Page 24 of 109

PIP-4048MS inverter

Posted: Wed, 15 Jul 2015, 06:03
by weber
Good News. The Giant Power IPS-4000WM (an Australian distribution of the PIP-4048MS) is now listed on the CEC website as having IEC 62109 approval, as you can see if you go to this web page and type "Giant" into the Manufacturer field and hit the Submit button. ... rters.html

PIP-4048MS inverter

Posted: Wed, 15 Jul 2015, 14:29
by offgridQLD
So what are the benefits of that Weber?

Is it that you are you now able to apply for rebates when installing it in some offgrid situations?

Or some other benefits?


PIP-4048MS inverter

Posted: Wed, 15 Jul 2015, 16:24
by weber
offgridQLD wrote: So what are the benefits of that Weber?
Is it that you are you now able to apply for rebates when installing it in some offgrid situations?
Or some other benefits?

That was discussed two and a half months ago here:
viewtopic.php?title=pip4048ms-inverter& ... 332#p56615
and clarified here:
viewtopic.php?title=pip4048ms-inverter& ... 332#p56655

In short, on 11-July it became illegal to connect a PV array to any Giant Power IPS inverter (or MPP Solar PIP inverter) in Australia (for safety reasons). Now it's legal to connect a PV array to an IPS-4000WM (a possibly modified PIP-4048MS). None of the lower power or lower voltage models are shown as legal on the CEC web site as yet.

I note that Giant Power now offer pre-wired lead-acid off-grid systems based on the IPS-4000WM.

PIP-4048MS inverter

Posted: Wed, 15 Jul 2015, 17:02
by offgridQLD
Ok, now I remember to meet the requirements to do with earth fault detection on the PV side.

last Friday My PIP4048 has developed a tick...tick...tick noise inside last time I used it. Sounded I bit like a electric fence. Not sure if it was just the stiff clear plastic shroud catching on something or flexing and making a tick as the fans created turbulence. I haven't used it again but will investigate it more and see whats going on.

One thing I can say is the PIP4048 doesn't have a precharge system as mentioned befor. I have tested it several times now (even when I let it sit for a long time befor reconnecting) I always get the spark when connecting the Anderson plug to it from the battery.

PIP-4048MS inverter

Posted: Wed, 15 Jul 2015, 19:26
by weber
Queensland electricity charges

Those of us living in Qld and considering building load-shifting systems around the PIP inverter should see the details starting on page 2 of this Qld Gazette. All prices are excluding GST. ... -2015.aspx
The tariffs of interest to most of us are T11 Residential, T12 Residential (Time-of-Use), T31 Night Rate (Super Economy), T33 Controlled Supply (Economy).

T31 and T33 are relay controlled, guaranteeing to be on for 8 hours and 18 hours per day respectively. They can only be had in addition to a residential or business tariff like T11 or T12.

I note that both T31 and T33 now say they can be used to supply "Electric Vehicles, at the discretion of the distributor". The distributor is Energex (SE Qld) or Ergon (rest of Qld).

But in the case of the Super Economy T31 it says, "... but not applicable ... if provision has been made to supply such apparatus ... under a different tariff during the restricted period", which sounds impractical for EV charging for most of us.

And I note that the off peak rate of T12 (Time-of-Use) is less than T33 (but greater than T31).

The daily rate is now the same for T11 (flat rate) and T12 (time of use). So you can now reduce your bill if you can use batteries and an inverter/charger to shift your loads from peak (4 pm to 8 pm Monday to Friday) to off peak (10 pm to 7am any day). Of course the amortised cost of the batteries may well outweigh the saving.

But if you were one of those early adopters who did the right thing for the environment back when PV cost $10 a watt, and are being paid 44c to 50c per kWh for your PV exports until 2028 under the Qld Solar Bonus Scheme, then your "peak" time is really during the day, say 8 am to 4 pm, when the sun is shining. That's because any consumption at that time costs you 50c/kWh in missed exports. So you'd want to shift your loads from that time to T12 off-peak (19c/kWh inc GST).

