers (for CV and Float), it's how BMSs normally operate. If this is the case, it is irrelevant to our issue.

I think you have some BMS terminology misunderstanding. Over voltage is a dangerous voltage you never wanna reach. Balancing starts at a balancing voltage (not OV) and it's totally different for the two methods of balancing (passive and active) with the latter being the better and newer technology, and which all new BMS products are starting to adopt it. 100% SOC is something defined in your BMS settings, though not all BMSs give you the choice to change this. I have this happening on both charging sources! @bratpit That's exactly what I've been trying to say since the very beginning, thank you Coulomb for joining in, finally! I have my BMS set at 3.5 for CV and 3.4 for float, but since I have the CV issue in my inverter, and my BMS comm is faulty and unable to connect to my inverter too, I got both on 3.5 V at the inverter level, I just hope to solve my problems soon enough to let my battery float, finally! Then, I'll even lower my CV to 3.450 and Float to 3.350, the best I've ever known! Side note: Unfortunately, I'll be discussing replacing the BMS or the whole battery under warranty with my seller next week, as my BMS communication board proved to be faulty, finally! Thanks for the rest of your explanation in your post, I always appreciate your advice. Dear @WallK, I really appreciate you trying to understand how everything works, but I think you really need to consider my words, at least! Trust me, lower the voltages you have. To gain yourself some more LFP batteries knowledge, go check out the Voltage vs SOC curve for LFP, this chemistry is crazy near the two curve knees! You never wanna go near them, that's were voltage is not stable anymore and you will struggle with balancing, and of course it's the danger zone! I'll end the discussion here because we went way too off-topic! I totally agree, and of course looking for a resolution for all of this!

My bad, you're totally right, sorry for the misinformation! I deleted it in my previous post. Sorry if I wasn't clear enough. The settings I shared do not exclusively apply to JK BMS, and I didn't assume that you have one, I just wanted to provide you with an example of good LFP battery settings. The suggested settings you're given by your seller are way too high for an LFP, I confirm again. A 3.6 volt per cell is near the max limit of safe LFP voltage, and is a recommended "over voltage protection" setting for LFP! I just hope you never try their "Max Absorption Voltage" (3.65 VPC)!!!! Anyway, if you're still not convinced and don't want to research, you better ask other battery experts about the values you got. I'm just trying to grab your attention for your own safety. I have no personal benefit out of this. Any BMS when connected to an inverter will change charging settings of the inverter to what is set in the BMS settings, it's how the inverter - BMS communication works. Whatever battery settings you set in your inverter will get changed as soon as you connect your BMS successfully to the inverter. Nevertheless, I will keep looking for a fix for that "not reaching CV stage" issue that I, you, and @bratpit have! I will mention you in any future findings in this regard!

Here are more HEX File Only Packages of (DSP / Main CPU / U1) firmware for VM III TWIN and VM IV TWIN. An updated list of available packages in this thread can be found in the very first post at: https://powerforum.co.za/topic/29708-collection-of-dsp-axpert-firmware-for-vm-iii-twin-and-vm-iv-twin/ CREDITS: Original uploaders of the attached firmware files: @Mohammad Jawish DSP version 60.10 dated 2024/06/14 Note: Quiet fans control logic is not available in this version + I copied this firmware's full package from the uploader's provided FTP link in his own post Link to original package post @Mohammad Jawish DSP version 71.03 dated 2024/03/12 Note: I copied this firmware's full package from the uploader's provided FTP link in his own post Link to original package post USE AT YOUR OWN RISK, NO GUARANTEES WHATSOEVER! DSP Reflash Tool is always available in the very first post of this thread at: https://powerforum.co.za/topic/29708-collection-of-dsp-axpert-firmware-for-vm-iii-twin-and-vm-iv-twin/ Detailed instructions of how to update/reflash firmware of Removable and Round Display inverter models, posted by @Coulomb at: https://forums.aeva.asn.au/viewtopic.php?p=87767#p87767 If you think your inverter is bricked, there are a number of helpful tips posted by @Coulomb at: https://forums.aeva.asn.au/viewtopic.php?p=59897#p59897 Enjoy! VM TWIN (III & IV) DSP 60.10 hex.zip VM TWIN (III & IV) DSP 71.03 hex.zip

It is low enough indeed, as long as it's 0.5C (half the total capacity) you're on the safe side for sure. I'm sorry to tell you this, but your settings are way above safe limits for LiFePO4 chemistry! Please search online for the recommended settings for a 48V LFP battery. I do prefer Andy's Off-Grid-Garage BMS settings (and I use them as-is for my JK-BMS), this guy is a genius! Check out his own settings at this link from his own website, or at least take some notes from the entire website, he has a great YouTube channel too, don't miss it. I believe that you'll solve the whole issue when you use the recommended LFP battery settings. Thank you for this info! We hate their bugs for sure!!!

Now, this at least proves my recent findings which I discussed in my previous post. You both had the same original MCU's major version number of 57.xx and you both have the push button display variant! @Khalil Koja I would really love to hear back from you after you reflash the recent MCU firmware I found for you a few posts earlier, hoping it solves your inverter's buttons issue!