Maximize my off-grid system?

[Doc]

New member to this forum, which looks like the most comprehensive I've seen.  So hopefully, one of you genius math savvy members might be willing to help me out. 

My current 10 year old off-the-grid system (self installed):
Trace 4024 inverter
Outback Flexmax 80 controller
6 - Solarex 24v panels @ 110w each
6 - BP 24v panels @ 140w each
2 - Solar One batteries @ 24v (1160Ah)

My goal here is to beef up this system to optimal capacity by purchasing additional panels and perhaps the MS Classic 200.  Currently, the panels are simply going to an OB combiner box at 24v.  What I'm starting to understand is that I can wire 3 panels in series strings to send 90v to the controller and have the controller then send 24v to the batteries.  Yes?  In this case, I would have a 1500w array as 4 - 90v strings.  My question is:  How many panels (watts) can I add to this configuration to maximize this system?  My math/brain is short circuiting!

Thanks much!

Doc

[Westbranch]

IS efficiency a concern?  If so ,and you do not have a long home run from PV to CC then you will be better off wiring the PV arrays to ~ 48 volts and down converting.  the greater the spread PV to battery the more heat created and less efficient the system will be....

More details are needed about all those panels please...

[tecnodave]

Doc,

I will agree with WB on the 2 panels per string as opposed to three. The higher the difference between PV volts at Pmax and battery volts the harder that your controller will work. As to the power I calculate that you have about 1500 watts now so you could add 800 watts to your system but you would be right the limit of your outback controller which I believe is 2300 watts. It will reduce life and cause more problems to run right at the edge, so I would recommend not more than 80% of maximum power to preserve life of your system so that's about 1850-2000 watts maximum input for a margin of safety, I run my Classic 150 at no more than 2300 watts where it is rated at 2700 watts at 24 volts.

td

[dgd]

Connecting the PVs in serial strings would give you a better headroom voltage range from the PVs to the OB controller. This would allow MPPT to work more efficiently.

If these are the SX series PVs then mpvs are 32.9v/110watt and 34v/140watt.
For a 24volt battery system I would be inclined to connect these in serial strings of 2 panels to give six strings in total. This should give you an input voltage between 45v and 65v over a range of weather conditions, enough to keep the controller mppt'ing.

The FM80 can make 80amps output and a 24v battery at 28v would mean about 2240 watts or if you want to maximise pv input in non optimal weather conditions then 2400 watts.
So you could add about another 900 watts of PV.
Another six of those mpv 34volt SX140 Pvs would probably be the best, in strings of two in series.

dgd

[Doc]

Wow guys!  Thank you so much for sharing your expertise.  My brain is relieved.  No wonder you are full Hero's!

Doc

[vtmaps]

My current 10 year old off-the-grid system (self installed):

You didn't mention a combiner box.   Do you have one?  Will it accommodate you upgrade? 

As mentioned, the decision on how to configure your panels depends somewhat on the distance from the combiner (near the panels) to the controller.  If they are far apart you may need the higher voltage strings (3 in series), but the lower voltage (2 in series) is preferable.

How many panels (watts) can I add to this configuration to maximize this system?

If you are buying more panels, make sure their Vmp matches the Vmp of your existing panels.  Most of the so-called "24 volt" panels that are produced today are NOT really 24 volt panels.  Their Vmp is about 30 volts.  True 24 volt panels have a Vmp closer to 36 volts.

Another upgrade to consider is a good lightning arrester.  Check out Midnite's SPDs.

--vtMaps

[Doc]

Thanks for your concerns, vtMaps.  I use an OB combiner box. The breakers in it are all 15's.  The run from the box to controller is only about 12 feet with #4 AWG wire.

The Vmp on the panels are all 34v.  And I do have lightening arrestors installed.

'dgd' suggested I add 6 similar 140w panels.  To conserve roof space, would 4 - 200/225w panels work just as well?

[Vic]

;Hi Doc,

The EXACT specs on the existing PVs are very important for we all to know,  or at least the exact part numbers.

Good that you have a Combiner,  but believe (without knowing the exact specs of your PVs),   that 15 A breakers are probably TOO HIGH a current rating.

dgd has surmised some of the specs for the PROBABLE PVs that you might be using.

So,   you are using real 24 V PVs -- those with about 34 or higher Vmp.  Good,  but please the SPECS or part number of all of them.

Many of the current generation PVs that are 200 W and greater have Vmps between 29 and about 31.5 V.  There are some that are in the 36 Vmp range,  but are less common.

Agree with each of those who have cautioned against running three 24-ish PVs in series on a 24 V system.  Even if three of your PVs are only about 34 V,  this is NOT 90 V total String Vmp.  It is in the range of 102 V.  And,  during Absorb,  as the battery charge current diminishes,  and especially in Float,  the Vin to the FM,  can approach 120 V.   Even at low currents delivered,  the FM will usually run HOT under these conditions -- particularly on a 24 V  battery.  If the ambient temperatures for the CC are warm,  the situation is worse,  if the ambients  are HOT,  worse yet,   IMO.

Further EDIT:  Plus,  while we cannot know what weather conditions that your system will face.   Cold weather will increase the PV Voc,  possibly to the point of damage to the FM-80 CC.   You should probably run your string configurations through a String Sizing Tool to see.   One more reason to run only strings of two real 24 V PVs.

