PIP-4048MS and PIP-5048MS inverters

Post by **weber** » Thu, 05 Feb 2015, 20:54

offgridQLD wrote:Back to the monolith......

Boy that sure sounds like a lot of restrictions based on movie themes and delivery transportation choice....if you do a monolith V2 your welcome to borrow my ute/trailer to open up more size options

Thanks Kurt.

But I think it turned out to be a great format, for reasons other than those I began with.

Compare the Black Monolith's 225 x 900 mm footprint, and 2025 mm height, with the Bosch BPT-S 5 Hybrid's 700 x 600 mm footprint, and 1700 mm height. And that's 700 mm out from the wall, not 600 mm.

The monolith's front elevation is pretty much exactly that of a doorway. A doorway through hyperspace? A doorway into a future free of fossil fuels? OK I'm getting a bit carried away now.

Post by **weber** » Thu, 05 Feb 2015, 21:48

lopezjm2001 wrote:I found a reseller in Australia on Ebay. The same unit is painted red and brand name is "GIANT".

GIANT Inverter

Thanks for that. I was aware of that option, but others may not have been. Almost double the price (excluding shipping), but with double the warranty period, and local support.

Post by **offgridQLD** » Thu, 05 Feb 2015, 22:01

The shop is on the sunshine coast a few min drive north of caloundra (walking distance from a charge point ev charging station at sunshine coast stadium to

)

Was going to pop in one day just to see what they had and know about them.

Kurt

Post by **coulomb** » Fri, 06 Feb 2015, 03:00

A late remark on the metalwork. Don't be tempted to leave out any of the triangular bracing plates that are provided [ edit: by the second starter pack; see next post ]. Weber somehow convinced himself that the structure wouldn't need bracing in one [edit: two] of the three dimensions, perhaps due to the cross members, the PIP inverter, and so on.

But when it was time for the cut polycarbonate to be fitted, we found that the structure was embarrassingly skewed. It was one of my jobs on the last day to attach the last set of triangular bracing plates, using a spirit level to check that it wasn't being braced into a skewed orientation.

These are the plates I'm talking about:

[ Edit: note that there are 8 plates there, sufficient for bracing one plane (two at the top, two at the bottom, and another 4 behind those four, for example. ]

BTW, that slotted angle system is pretty neat. The nuts and bolts don't need a washer by their design, and you can somehow get a spanner (open ended needed at times) where you need to, even though it gets pretty busy at times. The rails do bend a little under the weight of 8 180 Ah cells, but not badly, and not as much as I expected.

[ Edit: deflection not as much as expected. ]

Post by **weber** » Fri, 06 Feb 2015, 04:21

coulomb wrote: A late remark on the metalwork. Don't be tempted to leave out any of the triangular bracing plates that are provided. Weber somehow convinced himself that the structure wouldn't need bracing in one of the three dimensions, perhaps due to the cross members, the PIP inverter, and so on.

I fully endorse the spirit of these remarks, but just want to correct some details.

Not enough corner bracing plates are provided in the so-called Starter Pack, so you have to buy two starter packs, as I mentioned.

I convinced myself that it wouldn't need bracing in two of the three dimensions. It is still not braced in one of them. No horizontal plane is braced (the 225 x 900 mm faces and shelves), except by the stiffness of the vertical angles in that plane, and the friction with the floor and the underside of the blocks of cells, which are quite sufficient.

I initially braced it only in the planes parallel to the wall (the 900 x 2025 mm faces), which is the same direction braced by the PIP inverter.

I convinced myself it didn't need bracing in the vertical planes at right angles to the wall (the 225 x 2025 mm faces) because they would be locked up solid once the monolith was screwed to a wall. Indeed, now that it's installed, that bracing is redundant. But what I completely failed to understand, until the last moment, was that they are absolutely essential while cladding and transporting the monolith.

Many thanks to Coulomb for installing the extra corner bracing so rapidly on that last mad day before installation, so that none of the other work was held up by it.

Post by **weber** » Fri, 06 Feb 2015, 17:01

Reports from the customer after the first week are that the monolith and its PV array are performing flawlessly. The state of charge meter has never been below 90% any morning.

Of course a real test won't come until they get around to buying an electric fridge/freezer (currently gas) and a washing machine (currently by hand or laundromat) and they have an overcast week. Cooking will continue to be by gas. But they are very happy not to have to start the generator to do the vacuuming or ironing, and to be able to run a pressure pump for the tank water.

When I told the customer, while we were standing back looking at the monolith, that I had planned to paint the aluminium framing black too, but ran out of time. He thought for a second and said, "I don't think it needs it".

