Please help with sanity check on system design?

[Allen15]

Hello.

I'm trying to design a system that can be expanded if/when needed, and my requirements include split-phase 120/240 VAC.  Starting with a nominal 4 kW system, so far, I've narrowed down to the Apollo TSW4048 or the Magnum Energy MS4448PAE, because both of these units can be stacked for more power.

I'd also like to use Midnight Classic 150's as my charge controller of choice, & the string calculator on the website says I can safely run 6 strings of 3 panels and hit my targets safely without going over the ratings of a single Classic 150 until I exceed those first 18 panels.  Somewhere around that time, I expect I would need to also add another inverter too, and I am wondering what if anything I need to do to plan an installation so that I can drop in both another inverter & another charge controller down the road?

Each company has their own E-panel equivalent for combining/stacking their products, but Apollo's ISM doesn't seem to be a good fit for a Classic 150, and the same is true of the Magnum Energy product line...  If I use a Midnight Solar E-panel, how would I expand that later, and do I need to be concerned about the layout of the first one before it becomes a nightmare retrofit?  I like the Midnight Solar E-panels, as they seem to be a good fit for the Midnight Solar surge protectors.

I forgot to add this is intended for use on a 48 Vdc battery bank system, too, and NOT grid-tied.

Any thoughts, or warnings?  Thanks,

[Allen15]

If not that, can anyone point me to any wiring diagrams on stacking Classic 150's?

The manual describes how to network them with the communications cable, but doesn't show an example of a complete wiring diagram with a pair of them actually in a system, with power connections, & recommended/required breakers. 

If I can get more info on that, I can probably piece together the rest, with the docs already available about stacking the aforementioned inverters, and at least to my mind, the PV charging system is or probably should be independent of & separate from the load systems...

Thanks

[Vic]

Hi Allen..

These posts have been sitting here for a bit,  so,  will throw out my few cents worth.

Seems to me that you would choose the Inverter,  and then look at the choices in e-panels,  with the idea that there needs to be room for the expansions that you have mentioned.

The Charge Controllers do not necessarily need to be mounted onto a panel,  but that would make running wiring easier if you can mount them on the panel.  You will need breakers for the PV into each CC,  and battery breakers on the output of each CC,  as well.

Once you have found a panel that will accommodate basically doubling your system components,  or course,  the first Inverter,  and CC need to be located such that a second of each will fit.  And in trying to reserve space for possible future expansion,  there is some possibility that the Inverter manufacturer might change the mechanicals of the unit such that it might not fit when you get around to expanding.  Inverter manufacturers do have an interest in trying to keep things compatible,  as they,  too,  do not want to make a bunch of changes to panels/boxes to accommodate a newer inverter design (although it does happen on occasion).

You have not mentioned BATTERIES.  They are the items that are difficult to scale up later.   And if you have enough battery initially for your possible later demands,  you might lack the PV to keep them well charged and happy.   Adding new batteries to older ones usually does not work so well after the initial bank is 6-12 months old -- yes,  a generalization.

The only real "stacking" of CCs is in the Communication interface twix CCs.  The second CC needs its own,  separate PV array,  and it would normally feed the common battery bank -- shared with the first CC.

I do not claim to be an expert on e-panels,  but the oft seen review of MN e-panels is that they have more space to help easy wiring,  and MidNite tries very hard to offer the widest range of solution to the greatest number of customers.  Opinions,  I am  a real fan of MN products,  and their terific support,  but have no financial interest in them ...  just another happy customer.  Good Luck,  Vic

[Allen15]

Thanks for the reply.

I guess that my biggest question is in the area of connecting more than 1 CC, particularly whether or not to parallel the inputs, or if each CC should have its own input, and if the inputs are paralleled, do I need some sort of diode arrangement on the inputs.

The reason for this question is I'm trying to determine if I'd need more than one "home run" from the PV array combiner box(es), to be able to connect more than one CC.

