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
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 » Thu, 18 Jun 2015, 23:00

"Kurt, 3.0 V pc cell corresponds to a near-zero state of charge when the LiFePO4 cell is only under light load. So 48 V is too low for a low voltage cutout with 16 cells."

Yes light loads is something I overlooked when selecting 48v. I might bring that up a little (Selectronic has more adjustment scope for the LVD) What do you think 50V ? Minimum load is about 2A.

I don't have any objection to using 15 cells if it helps you avoid preset limitations with the PIP. I guess I was under the impression that 48V wasn't a issue for LVD of the inverter. I will do some discharge tests tests from 3v down to 2.5V and see what the reserve capacity is (more for my own knowledge).

You would loose 1/16 of the capacity, perhaps 1.3kwh of storage on a 400ah bank but that's no big deal.

No abjection's just didn't think it was necessary.

Kurt

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

PIP-4048MS inverter

Post by solamahn » Thu, 18 Jun 2015, 23:30

[Moderator note: Preceding exchanges with solamahn at this time can be found here.]

That's what I like to hear. I have no experience with lifepo4 but from what I have read, it is best to keep them in the flat part of their charge discharge volts v soc graph. Also read that unlike pb, where 100% soc is good for them, 50% for lifepo4 so the best setup would be to keep them as close to 50% so you might charge to 80% and discharge to 20%. In a real life situation, this would be difficult because of varying daily power usage and insolation but the point is that you dont have to strive for 100% soc. In fact 100% soc is detrimental to the life of lifepo4. Just charge as much as you think you are going to need and not discharge too much and the lifepo4 should last even longer. Trial and error. Different voltage settings for different times of the year. I am looking forward to experimenting with lifepo4.
Sounds like you do not need to make the lvd setting load dependent because the lifepo4 voltage does not vary much with load. I tested a 4048 last night with 4 x 12v250Ah AGM starting at 100% soc with 2,200w load and voltage at the battery dropped from no load 52v to 49.3v so pb lvd definitely needs to be load dependent. Actually it is the inverter lcd screen voltage reading that is important here but I think it was about the same 49.3. Will be interesting to try this test with lifepo4.
Solamahn PNG

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

PIP-4048MS inverter

Post by solamahn » Thu, 18 Jun 2015, 23:58

I have had no problems with the 4048ms scc even with 15 x 280w connected 5 x 3 and all facing the same direction. I did not do any proper testing of 2s v 3s but 3s feels like its working better plus less cable and dc breakers.
Solamahn PNG

User avatar
weber
Site Admin
Posts: 2621
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Fri, 19 Jun 2015, 00:21

solamahn wrote: I have had no problems with the 4048ms scc even with 15 x 280w connected 5 x 3 and all facing the same direction. I did not do any proper testing of 2s v 3s but 3s feels like its working better plus less cable and dc breakers.

That's good to know. Are they 54-cell panels (6 x 9) or 60-cell panels (6 x 10)? Either way, 3S is best (162 or 180 cells total). 2S of these would be too low (108 or 120 cells). They would fall below battery voltage when the panels are hot.

But I'm using 3S of the old-style 72-cell panels (6 x 12) which is definitely pushing things. That's equivalent to 4S of 54-cell panels (216 cells total). But 2S of 72-cell panels (144 cells total) is getting close to battery voltage when hot too. Although not so bad with 15 (battery) cells instead of 16.

I agree with all you say about Li-ions preferring 50%. 20% (or even 30%) is detectable by voltage (compensated for current times internal resistance (temperature-dependent)). But it's impossible to distinguish 80% from 90% by that means, and coulomb counting can get out of sync, so we normally end up going to 95% (3.36 V rested).

