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.
Post Reply
User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Mon, 02 Feb 2015, 19:49

Thanks Johny.

I know people are interested in the cost. So here's the original quote for this customer, verbatim. I note that while this system has only half the monolith's design battery capacity (because it is for a frugal off-grid household of only two people), it has just over the maximum non-clipping PV array power for a single PIP-4048MS inverter, because the location has massive shading problems that the customer was unwilling to address, for quite valid reasons of aesthetics and thermal comfort.
Summary: 
PV array: 3510 watts (7 times existing power)
Inverter: 4000 watts (2 times existing power), surge 8000 watts for 5 seconds
Battery: 8000 watt-hours (similar to existing capacity [when it was new], but able to be cycled twice as deep every day), Lithium Iron Phosphate cells

Itemised:
 $3500  18 x PV modules, 195 W, Suntech STP195S-24/Ad+
 $1800  18 x PV tilt frames for corrugated steel roof, 
 $1800   1 x Inverter/Genset-charger/PV-MPPT-charger, 4000 W, 48 V, MPPT-Solar PIP-4048MS
 $4800  16 x LiFePO4 cells, 160 Ah, 3.2 V, Winston Battery LYP160AHA
  $600  Battery Management System, LyteFyba 
  $300  Battery shelving 
 $1600  Cables, lugs, connectors, fuses, circuit breakers
  $400  Conduit, cable ties, fasteners, sealant
 $3500  Labour, travel, freight
-------------
$17900  Total

I will provide a 5 year warranty on the whole system. In addition, the manufacturer of the PV modules (solar panels) gives the usual 25 year warranty for 80% of original capacity, as you can read in the datasheet below.

Here are data sheets for the major components:
http://www.mppsolar.com/v3/catalogs/PIP-MS.pdf
http://www.solar360.com.au/files/Suntec ... 0Specs.pdf
http://en.winston-battery.com/index.php ... ory_id=176

I have listed Winston Battery but I may instead use cells made by CALB or Sinopoly if they can get them to me sooner. These have slightly higher capacity, but I would charge you the same price.
In fact I supplied CALB cells because Rod Dilkes (EVPower, WA), who always responded to my emails the same day, actually had them in stock in Australia, and because I suspect they might be higher quality than the others (we're happy with their "Sky Energy" brand precursors in the MX-5), and because the Winston 160 Ah cells don't actually fit the monolith shelves. Apparently they changed them from 183 mm wide to 210 mm wide in 2010. But I should have charged more for the CALBs.

Sinopoly were a dead loss. Sinopoly China took a week to respond to my email request for either a quote or contact details for an Australian agent, and then when they passed my quote request on to this supposed agent (Cain Dixon of Enerdrive), and also when I emailed him direct, I got no response whatsoever. Zero, zilch, zip, nada.

Because the PIP-4048MS is such an unknown, I factored in the possibility of having to replace it once within my 5 year warranty period. Its manufacturer only gives a 1 year warranty. This is the biggest risk of the whole quote since replacing it with a known reliable brand such as a Selectronic SP Pro plus separate MPPT charge controller, would cost around $7000.

And of course the labour charge does not include all the design, research and development work, as I wished to retain all the IP rights and open-source them, since I figure this is my best way of contributing, in the war that those psychopathic artificial intelligences, the fossil fuel corporations, and their puppets in government, have declared against us all, and against the planet.

And perhaps more importantly, when you know I'm contributing for free, Coulomb, and all you other wonderful people are happy to contribute for free. That sort of willing co-operation brings tears to my eyes.
Last edited by weber on Sun, 28 Jun 2015, 09:25, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Tue, 03 Feb 2015, 00:17

You may have noticed, I wrote inside the monolith, "Warning short circuit current: 9000 A". This is [Edit: possibly] wrong. It was the result of a sloppy mental calculation on site, at the end of a long day. It's "only" about 3500 A, or 20C, by extrapolation from measurements of droop under load, which give an internal resistance for each 180 Ah cell of about 1 milliohm.

It's a shame the battery manufacturers don't give this figure. It's important for sizing the unprotected part of the battery cable and the links. That is, the part of battery circuit up to the fuses. The standards require this to withstand a 1 second short circuit without compromising the insulation. With common 90° PVC insulation this requires about 1 mm^2 of copper for every 100 A of short circuit current.