I have to assume that the outputs of each CC should be connected to a common +POS bus, so if that assumption is wrong, please inform also

I'm not terribly worried about the batteries, because long-term, I'm hoping to have moved on to some version of NiFe batteries in the future, and while they're more expensive, I am led to believe that one can expand such a battery bank at any time, without the issues of doing so on a lead-acid battery system.  They also will likely last longer than I'll be alive with proper care

I haven't found an E-panel that is a ready fit for 2 inverters of any kind from any source, but both the Apollo & the Magnum Energy ones make modular equivalents that appear to be stack-able with each other (they just don't seem to be a good fit for adding MS Classic 150 CCs to).  I may end up resorting to making my own or mounting the CCs to the wall, and just running conduit to their modular solutions.

I had narrowed my choice of inverters to 2, but neither of the E-panel choices from Midnight Solar seem to be what I had in mind (which could be my ignorance).  I've seen a few examples where they've shown a pair of Magnum Energy inverters with a pair of MN Classics, but these were all shown with 2 separate E-panel boxes, 1/inverter.  I could even live with that, but I'd like to see a wiring diagram of that setup too, if possible, to see how they wired up the Classics.  So far, the Midnight E-panels seem to be the most compatible with the MNSPD and the MN Classics, which I wish to use, just not designed to add 2 of them (Classics) to a single box.

How did you do yours?

Thanks,

Allen

[Vic]

Hi Allen,

Just to be clear;  Each CC really must have its own PV array.  Even "identical" arrays,  on the same roof,  with identical tilt & view of the sky,   will have somewhat different string Vmp,  and therefore somewhat different MPPT voltage.

It is really better to look at mulitple CCs as redundant systems.  If there should be a problem with one array,  for example,  there can still be another  PV array to do some charging,  and so on.

Believe that you will be happier with separate PV arrays for each CC.

If the panel/DC Conduit box that you are using does not have one,  you can add a Battery Busbar,  to land the CC output breaker conductors onto.  MidNite has several options for this,  and believe that there is a place on most of the larger e-panels for such busses,  and if not,  you could drill a couple of mounting holes for one.

On the system here that has two Classics,  the CCs are mounted off the main mounting plate,  and there are EMT conduit runs twix the Combiner (also mounted sepaerately),  the CCs,  and the DC conduit box,  which has the four circuit breakers for the two CCs.

With two CCs,  if you do run your own conduit,  vs mounting them on a panel,  you need to plan for large enough conduit to meet the maximum "fill" requirements -- maximum number of conductors -- and plan for derating those conductors,  if there are more than 3 current carrying conductors in one conduit,  and so on.  Usually,  using panels on conduit boxes adjoining the CCs,  these issues do not apply.  YMMV.  Vic

[Vic]

And,  Allen,

I know NOTHING about Apollo TSW inverters,  or any of their products for that matter,  but,  you might want to look at this Thread on another site:

http://www.wind-sun.com/ForumVB/showthread.php?12555-Opinion-of-Apollo-Solar-TSW-Inverter

I will say  that Apollo almost never comes up in the discussion of what Inverter to buy.

Just FWIW,   Vic

[Allen15]

Thanks for all the info.  It isn't really THAT much of a problem to run separate home runs for each PV array, but it would be good for me to know in advance, so I know what size conduit to lay in the trenches.  I've got an old copy of the NEC handbook around, and I'm pretty familiar with the differences are between it and local codes, but the fill rate tables haven't changed much, so I should be ok on that part if I can plan for it ahead of time, & the difference in conduit costs aren't really worth not going larger to start with.

The greenhouse service entrance feed is roughly 200 feet from the house main breaker panel & service entrance, but since the greenhouse will be heated in the winter too, & the room for the 31 PV panels is out at the greenhouse (big greenhouse), it might make more sense to build a heated addition to the greenhouse for batteries & E-panels, etc, and just feed 120/240 VAC back to the house to a transfer switch there (that way, when I move in 5-7 years, I can return the house to full utility power, & take my PV system with me).

