Page 66 of 98

Re: PIP-4048MS inverter

Posted: Fri, 27 Oct 2017, 13:56
by paulvk
weber wrote:
Fri, 27 Oct 2017, 08:25

Just in case anyone doesn't know, our patched firmware for the PIP also fixes that problem.
Not totally it is indeed much better, but it will cut off if a large load is applied to the inverter for a period of time thus spending too short a time in bulk , whereas the eSmart remembers its not been at bulk volts for long enough and picks up the ball and finishes the job. My battery performance has been better since adding the eSmart charger. I have two eSmart MPPT chargers the older version with two digit LED display running on my small one inverter set up (had to modify it to work with 3 x 24v as it cut off at 100v PV input) and the new one with LCD display on the large two inverter system this comes in a 130v PV input version

Re: PIP-4048MS inverter

Posted: Sat, 28 Oct 2017, 08:31
by T1 Terry
Are there updates for the 4000w 24v unit that stop the fan running 24/7? I've turned the fan over so it now blows up and that has greatly improved the noise and cooling, but the fan running 24/7 just wastes battery storage that isn't really necessary.

If this should have its own thread, please move it so it doesn't distract from this 48v version thread

T1 Terry

Re: PIP-4048MS inverter

Posted: Mon, 30 Oct 2017, 18:14
by vulcanescu35
hello guys,

i am new to the forum and i have found here a great documentation for MPP, i own a PIP 4048 MSD and i wanted to know if the modified firmware is suitable and for this version because the bulk current with the original firmware configuration is just some seconds,

thnak you in advance for you answers,

Re: PIP-4048MS inverter

Posted: Mon, 30 Oct 2017, 19:55
by weber
vulcanescu35 wrote:
Mon, 30 Oct 2017, 18:14
i own a PIP 4048 MSD and i wanted to know if the modified firmware is suitable and for this version because the bulk current with the original firmware configuration is just some seconds
Hi vulcanescu35. Thanks for your kind words. Sadly no, we do not have access to a firmware update file for the MSD or MST models, so we are not able to patch them. Can you please tell us what the "U1" firmware revision number is for your PIP?

Re: PIP-4048MS inverter

Posted: Tue, 31 Oct 2017, 00:23
by vulcanescu35
weber wrote:
Mon, 30 Oct 2017, 19:55
vulcanescu35 wrote:
Mon, 30 Oct 2017, 18:14
i own a PIP 4048 MSD and i wanted to know if the modified firmware is suitable and for this version because the bulk current with the original firmware configuration is just some seconds
Hi vulcanescu35. Thanks for your kind words. Sadly no, we do not have access to a firmware update file for the MSD or MST models, so we are not able to patch them. Can you please tell us what the "U1" firmware revision number is for your PIP?
hi again,

Yes, of course,

MAIN CPU Vers: 75.20
SCC1CPU :1.02
SCC2CPU: 1.02

regards,

Re: PIP-4048MS inverter

Posted: Wed, 01 Nov 2017, 23:07
by frnandu
I've been playing with my PIP, trying to charge from AC (utility), I have no solar connected to it at this moment.
I have to change not only "Charger source Priority" to a setting that includes Utility, but also I have to change "Output Source Priority" to Utility otherwise it will not charge the battery from AC. That makes some kind of sence, because I guess it isn't able to do simultaneously charging and discharging battery. Which means that either the DC->AC is working or the AC->DC is. Battery mode or Line mode, one or the other, both it's not possible.
My problem is that when I decide that it's enough charge from AC and want to change back to battery, changing "Output Source Priority" to SBU doesn't make the PIP change back to Battery Mode, it stays in Line mode. Only after turning power button off and on will it come back to Battery mode.
Did you guys notice this same behaviour?
Is there any workaround?
I'de like to remotely/programatically be able to charge from AC if the battery is below certain SOC during night cheap tariff.

