PIP-4048MS and PIP-5048MS inverters

non-EV Solar, Wind and other renewable power sources
non-EV batteries and other energy storage stuff
Forum rules
Important!
This forum is for discussion of Non-EV matters.
User avatar
coulomb
Site Admin
Posts: 3678
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by coulomb » Tue, 20 Sep 2016, 17:42

Lopezjm2001 has an unusually high parasitic load on his battery. Weber determined that he needs a slightly different charge terminating condition, and sent him a modified LiFePO4 patched firmware (a patch on a patch, if you like).

Alas, this firmware update was interrupted, and now his machine doesn't seem to respond to the firmware reflash program. It happens that this is the second "bricking" that I'm aware on in the last week. One thing that John (Lopezjm2001) mentioned was the fact that after turning on his machine again after some failed reflashes, it does not even turn on the LCD backlight. This has the effect of making the machine appear "extra dead", not even doing the very basic task of turning on the LCD backlight. I imagine that this is likely to make the user, already dreading what has happened to his inverter, to feel even more dread.

It turns out that turning on the LCD backlight is a non trivial thing; it's not just flipping one IO bit as you might expect. The LCD backlight seems to be one of several things allocated to some sort of "IO expander", which involves an IIC (Inter Integrated Circuit protocol) port. The boot code doesn't seem to call this code, although interestingly the code to turn the light on or off seems to be in the same flash segment as the bootloader code.

At startup, the firmware does a simple sanity check on the firmware; if it doesn't appear to be complete, it won't attempt to run the firmware at all. This is a good thing, because if it did, it would completely crash, and you'd have no way of recovering from the firmware crashing every time it started. Instead, it runs a program called the bootloader, which attempts to "pull the inverter up by its bootstraps" by loading new firmware over the serial port. Of course, it needs a PC with all the right software waiting to send that firmware. This bootstrap loader is small and not at all fancy; the reason for this is that this part of the flash program area (one eighth, one "flash segment") doesn't get erased or updated by the normal reflash (firmware update) process. So they don't put anything fancy in there, because anything fancy will have bugs in it that need updating.

So the message here is: if only the bootloader is running, then don't expect the LCD backlight to come on, or anything to display on the LCD at all. Don't panic; just ignore the fact that you're "running blind" and keep trying to re-flash the firmware. As soon as you succeed, the full firmware should take over, which should give you full operation, including LCD and backlights.

[ Edit: in fact, not seeing the backlight come on is a good thing. It means that the bootloader is running, not the failed firmware. If you get the backlight on and it won't respond to firmware updates, then you have crashed firmware, and need to time your inverter time-on; see if you think your inverter is bricked. ]

Unfortunately, it's difficult to distinguish having just the bootloader running from a machine that has completely crashed. The sign of a healthy bootloader would be bytes coming back from the serial port, responding to bytes sent to it from the reflash program. You could see this with an oscilloscope or with a gadget that displays serial data on LEDs. Most people won't have either of these, so just trust that the bootloader is running, and retry the firmware update. If after several attempts it's obvious that it's not going to work, then the inverter really is bricked (not fatally though), and you'll need to read about JTAGing or replacing your control board (this is a small, easily replaced board that has the now-failed processor on it). JTAGing will almost certainly bring back the processor, but it's very technical and you need some hardware that isn't completely off-the-shelf.

Edit: more discussion on the reflash process immediately after the image in the LiFePO4 firmware post, and here (If you think your inverter is bricked).

[ Edit: rewrote to make it clearer that you don't need an oscilloscope or LED box if the backlight doesn't come on; just retry the flashing process. I also had an inadvertant word "not" in there that made it very confusing; sorry! ]
Last edited by coulomb on Wed, 21 Sep 2016, 05:10, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

andys
Groupie
Posts: 75
Joined: Wed, 20 Jul 2016, 19:44
Real Name: Andrew Snow
Location: Sydney

PIP-4048MS inverter

Post by andys » Wed, 21 Sep 2016, 14:15

Hi PIPsters,

I received this from MPP Solar support, regarding PIP charging logic.

Image

- Andrew

User avatar
coulomb
Site Admin
Posts: 3678
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by coulomb » Wed, 21 Sep 2016, 14:51

andys wrote: I received this from MPP Solar support, regarding PIP charging logic.

That's really interesting, thanks for posting.

I've been trying to reconcile what it says above with what Weber and I observed last year, and what I read in the firmware. I'm unable to do so.

I think what happens is that they get some query about how the PIPs charge batteries, do a quick search among the manufacture supplied documentation, and find nothing. Snap! Oh but wait, the PIPs have the same SCC as these stand alone units, and these have charge algorithm documentation; surely they must operate the same. So we'll send the info we have on those.

The only problem is that the "surely" part of the logic is wrong, at least as far as I can tell.