The monolith uses 35 mm^2 battery cable with a 100 A fuse in both legs (as no leg is earthed).
Last edited by weber on Sat, 07 Feb 2015, 10:01, 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 » Tue, 03 Feb 2015, 00:37

"The monolith uses 35 mm^2 battery cable with a 100 A fuse in both legs (as no leg is earthed)."

I used 95mm2 for the calb 400ah up to the 250A fuses.

Speaking of shorting spanners on the terminals Image

Great example of the awesome power potential in a single calb 180 grey cell when (essentially) shorted giving 2200A for a 20mm spanner.
https://www.youtube.com/watch?v=lyYfFeWwd9I

Kurt
Last edited by offgridQLD on Mon, 02 Feb 2015, 13:40, edited 1 time in total.

T1 Terry
Senior Member
Posts: 951
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

PIP-4048MS inverter

Post by T1 Terry » Tue, 03 Feb 2015, 00:49

I was thinking the 9,000 amps was a little under rated actually, the initial short circuit current would be higher than 20C before the load starts to drag the current down wouldn't it? It's also not an instant flash and it's all over, the single cell short circuit tests continue to smoke the double conductor 2/0 cables for 10 mins in the You Tube videos, and that is only at 3.2v nominal. After vaporising the head off an 8mm bolt on a 360Ah 12v battery by not paying attention, I have a very healthy respect for the potential available

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Tue, 03 Feb 2015, 02:46

T1 Terry wrote: I was thinking the 9,000 amps was a little under rated actually, the initial short circuit current would be higher than 20C before the load starts to drag the current down wouldn't it?

When I extrapolate from Damien Maguire's excellent experiments (thanks Kurt) I get a short-circuit current of around 6900 A or 39C for the 180 Ah grey CALB.

This is because the cell terminal voltage droops instantly from 3.33 V to 2.35 V (a droop of 0.98 V) at a current of 2050 A, so that's 0.98 V / 2050 A = 0.48 milliohms, and 3.33 V / 0.48 mR = 6900 A.

I'm guessing that by extrapolating from the droop I measured with only tens of amps, I was ignoring the pretty-much-instantaneous heating of the electrolyte that would happen with a much higher current.

So with a real short circuit, say 0.05 mR (Damien's spanners are nowhere near a short circuit when applied to a single cell) it might drop the internal resistance even further and peak at 9000 A or more. Or it might boil the electrolyte and thereby raise the internal resistance so the current doesn't get much above the 2050 A. We won't know until someone actually does it. Should be the manufacturer.

In any case, it looks like I need to double up my unfused cable, or I could add another 100 A fuse, bolted inline, in the cable at the middle of the battery -- the one that goes between rows on the left side -- so there effectively is no unfused cable.

I note that a real short circuit, on a single cell, won't get anywhere near as hot as Damien's spanners. Most of the heat will be dissipated inside the cell itself, and will certainly damage it.
It's also not an instant flash and it's all over, the single cell short circuit tests continue to smoke the double conductor 2/0 cables for 10 mins in the You Tube videos, and that is only at 3.2v nominal.
I wasn't saying the short circuit current would only last 1 second. That's just what the insulation on the unfused part of the cable has to withstand according to the standards. A short can only happen during installation or maintenance because at other times the terminals should be inaccessible, needing a tool or a key to access them.

If it goes on for 10 minutes, i.e. 1/6 of an hour, then by definition it can't be more than 6C (1080 A for a 180 Ah cell).

Sorry to preach, but people's attention will always wander. That's why we should only use insulated tools. It doesn't take long to wrap a spanner in a double layer of electrical tape.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 951
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

PIP-4048MS inverter

Post by T1 Terry » Tue, 03 Feb 2015, 03:55

The fuse in the pack linking cables is probably the best solution, even if it's 500 amp mega fuse as it's only intended to protect against a catastrophe, not as service protection device.
Blueseas make a fused terminal contact that would suit such a purpose as well https://www.12volt.com.au/redirect.html?a=/General Htmls/webcat2003/breakers.html about 1/3 way down the page.