Re: PIP-4048MS inverter

Posted: Thu, 02 Nov 2017, 06:08
by weber
Hi Frnandu,

1. When you change output source priority (parameter 01) from UTI to SBU (or SOL), the PIP will only change back to battery mode when the battery voltage has been higher than or equal to the back-to-battery setting (parameter 13) for 10 minutes. In the next release of our patched firmware (72.70c) we will reduce that time to 1 minute.
2. Setting charger source priority (parameter 16) to OSO (only solar) will stop utility charging while still in line mode. The serial command is PCP03. You are correct that utility charging can never occur in battery mode because the utility charger uses the same hardware as the inverter, operating in the reverse direction.
3. By setting the back-to-utility voltage (parameter 12) appropriately, you can leave the output source priority set to SBU, and then you only need to send PCP02 at the start of the cheap off-peak tariff and PCP03 at its end. I am currently doing this.

Re: PIP-4048MS inverter

Posted: Tue, 07 Nov 2017, 11:25
by weber
Beta-versions of Patched Firmware 72.70c with Dynamic Current Control

[Edit: Superceded by version 72.70c, but the following is the only description of these bug-fixes and features, which are retained in later versions.]

After 5 weeks, and about 18 cycles of analysis, design, programming and testing, Coulomb and I are finally making the result of our latest PIP-patching effort available for those brave and kind souls who are willing and able to test it. We are of course running it in our own systems, but these are of limited variety, both being single-inverter systems with LFP (Lithium Ferrous Phosphate) batteries.

The new version has all the features of version 72.70b, plus the following additional features:

1. We have fixed the bug found by paulvk where slave inverters in a parallel or 3-phase setup would still count time spent well below the absorb voltage setting (parameter 26), e.g. due to heavy loads, as if it was absorb time, after having once been within 0.5 V of the absorb voltage setting.

2. When the inverter switches to line mode (utility mode), due to the battery voltage falling below the back-to-utility setting (parameter 12), there is a minimum time that it must stay in line mode, even though the battery voltage may reach the back-to-battery setting (parameter 13) sooner. We have reduced this time from 10 minutes to 2 minutes.

3. We have improved the accuracy of the battery charge and discharge current readings, particularly for low currents, as displayed on the LCD, and as given in response to the QPIGS and QPGS serial commands. There is more explanation of this in the Dynamic Current Control manual linked below.

4. We have made it work with SCC firmware revisions from 1.24 to 4.10 (and probably later), irrespective of the setting of maximum total charge current (parameter 02), so there is no longer any reason to upgrade your SCC firmware, which is a difficult and dangerous process.

5. While the version number shown on the LCD, for the LFP release version, will still begin with "LF", the version number for the non-LFP release version will no longer begin with "Pb". It will instead begin with "LC", which stands for both Lead-aCid and Lithium Cobalt-blends (LCO, NMC, NCM, NCA, LTO). But for these beta versions, a "B" has been substituted for the "L" in both cases.

6. And finally, for lithium battery users, what this version is really all about: We have succeeded in implementing (drum roll please) Dynamic Current Control. :) You can read all about this in its own (3 page) manual linked below. This manual is also included in the zip files linked below.

Dynamic current control.txt
(9.08 KiB) Downloaded 118 times
[Edit: Updated to release version]

Here are the zip files with all the software you need, to reflash your PIP-4048MS or Axpert MKS 5K-48 with these beta-test versions, and to revert to standard 72.70 firmware if required. The instructions are the same as for version 72.70b except that upgrading the SCC firmware is no longer recommended.

[Edit: You can now download the release versions here.]

For lithium ferrous phosphate (LFP) (15S or 16S)
dsp_BF1_72.70c.zip [beta version no longer available]


For lead acid (24S), lithium cobalt-blends (LCO, NMC, NCM, NCA) (14S) and lithium titanate (LTO) (21S)
dsp_BC1_72.70c.zip [beta version no longer available]

Please let us know if you have used it for a week with no problems.
Any problems you find should be reported to me via this forum's private message facility. Please give as much detail as possible to enable us to reproduce the problem, including the versions of both the main and SCC (u2) firmware as displayed on the LCD, and the values of parameters 01 02 11 12 13 16 26 27 29 31 32 and any others you think might be relevant. Also the type and capacity of your battery and PV arrays and the number of inverters and their configuration. Thanks.

[Edit: Changed feature 2 description from: about 1 minute after above param 13, to: a minimum of 2 minutes after below param 12.]