Just got my utility bill, and I'm paying just over 13.85 cents/Kwh, and by my calculations, I can provide my own for about 12 cents or less, if I do the work to install the PV system, including the costs of buying expensive batteries every 5 years...

I'm also certain that while I WILL need more than one charge controller, I might be able to get away with only one inverter, so I'd like to be able to add on or upgrade those parts independently.  Right now, I'm leaning toward moving to a Magnum Energy 4448-PAE, starting with a single Midnight Solar Classic 150, making my own panels from scratch (raw 4.8 watt solar cells are $1.92/ea), & buying multiple 250 ah 12 volt batteries, 'cause I could get started pretty quickly for about $14K, and that would have an ROI of 50 months, based on my current electric utility cost of $280/month (not counting labor) IF THE PRICE OF ELECTRICITY DIDN'T GO UP...

Since such a system could reasonably be expected to last at least 5+ years without substantial additional costs, it would be about a 20% cut in my electric costs, if I had to completely replace it after 5 years, and odds are I could get 15-25 years out of the solar panels if they were carefully made, & I can probably also get a few more years out of the batteries too, which would all be like free power, or at least pay me back $280/month for each month after the first 4 years, which should cover labor at $3360/year for 2+ years...

The Apollo inverter looks good on paper, & I've read through the online manuals for both, but it's close enough in price & capacity to the Magnum Energy PAE 4448 to warrant serious comparison.  I do find it a bit alarming that Apollo doesn't appear to be as well supported, even though it looks to be more capable of expansion (it can stack to at least 7 units, vs only 4 for the ME).  I don't think my need for power will expand enough to have to go there (unless I do build the bowling alley & Olympic pool;) ), but by then, I suspect it won't matter, because cost won't be an issue by the time I can afford to do something like that

[Vic]

Hi Allen,

I do not want to be too negative.

However,  there is no way that you want to build your own PV modules.  Good PV do sell currently for about $1.00/  watt.  The asolute best PV are about $1.30/watt from a reputable reseller (altho,  freight is extra).  PV module mfgs  buy or make the good cells,  and the rest go on fee-bay or some other online outlet.  You could not make a competitive  PV module even if some PV mfg gave you their best cells (and they will not).  This is nothing against you.  This requires specialized tools,  materials,  knowledge and time.  It is a cut-thoat biz these days,  with PVs being dumped on the market,  quite possibly below mfg cost.

I would also be surprised if you could produce power that is even close to the low cost of Grid power.  Most of those who have done the math say that the very lowest practical cost for off-grid power is $0.60/Kwh,  others say it is close to $1/Kwh.  Batteries are the main consumable item,  and they are expensive.

I believe that you might want to look around on the Wind-Sun site,  as this topic has come up many times,  there,  and you will find detailed analyses,  or post the qestion:

http://www.wind-sun.com/ForumVB/forum.php

Know that I am sounding very preachy,  and I apologize for that.   Just opinions,   Vic

[Allen15]

I certainly don't think your advice and input to be negative at all, and I appreciate it very much.

Thank you very much.

In that same vein, if you know where I can buy panels at those prices, please inform, as I'm apparently not shopping in the correct places.  I do have a local source (pick up, no shipping, and check before walking off with product) where I can buy grade A new polycrystalline 6" x 6" cells at $1.92/ea, but haven't yet spotted anywhere I could buy top quality commercial panels at $1.30/watt.

At $0.40/watt, it is appealing to make them, because I do have some skills and prior experience at assembly, plus I already have all the tools to do so.  I do value my time also, so I'm open to reasonably priced options, even if they cost more than the raw materials.  I also have help to do this too, with a vested interest in success so some of the labor costs are free.

I've seen some places that will sell me panels in bulk at what appeared to be good prices, but usually in quantities larger than I wished to buy at one time...

If I were going grid-tie, then I would just buy everything off-the-shelf, because it would make the building inspectors much happier, and one must make them happy to connect to the grid, but I'm not, so I have opportunity to enjoy some possible savings.  I also realize that if I don't make my panels properly, they WON'T last 20+ years, and I'd have to do it all over again sooner than hoped, but I'm willing to make the attempt if feasible.