But it seems that temperature is way more important than time spent near 100%, in terms of aging the cells.
Last edited by weber on Thu, 18 Jun 2015, 14:23, edited 1 time in total.
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 » Fri, 19 Jun 2015, 01:37

72 cells. 36v panels. about 40 Voc. I used to use 2s because I thought that the Vmp should be just a bit above the battery voltage. I find 2s runs at about 60v so you would think that was right but 3s at 90v seems to work better. MPPSolar advised me to use 3s. I still use 2s or 1s on hs models. 1s for 812hs with 36 cell panels. 2s 36 cell for 1624hs and 2424hs, 2s 72 cell for 2424msx. 3s 72 cell for 2448ms, 3248ms and 4048ms. I dont have any problem with cold mornings increasing Voc but no doubt the hot temperature here is shortening the life of the agm's and the lifepo4 when I get them. I like talking about solar power so this forum is good for me.
Maybe 3s 60 cell would be better for you. 3s 72 cell for me because it is hotter here. I just changed a 2s to a 3s today. We do that whenever doing any work on an existing site. Changed 6 x 2 72 cell to 4 x 3.
Solamahn PNG

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

PIP-4048MS inverter

Post by solamahn » Fri, 19 Jun 2015, 02:10

I am just wondering how high solar panel Voc on a cold morning could cause a problem with a scc. The scc would load the panels and prevent them from exceeding the scc Vmax
Solamahn PNG

User avatar
weber
Site Admin
Posts: 2621
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Fri, 19 Jun 2015, 04:09

solamahn wrote: I am just wondering how high solar panel Voc on a cold morning could cause a problem with a scc. The scc would load the panels and prevent them from exceeding the scc Vmax

Apparently it doesn't load them immediately. When I asked Eric to clarify, he apparently checked with an engineer and wrote:

"If there's a chance on site where temp could fall so low that the PV could trigger 145Voc or more, fault will result, but that only acts a as warning as the circuit's still connected so there's a chance of damaging bus."

I agree you wouldn't have a problem with that in PNG. I see it's about 22°C to 32°C all year round. It's good to know it's working fine for you with the 3S of 72-cell configuration that I chose too.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
weber
Site Admin
Posts: 2621
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Fri, 19 Jun 2015, 04:34

offgridQLD wrote:Yes light loads is something I overlooked when selecting 48v. I might bring that up a little (Selectronic has more adjustment scope for the LVD) What do you think 50V ? Minimum load is about 2A.
Yes. 50 V sounds about right. If you're going for a 30% SoC cutout then maybe even 51 V, provided it doesn't nuisance trip due to sag under load when you boil the kettle first thing in the morning when the cells are cold after a cold winter's night.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
weber
Site Admin
Posts: 2621
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Fri, 19 Jun 2015, 04:43

But don't you have an autostart genset, Kurt? If so, then for 16 x LiFePO4, that 50 or 51 V is when you'd want the genset to come on, and maybe charge back up to about 54.7 V (about 70% SoC under heavy charge).
Last edited by weber on Fri, 19 Jun 2015, 08:34, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

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 » Fri, 19 Jun 2015, 14:44

Yes I do have a auto start Generator (though it's not fueled or connected) Fuel gums up and cranking batterys go flat as we just don't use it. I will take note of the voltage tomorrow morning like you said when it's cold. The loads around breakfast time can get crazy in our house.

With espresso machine, grill and microwave on at the same time. Surprising how common it is to be making coffee while my wife is grilling a toast/cheese melt and my daughter puts some quick oats in the microwave within the same 5 min span. 2400w + 3200w + 1000w + base loads that could be another 1000w due to sub irrigation pump getting rid of the mornings shower water under the lawn. 7600w perhaps more if some one forgot to run the dishwasher the night befor and hits the button Image

Edit: I guess voltage sag depends a lot on how big ones battery bank is in relation to the load. A 200A load (10,000w) is only a 0.5C load on my 400AH bank.

Thinking about it I think the Selectronic has load dependent LVD that can be based on voltage and SOC readings.Will have to dig through the menu this weekend and se if I can tweak it a little.