Patched Firmware 72.70c

Posted: Wed, 08 Nov 2017, 03:34
by JvdSpoel
Hi, I am running 3 x inverters in parallel. They are currently running 72:70B - (4:10) Running on lead acid batteries. Happy to test new version. I assume it is the dsp_bc1?
Also running Inverter control Centre on my system. Anything I need to check before upgrading.

However what will be the main benefit for doing the upgrade? My system runs in SOL mode.
Regards,
Johan

Re: Patched Firmware 72.70c

Posted: Wed, 08 Nov 2017, 07:59
by weber
JvdSpoel wrote:
Wed, 08 Nov 2017, 03:34
Hi, I am running 3 x inverters in parallel. They are currently running 72:70B - (4:10) Running on lead acid batteries. Happy to test new version. I assume it is the dsp_bc1?
Also running Inverter control Centre on my system. Anything I need to check before upgrading.

However what will be the main benefit for doing the upgrade? My system runs in SOL mode.
Regards,
Johan
Thanks Johan, for being willing to test. Yes, it is the dsp_BC1 for lead acid. Since you are already running 72.70b and 4.10, there is nothing you need to check before upgrading. The main benefit for you will be the bug-fix for the absorb time for the slaves, assuming your slave inverters have PV arrays attached.

After upgrading, you should re-set parameter 02 on all inverters. It will have been reduced by 10 amps.

For the benefit of other readers with similar questions: Counting backwards through the new features, as listed in my previous post:

6. Lead acid users will not benefit from the new dynamic current control feature. That requires a battery management system (BMS) to send commands to the serial input of one of the inverters. BMS are typically only used with lithium ion batteries. The BMS will need to have its software modified, or perhaps have an add-on controller, such as an Arduino, Raspberry Pi or Beaglebone Black, programmed to send the commands.

5. Changing the battery type code in the version number from "Pb" to "LC" (temporarily "BC" for the beta version) is of no benefit to anyone. It's just something we needed to tell you about. But maybe it saves some users of Lithium Cobalt-blend batteries having to explain why their version number has lead (Pb) in it. While Lead aCid users can now explain that the authors of the patches apparently can't spell, since they seem to think "acid" starts with a "C". ;-)

4. Since you're running 4.10 in your SCC, this will be of no benefit. But those running earlier versions will benefit from being able to set parameter 02 higher than 60 amps, for when solar and AC charging are happening at the same time.

3. Everyone will benefit from the greater accuracy of charge and discharge current reporting. It's refreshing to have some honesty from the inverter for a change. ;-) e.g. When operating with no load, it now admits that it is still drawing 1 amp from the battery.

2. On-grid users may benefit from not always charging the battery from the AC input for a minimum of 10 minutes despite having reached the back-to-battery voltage much sooner.

1. Users of parallel or 3-phase systems will benefit from fixing the absorb-time bug for slaves with PV arrays.

Re: PIP-4048MS inverter

Posted: Fri, 10 Nov 2017, 07:33
by dinu_tiberiu_george
Hy,
I have connected 2 5048ms inverters in parallel, and i need to set up the charging current of the inverters. Maximum charging current for my batteries is 40amps. I need to set up each inverter to 20amp resulting a total 40amp? Or only set up the master unit to 40amp? Thx in advance

Re: PIP-4048MS inverter

Posted: Fri, 10 Nov 2017, 08:02
by weber
dinu_tiberiu_george wrote:
Fri, 10 Nov 2017, 07:33
Hy,
I have connected 2 5048ms inverters in parallel, and i need to set up the charging current of the inverters. Maximum charging current for my batteries is 40amps. I need to set up each inverter to 20amp resulting a total 40amp? Or only set up the master unit to 40amp? Thx in advance
You need to set the maximum charging current separately on each inverter, so that they total 40 amps.

Re: PIP-4048MS inverter

Posted: Fri, 10 Nov 2017, 13:59
by sjbw
Will this work with the hybrid-v series?

Re: PIP-4048MS inverter

Posted: Fri, 10 Nov 2017, 14:15
by coulomb
sjbw wrote:
Fri, 10 Nov 2017, 13:59
Will this work with the hybrid-v series?
No. These operate very differently, so they can't use the same firmware.