If you want to change from tariff 11 to tariff 12 you will need to phone your retailer (e.g. Origin) and request the change. I've just done it. Don't be surprised if you have to tell them what tariff 12 is. On second thoughts, do be surprised! And if you have an EM1000 meter (they didn't know what that was either),
don't let them tell you that you will need to hire an electrician, or that you will need a new meter, based on some cockamamie theory whereby they get you to look at the back of your bill and if there are less than 3 lines that begin with your meter number, you will need a new meter! Respectfully say that you do not believe them and ask them to please check with your distributor. When they did, they learned that I had an EM1000 meter, which can definitely be reprogrammed for tariff 12. There is however a charge of $85 for the reprogramming, passed through from the distributor. And it may take up to 20 working days. [Edit: They also confirmed that changing to T12 would not affect my Solar Bonus Scheme entitlements.]

PIP-4048MS inverter

Posted: Wed, 15 Jul 2015, 20:03
by weber
Apparently Energex managed to convince the regulator to let them hit us with some additional charges not listed in the above Gazette. They will come into force on 1-August-2015. See ... harges.pdf

I read this as saying that on 1-Aug the cost of having your meter reprogrammed for tariff 12 will go up from $85 to $100 (inc GST).

Note also the daily metering service charge of 10.6c/day which will presumably be on top of the gazetted service charge of 117c/day. And those of us with solar inexplicably get hit with an additional 7.4c/day on top of that, even though there's still only the one meter.

These are new charges from distributor to retailer (e.g. Energex to Origin) which will almost certainly be passed on to consumers.

PIP-4048MS inverter

Posted: Fri, 17 Jul 2015, 01:22
by Scott
Does anyone know the pinout of the RJ45 communication port by any chance? Is there any power available on there?

I see there's a remote display panel available that doesn't need external power as long as the inverter has dry contacts. My hunch is that there's 12V on there but want to see if there's any concrete info before I go poking around.

PIP-4048MS inverter

Posted: Fri, 17 Jul 2015, 13:47
by coulomb
Scott wrote: Does anyone know the pinout of the RJ45 communication port by any chance? Is there any power available on there?
Yes and yes. It seems that while there is a perfectly good standard for RS232 on RJ-45 EIA/TIA - 561, the PIP does NOT use it. So this is a non-standard cable.

There is power, but it is not used by the provided cable. The provided cable uses only 3 wires, ground, TX and RX.

I could not find specifications on this; this was found by examining and buzzing with a multimeter:

[ Edit 2: This seems to be a 2013 model. The 2014 model seems to be different, with power on pin 4; see next post and Weber's confirmation. ]

| Function. | RJ-45 pin # | D9 pin #       | Function |
| (PIP)     |             |(provided cable)|(Computer)|
| SGND      | 8 --------- | 5              | SGND     |
| TD        | 1 --------> | 2              | RD       |
| RD        | 2 <-------- | 3              | TD       |
| +12V      | 7 (or 4)    | NC             |          |

[ Edit: Weber pointed out in email that the TX and RX with respect to the DTE are confusing; he offered the above as making more sense. ]

The PIP is transmitting on D9 pin 2. At idle, you can measure ~ -11 V on D9 pin 2 with respect to D9 pin 5. The provided cable has a female D9 connector.

[ Edit: the RS232 port is isolated, so the "Ground" mentioned above is really a signal common. It is also the common for the 12 V power. I measured 11.9 V pin 7 to pin 8. ]

Note that the above could change at any time. Some newer PIPs seem to have a USB port as well as the RS232 on RJ-45 connector. Note that other (lower power) PIP models have *only* a USB port, no RS232 port at all.