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Tue, 03 Feb 2015, 04:10

T1 Terry wrote: The fuse in the pack linking cables is probably the best solution, even if it's 500 amp mega fuse as it's only intended to protect against a catastrophe, not as service protection device.
Blueseas make a fused terminal contact that would suit such a purpose as well https://www.12volt.com.au/redirect.html?a=/General Htmls/webcat2003/breakers.html about 1/3 way down the page.

Good point. It should have a higher rating than the others. But that URL doesn't work for me.
Last edited by weber on Mon, 02 Feb 2015, 17:11, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 951
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

PIP-4048MS inverter

Post by T1 Terry » Tue, 03 Feb 2015, 17:26

Green but want to learn

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 » Tue, 03 Feb 2015, 17:45

No that fuse would be useless

Specifications
Amperage     300A
Maximum Voltage     58V DC
Interrupt Capacity      
10000A @ 14V DC
5000A @ 32V DC
2000A @ 58V DC
Weight     0.06lb (0.03 kg)

Notice how the interrupt capacity gets smaller as the voltage is increased.

That's the kind of fuse that guy on u tube ( Who isn't open to constructive criticism) keeps recommending without knowing if the specifications meet his application.

This is the kind of fuse you want to interrupt the battery. Notice how high the voltage rating is 500V at 120ka and 660v 50ka !

At $7 you cant go wrong.
http://www.jaycar.com.au/productView.asp?ID=SF4172

Kurt
Last edited by offgridQLD on Tue, 03 Feb 2015, 07:08, edited 1 time in total.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Tue, 03 Feb 2015, 19:08

No. But these do.
https://www.outbackmarine.com.au/Fuse-B ... a-5191.667
https://www.12volt.com.au/redirect.html ... akers.html

I think what you're doing wrong is putting an equals sign (but no url) after the word "URL" inside the square brackets.

See faq.php?mode=bbcode for usage examples of the two BBCodes [ URL ] and [ URL= ].

Thanks for the suggestion, but Kurt's right. That's another reason we need to know the short-circuit current of our LiFePO4 cells -- to ensure we get fuses with sufficient breaking capacity (interrupting capacity or interrupt rating).

But that Jaycar fuse is no good as a mid-pack fuse for the monolith either, as it needs a huge $129 3-pole fuse holder.

I could bolt one of these inline
http://www.evworks.com.au/bussmann-fwh- ... -acting-se
The much higher price of the next size up ($73 vs $42) is a good reason to stick with 100 A for this even though it gives no discrimination relative to the two 100 A 22x58 mm fuses in the battery switchboard.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 951
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

PIP-4048MS inverter

Post by T1 Terry » Tue, 03 Feb 2015, 19:19

Fair enough, the holders for the HRC fuses in the link you supplied are a tad on the expensive side, maybe a mega fuse and holder would be a better deal http://www.jaycar.com.au/productView.asp?ID=SF1986 and a holder
The ANL fuses tend to smoke when running close to capacity as the fuse section glows red and starts to melt the casing. I had a cloud of smoke and associated stink from a 300 amp ANL fuse running at 340 amps or so, a switch to a 500 amp mega fuse and holder seems to have solved the problem. With inverters, it's not an over load protection you need, it's the short circuit protection if something goes pear shaped

T1 Terry

EDIT: Fixed the link, when the URL button is used and the link pasted between the bracket the forum software adds an = something something 20% nofollow inside the first url bracket, but that doesn't appear until after it's posted but appears in the edit version. Once I deleted that bit the link works http://www.jaycar.com.au/productView.asp?ID=SF1980

Edited by admin to fix links
Last edited by rhills on Thu, 06 Jul 2017, 00:36, edited 1 time in total.
Green but want to learn

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 » Tue, 03 Feb 2015, 19:27

"But that Jaycar fuse is no good as a mid-pack fuse for the monolith either, as it needs a huge $129 3-pole fuse holder."

The isolator/enclosure in the Jcar catalog is overpriced you can get brand name ones for a lot less. You can get 2 gand and single gange enclosures.

I agree the fuse you linked to is ideal but a question for my own use...

Any reason if you don't need the pull down isolation feature. Why you can't just drill two 10mm holes in the bars On the NT1 at each end? I was considering this as I have a stack of NT1 fuses laying around. The bar is about 5mm thick and 25mm wide. Other than insuring the terminals are not exposed.