If you are asking could something similar be done with the V series, I suppose that's possible, but Weber and I won''t be doing it.

Re: PIP-4048MS inverter

Posted: Fri, 10 Nov 2017, 23:15
by ghatikar
I have a PIP MS4048 x 2 parallel setup and a Spare 4048 :D

Do I have to sacrifice the USB( serial port) to BMS connectivity? Or Can I connect BMS to one inverter and ICC to another inverter.

Does the BMS have to connect to master only or can be connected to slave - The master and slave in Axpert change dynamically depending which one was switched first

Re: PIP-4048MS inverter

Posted: Sat, 11 Nov 2017, 07:52
by weber
ghatikar wrote:
Fri, 10 Nov 2017, 23:15
Do I have to sacrifice the USB( serial port) to BMS connectivity? Or Can I connect BMS to one inverter and ICC to another inverter.

Does the BMS have to connect to master only or can be connected to slave - The master and slave in Axpert change dynamically depending which one was switched first
Thanks ghatikar. Those are excellent questions. Although we have not tested it [Edit: We have tested it now, and it works fine], we have every reason to believe that it does not matter which inverter the BMS is connected to, for the purpose of dynamically controlling the charge current. It can be the master or any of the slaves. That is certainly our intention, for the very reason that you gave. If this turns out not to be the case, this will constitute a bug in our patches that we will fix.

I don't know what an ICC is, but assuming it too doesn't care whether it is talking to a master or a slave then yes, you can connect the BMS to one inverter and the ICC to another.

I note that every MNCHGC command must contain a machine number (a single digit between 0 and 8). The machine receiving the MNCHGC command on its serial port, compares that with its number. If they are the same, it deals with it locally. If different, it sends a message on the CAN bus to all the other inverters who then do the same comparison and either deal with it or ignore it.

For example, in a parallel system with two inverters, the BMS will need to send both a MNCHGC0<nnn> command and a MNCHGC1<nnn> command. If the BMS wants to reduce the total current to 2 amps towards the end of charge, it could send MNCHGC0501 and MNCHGC1501. To reduce the total current to 1 amp, it could send MNCHGC0500 and MNCHGC1501.

The dynamic current control manual mentions that the current can't be changed more often than once every 2 seconds. But that is for a given inverter. It is not a limitation on how often a MNCHGC command can be sent. So in a system with two inverters, a MNCHGC command might be sent every second, provided the machine number alternates between 0 and 1, and provided each command is acknowledged before the next is sent.

If MNCHGC commands are addressed to the same inverter more often than every 2 seconds, nothing bad happens. Some of the MNCHGC commands will simply have no effect, even though they may be acknowledged. The 2 second limitation is due to the internal communication between an inverter's main processor and the secondary processor in its solar charge controller, and our decision not to patch the SCC firmware.

Some serial commands, such as the voltage setting commands PCVV, PBFT and PSDV, do not take a machine number as an argument in the way that MNCHGC and QPGS do, but they automatically get applied to all machines, no matter whether the machine receiving them is a master or a slave.

Other commands, such as QPIRI, that do not take a machine number, only return information about the inverter whose serial port the command is received on. But it doesn't matter whether that machine is a master or a slave. The issue in that case is not master versus slave, but local versus remote.

Re: PIP-4048MS inverter

Posted: Sun, 12 Nov 2017, 10:27
by paulvk
With parallel inverters the current settings are not linked I have to set charge settings on each individually in a way I can see some reasoning behind this
The SCC may have different numbers of panels
The AC chargers may not all have grid/generators attached
But it could have been another command for each that sets them all the same.
I think there is enough work to get the bugs sorted without going to more commands.

Re: PIP-4048MS inverter

Posted: Sun, 12 Nov 2017, 11:53
by andys
Thanks for the hard work. I am looking forward to trying it out.

My usage is likely to be to default the current to zero (for safety / full battery), and have the BMS set maximum possible charge current when the battery is not full, and not really anything in between.