I guess I haven't had any issues as I haven't ever seen below 50% SOC (that's after playing with EV charging at night) So voltage is always in the 53 to high 52v overnight.

Kurt
Last edited by offgridQLD on Fri, 19 Jun 2015, 05:15, 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 » Fri, 19 Jun 2015, 15:30

Solamahn,
        One thing I wanted to ask is . Have you had any issues with voltage regulation. Have any of your PIP units built in SCC strayed beyond the voltage set points when absorb or float or have they behaved them self?

I think this was just a issue with some of there external MPPT SCC but with 70 PIP units installed if this issue was a issue carried forward into other products like the PIP's SCC you would most likely have seen it.

Have you had any issue's when paralleling two PIPs. I have noticed a few posts online where people talked about there two pips being damaged due to there electrician wiring something incorrectly when paralleling the two pips.

Is there something special to take note of when wiring them for parallel mode? Something that might catch people out?


Kurt






Last edited by offgridQLD on Fri, 19 Jun 2015, 05:32, edited 1 time in total.

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

PIP-4048MS inverter

Post by solamahn » Fri, 19 Jun 2015, 16:04

I had a problem with an early modem 4048ms. If batteries were full, and a high load was turned off, the battery voltage would climb to 60 and inverter would turn off and then turn on again and everything would be ok untill a high load was turned on and off. Later software fixed this problem. I could not download the software, just changed the control board.
I might be having this problem with other sites which is causing the mosfets to blow but i have not been there to witness whats going on. By the time I get there it is just 09 alarm.
We have done about 10 parallel jobs, 4048ms and 3248ms. I connect battery + and - of the inverters using 10mm cable. 2 + and 2 - x 16mm from each inverter to batteries. Have the batteries close to the inverters. I connect AC out active together on one side of one circuit breaker. If I use a separate circuit breaker for each inverter, it causes a fault if one breaker is off. I know it should not be done this way but I have to make the sites idiot proof. I am using 2 x 4048ms now. 24 x 280w and 24 x 12v200Ah AGM with mains backup. Works a treat.
I have one site with 3 x 4048ms with 45 x 280w and generator backup.
I would like to find out what is causing the mosfets on hs1 to blow. Is it the 4 caps, high DC voltage, low DC voltage or what. One the other day only lasted 4 hours. It was a 4048ms with a new replacement main PCB. Could not have been low DC. One of the 4 caps is swollen.
I am flying to Daru now to do a vsat job up the fly river so no internet till I get that going.
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 » Mon, 22 Jun 2015, 16:10

Considering it was only 4hrs of service then I would guess high voltage if the cap is bulged.

My electronics knowledge is very basic. So others might have better idea.

One thing I did notice when playing around with My PIP4048 on the weekend. I was charging a small Lifep04 bank 40ah x 16 cells. I just had 3 x 190w panels connected to it's pv input. I had the float voltage set at 53.8v and it had reached float and was doing a great job at regulating the voltage spot on 53.8v. I connected a 1000w vacuum cleaner as a AC load and switched it on. Obviously with a 1000w load only 570w of pv connected the PIP just gave it all the PV output it could (about 10amps) and the batterys made up the 500w slack. The voltage dipped a little on the Lifepo4 bank (perhaps 52.5V)

The interesting part was when I switched off the load the PIP continued to pump 10A into the battery and the voltage went up on the Lifepo4 bank 53v, 53.6v, 53.8v, 54V then continued to climb to around 54.3v then slowly the amps started to come down on the pips display (S l o w l Y) is the key word. It did settle back down to 53.8v after a while but not what I was expecting.

In comparison to my other charge controllers that seem to have instantaneous response when regulating voltage. Minimal to no voltage overshoot. When you set a float voltage it's strictly kept to that voltage. Perhaps dipping under at times as the charge controller reacts to load and brings the battery voltage back up to float voltage but never over shooting like the pip when loads are switched off.