I note that the above refers to a 12 V battery. You could say "oh they mean each 12 V module, just multiply the numbers by 4 for 48 V nominal". (Or divide by 4 for per-cell LiFePO4.) But I think it shows that this is documentation for a 12 V stand alone MPPT charger, which has similar hardware (scaled up for 48 V operation), and similar firmware. But one of the differences in the firmware says in essence "except if the DSP overrides, disregard all this other logic and do what the DSP says". In a PIP-4048, we do have a DSP overriding, and the above documents the "other logic" that is being disregarded.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

andys
Groupie
Posts: 75
Joined: Wed, 20 Jul 2016, 19:44
Real Name: Andrew Snow
Location: Sydney

PIP-4048MS inverter

Post by andys » Wed, 21 Sep 2016, 14:58

Yeah, it makes no sense to me either (based on what you found, and what I observed in my own unit).

I think its become too complex for them to make changes easily.

It would be much simpler to wipe the slate clean, and only cater to Lithium packs..   we don't need all this complexity, just a normal 3 stage charger?

User avatar
lopezjm2001
Groupie
Posts: 85
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

PIP-4048MS inverter

Post by lopezjm2001 » Sat, 24 Sep 2016, 00:57

coulomb wrote: Lopezjm2001 has an unusually high parasitic load on his battery. Weber determined that he needs a slightly different charge terminating condition, and sent him a modified LiFePO4 patched firmware (a patch on a patch, if you like).

Alas, this firmware update was interrupted, and now his machine doesn't seem to respond to the firmware reflash program. It happens that this is the second "bricking" that I'm aware on in the last week. One think that John (Lopezjm2001) mentioned was the fact that after turning on his machine again after some failed reflashes, it does not even turn on the LCD backlight. This has the effect of making the machine appear "extra dead", not even doing the very basic task of turning on the LCD backlight. I imagine that this is likely to make the user, already dreading what has happened to his inverter, to feel even more dread.

It turns out that turning on the LCD backlight is a non trivial thing; it's not just flipping one IO bit as you might expect. The LCD backlight seems to be one of several things allocated to some sort of "IO expander", which involves an IIC (Inter Integrated Circuit protocol) port. The boot code doesn't seem to call this code, although interestingly the code to turn the light on or off seems to be in the same flash segment as the bootloader code.

At startup, the firmware does a simple sanity check on the firmware; if it doesn't appear to be complete, it won't attempt to run the firmware at all. This is a good thing, because if it did, it would completely crash, and you'd have no way of recovering from the firmware crashing every time it started. Instead, it runs a program called the bootloader, which attempts to "pull the inverter up by its bootstraps" by loading new firmware over the serial port. Of course, it needs a PC with all the right software waiting to send that firmware. This bootstrap loader is small and not at all fancy; the reason for this is that this part of the flash program area (one eighth, one "flash segment") doesn't get erased or updated by the normal reflash (firmware update) process. So they don't put anything fancy in there, because anything fancy will have bugs in it that need updating.

So the message here is: if only the bootloader is running, then don't expect the LCD backlight to come on, or anything to display on the LCD at all. Don't panic; just ignore the fact that you're "running blind" and keep trying to re-flash the firmware. As soon as you succeed, the full firmware should take over, which should give you full operation, including LCD and backlights.

Unfortunately, it's difficult to distinguish having just the bootloader running from a machine that has completely crashed. The sign of a healthy bootloader would be bytes coming back from the serial port, responding to bytes sent to it from the reflash program. You could see this with an oscilloscope or with a gadget that displays serial data on LEDs. Most people won't have either of these, so just trust that the bootloader is running, and retry the firmware update. If after several attempts it's obvious that it's not going to work, then the inverter really is bricked (not fatally though), and you'll need to read about JTAGing or replacing your control board (this is a small, easily replaced board that has the now-failed processor on it). JTAGing will almost certainly bring back the processor, but it's very technical and you need some hardware that isn't completely off-the-shelf.
Thanks for the info. At the moment when I power it up I just hear one click. Probably the sound of a mini relay switching on. And that's it, no lights, nothing. Just like if it was dead.
I tried to re-flash program again but it did not work. The program only went as far as "COM OPEN" and that's it. So I guess it must be bricked. Anyhow I have removed the inverter processor board and taken a few photos for the benefit of others.

Image

Image

Image

I would love to do the Jtag thing but it sounds like it is way over my head for an electrician. Anyhow I have asked mppsolar for a price for a new board already flashed and working. Needed to take the processor board out so I can send the photo to mppsolar for replacement.
Last edited by lopezjm2001 on Fri, 23 Sep 2016, 15:30, edited 1 time in total.