Re: PIP-4048MS inverter

Posted: Mon, 13 Nov 2017, 16:06
by coulomb
weber wrote:
Fri, 06 Oct 2017, 08:00
hennejg wrote:
Thu, 05 Oct 2017, 04:03
Somewhere you mentioned that you sniffed the traffic between SCC and main controller. Is this data available anywhere?
No. That method was long ago superseded by coulomb reading the disassembled code, for both the main controller and the SCC, to find out what they send to each other.
Actually, for debugging our dynamic current control code we found it necessary to re-create the sniffing hardware. Here is the schematic for what we used. The sniffing circuitry that we added is inside red rectangles.

SCC IO.png
SCC IO.png (32.54 KiB) Viewed 5072 times
[ Edit: where lines cross without a dot, they do not connect. I try to avoid four-way joins to prevent confusion.]

The IFO LEDs mentioned above are Industrial Fibre Optic connectors, as used in our LyteFyba Battery Management System.

In case you're interested in what the traffic looks like, here is a short sample, from DSP to SCC:

Code: Select all

RSTq3                                                                                     
RSTq3                                                                                     
(PI99äé                                                                                   
(ACK9                                                                                     
(RI048.0 12.0 04 060.0 055.2 053.7 060.0 065.0)å                                          
(GS00 000.0 051.5 000 0000 0001 0 10©                                                     
(GS01 000.0 051.4 000 0000 0001 0 10Åí                                                    
(GS01 000.0 051.4 000 0000 0001 0 10Åí
The above is a startup sequence. RST is the "reset" command from the DSP; the remaining traffic is responses to SCC commands. The first "(ACK9" tells the DSP that the SCC has received the RST, and is powered up. "(PI99" is a response to the "QPI99" command. The "(ACK9" (acknowledgement with partial CRC) is a response to the VERFW command, which contains the SCC's firmware version. The "(RI" response tells the SCC some settings. Similar RI responses, in place of later GS responses, are use for dynamic current control. Finally, the remainder of the traffic will be GS responses, which contain measurements. [ Edit: NOTE that these commands can't be sent from the PC to the DSP; these are from the DSP via a second RS232 port directly to the DSP. You can't talk to the SCC directly from the PC (i.e. via the external RS232 / USB connection). ]

Here is a startup sequence from the SCC to the DSP:

Code: Select all

(ACK9
QPI99Q
VERFW:0004.10
QRIÏ+
QGS092.8 54.06 013.9 000 0026 0026 0 11 000000000Q
QGS090.7 54.12 015.6 000 0026 0026 0 11 00000000Ü%
QGS087.3 54.19 016.9 000 0026 0026 0 11 00000000zu
[ Edit: Mentioned what IFO LEDs are and linked to LyteFyba topic. ]
[ Edit June 2019: Added ACK after RST, more double quotes for literal commands and responses. ]

Re: PIP-4048MS inverter

Posted: Mon, 13 Nov 2017, 21:46
by weber
Referring to the DSP <-> SCC comms in Coulomb's post above: We implement dynamic current control by occasionally sending an unsolicited RI packet in place of a GS packet. But as well as sending the required current setting in the appropriate field of the RI packet, we have to put a deliberate bad value in one of its later fields to exploit a loophole in the SCC code, to make the SCC accept the new current setting without stopping charging. And to obtain currents below 10 amps we also have to fiddle the current measurement sent in the GS packets, and set some flags to tell the SCC that AC charging is on, when it isn't really.

And that isn't the end of the packet fiddling. I'll spare you the details, but suffice it to say that during development it seemed like every time we fiddled a field in one of the packets, we'd be sure that it would do the job, and then on testing we'd despair when it still didn't work. It seemed like the software gods were tormenting us. But we were determined not to patch the SCC code, only the DSP. We were close to giving up, several times, but then we'd analyse the SCC code some more and realise we could fix the latest problem by fiddling some other field of either the RI or GS packet to compensate. But on testing, we'd find it still didn't work fully. This sequence of events was repeated 5 times, over a number of weeks, before it finally worked in all cases!

In the end we had written 16 short subroutines, called from 20 locations in the original code, just to implement dynamic current control. Then, to allow a zero EEPROM setting of current, and to improve the accuracy of current reporting, required another 8 subroutines, and a table of allowed current values, patching another 15 locations in the original code.