PIP-4048MS inverter

Posted: Sat, 18 Jul 2015, 00:04
by Scott
coulomb wrote:
| Function_ | RJ-45 pin # | D9 pin #       |
|           |             |(provided cable)|
| Ground    | 8           | 5              |
| RD (DTE)_ | 1           | 2              |
| TD (DTE)_ | 2           | 3              |
| +12V      | 7           | NC             |

Thanks very much Coulomb.

I had another look, it seems my inverter has a different comms board to one I've seen on this thread. It's similar, but mine is double sided with some jumpers for setting the relay/power switch switch behaviour.

Worst part is the +12V line is on pin 4 on my RJ45. Ugh. I guess I'll just pull power through diodes on both options.

The USB versions still worry me. I was told that the RJ45 on the USB equipped inverter is labelled as the remote display port. They probably just put a USB-Serial chip in there, but if anyone comes across a photo of the board it would be helpful.


PIP-4048MS inverter

Posted: Sat, 18 Jul 2015, 00:31
by ChrisHobson

9:45am -Time for a cup of coffee. So I put the kettle on and it is a good time to see what the panels are doing.

P = V x I (91.8V x 36A) = 3304.8 W. But I only have 3kW of panels and it is the middle of winter (OK today's temperature could be making the panels more efficient).

The power being consumed is only 2415W so where is the unaccounted for 800 odd Watts (plus we are drawing 600 odd Watts from the battery - even worse - now we are looking for 1400W).

Does anyone have a reasonable explanation?

I do not think the figures on the PIP are accurate and will do the same experiment once I have my battery monitor set up.

PIP-4048MS inverter

Posted: Sat, 18 Jul 2015, 01:43
by offgridQLD
While most of the pv power would be going to the Ac loads.The missing power is most likely going to your battery charging it.


PIP-4048MS inverter

Posted: Sat, 18 Jul 2015, 17:21
by PlanB
PV input current is measured post the MPPT Chris, so at the battery voltage, panel current at the MPPT 91.8v input is actually much less since the MPPT part of the pips steps the solar voltage down & the current up. So your actual solar power is 52.1v x 36A = 1876w.

PIP-4048MS inverter

Posted: Sun, 19 Jul 2015, 00:56
by ChrisHobson
Ah PlanB

That makes sense.

Yesterday I received a tip from a Saffa user. The SOC reading becomes accurate if you set the battery cut-off to the default (21V or 42V). The battery cut-off values are so low that they are in fact meaningless (even at the maximum (24V or 48V). So by setting program 29 to the default the software SOC is accurate and this may influence other parameters. This may account for some of the trouble that users are experiencing. One then has to use the dry contact or a battery monitor to activate a contact to control load instead of using the battery low cut-off.

PIP-4048MS inverter

Posted: Sun, 19 Jul 2015, 15:36
by coulomb
Scott wrote: I had another look, it seems my inverter has a different comms board to one I've seen on this thread.
Ah. Yes, the PIP I have here (borrowed) is a 2013 model. The serial interface board looks like this on the left, with your apparently 2014 board to the right:

Image     Image

The 2013 model doesn't seem to have the dry contact at all. It does appear that they changed the position of +12V and even ground. So beware if anyone else tries to plug anything into that RJ-45 other than the supplied cable.

The power switch on this older model is totally separate, and appears at the bottom right of the front panel, not underneath (as normally installed). [ Edit: Duh! This was a mod to the unit I've borrowed. The switch is underneath on all models, as standard. ]

I hope you're right about the USB connector being or a remote display. That would indicate that the RS232 port won't disappear in future versions. Presumably there would be a 2015 model out now, with the hardware for the leakage testing of the PV array, as required by the latest standard.

PIP-4048MS inverter

Posted: Sun, 19 Jul 2015, 16:19
by coulomb

after looking at your photos again, I wondered why there was a relay on my board, and no DC/DC. So I looked underneath, and realised that what I assumed was a small relay is actually a tiny power supply. I don't know if it's DC/DC or AC/DC.