Kurt
Last edited by offgridQLD on Tue, 03 Feb 2015, 08:30, edited 1 time in total.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Tue, 03 Feb 2015, 20:49

T1 Terry wrote: ... maybe a mega fuse and holder would be a better dealhttp://www.jaycar.com.au/productView.asp?ID=SF1986and a holder
But Terry, the datasheet linked from that Jaycar item clearly says these are only good for 32 V and 2 kA interrupt!
The ANL fuses tend to smoke when running close to capacity as the fuse section glows red and starts to melt the casing. I had a cloud of smoke and associated stink from a 300 amp ANL fuse running at 340 amps or so, a switch to a 500 amp mega fuse and holder seems to have solved the problem. With inverters, it's not an over load protection you need, it's the short circuit protection if something goes pear shaped

Fuses are to protect cable too, and stop it causing fires. You can't guarantee the inverter wont fail in a cable-overloading state rather than a short-circuit state. So I hope you upgraded the cables to take 500 amps continuous when you upgraded the fuse -- say two parallel 95 mm^2.

As I wrote 5 months back
viewtopic.php?title=new-to-the-forum-ne ... 714#p52787
I personally think that, for domestic power systems and EVs, if you need more than a 250 amp fuse, and therefore more than 2 x 50 mm^2 (2 x 0 gauge) cables, then your voltage is too low. It's a tradeoff between the cost of the conductors (copper and silicon) versus the cost of the insulation/isolation/protection. But also the difficulty of bending the cable.

Kurt, I agree, drilling the blades of the NT1 fuses should be OK.
Last edited by weber on Tue, 03 Feb 2015, 09:50, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Wed, 04 Feb 2015, 04:05

There were a couple of things about the PIP-4048MS that I didn't realise until David Chaplin and I actually tested it on site, with a generator.

It became obvious, first to David Chaplin, and then to me, that I'd been thinking the wrong way about parameter 01, the AC output source priority. This parameter supposedly determines which of three sources, solar, battery or "utility" (i.e. AC input, either grid or generator) are used, when more than one is available.

One of the three options displayed is "SbU" which stands for Solar then Battery then Utility. The mixture of upper and lower case is because it's only a seven-segment display. So you might think there would be six options, SbU, bUS, USb, UbS, bSU, SUb.

But you can't feed solar energy to the loads if the battery is not available (is low voltage), so "S" and "b" aren't really independent. In effect, whenever you have both S and b, S will feed the loads before b, so you really have a fixed priority pair (Sb). So then you'd think there would only be two options, (Sb)U and U(Sb).

Well there are those two, but they are called SbU and UtI (utility first), and for some dumb reason I had it in my head that when the utility isn't normally present (off-grid) you'd want to give solar/battery the priority and so set it to SbU, but as David Chaplin soon made me realise, it's the other way 'round. When you're off-grid you want to give the "utility" priority, because if someone goes to the trouble of starting the generator, the system had damn well better make full use of it. So UtI mode is what you want when you're off grid.

Whereas, when you're on the grid, and "utility" is always present, you want to give solar and battery the priority, so you only use the grid when the battery gets low. So SbU mode is what you want when you're on grid.

Probably seems obvious, right? My confusion must have been because I was short on sleep. Yeah, that's it. Image

But what still isn't obvious is what the hell the third option is about. It's called "SOL" (solar first). As far as I can tell it behaves exactly the same as SbU mode except in the case where you have no solar input but you have both utility and battery. In that case it will use utility where SbU would use battery.

So SOL is SUb priority, right? Well, not really. Because that would imply that if the battery was low and the choice was between solar and utility, it would use solar. But it doesn't. It uses utility. So "SOL" mode can't be described as a priority ordering of the three sources at all!

One can also say that SOL is the same as UtI except in the case where all three sources are available. SOL chooses solar instead of utility in that case, as you might expect from its name.
Avail.  | Source, in mode:
U  S  b | UtI SOL SbU
--------|-------------
0  0  1 |  b   b   b
0  1  0 |  S   S   S
1  0  0 |  U   U   U
0  1  1 |  S   S   S
1  1  0 |  U   U   U
1  0  1 |  U  [U   b]
1  1  1 | [U   S]  S
You really needed to know that, right? Well, it was driving me nuts that I didn't understand what SOL mode was. And you don't really understand something until you can explain it to someone else.