Kurt

Last edited by offgridQLD on Mon, 22 Jun 2015, 06:21, edited 1 time in total.

antiscab
Senior Member
Posts: 2479
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

PIP-4048MS inverter

Post by antiscab » Mon, 22 Jun 2015, 18:59

offgridQLD wrote: The voltage dipped a little on the Lifepo4 bank (perhaps 52.5V)

The interesting part was when I switched off the load the PIP continued to pump 10A into the battery and the voltage went up on the Lifepo4 bank 53v, 53.6v, 53.8v, 54V then continued to climb to around 54.3v then slowly the amps started to come down on the pips display (S l o w l Y) is the key word. It did settle back down to 53.8v after a while but not what I was expecting.


Sounds a lot like your normal 3-stage charge profile normally used for lead acid batteries - CC-CV-CV

the voltage dipping to 52.5v would have put the charge sequecnce back to the CC stage.
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
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 » Mon, 22 Jun 2015, 19:17

Not sure what mode it stayed in or didn't stay in. As there is no visual indication of it on the display.I dont think It switched back to bulk as the voltage would have kept on going up to the bulk charge setting.

All I can say is float was set at 53.8v and it overshoots it. It's not to bad as it doesn't overshoot by much or for long and only under the conditions listed in my first post.

I just noticed it didn't do as good a job of adhering to the 53.8v (RULE) I set for float voltage when transitioning from covering a load to idle float. My dedicated charge controllers on my house system do a better/ faster job of reacting without the overshoot.

In reality would it matter. I guess not a few min at 1v or so higher than your float setting after a large load is taken off might not do much at all.

kurt
Last edited by offgridQLD on Mon, 22 Jun 2015, 09:19, edited 1 time in total.

User avatar
weber
Site Admin
Posts: 2621
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Mon, 22 Jun 2015, 20:45

Kurt, what was your PIP's bulk/absorb voltage setting (parameter 26)?

Yes, solamahn's PIP failure after 4 hours sounds like it is due to under-rated DC bus capacitors exposed to excessive voltage while the MOSFETs were switching maximum current the whole time -- a combination of temperature and voltage. Once a capacitor goes, the MOSFETs soon follow. I don't think having the fans trying to _suck_ air down past the caps to cool them would be much help either.
Last edited by weber on Mon, 22 Jun 2015, 10:46, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

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 » Mon, 22 Jun 2015, 21:28

"Kurt, what was your PIP's bulk/absorb voltage setting (parameter 26)?"

56 volts

Kurt.

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

PIP-4048MS inverter

Post by solamahn » Mon, 22 Jun 2015, 22:14

Back from the crocodile infested country. One thing I did find out about pips is that if the battery voltage is above 50 at the start of the day, then they go straight into float and do not do bulk/absorb stages. I tested on 2424msx and 2448ms and found this to be the case. I have not noticed this on a 4048 but maybe they are the same. Just killed a Papuan black with a bushknife.

[Moderator note: For those who may have been somewhat alarmed, that's a Papuan black snake. :)]
Solamahn PNG

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

PIP-4048MS inverter

Post by solamahn » Mon, 22 Jun 2015, 22:32

I am thinking of changing all 4 caps on new inverters. Same as I used to have to pre delivery modify the fans and Dc cable screw tightness. I have been wondering what caused those mosfet failures. I thought it could have been overloaded but my system gets overloaded every now and then (not by me) and it just turns off and on again. I thought low battery voltage but changing 29 to 47 did not stop it. Same with charging current. Changing prog 02 to a lower value does not fix it. I tried less panels and 2 inverters with less panels on each inverter. This slowed down failures but did not stop them. All inverters with 3 panels in series have not failed. I thought it must be something I did or my configuration or something the customer did. All my latest installs have 3 panels in series and no failures so far. I also change older installs if doing any maintenance or upgrades. Maybe its a combination of too many panels, 12 x 280, 2 in series and the caps. I have a 2448 with 15 panels 5 x 3 supplying 30% load during the day and no hiccups. Mie is 2 x 4048 with 2 x 6 x 2 panels, 6 x 2 into each inverter. Might have something to do with the loading. Less load means more failures. The 3 x 4048 in parallel runs at 50% load on each inverter 6 days a week. Been humming along with 45 x 280w for a year now. I have started putting 2s on 2424msx. Used to put 1s.
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 » Mon, 22 Jun 2015, 22:32