[ Edit: it actually is a relay; see below ]

So now I have an under the board photo to compare with yours. Again, the 2013 model is left/first, and the 2014 model is right/second.

Image     Image

PIP-4048MS inverter

Posted: Sun, 19 Jul 2015, 23:07
by Scott
coulomb wrote: after looking at your photos again

Thanks for the photos, they're very helpful.

That black box is a relay (see here) and it looks to me like they're routing it out to the RJ45 on pins 3,4,5 although it's difficult to see clearly with all the silkscreen.


I haven't measured but I'm pretty sure the HFPW pins are 'high frequency power' which drive the primary side of the isolation transformer. The secondary side would the be center tapped and the two zener diodes clamp it to plus and minus 12VDC for the RS232 voltages. The two caps would then smooth the two supplies.

I managed to find a destructive test report that had photos of the '2012' board. It doesn't use the separate 2 way HFPW connector but still uses HFPW+ on the other 6 way connector. I assume that the + line wrt ground will generate enough power to drive the RS232 line, but not enough to drive an external display. It doesn't look like there's any power line on the 2012 board's RJ45 at all.

Here's the photos I have of all the boards, the dates on the transformers match up.
boards2.jpg (223.46 KiB) Viewed 1633 times
Comms board 2017.jpg
Comms board 2017.jpg (45.58 KiB) Viewed 880 times
EDIT: Added photos of 2015 board
Edited Coulomb: added photo of 2017 board

[ Edited Coulomb: Replaced the above dead image with a saved one. Also added the 2016 board below. It's from a 3 kVA machine, which means it doesn't have the USB port. Thanks to RiaanH for the images. ]


PIP-4048MS inverter

Posted: Mon, 20 Jul 2015, 00:30
by ChrisHobson


MY PIP/Axpert is a June 2015 build. I took a photo of the board this morning. Hope it helps.


PIP-4048MS inverter

Posted: Mon, 20 Jul 2015, 09:53
by Scott
That's great, thanks very much.

The inputs to the board look the same, and it still looks like it drives RS232: the RJ45 connector, optoisolators for the serial lines and two capacitors for the dual supply which wouldn't be needed if it were USB only.

The 6MHz crystal is probably for a USB-serial chip on the bottom of the board.

If anyone ever has a chance to see underneath the board then please post pictures.

PIP-4048MS inverter

Posted: Mon, 20 Jul 2015, 16:41
by coulomb
ChrisHobson wrote: MY PIP/Axpert is a June 2015 build. Hope it helps.

It sure does, thanks Chris. Interesting that there only seems to be one relay on your board. So it seems that the relay on the 2013 board is the dry contact one, even though the signal doesn't seem to be named like the others on the main connector.

PIP-4048MS inverter

Posted: Wed, 22 Jul 2015, 22:56
by dockarl
Sounds like you folks have made some awesome progress here - congrats to Weber and Coulomb for a job well done..

LOL Coulomb - the power switch you so describe on the front panel of the 2013 model doesn't come out of the factory that way - but I did modify mine that's with Weber at the moment so that the switch was more accessible...

I've been pretty hard to get in contact with over the last while, but I'm guessing this means that Weber was successful in getting my blown board up and running again? Woot! If that's the case, awesome news.

We're doing some work with Aboriginal communities up in Central Australia and I'd dearly love to use this unit as a backup for our genset, so if my hunch is right and between the two of you you've pulled off a miracle and got it kicking again - would be appreciated.



PIP-4048MS inverter

Posted: Thu, 23 Jul 2015, 21:52
by dockarl
{duplicate removed}

PIP-4048MS inverter

Posted: Thu, 23 Jul 2015, 21:54
by dockarl
{duplicate removed - anyone else noticed the forum is taking 5 mins to load a page today? These duplicates came after the forum threw errors when I hit submit..)