But I still don't know why you'd want to use SOL mode.

Now what about the other thing? That's a lot easier to describe. We set it to UtI mode and started the generator and, ... nothing happened. The PIP displayed the voltage and frequency of the generator, which were somewhat variable -- not surprising since it was completely unloaded -- but easily within the specified range for the PIP. But the PIP acted as though there was no generator there. It refused to feed it to the loads or charge the battery from it, but just kept running the inverter off the battery. I could only exchange stunned and mystified looks with David.

It was dark by this time and I couldn't think with the noise of the generator. So I walked around to the side of the house opposite the generator. Then I came back and stepped thru the PIP's parameters. Only one seemed like it could possibly have any bearing on the matter.

Parameter 03 is described as "AC input voltage range". The two options are APL (appliances, the default, 90-280 Vac) and UPS (uninterruptible power supply, 170-280 Vac).

I had set it to UPS, thinking I didn't want the PIP dragging the voltage of the poor generator down below 170 V if an underpowered generator was connected. But now I set it to APL, and it immediately switched the generator thru to the loads, and then it started ramping up the charge current. The generator stopped "hunting" and stabilised. Beautiful!

So apparently the UPS setting is not just about the allowed AC input voltage range, but also makes the PIP far more fussy about the frequency and voltage stability of the AC input.
Last edited by weber on Tue, 03 Feb 2015, 17:31, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 951
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

PIP-4048MS inverter

Post by T1 Terry » Wed, 04 Feb 2015, 04:21

weber wrote:
T1 Terry wrote: ... maybe a mega fuse and holder would be a better dealhttp://www.jaycar.com.au/productView.asp?ID=SF1986and a holder
But Terry, the datasheet linked from that Jaycar item clearly says these are only good for 32 V and 2 kA interrupt!
The ANL fuses tend to smoke when running close to capacity as the fuse section glows red and starts to melt the casing. I had a cloud of smoke and associated stink from a 300 amp ANL fuse running at 340 amps or so, a switch to a 500 amp mega fuse and holder seems to have solved the problem. With inverters, it's not an over load protection you need, it's the short circuit protection if something goes pear shaped

Fuses are to protect cable too, and stop it causing fires. You can't guarantee the inverter wont fail in a cable-overloading state rather than a short-circuit state. So I hope you upgraded the cables to take 500 amps continuous when you upgraded the fuse -- say two parallel 95 mm^2.

As I wrote 5 months back
viewtopic.php?title=new-to-the-forum-ne ... 714#p52787
I personally think that, for domestic power systems and EVs, if you need more than a 250 amp fuse, and therefore more than 2 x 50 mm^2 (2 x 0 gauge) cables, then your voltage is too low. It's a tradeoff between the cost of the conductors (copper and silicon) versus the cost of the insulation/isolation/protection. But also the difficulty of bending the cable.

Kurt, I agree, drilling the blades of the NT1 fuses should be OK.
Sorry, forgot you were using a 48v pack, my bad. I can't see an instance where an inverter could fail to a constant 500 amp load without the internal safety fuses failing well before that, but interesting that you brought that up.

T1 Terry
Green but want to learn

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, 04 Feb 2015, 04:22

Yes all the prioritizing modes had my head spinning a little and although I didn't intend to use a generator with my unit I was thinking it worked just as you did until you proved it wasn't so.

I was able to start the pip with a battery and PV connected then run a AC load (TV) then while running I remove the battery and the unit ran the TV for 30min with no battery. Then I connected a variable speed polisher to it and slowly ramped up the load. As expect as soon as the load exceeded the two 200w panels I was testing with it shut down.

Not that it's a feature just interesting that it ran the ac load with no battery connected.



Kurt
Last edited by offgridQLD on Tue, 03 Feb 2015, 17:29, edited 1 time in total.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Wed, 04 Feb 2015, 04:28

offgridQLD wrote:I was able to start the pip with a battery and PV connected then run a AC load (TV) then while running I remove the battery and the unit ran the TV for 30min with no battery. Then I connected a variable speed polisher to it and slowly ramped up the load. As expect as soon as the load exceeded the two 200w panels I was testing with it shut down.