"One thing I did find out about pips is that if the battery voltage is above 50 at the start of the day, then they go straight into float and do not do bulk/absorb stages. I tested on 2424msx and 2448ms and found this to be the case. I have not noticed this on a 4048 but maybe they are the same."

I haven't played with the PIP4048's PV charge controller side of things much. I cant rule out if the pip4048 suffers the same.

I don't think I have seen the absorb/bulk stage (or at least I have not witnessed a period of time with it siting at absorb voltage (56v) I just assumed it had woken up when the sun hit the panels on Sunday and went through the bulk/absorb stage without me and was in float by the time I got to it. It may not have happened.

My pack voltage would have been above 50V (more or less always is). I have the PIP and battery with me at the moment but no PV on hand at this house to confirm. Not until the weekend. Though I would have thought any behavior as dramatic as (no bulk/absorb if battery voltage is over 50v) that Weber would have picked up on it when testing his unit?

Kurt

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

PIP-4048MS inverter

Post by solamahn » Mon, 22 Jun 2015, 22:39

This 50v thing is not a fault. They are designed that way. The idea must be that if the batteries are above 50v at the start of the day, then they dont need bulk/absorb. I only did the test with 2424msx and 2448ms on the bench using ac charge. I just connected 4 batteries and took their voltage down below 50, then charged and then tried it with a starting voltage above 50.
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 » Mon, 22 Jun 2015, 22:46

The thing is 16 x Lifep04 cells will always be over 50v (unless under a very big load) So they would never trigger bulk absorb if that's how the units behave.

If the AC/DC charger act's the same then I can test that today.
EDIT: I can confirm when charging from the built in AC/DC charger in the PIP4048 and starting with a voltage above 50v (53.3V) It went to bulk as expected. I will test the PV charger on the weekend and confirm that to.

Kurt
Last edited by offgridQLD on Mon, 22 Jun 2015, 12:58, edited 1 time in total.

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

PIP-4048MS inverter

Post by solamahn » Mon, 22 Jun 2015, 23:00

I dont think there is any difference with the charge profile if you are using solar charge or ac charge. I will test a 4048 also.
Solamahn PNG

User avatar
weber
Site Admin
Posts: 2621
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Tue, 23 Jun 2015, 04:47

According to page 12 of the manual,
http://www.mppsolar.com/manual/PIP-MS%2 ... manual.pdf
you can tell it's in float mode from the LCD by the word "charging" combined with all four bars of the battery symbol being on solid, none flashing. If you see the top bar flashing while the battery voltage is higher than the float setting, then you're in absorb.

I'm pretty sure I saw absorb stages despite the battery never falling below 50 volts. But I suppose it might have just been the overshoot you describe, Kurt. I have absorb 55.2 V, float 53.8 V with 16 x LiFePO4.
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 » Tue, 23 Jun 2015, 10:22

If the middle green led is solid, that is float. I am just thinking about these inverter failures. Could be a combination of full battery, maximum w of panels and the load. Full battery will not regulate the DC voltage as well as a not full battery. Maximum w of panels has the ability to supply a lot of current. Fluctuating load might set things off. 2s panels might also have something to do with it. Really need to log the fault codes continuously to get more information of what happened just before a failure. Lately, I have been setting bulk to 57, float to 54.8 and LVD to 47.
Solamahn PNG

Post Reply