PIP-4048MS inverter

Posted: Thu, 23 Jul 2015, 22:04
by dockarl
Solamahn - fellow previous PNG expat here.. I worked on the LNG project based in POM and previous to that around Madang. I'm a comms eng currently out of work, or working for myself which means working for free :)

I thought I'd pipe up here - just replying to your comments quite a few pages back with regards to your mosfet failures always seemingly occurring when the batteries are at a high state of charge, and usually after a load.

I can confirm that's the same behavior I noted with my PIPS - I had 3 failures in 12 months - all up it was only in service around 6 months the remainder of the time was either waiting for parts from the manufacturer to manually replace myself, or waiting for new boards to arrive..

I also noted that these failures often occurred when we were at a high SOC and the sun came out from behind a cloud - but the commonality was that it was always when batts were at mid to late absorb and high SOC. I was using AGM batts, about 375Ahr total 3 strings of 12V 125Ahr in series parallel.

These issues initially occurred (as you described with the PIP itself), without any external SCC (just using the internal scc - see my video here, which I captured and sent to MPP) and after some weeks working back and forth with MPP they issued a firmware upgrade early 2014 which made the PIP scc a lot more stable. Unfortunately though I continued to use their PCM5048 SCC in parallel with the inverter. These exhibited similar behavior with 'overvoltages when batteries at high SOC' which kept causing the bus caps and fets to blow on the inverter - here is an example I caught on camera and sent to MPP. It was very, very frustrating, particularly given the high cost of returning boards etc back to Taiwan for changes that never seemed to fix the problem.

I was a very early adopter (they told me my unit was the first into Oz). After 12 months of patiently navigating the difficulties of the whole experience, and being sent up a lot of dry gullies, when they couldn't find a solution they stopped replying to my emails. In the last email I received, they blamed my failures on 'high inductive loads', and from memory said the failures were on the AC side, which just wasn't the case. The failures were always on the DC side and always occurred during these DC overvoltage events at high SOC. I found myself with a bunch of useless fried hardware, including a broken inverter board only 3 months old. This disappointed me a lot as I had thought Eric and I had developed a pretty good rapport through hundreds of emails. Coming from an R&D background myself, I understood that product development is an ongoing thing, so I'd also invested a lot of patience, time and effort into helping him sort out the teething issues. It was certainly a very disappointing outcome.

Ultimately after lots of unanswered emails I got pretty insistant and warned him that if he failed to address the problem and honor the warranty, my only choice would be to let others know. I got no reply so I made a series of videos detailing my experiences (see the first one here). Ultimately the whole experience left me with a very bad taste in my mouth, a lot of wasted time and a much lighter wallet.. That is probably one good example of why it may be better to deal with Giant as at least then you'd have consumer protection laws to assist in the event that something similar happened.

Glad to hear things appear to have been more responsive for you with warranty, and interested to hear that they've seemingly changed the fets now..

Luckily a friend (as above) offered to have a look at the inverter and we're still hopeful I might get mine kicking again, and I feel fairly confident that with the combination of better fets and caps, and different charge controllers, my experience may be better this time.

For those considering using the MPP gear, one thing I definitely would suggest is be wary of their charge controllers. There are numerous other examples around the interwebs of people having similar experiences to me with the MPP Charge Controllers going haywire and failing to do that most basic task - controlling charging. Prolonged overvolts to 63V and higher seem to be a 'thing' with them. Personally I think it's because their PIDS are software based and under certain conditions the nyquist rate of the underlying signal exceeds the sampling rate of the gear - and control is just completely lost. I thought this may only be an issue with AGM's but I've since had one report to my youtube channel of it happening with straight LA's as well.

I log all my gear live online at 10 second intervals and I definitely have not seen the same overvoltages since I've moved to a different manufacturers charge controller and (by necessity) inverter - everything else in my system, batteries, cabling, panels etc - is all the same.. so it tends to indicate pretty strongly to me that there is something with the MPP SCC's that doesn't play well with some battery chemistries, and that it could be worth considering using a different brand charge controller if you choose to use the PIPs.