Cool experiment. Thanks for proving part of my theory wrong.
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 » Wed, 04 Feb 2015, 04:35

I made a small video showing part of that experiment a while ago when I first got the PIP.

https://www.youtube.com/watch?v=pW2V8AbVco0

Kurt
Last edited by offgridQLD on Tue, 03 Feb 2015, 17:35, edited 1 time in total.

DavidC
Noobie
Posts: 7
Joined: Wed, 18 Dec 2013, 15:54
Real Name: David Chap
Location: Elimbah QLD

PIP-4048MS inverter

Post by DavidC » Wed, 04 Feb 2015, 12:54

The sol setting is not to do with changing from grid to solar when the battery is full . A setting for when the inverter is used as a ups ( ensure battery is full then feed loads instead of wasting solar input to nothing ) may be a useful setting when used in a grid feed home.
Last edited by DavidC on Wed, 04 Feb 2015, 01:56, edited 1 time in total.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Wed, 04 Feb 2015, 20:28

DavidC wrote: The sol setting is not to do with changing from grid to solar when the battery is full.
Right. Because both SbU and SOL modes will do that. Compare the last and third-last lines of my table below.
A setting for when the inverter is used as a ups ( ensure battery is full then feed loads instead of wasting solar input to nothing ) may be a useful setting when used in a grid feed home.

Aha! Thanks.

I didn't really explain my table, so I've annotated each line below.
Avail.  | Source, in mode:
U  S  b | UtI SOL SbU
--------|-------------
      b |  b   b   b    These three lines are unnecessary because
   S    |  S   S   S    of course, if you only have one source,
U       |  U   U   U    that's the one you will use, in any mode.
   S  b |  S   S   S    If no utility, then solar before battery, in any mode.
U  S    |  U   U   U    If battery low, then utility before solar, in any mode.
U     b |  U   U   b    If no solar, then utility before battery, except in SbU mode. 
U  S  b |  U   S   S    If all available, then solar before utility, except in UtI mode.
So now, with what David has pointed out, I can summarise:

UtI mode is for off-grid use because it will make use of the generator whenever it is going.

SOL mode is for on-grid use as a solar UPS. It will not use the battery at all*, keeping it fully charged, until you lose both solar and grid at the same time. It will reduce but not minimise grid usage. It will maximise the life of a lead-acid battery. Not so good for Lithium.

[* Correction Jan 2019: SOL mode will use the battery whenever the solar is not sufficient to cover the loads, provided the solar is not zero. This will typically occur briefly in the early morning and late afternoon. But as soon as the sun is gone, it will top the battery up to 100% SoC using the utility, provided the charge source priority [16] is not set to only-solar (OSO). Thanks to @Coulomb for pointing this out and conducting experiments to prove it.]

SbU mode is for on-grid use where you want to minimise grid usage. It will use the battery whenever solar is not available. It will only use the grid when the battery gets low.

[Edit: On the input side of the table, changed zeros and ones to blanks and U, S or b as appropriate, to make the table more readable. Corrected some grammar.]
Last edited by weber on Mon, 21 Jan 2019, 14:24, edited 2 times in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Thu, 05 Feb 2015, 00:14

I had a question by email about the load sourcing priority matrix. I thought others might benefit.
In the case where the battery is low, but you have utility and solar, then it will use utility? Surely that's not what you want for on-grid.
I agree it seems wrong at first thought.

But the manual makes it clear that (in all modes) if the battery is low and the utility is present, then the loads will be supplied from the utility. There's no mention of varying that depending if solar is available or not.

And when you think about it a bit more: If you have solar and yet the battery has fallen below the low threshold, then clearly that solar has been inadequate -- unable to keep up with the loads for quite some time -- so it's time to use the utility to feed the loads, until the battery gets back above the high threshold.

I note that the low threshold can be set in one volt steps from 44 V to 51 V and the high threshold from 48 V to 58 V (or "FUL" which means "gone to float mode").