Post by **Adverse Effects** » Fri, 06 Feb 2015, 17:12

weber wrote:that I had planned to Paint the aluminium framing black too, but ran out of time. He thought for a second and said, "I don't think it needs it".

it depends on if you want to to disappear or be a accent pice i like the shiny outline

Post by **offgridQLD** » Fri, 06 Feb 2015, 17:41

Having recently had to acquire a new set of white goods for our Off Grid home. I did a lot of research before selecting each product. It really is amazing how the consumption can vary so much between product of the same capacity and how it can all add up if your make the wrong choice.

I settled on Electrolux ETM5200S as the fridge/freezer 517lt and only 356kwh/pa (large freezer section to) Though the even more frugal Electrolux ETM 4200S 417lt at just 318kwh/PA would be a good pick for practical small home fridge/freezer.(Nextdoor off grid house has this model 2 adults)

The 2nd big consumer the washing machine. I picked the ASKO W6444 7kg at just 180kwh pa. Usually a expensive brand but this basic model they offer is affordable and is actually there most efficient offering.

After the two big consumers mentioned above the returns on efficiency became less apparent. Though It didn't stop me going nuts trying to find a 40" led tv that used less energy than a typical 19" computer monitor. Lighting was another area, particular lighting my shed.

Anyhow just thought if there in the market for a new fridge freezer and a washing machine. The above mentioned to models was the best I could find that's easily available in retail stores. I found the best price online in Melbourne and just had a local retail store match it (at a significant discount)

Kurt

Post by **weber** » Sun, 08 Feb 2015, 01:33

I agree, Adverse. Thanks for that, Kurt. I'll pass it on.

Just a few more details about the frame construction. In particular the three battery shelves.

The lowest shelf is bolted through the third slot up from the floor. This is to allow for the possibility that skirting board may have to be left in place and the case cut around it. In that case the rear legs would step in by 50 mm below the lowest shelf and would be braced with more corner plates.

Shelves are 400 mm (or rather 8 slots) apart vertically. The short horizontals have their bolts pushed to the bottom of the slots in the verticals and the long horizontals sit on top of the short horizontals (but also bolt only to the verticals).

The framing around the top is a bit like an upside down version of a battery shelf except the longs still sit on top of the shorts.

Each actual shelf has a third short horizontal bolted upside-down under its middle, to stop the long horizontals from twisting or spreading under the weight of the cells. The cells end up sitting on top of the 2.5 mm high cup-heads of the bolts holding this middle cross-tie, but the slight bow in the shelf means that the clamped block of cells does not rock on them.

When bolting up the frame you need to adjust the dimensions between the outsides of the verticals to match the inside of the cladding minus the bolt heads and minus 1 mm clearance/margin-for-error. So that's 873 mm and 204 mm horizontally.

With the plastic feet in place, the distance from the floor to the highest parts of the frame (the top of the highest long horizontals) should be set to 2014 mm if there will be no aluminium framing on the floor (framing the polycarbonate), and it should be set to 2015.5 mm if there will be aluminium framing on the floor.

Here's some detail of the 7 x 7 mm notches needed for mounting the switchboards. Those height markings are 150 and 280 mm (up from the horizontal below them).

BTW, Here's how 17 Suntech 195 W PV modules fit in the back of a Prius. Note that they are very thoroughly restrained, but I drove, and braked, very conservatively so as not to have to test the restraints.

Post by **weber** » Sun, 08 Feb 2015, 02:15

I've been thinking about how I might fit two PIP-4048MS inverters inside the standard monolith, since there's an optional paralleling comms board that allows them to work together as a single 8 kW inverter with a single battery, and a single AC in and AC out, but different PV array MPPT inputs.

They would have to go side-by-side with a space between them, and the switchboards would have to go below them, except for the 8-pole which could fit between them.

To make more height available for the switchboards, battery shelves could be lowered, and be spaced only 7 slots (350 mm) apart instead of 8 slots (400 mm). And the PIPs could be raised by one or two slots. A PIP is 300 x 140 x 540 mm.

The top and sides of the case would have to be pretty much all vent. Of the 145 x 245 mm louvred vents, we'd need 3 in the top and 6 or 7 in each side. Or fan-force the vents themselves.

The original ventilation method planned for the monolith assumed it would be placed indoors, which is why it had to look like a piece of modern sculpture in the first place. In this scenario it would need to be mounted against the inside of an outside wall and the ventilation and conduits would all pass through the wall. The shutdown button would be on the outside wall and the state of charge meter would be wireless.

The louvred vents through the wall, top and bottom, could be of the kind used for clothes driers. http://www.bunnings.com.au/exhaust-clot ... f_p0813594 and would need to be fan forced.

But before I get too carried away with such ideas, I need to provide some schematics, as promised to Plan B some weeks ago.