Re: PIP-4048MS inverter

Posted: Tue, 21 Nov 2017, 00:23
by PurePower
weber wrote:
Tue, 07 Nov 2017, 11:25
Beta-versions of Patched Firmware 72.70c with Dynamic Current Control
...
Please let us know if you have used it for a week with no problems.
It's been almost 2 weeks... any issues? I'm keen to upgrade to the latest firmware for my LifePO4 batteries :)

Re: PIP-4048MS inverter

Posted: Tue, 21 Nov 2017, 06:39
by weber
Hi PurePower. Yes, it has been 2 weeks now. There have been more than 10 downloads of the LFP version and the same for the Lead-acid/Lithium-cobalt-blend version. No one has reported any issues. In fact, no one has reported anything, but I think we can take it that no news is good news. :)

Coulomb and I did however agree that it's still a little too difficult to exit the absorb stage and go to float with our LFP version, and so we will reduce the divisor that we use to calculate the low current threshold for LFP batteries, from 15 to 12. The threshold is calculated by dividing the maximum total charge current setting (parameter 02) summed over all parallel machines, by this divisor, and clamping the result up to a minimum of 5 amps times the number of parallel machines. But for various reasons, the charge current, as measured by the PIP/Axpert in whole amps, has to fall to 2 amps less than this threshold, and stay there for 30 seconds, before it will exit the absorb stage and go to float. This applies only when absorb time (parameter 32) is set to "Aut".

Re: PIP-4048MS inverter

Posted: Sat, 25 Nov 2017, 14:59
by Howard
Hi Guys,

Thanks so much for the new 72.4 firmware for Pb from old 52.3. Instructions work fine --I was a bit stressed.Had to make serial cable ..but thanks to paukvk was to get the pin outs for D9 and Mod Plug because I don't think I received cable with my Inverters. Anyway "Install success" with trumpet call with old XP machine with serial Fem. D9! Thanks!
Still working up to 2 PIP4048 IN PARALLEL.Changed caps as recommended in past posts -works fine thank you.
I thought 72.4 firmware had new setting 38 for dry contact Neut. earth and AGM Pb bulk charge mods?
The dry contact is the Gen.contact relay connection in this Inverter to switch my contactor neutral -correct?

Question is can I now upload 72.7 ..? for those fixes an waht version is the working version to date?

Thanks for All your work,
Cheers

Re: PIP-4048MS inverter

Posted: Sat, 25 Nov 2017, 16:21
by coulomb
Howard wrote:
Sat, 25 Nov 2017, 14:59
Instructions work fine --I was a bit stressed. ... Anyway "Install success" with trumpet call with old XP machine!
Glad to hear you were able to successfully update your firmware.

Actually, I think you'll find the final message is "Updata sucess" [sic]. Weber and I have seen that message many times :o
I thought 72.4 firmware had new setting 38 for dry contact Neut. earth and AGM Pb bulk charge mods?
No, settings 38 and 32 (the two you mention above) only appeared in 72.70, and don't exist in 72.40.
The dry contact is the Gen.contact relay connection in this Inverter to switch my contactor neutral -correct?
I've never used it, and I've not checked the firmware, but that's my understanding. If setting 38 is enabled (ENA instead of dIS), then the one "dry contact" relay is re-purposed for the neutral to earth connection instead of its usual function, which is to turn on when the battery is low.
Question is can I now upload 72.7 ..? for those fixes an what version is the working version to date?
My understanding is that there is no difference in applicability to hardware for 72.70 versus 72.40. In other words, if 72.40 is working for you, then 72.70 should also work just as well, and you will have these additional settings. That's why Weber and I dropped support for the 72.40-based patched firmware, and only update the 72.70-based (or later) patched firmware now.

I've not seen any firmware later than 72.70 in the 72.XX series. New (as of 2017) inverter chargers with single SCCs come with 73.XX, and dual/triple SCC models seem to come with 75.XX. The latter do NOT appear to be compatible with single SCC models, but I'm not totally sure on this point. [Edit: Delete: It is safe to "revert" new single SCC models that come with 73.XX to 72.70-based patched firmware.] 75.XX firmware (for dual and triple SCC models) do NOT contain settings 38 and 32 mentioned above.