With a LiFePO4 battery, the only usable low threshold is the highest value, 51 V. And that's really a bit too low as it corresponds to about 3.19 V per cell, which corresponds to about 10% SoC at low loads. You're really relying on a steady heavy load coming along (like boiling a kettle) so it trips off at more like 25% SoC. You're also relying on your cells being balanced at the bottom (which means equal in capacity, if you have top-balanced), so you don't have one going below 0% SoC while the others are at (51-2.8)/15 = 3.21 V (20% SoC for low loads).

Here's the paper Johny found, with the best SoC vs voltage curves.
www-personal.umich.edu/~hpeng/DSCC2013_Weng.pdf

So to minimise grid use, you could set the low threshold to 51 V (3.19 Vpc) and the high threshold to 52 V (3.25 Vpc). If this resulted in too much switching, you could bump the high threshold up to 53 V (3.31 Vpc).

There is also another parameter (16, Charge source priority) with which you can disallow charging of the battery by the utility (by setting it to "OSO", Only SOlar). So if the battery goes low, the PIP will feed the loads from the utility while charging the battery only from solar. Of course if it's night time, the battery will stay below the low threshold until the sun comes up the next morning. When the battery gets enough solar to get above the high threshold, the PIP will switch the loads back to solar plus battery.

A very flexible system. Just not designed for Lithium cells or battery monitoring systems.
Last edited by weber on Wed, 04 Feb 2015, 13:21, edited 1 time in total.
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 » Thu, 05 Feb 2015, 00:29

Hello All,

I have just joined this forum after reading this great thread. I bought this PIP4048MS two weeks ago and have asked Maximum Solar the following.

Dear maximum_solar,

You need to improve the design of this item to be used with Lifepo4. To function properly with Lifepo4 battery bank the inverter/charger needs two digital inputs that can be connected to a BMS relay output.

1. One input for LVD (low voltage disconnect to make inverter go into LINE MODE).

2. the other digital input for HVD (high voltage disconnect to turn off charging).

At the moment the user has to use a 200 amp DC contactor to isolate the DC battery input to inverter/charger which turns everything off leaving the user without a 240Vac output.

I hope you improve design. Thanks.


Thanks for the effort you guys are putting in. I am not a fan of using serial comms to do this. Image
Last edited by lopezjm2001 on Wed, 04 Feb 2015, 13:37, 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 » Thu, 05 Feb 2015, 01:30

I just got a reply from Maximum Solar:


Hi John
thanks for your feedback. that's right as you already know the challenge is that inverters run on DC so disconnecting battery will inevitably trigger a fault and possibly lead to shut down of inverter. I don't know how difficult it is to implement the features you proposed but I'll forward your suggestion to the engineering team of the manufacturer so they can evaluate it.
thanks again and take care
Shao


Last edited by lopezjm2001 on Wed, 04 Feb 2015, 14:31, edited 1 time in total.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber » Thu, 05 Feb 2015, 01:32

Great to hear from you, lopezjm2001. Thanks for the kind words. Good on you for encouraging them to better support LiFePO4 and BMS. I note that Maximum Solar are only the authorised reseller. The manufacturer is MPP Solar Inc, Taiwan. [Edit: Wrong. It's Voltronic Power, Taiwan]
lopezjm2001 wrote:I am not a fan of using serial comms to do this.

Can you explain why?

Because if you ask the question, "What is the easiest (and therefore most likely to happen soon) way for MPP Solar to implement this?", it would have to be by changes or additions to the serial commands, since that's purely a software change.

But it's also worth asking, "What would be the best implementation, to be able to make optimum use of all the many different BMS out there?"

I can't help still thinking that the answer is serial commands. If you need two digital lines (some BMS only have one digital line, and some have full max and min voltage and temperature information) then you only need someone to program up a $10 TI-Launchpad with Energia (or an Arduino if you want to slum it with an 8-bit processor on a board that costs more Image) to convert between whatever you've got and the serial comms.
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 » Thu, 05 Feb 2015, 01:52

Hi Weber,

Simply because my BMS (brand is miniBMS) does not have serial comms. The other reason is that at the moment the unit is losing and restoring communications between my PC and PIP4048MS, I would not want a HVC event occur whilst it keeps charging/cooking my Lifepo4 cells because it lost communications. Image.
Last edited by lopezjm2001 on Thu, 05 Feb 2015, 01:29, edited 1 time in total.

Post Reply