Post by **offgridQLD** » Sun, 08 Feb 2015, 02:32

I'm just thinking with two PIP4048's that would be 2 x idle consumption. 2.4kwh a day befor they make any AC power.

Would there be a way to have one PIP always on to keep the devices that need 24/7 power happy and then have the other one in power saver mode and it just kicks in when you need it.

Or do you just throw another PV panel on and be done with it?

Kurt

Post by **weber** » Sun, 08 Feb 2015, 18:40

offgridQLD wrote: I'm just thinking with two PIP4048's that would be 2 x idle consumption. 2.4kwh a day befor they make any AC power.

Would there be a way to have one PIP always on to keep the devices that need 24/7 power happy and then have the other one in power saver mode and it just kicks in when you need it.

Or do you just throw another PV panel on and be done with it?

It is a scary thought, and I don't know the answer to your first question, although I suspect there is no way. In south-east Queensland it requires about 300 watts of PV to supply the standby power of one PIP-4048MS.

Post by **weber** » Mon, 09 Feb 2015, 08:02

Let's start with a block diagram that lists major components and wire sizes.

Post by **PlanB** » Thu, 12 Feb 2015, 23:10

I'm confused. Does anybody know the difference between 'Total output active power' & 'AC output active power'?

Post by **weber** » Thu, 12 Feb 2015, 23:14

PlanB wrote: I'm confused. Does anybody know the difference between 'Total output active power' & 'AC output active power'?

Some context would be useful. Where is this coming from?
[Edit: Punctuation]

Post by **PlanB** » Fri, 13 Feb 2015, 02:01

Pages 11 & 12 of the manual. Value K AC output active power & value S total output active power.

Post by **weber** » Fri, 13 Feb 2015, 04:41

PlanB wrote: Pages 11 & 12 of the manual. Value K AC output active power & value S total output active power.

Is your middle name "Terse" or what? Fortunately I took a course in advanced mind-reading and so I immediately understood that you meant the Protocol Manual and not the User Manual.*

So now I can confidently answer that R, S and T are the "Total" counterparts of J, K & L respectively. In other words, I haven't a clue.

OK. I lied. I do have a clue. The clue is that this "QPGSn" command (where n is the parallel machine number, 0..5) is called "Parallel Information inquiry". So presumably these two sets of numbers will be different only when you have multiple PIP-4048MS operating in parallel, as described in the Parallel Guide.

*I lied about that too.

[Edit: Added links to the three manuals]

Post by **PlanB** » Fri, 13 Feb 2015, 14:08

That makes sense actually Dave. For 2 in parallel load sharing equally I was planning on doubling the AC output active power of the one the Pi is on to get the total but maybe they share load data & total output active power is the sum?

Post by **weber** » Fri, 13 Feb 2015, 18:38

PlanB wrote: That makes sense actually Dave. For 2 in parallel load sharing equally I was planning on doubling the AC output active power of the one the Pi is on to get the total but maybe they share load data & total output active power is the sum?

Yes. That's my guess.

While we're on the subject of interpreting that protocol manual (and I use the term loosely), I should report what Mike and I learned about another command after much agony and wasted time, while staring down a looming deadline.

The command described on page 13, first as

"PE<XXX>/PD<XXX><CRC><cr>: setting some status enable/disable"

and then 3 lines later described as

"PExxxPDxxx set flag status. PE means enable, PD means disable"

followed by a table indicating that each "x" may be replaced by various uppercase letters for various control (not status) settings,
is in fact two completely separate commands

PE<xxx><crc><cr>
and
PD<xxx><crc><cr>
where <xxx> can be replaced by only one character, which must in fact be lower case.

Their use of angle-brackets suggested they were using some dialect of EBNF to describe their command syntax. But after struggling in vain to divine what it is -- is the slash a literal character that we must include or is it equivalent to the vertical bar as used in most EBNFs, and if so, have they omitted some parentheses, etc etc -- I can tell you it seems more like a Cargo Cult EBNF where the angle-brackets are akin to coconut headphones.

They didn't even need to use EBNF, if they had just given one example of a valid form of this command ...

Post by **weber** » Sat, 14 Feb 2015, 07:30

Here's a diagram showing the BMS communications topology.

"IMU" stands for "current monitoring unit", based on "I" being the quantity symbol for current, since we already have "CMU" standing for "cell monitoring unit".

In the MX-5, the IMU also performed insulation testing, but in this solar application the reed relays were replaced with MOSFETs and their drivers, for switching contactors.

In this solar application the IMU acts as the BMS master, and the system controller. It accepts regular status bytes from the string of CMUs twice a second, and acts to protect the cells by turning off load or source contactors if required.

The IMU tells the CMUs what the current is, twice a second, so they can compensate for their cell's internal resistance in calculating undervoltage or overvoltage stress. All the communication mentioned so far is at 9600 b/s.