User avatar
lopezjm2001
Groupie
Posts: 85
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

PIP-4048MS inverter

Post by lopezjm2001 » Sat, 24 Sep 2016, 01:40

It's downloading. LCD has come back to life. I have put the original processor board back in and applied 240v to ac input only. Nothing else except Ethernet plug for comms. Beauty.. Image
Now I will try out Webers new Lifepo4 Patch div24
Last edited by lopezjm2001 on Fri, 23 Sep 2016, 16:09, edited 1 time in total.

User avatar
weber
Site Admin
Posts: 2595
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Sat, 24 Sep 2016, 22:17

That's great news, John. I look forward to hearing how the modified version goes.

Do you have any suggestions for improving the "What to do when you think you've bricked it" post?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
lopezjm2001
Groupie
Posts: 85
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

PIP-4048MS inverter

Post by lopezjm2001 » Sun, 25 Sep 2016, 02:10

I don't know. Maybe Coulomb can make sense out of my experience. He seems to have it all covered.

User avatar
lopezjm2001
Groupie
Posts: 85
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

PIP-4048MS inverter

Post by lopezjm2001 » Sun, 25 Sep 2016, 05:08

weber wrote: That's great news, John. I look forward to hearing how the modified version goes.

I have started using a Goodwe GW2500-BP dc-dc converter to charge the battery pack so I will no longer be using the patch. On the GW2500-BP all charge parameters are configurable without having to communicate with a BMS. I'm still using the PIP4048MS but without PV connected. Thanks.

EDITED:26 September 2016 : I have gone back to using the PIP4048MS for charging so I am now using the new lifepo4 div 24 patch.
Last edited by lopezjm2001 on Mon, 26 Sep 2016, 04:38, edited 1 time in total.

Northland
Groupie
Posts: 102
Joined: Tue, 19 Jan 2016, 21:03
Real Name: Paul
Location: NZ

PIP-4048MS inverter

Post by Northland » Sun, 25 Sep 2016, 17:31

I have several of these, cheap and very good quality clone. Plus it has serial out, no crc requests required
http://s.aliexpress.com/vuUV7fyQ

[ Edited Coulomb: removed spaces from URL (prevents it from working) ]
Last edited by coulomb on Mon, 26 Sep 2016, 04:58, edited 1 time in total.
My YouTube Channel DIY TESLA POWERWALL

User avatar
lopezjm2001
Groupie
Posts: 85
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

PIP-4048MS inverter

Post by lopezjm2001 » Sun, 25 Sep 2016, 17:50

Thanks. But your shortcut does not work. The BP series only came out recently.

User avatar
Adverse Effects
Senior Member
Posts: 1078
Joined: Sat, 01 Jan 2011, 03:30
Real Name: Adverse Effects
Location: Brisbane

PIP-4048MS inverter

Post by Adverse Effects » Sun, 25 Sep 2016, 22:16

Northland wrote: I have several of these, cheap and very good quality clone. Plus it has serial out, no crc requests required
http://s.aliexpress.com/vuUV7fyQ


fixed the link

12 V 24 V 48 V 60A MPPT controlador de carga solar MPPT150I60A com verdadeiro MPPT fuction

User avatar
lopezjm2001
Groupie
Posts: 85
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

PIP-4048MS inverter

Post by lopezjm2001 » Tue, 27 Sep 2016, 22:10

weber wrote: That's great news, John. I look forward to hearing how the modified version goes.

Image
I think it is safe to say my PIP4048MS went into float mode. Thanks.

solamahn
Groupie
Posts: 121
Joined: Thu, 18 Jun 2015, 00:09
Real Name: Julian Leach
Location: PNG

PIP-4048MS inverter

Post by solamahn » Sun, 02 Oct 2016, 19:02

Latest Job
48 x 300w MPPSolar PNG panels
16 x 10A DC breakers
2 x 20A RCBO, 2 x 10A mcb
Generator connection
2 x 4048MSD inverters in parallel
32 x 400Ah CALB Lifepo4
28 x 18w 1200mm LED tubes
4 x 50w LED spotlights on daylight sw
14 double GPO's
12 switches
Earth stake

Solamahn PNG

User avatar
weber
Site Admin
Posts: 2595
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Sun, 02 Oct 2016, 19:23

Sounds great. What about battery fusing and isolation?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

solamahn
Groupie
Posts: 121
Joined: Thu, 18 Jun 2015, 00:09
Real Name: Julian Leach
Location: PNG

PIP-4048MS inverter

Post by solamahn » Sun, 02 Oct 2016, 19:26

None
Solamahn PNG

User avatar
offgridQLD
Senior Member
Posts: 1825
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD » Sun, 02 Oct 2016, 21:12

Yikes....Hopefully for the unlucky person it's a quick death then with no interruptions.

