Mixing panels - matching string voltage

Earthandsol

Greetings Sun seekers,
After what has been the coldest , least sunny winter in the last 8 years of my off-grid solar adventures , I am getting around to adding a few new panels to my array. I have a 24 volt battery bank ( 4 Surrette CS-PS-25s -880 ah ) and my array is currently 12 Siemens SR-90s wired in pairs ( 17.0 Vmpp @5.4 Impp- per panel -34.0 volts @5.4 amps per string ((6 strings ))and 4 Shell SQ-85 s also wired in pairs( 17.2 Vmpp@ 4.95 Impp per panel - 34.4 Vmpp@ 4.95amps per string ((2 strings)) for a rated total of approx. 42.3 amps @ 34 volts . A little undersized for my battery size ( C20 charging rate ) but manageable with conservative use and liberal generator run time. The panels are fed thru a trusty Outback MX-60 charge controller . The output is run thru a pair #1 copper wires aprox. 45 ft to the controller and batteries- an extra pair of #6 copper is run in the same conduit out to the pole mounts and combiner box( a now discontinued Pulse TCB-10 which has been a bit of a nuisance).
The panels are mounted on 2 Two Seas racks on 6" pipes. the racks were bought as sized for 8 of the SR-90 panels but for some reason are slightly oversized ( longer rails ) so there is room for 6 Sr-90s and the 4 SQ-85s on one rack and the other rack has 6 Sr-90s and room for some more.
I have 4 Evergreen ESL Spruce 190 watt panels ( 26.7 Vmpp @ 7.12 amp) left over from some grid-tie projects and have come up with what may be a good idea to tie them in to my system. By removing 2 Sr-90's and spreading the rails on one rack -( I have to redrill mounting holes on the leftover 4 -Sr-90s) I can fit the 4 Evergreens on the less filled rack.
My plan is then to wire the Evergreens in pairs ( 53.4 volts Vmpp @ 7.12 amps--2 strings ) and wire 3in series SR-90s( 51 volts Vmpp@ 5.4 amps --2 strings ) and then run these 4 strings thru the MX-60 on the unused #6 copper wire( 51 volts @ 25 amps . The remaining panels would remain wired in pairs 4 Sr-90s and 4 Sq-85s ( 34 volts @20.7 amps ) and be run thru the #1 copper into new charge controller - a Rogue 3024 or a Blue Sky 3024.
Comments, criticism or cudos welcome.
Chris

BB.

Re: Mixing panels - matching string voltage

Double check the controller Vpanel maximum (panel Voc Cold)--Only the MorningStar Tri-Star MPPT (45 or 60 amp), Xantrex XW MPPT 60 amp, or the Outback FM/MX series will be capable of the over 60/70 VDC that your panels can output.

If you cannot series/parallel connect your panels (match Vmp/Imp within ~10%)--Then each "oddball" string would need its own charge controller.

-Bill

Earthandsol

Re: Mixing panels - matching string voltage

The Evergreen strings( Voc 32.8-so 64.8 Voc per string) and the tripled up Siemens SR-90 strings ( Voc 21.6 -so 64.8 Voc per string) are to be run thru my existing Outback MX-60 - it can handle up to 125 Voc so that is okay.
I understand that Vmpp needs to be within 10% - it is 53.4 V for Evergreens vs 51V for Siemens Sr-90s but the Imps are very different- (7.12 a for Evergreens and 5.4 for Srs) . I didn't think the amps needed to be similar - I assumed the Mx-60 would just operate each string at an equal voltage - which would see some small loss in the 2 Evergreen strings.

mbzebro

Re: Mixing panels - matching string voltage

Earthandsol wrote: »

The remaining panels would remain wired in pairs 4 Sr-90s and 4 Sq-85s ( 34 volts @20.7 amps ) and be run thru the #1 copper into new charge controller - a Rogue 3024 or a Blue Sky 3024.

my comment on your possible choice of a blue sky 3024:
I had a blue sky 6024 running for 2 years and found it has serious flaws; I assume the 3024 is basically the same technology.

Quote from the 6024 user manual:
"The unit requires a minimum PV input current of approximately 2.5A for the MPPT system to operate and raise PV voltage to the
optimum value where the PV array can produce maximum power"

This unfortunately means that the 6024 outputs his input amps directly (subtracting selfconsumption of 0.2A) without "multiplying" by the input/output voltage factor (I checked this many times with current clamp)

For example, when getting 2A PV input @48V, the 6024 outputs 1.8A @24V instead of 5A (assuming 25% mppt gain)
With the same PV input (wired to 24V), any cheap charge controller would output 4A @24V, more than the double.

In a small system, for example 12A max. PV input @48V, in <=20% light conditions (common on cloudy day), the 6024 charge efficiency drops to 35% or less (in the case of 20% light condition: 2.2A output instead of 6A (2.4 * 2 + 25%)).

(Another flaw my unit had was an annoyingy loud fan (vacuum cleaner like); when the temperature reached about 24 celsius, the fan switched on and would not stop until night, even when outputting only a few amps in float mode)

Of course in your case (20amps max input), this "low input flaw" is a bit less serious because your input would be below 2.5 amps less often.

(corrected: I originally wrote blue sky 6048 instead of 6024)

Earthandsol

Re: Mixing panels - matching string voltage

Thanks for the comments on the Blue Sky controller. I installed one for a customer about 5 years ago ( it was a 50 amp 24 volt model as I recall) and it has worked well for her. I am sensitive to noise and I really dislike the idea of losing power at all esp under low light conditions like this winter ( Yuck). So I may look at other options.
I was stung once by the BZ product and have been happy with several Outback MXs and even a new Xantrex XW for a recent client. The new Morningstar Tristar MPPT looks good but there is a dearth of medium sized ( 500 -1000 watt or 30 amp)MPPT controllers. Like many I eagerly await Midnite Solar's new large high voltage MPPT controller and have the fantasy that they might make a 20 -30 amp controller as well.

BB.

Re: Mixing panels - matching string voltage

The Rogue controller has been used by some people here and they find it to be a nice MPPT 30 amp controller (12/24 volt bank). (for lower voltage arrays).

-Bill