The IMU also sends commands to the PIP-4048MS inverter/charger at 2400 b/s. Initially it was planned to run a PI-control loop for charging, based on the stress level of the cell with the highest voltage. And indeed this was successfully implemented for charging from the AC input. However it was discovered that this was not possible with the MPPT charging from the PV array.

So the only things the IMU sends to the PIP now are commands for various non-default settings, on startup, since it seems the PIP will forget its settings and return to factory defaults if it is turned off for long enough.

It's quite tricky how the IMU can send 2400 b/s commands to the PIP and 9600 b/s commands to the CMUs because in fact it only has one serial output. The two Industrial Fibre Optic LEDs on the IMU are simply driven in parallel with 10 ohm current sharing resistors. So the CMUs see all the 2400 b/s data sent to the PIP and the PIP sees all the 9600 b/s data sent to the CMUs.

It turns out that, to a 9600 b/s receiver, 2400 b/s data can appear as one of only four different byte values, none of which need to be used by the CMUs and so CMU-1 simply deletes them.

To a 2400 b/s receiver, 9600 b/s data can look like any value, however the probability that they will accidentally compose a valid PIP command, let alone one with a valid CRC-16 followed by a carriage return, is small enough to ignore.

Post by **weber** » Sat, 14 Feb 2015, 07:43

Ths photo shows how simple it was to add a fibre-optic input to the PIP-4048MS. The Industrial Fibre Optic (IFO) phototransistor is simply soldered in parallel with the phototransistor (the inboard side) of the PIP's input opto-coupler.

The IMU had no way to read ACKs or NAKs coming back from the PIP. But in any case, we found that although an ACK might come back almost immediately, there was still some time after that before it would accept a new command without corrupting it. This delay was typically one to two seconds, depending on the command. I asked MPP Solar if there was any way to predict this, and have received no reply.

This photo shows the IFO receiver held in place with neutral cure silicone.

Post by **PlanB** » Sat, 14 Feb 2015, 17:47

The CoconutHeadphones what a delicious turn of phrase. You're right the occasional example wouldn't go astray. We've done a lot of trial & error here also,reminds me of this:

You do nice BMS drawings too what app is that you use? Some of us are stuck in a time warp with Corel, don't know what I'm gonna do when my last XP machine dies?

Post by **weber** » Sun, 15 Feb 2015, 01:13

PlanB wrote: The CoconutHeadphones what a delicious turn of phrase. You're right the occasional example wouldn't go astray. We've done a lot of trial & error here also,reminds me of https://www.youtube.com/watch?v=czro0qB72Ng

You do nice BMS drawings too what app is that you use? Some of us are stuck in a time warp with Corel, don't know what I'm gonna do when my last XP machine dies?

I try. Thanks for the kind words. They make it all worthwhile.

I use Microsoft Word for drawings. Currently Word 2007. Frequently used commands are Insert/Illustrations/Shapes and Home/Editing/Select/Select_Objects. And when a graphical object such as a text-box, rectangle, line or arc is selected, the "Format" set of commands becomes available on the ribbon. And of course you have context-menus available via right-click. The most annoying thing is how it keeps taking you back to word-processing mode, and you have to recognise the different highlighting of objects and the different cursors that tell you which editing mode you're in.

If anyone knows how to convince Word that you're never going to do any actual word-processor-style text-editing in a given document -- that you're only ever going to edit text in text-boxes -- then please let me know.

[Edit: Thanks for a blast from the past with that cartoon. Duh, open sasquatch?]

Post by **rhills** » Tue, 17 Feb 2015, 07:32

Hi PlanB,

PlanB wrote:You do nice BMS drawings too what app is that you use? Some of us are stuck in a time warp with Corel, don't know what I'm gonna do when my last XP machine dies?

I use Inkscape quite a bit. It's Open Source (ie free, as in beer), available for Windoze, Mac and Linux.

Like any powerful drawing program, it takes a bit of learning, but there's lots of good manuals, tutorials etc online for it. It also lends itself well to building libraries of drawing objects (eg electronic component symbols) that you then just drop onto a drawing and push around into place.

Probably harder to learn than Word Draw, but a whole lot more powerful also.

HTH,

Post by **weber** » Fri, 27 Feb 2015, 05:20

Some great real world experience of LiFePO4 cells in a solar power system here, with recommendations on boost and float voltages.

http://www.technomadia.com/2015/02/livi ... ry-update/

Even more reason for me to tweak that float voltage down from 3.375 Vpc (54.0 V) to 3.3625 Vpc (53.8 V) as per Kurt's friend's recent research. The boost to 3.45 Vpc (55.2 V) can stay, with bypass resistors turning on at maybe 3.44 V.

[Edit: Added bypass voltage]