PIP-4048MS inverter

Posted: Sun, 26 Jul 2015, 02:21
by coulomb
dockarl wrote: Sounds like you folks have made some awesome progress here -
Errrm... Image
...I'm guessing this means that Weber was successful in getting my blown board up and running again? Woot! If that's the case, awesome news.
Actually, I was just borrowing the PIP unit, because it wasn't being used much, for protocol research.

But now that you remind us, it is rather a long time since Weber's first unsuccessful attempt at fixing it, so we decided to make testing it a part of Project Day.

Weber has a PIP repair manual from somewhere, so we decided to start by following the testing procedure. Some parts of it were a bit silly, where resistance values obviously depend on the multimeter in use (and we obtained three wildly different readings from three multimeters for some measurements). But overall, it was still pretty useful, and we discovered a multitude of failed parts:


Ok, I'm being a little melodramatic here...

The row of capacitors were pre-emptively removed from Kurt's unit, but there was a similar number each removed from both of dockarl's inverter main boards. Several of these had bulging tops, which would have looked more impressive, but alas weren't kept.

The bottom row of MOSFETs were replaced with higher voltage versions, even though only one set of four was faulty from each board. You can see the right hand two let go with gusto, and a few of the others were a bit blackened. That's only one PIP board's worth of battery-side MOSFETs, by the way.

In the middle row, we have IGBTs from the two boards, on the high voltage side of the high frequency transformer. "Only" four are blown, which means we can make one good board from two bad ones. But on the right end of the middle row, there are three faulty high voltage MOSFETs, from the buck stage. There are only four of these (from two boards), so alas we could not get one board working after all. There is a last stage, another full bridge of IGBTs, and they seem to be OK, at least so far. (Perhaps this last stage was protected by the failures earlier down the chain).

It might be worth mentioning the amazing chain of power conversion in the PIP inverters. When inverting, it starts with the MOSFETs on the nominal 48 V bus; these are the 16 MOSFETs shown at the bottom of the photo (four sets of four paralleled devices). The resulting square wave feeds the high frequency transformer, which then connects to the full bridge of IGBTs (left middle row). This seems to create a high voltage bus, perhaps 400 V. There is then a buck stage, involving strangely either a pair of MOSFETs or a pair of IGBTs. These boards have two MOSFETs, and they do seem to be paralleled (source to source and drain to drain). Finally, there is the full bridge at the 230 V AC output, using four IGBTs.

Note that all of these stages (with suitable gate drive!) are power flow reversible, and this seems to be how the PIP does its mains to battery charging.

So dockarl, we gave it a good go today, but fell short. Weber will order replacement parts soon, with the BMU prototype parts. Hopefully after those parts are replaced, the PIP boards will be working again.

[ Edit: reworded to mention Weber's original attempt to fix both boards; 230 V -> 230 V AC ]

PIP-4048MS inverter

Posted: Sun, 26 Jul 2015, 13:57
by solamahn
Yes. Saw your videos. I don't think it is the number of panels or too much inductive load that has been causing the failures of mosfets on hs1. I have one site with a 2448ms running 15 x280w and 4 500 ltr deep freezers and no problem. It has 12 x 12v250ah AGM's. I just had another failure last week from a 2448 with 9 of those panels running 1 deep freezer but only has 4 batteries. I will update the software in all my sites. About 75 sites. Over time by I will end up changing all the main pcb's I suspect. I have changed more than 20 so far. Last one was a 2424msx. I also sell mppsolar ups from 2.4kw to 8kw with no failures. Hopefully the ongoing improvements to the pip range will result in reliable units also. We install from Aroma to Mekeo so far. Apart from the failures, I think pips are a terrific inverter/charger for the price. I deal with James at mpp. I have a 4048ms in service for more than 2 years without failure. I have never been back to that site. It has 16 x 280w and 16 12v200 AGM