Edit: 73.XX does retain settings 32 and 38. The latest official firmware for single SCC models is 73.XX. [Edit: Delete: In my admittedly biased opinion, patched firmware based on 72.70 is the most advanced firmware to be running].

[ Edit: I had mixed up 73.XX and 75.XX version numbers. ]

[Moderators note: Howard's response to the above has been moved to the PIP repairs and hardware modifications topic.]

Re: PIP-4048MS inverter

Posted: Fri, 01 Dec 2017, 15:51
by weber
Patched Firmware 72.70c with Dynamic Current Control and AussieView™

[Edit: This has been superceded. Please see the Firmware section of the index post to find the latest patched firmware for your machine.]

Our thanks to those of you who tested the beta version. We haven't heard anything from any of you, but we're assuming no news is good news. :) We did however have some helpful feedback from @PurePower regarding version 72.70b, which contributed to the second of the changes below. Thanks PurePower!

This release version has all the features of 72.70b and β 72.70c, with the following minor changes:

1. For the LFP version only, we have reduced the the divisor that is used to calculate the current threshold for going from absorb to float, from 15 to 12. The threshold is calculated by dividing the maximum total charge current setting (parameter 02) summed over all parallel machines, by this divisor (12), and clamping the result up to a minimum of 5 amps times the number of parallel machines. The charge current, as measured by the PIP/Axpert in whole amps, has to fall to 2 amps less than this threshold, and stay there for 30 seconds, before it will exit the absorb stage and go to float. This applies only when absorb time (parameter 32) is set to "Aut". This divisor remains 5 in the lead-acid/lithium-cobalt version.

2. For the LFP version only, we have lowered the voltage threshold for going from float to bulk, from 1.0 V below the float setting to 1.2 V below the float setting. This figure remains 4.0 V below the float setting in the lead-acid/lithium-cobalt version.

We've also added seven "easter eggs", that we collectively call "AusieView". We'll award coloured stars to posts that describe any of these easter eggs in a way that benefits others. They aren't exactly hidden. We're just too lazy to document them ourselves. :D We figure we've done enough work just implementing them. :geek: :ugeek: [Edit: I've now documented them in this AussieView Manual.] And here's the other manual,
Dynamic current control.txt
(9.24 KiB) Downloaded 148 times

Actually, it would be great if someone would write a manual for them, with pictures, to replace pages 14 thru 28 of this manual. A patched manual to go with the patched firmware. :) And while you're at it, please eliminate that confusion where they refer to "bulk" when they really mean "absorb" (twice). :o

By the way, if you find a parameter value that isn't in the manufacturer's manual, it's safe to assume it has the same meaning as the value that differs by one letter.

Here are the zip files with all the software you need, to reflash your PIP-4048MS or Axpert MKS 5K-48, and to revert to standard 72.70 firmware if required. The installation instructions are the same as for version 72.70b except that upgrading the SCC firmware is no longer required or recommended.

Edit note: This firmware is only suitable for the models with a single low-voltage MPPT and a power factor of 0.8 (4 kW / 5 kVA), not those with dual or triple MPPTs or a power factor of 1.0 (5 kW / 5 kVA), and not those with a maximum PV array open circuit voltage greater than 145 V DC. Machines that are paralleled all need to run the same firmware. [ Edit: not quite true; see this post. ]

For lithium ferrous phosphate (LFP) (16S or 15S)
dsp_LF1_72.70c.zip
(1.5 MiB) Downloaded 197 times

For lithium cobalt-blends (LCO, NMC, NCM, NCA) (14S), lithium titanate (LTO) (21S) and lead acid (24S)
dsp_LC1_72.70c.zip
(1.5 MiB) Downloaded 260 times

Note: Lithium ferrous phosphate (LFP) is the same as lithium iron phosphate or LiFePO₄. The name "lithium ferrous phosphate" is preferred, as it avoids the confusion between "iron" and "ion".

[Edit: Note: We no longer recommend putting version 72.xx (not even our patched versions) into inverters that are supplied with 73.00. We recently discovered that this causes a problem with a relay that has been added inside these new machines, that can connect Neutral to Earth on the AC output. See the Firmware section of the index post to find the latest patched firmware for your machine.]