You should NOT run strings of three of your PVs on a 24 V system.

And,  there is a bit of technicality of just what is the rated Vmp of each of the PVs that you are using and planning to add.   The Vmp of each of these PVs should really be within 10%.  Ideally they should be within 5% Vmp of each other.   Opinions differ a bit on this,  but many of the fine details DO matter in this area.

As a bit of an aside,  your HUP Solar One batteries are speced for 10% of 20 Hr Capacity as the minimum charge rate.  Even when you add enough PVs to use the full capacity of the FM-80,   you will be a bit short.   Have not looked recently at the HUP specs,  but a few years ago,  they did moderate this requirement,  a bit,   but still,  you might want an additional CC and yet more PVs,  particularly if you try to use the ability of the HUPs to accommodate discharges much below 50% State Of Charge.

Opinions,   Thanks in advance for the specs on the PVs that you have.     Vic

[Doc]

Hi Vic -

Thanks for your detailed advice.  Attached are photos of the panel specs.  The Solarex are each split 12v, wired as 24v.

The strings of 3 idea came from "advice" I recently got from a major wholesale solar retailer in California.  Being a bit math retarded, that "advice" prompted me to consult this forum. 

I live off-the-grid in the mountains in S. Cal at 3600' altitude.  I'm a recently retired sound engineer.  The weather here can range from 10°F to 105°.  I have an EU6500 Honda generator tied into the inverter.  Now that I finally get to live here full time, I'm willing to invest into correcting this system as y'all might advise.  This forum is invaluable!

The HUP's are rarely discharged below 30%.  Never below 50%.  We've been topping off the charge almost daily in the winter with the gennie.  It's that gas usage I'm trying to minimize.

Speaking about gas, I've installed a 2nd system for a rental house on the property.  An OB twin inverter system as outlined by a seller in New Mexico.  It all works fine, but without 'Gen Support' (allocating x% for load and x% for charging, while the array gets shut down), it's a gas guzzler to charge with the generator.  But that's another story.

Thanks again to you all for the help!

Doc

[Vic]

Hi Doc,

Thanks for the detailed reply.

First,  I was wrong on a 15A string breaker being too large.  So you did fine there  ...   sorry.

And on the FM-80 running 100 V or so string Vmp on a 24 V bank,   this will work,  but  I sure would not do it.  Here,  we run  a string Vmp of about 106 V on a 48 V battery.  This is a bit of a stretch on an MX-60 or the more-recent Classic CC.   We DO use A/C in the power rooms to keep the batteries and electronics as cool as possible.   Routinely see Vin to the CC in the 122 V range,  even on warmish days,  when in Float and light loads.   But if the batteries here were 24 V,  would run something around 70 - 75 V strings.

You might want to run the three-per-string configuration through the OB String Sizer,  with your lowest temperature to see how that fits.   You might be just OK,  but seems that the weather gyrations are increasing,  so you could possibly see some somewhat colder nights than previously.

The HUP is a good stout battery.   It is good that you have that Honda generator -- we use one here,  on occasion.   It is a very nice genset,  and will help save some fuel if you choose to run it to charge batteries into the Absorption stage,  where the charge current tapers.

There are some current generation PVs with Vmps of about 36-ish V,  but that seems to be right on the edge of maximum Vmp difference between what you now have.  You could use a different string voltage on an additional CC  and have a much broader selection of relatively inexpensive PVs.  These PVs often have Vmps of 29 - 32V.   You could runs strings of two of these and be fine.

Great that you are an Engineer.   RE power systems and battery maintenance is all about the details.  And observation of what is happening with a system is beneficial to system performance,  IMO.

You MIGHT be able to run one of the 60-cell (30-ish Vmp) PVs in a string with some of your existing PVs,  in which case the Imps should match withing 10% Max -- 5% ideal.   This can be a bit of a mess to rack and wire,   as these 60 cell PVs are considerably larger than your existing PVs.

Mostly opinions,    Thanks,   Vic

[Doc]

WOW!  So my questions were accurately answered - and then some!  And the bonus is a personalized education as well.  I feel like I "get it" now, and will proceed accordingly.  And hopefully this info will help others.

Thank you all for the brain-storming!

Doc

[dgd]

Doc,

Those are fairly ancient Solarex PVs you have. I remember having several of these about 1993/4.

The Vmp you currently have is just on 34v. You will have difficulty getting more PV at this Vmp as Vic has said the 72cell types tend to be 36v+ and the 60 cell types are around 30v.

If you cannot get fairly close Vmp matching panels then the closest would be the 36v types.
Or you may be able to source 17.x volt panels and serial pairs together to get closer to 34v.

The mppt controller will sort out where the maximum power voltage setting is if you have groups of 34v to 36v pvs paralled together.
They are close enough not to be concerned and there will be very little loss of potential power from this slight mismatching.

OTOH the 29/30 volt types have a 4 to 5 volt difference and IMO thats just getting too divergent and more likely to have power lost.

dgd

[Doc]

Now with all the good advice from this forum, I'm finally ready and able to order more panels.  It sounds like my best route is to order a second CC (I always wanted that MS Classic) and 6 more (36v) panels (3 strings) for an additional 1800 watts, or so.  Am I correct in this thinking?

Thanks!  Doc