User avatar
coulomb
Site Admin
Posts: 3678
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by coulomb » Mon, 03 Oct 2016, 02:13

The PIPs do have a 200 A fuse in them. I've now heard of one of them blowing [ Edit: but that might have been an external fuse ] (on the South African PowerForum inverters forum), so there is that protection.
Last edited by coulomb on Sat, 22 Apr 2017, 05:21, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

andys
Groupie
Posts: 75
Joined: Wed, 20 Jul 2016, 19:44
Real Name: Andrew Snow
Location: Sydney

PIP-4048MS inverter

Post by andys » Wed, 12 Oct 2016, 01:46

There seems to be a minor bug/issue with the charge logic:

Given these conditions:

1. recharging is nearly complete or already completed and on float, and
2. high AC load (3-4kW), and
3. high PV input (3-4kW)

If the load suddenly gets removed, the charge voltage seems to spike really high as it dumps kilowatts of excess PV input into the already-charged battery.

It takes 10 seconds to ramp down the PV input back to low levels and during that time the battery voltage goes high above float.

I don't really have a problem because I'm on LiFePO4 and I only charge to 90-95% so can easily soak up the excess charge.

But I imagine this would be bad if it happened repeatedly, or if you had a Lead Acid pack which had high resistance due to a recently finished charge cycle, the voltage must skyrocket?


Northland
Groupie
Posts: 102
Joined: Tue, 19 Jan 2016, 21:03
Real Name: Paul
Location: NZ

PIP-4048MS inverter

Post by Northland » Wed, 12 Oct 2016, 17:54

LEAD ACID PARADOX

So I was having a debate with someone on another forum.

He was charging his agm bank at night using off peak (no solar), then running his pip during the day. I explained that this was costing him money instead of saving it.

He strongly disagreed, explaining to me that agm was 100% efficient, if he put 100Ah in, he can get 100ah back out.

Knowing this to be false, I showed the math. I tried to find research to disprove the amps in vs amps out theory. But it seems this conversation is a first. I could not find anything one way or the other. Then by accident some time later I came across this scientific study
See the highlighted areas below

Image

If you operate your batteries between 100% and 80% SOC, then you will extract LESS THAN 50% of the energy put in to them.
Last edited by Northland on Wed, 12 Oct 2016, 07:19, edited 1 time in total.

User avatar
offgridQLD
Senior Member
Posts: 1825
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD » Wed, 12 Oct 2016, 18:05

Yes shallow cycling a big led acid bank gives terrible round trip efficiency.

From memory (need to check as I haven't checked it in a long time)After a lot of fiddling and trial and error I settled on something like 98...99% efficiency setting on my lithium bank.

Kurt

antiscab
Senior Member
Posts: 2406
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

PIP-4048MS inverter

Post by antiscab » Wed, 12 Oct 2016, 18:10

You'd have to include the inverters own round trip efficiency

yeah, charge off peak to support peak loads is a bad plan
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

PIP-4048MS inverter

Post by Johny » Wed, 12 Oct 2016, 18:22

antiscab wrote: You'd have to include the inverters own round trip efficiency

yeah, charge off peak to support peak loads is a bad plan
Currently true, however most of the electricity suppliers have put forward schemes to attempt to cut peak load by heavily penalising users who exceed their average usage by (insert formula here). The Victorian Minister for Energy has only agreed if it is an "opt in" scheme - which is kind of useless.

Anyway - in this case, charging a modest bank off-peak and supplementing the 5PM->8PM heavy use would be encouraged by suppliers.

The Grattan Institute published a paper on this:
https://grattan.edu.au/report/fair-pricing-for-power/

User avatar
offgridQLD
Senior Member
Posts: 1825
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD » Wed, 12 Oct 2016, 18:27

I was talking about the battery charging metering eficiancy measured from the Solar charge controllers. As in take a measured 2.5kwh out from the bank (DC) then return 2.5kwh DC.

(not DC kwh in from PV vs AC out from a inverter) That gets nasty even with the high efficacy of lithium (particular with big inverters running small loads)


Northland
Groupie
Posts: 102
Joined: Tue, 19 Jan 2016, 21:03
Real Name: Paul
Location: NZ

PIP-4048MS inverter

Post by Northland » Wed, 12 Oct 2016, 18:44

Let's say you want to use 1kw to charge. Just by having the pip connected, you waste 50w. So the actual consumption is 1050w.

Let's assume the charger is 90% efficient. Your 1000w is now 900w.
Your batteries store 50% (being generous) of that 900w. It's now 450w.
When you run the inverter from the batteries, you again pay a 50w penalty. You now have 400w.
The inverter is about 90% efficient. It's now 360w.

1050w in, 360w out. 34% efficiency.

That's assuming no large loads which will give a much worse result. Also ignored is the life used up in the batteries and inverter, which typically costs between 10c and 15c per kwh.
Last edited by Northland on Wed, 12 Oct 2016, 07:47, edited 1 time in total.

Post Reply