Where would you recommend shopping for panels?

I have run the numbers, and believe I've accounted for nearly all of the costs to do so on my spreadsheet, so at least for this area, I believe I *can* compete with the utility company.  I do believe that I can make power for somewhere between 8-12 cents a kilowatt-hour, though I didn't used to think I could compete back when I naively thought I was only paying 9-10 cents a kilowatt-hour from the utility company, but this last bill went under close scrutiny, and I found that my real cost of electricity is much closer to 14 cents/kilowatt-hour, and subject to go higher as they choose to jerk up the rates whenever they please...

In any case, I believe that it is close enough to be worth striving for energy independence, and if I look back on this period 5 years from now, I suspect I will regret not making the attempt.

Our local utility company (Greater Chicago market) charges us some of the highest costs in the country, & they also have watered down their net-metering program to stack the deck against anyone who might try to use it by limiting it to 2Kw solar arrays or smaller, and overall, they don't expect one should really ever do more than decrease their current utility consumption by more than 15% with this...  Now, if I'm going to spend 15-25K$ on a solar installation, I'm not patient enough to wait 50-75 years for it to pay for itself, so I'm not willing to qualify for that program.  I'd just as soon go out and cut my bill by 100%, and make my system pay for itself in less than 10 years, while it still has some chance of running my cost down to below what I'd have paid by continuing to suck out of the grid.

The added benefit of having power even when our neighborhood grid power is dead is just icing on the cake, & we do have statistically a higher percentage of outages than most of their service area because they didn't reinvest in the infrastructure in the older neighborhoods.

I'm also open to recommendations for batteries too, but I suspect I'm going to want to start out with 1000 Ah of 48 V, but I have found a distributor that (currently) has some free shipping options for their larger batteries.

Thanks for the dialog, I really do appreciate the sanity check.

Allen

[Vic]

Hi Allen,

I've done business with the two following companies.  Northern Arizona Wind & Sun's page for PVs:
http://www.solar-electric.com/hiposopa.html

And another very good company is Colorado Solar AKA teh Solar Panel Store (SPS):
http://www.solarpanelstore.com/solar-power.large-solar-panels.html

I do not do a lot of shopping on the web.  The two above companies are top notch,  have been around for a decade or two.  If there is a problem,  there is someone there to make it right.  I have no financial or other interest in either of the above companies,  just have been doing business with them for nearly 10 years,

If I were in the market for a large battery bank today,  I'd look at Forklift batteries.  There are only a couple of disadvantages to them;  they are usually a single HEAVY unit,  often weighing  a ton or more -- difficult to move or place without a tractor or,  oh yea,  a forklift.  They have somewhat higher self-discharge rates,  and some say may need higher charge currents.  They may well use somewhat more water than a more conventional RE type deep cycle.  BUT,  they are tough,  can be discharged to 20 % State Of Charge (but only if they can be quickly recharged),  and there are reports of some lasting 20 or so years in RE service.

ENUF for now.  You might consider becoming a menber of the NAZ Forum,  or the SPS Forum.  Lots of general Q&A handled on those sites.  Thanks,  Vic

[Robin]

The Magnum 240 E-Panels can be stacked when you add another inverter. We have a kit that turns the AC bypass switch into a 2 inverter bypass switch. It is pretty easy to rewire. The Magnum version has some flaws. Pretty bad ones if you are in the business of design, which I am. They cost more too.
I would stay away from Apollo. Nice people, but don't go there.
Do not parallel the inputs on the Classics. boB is finishing up his networking stacking code. I saw it working tonight where all the classics go to absorb and float and EQ together. I think this code will be released very soon. You can always update in the field, so no big reason to wait.
I think boB will be devoting a bunch of time to our inverter project after this release of code. He will be continuing Classic development from his easy chair at home each night. I believe that speech is his next project. That will be fun. I'm sure you will all have fun helping boB figure out what to make it say.