project e-trailer w solar charging - electrification phase

[aum]

ave, gratis plenum

i am making an e-trailer with solar charging, here is link to design and making of pre-electrification phases:
https://endless-sphere.com/forums/viewtopic.php?f=1&t=108424&p=1585634#p1585634

now i am the electrification phase, and have some questions, which, i decided, i will post on-the-need basis, in order to not clutter this whole thread.

q1: first we need to mount & connect solar panels; there are
(both flex) sunpower 3 x 110 w (voc = 22.8 v, isc = 6.3 a) & 3 x 50 w (voc = 21.5 v, isc = 3.1 a) solar panels
and
mpt-7210a mppt boost charge controller (in v: dc 12-60 v; out v: dc 15-90 v; out a: 0-10 a; out power: 600 w max)
and
grin technology (new but half-melted) solar current sense 16 a
and
lifepo4 48 v 28 ah battery

all together:


solar panels are mounted on a trailer like this:



from the beginner's and amateur's point of view, wiring is, after some homework regarding serial and parallel wiring, very simple:
connect in serial panel a-110w and b-50w, which gives 22.8 + 21.5 = 44.3 v, 3.1 a (weakest link (current) dominates, ja ?)
then, following same logic, connect panels c-110w & d-50w, next e-110w & f-50w
this gives 3 x 44.3 v, 3.1 a (serial) panel groups; connect those in parallel, which gives 44.3 v, 9.3 a (3 x 3.1 a); this is, from beginner's and amateur's point of view, satisfarily enough below-the-rated specs of involved elements (grin t. solar current sense 16 a is weakest link, i believe ?)

but, from not-beginner's and not-amateur's point of view, there are some issues:
1) panels are obviously mounted on either side of the trailer (group c-d, & e-f, plus on top and back of it (group a-b); this, at any point of time, makes one side of trailer being shaded (ie c-d (= left) serial panel group); this shuts off 1 sol panel group, and, if not mistaken, significantly reduces overall (that is, all 3 serial-sol-panel-groups, ie a-b, c-d, & e-f group) power (voltage and current) output to mppt controller. ja ? also, the back panel could be shaded (if sun is in front of driver / trailer), which could also shut off top-back (a-b serial) sol panel group (though this now not matters, since either side is already shaded by now)

of course i did some more homework, and found out that not 1, but 2 or even more mppt controllers would be a possible solution; this is not clear to me, but as i understand this, i could:
a) wire group a-b (top-back sol panels) to 1 mppt controller, then perhaps group c-d (left panels) to 2nd mppt controller, and group e-f (right panels) to 3rd mppt controller, and then somehow connect those 3 mppt controllers. how ? in parallel, is my 1st guess, if in series, voltage will (potentially) increase way over, resulting in battery explosion of world-cataclysmic proportions. we do not want that. yet.
b) i could use some sort of frankenstain switch array (image below),

and i) permanently connect a-b group (top panel will always receive sun, back panel i could 'open' even when riding);
ii) then as need dictates, i could frankenstein-switch-connect either left (c-d) group, or right (e-f) group, depending on the direction of travel. does this make sense to experienced and knowledgeable counselors ?
iii) when either side (c-d or e-f group) is shaded, i could remove it from solar grid, simply by turning off the switch (be it frankenstein or something modern, spacey, ie used, 2nd hand iss main rocket booster switch, lol)

please keep in mind, the main purpose of this question is for me to connect the solar panels in a safe way, so i do no smoke-vanish any electrical component.
max power input into battery is NOT main goal - safety first. nor am i interested in max power harvest over the course of the day. nor in top speed all the time, au contraire, easy, safe and therefore enjoyable ride is the goal. and i am prepared, and am sure that it will happen, to stop for a few days, if overcast sky dictates, and charge the battery, then continue the travel.

trailer's e-powered motors are by default meant for help uphill, and to help save brake pads, by using regen braking downhill.

what is also important, and perhaps i should mention this before, while traveling, i will take brakes, for rest and meals, and during those rests i will / can 'open' all 5 (6) panels at once, and therefore 'quick' charge the battery.

uuups, i forgot to tell that all side (5 pcs) panels are on hinges, therefore they can be 'opened', ie when not traveling. also, top panel (110 w) is on top cover, which can also be turned towards the sun, but this would open and expose the interior, the precious oh precious cargo.

please advise, how would you connect those 6 panels ?

have fun

aum
s love nia

 

[ZeroEm]

I have one of them Green mpt-7210a mppt boost charge controller. It lasted about one summer. Hope yours last longer. Grin has one, that's marked for clearance. I may get it myself.

Had my panels connected in series. Shade on anyone of them will drop the output. Did not have mine on a trailer.

 

[JackFlorey]

a) wire group a-b (top-back sol panels) to 1 mppt controller, then perhaps group c-d (left panels) to 2nd mppt controller, and group e-f (right panels) to 3rd mppt controller, and then somehow connect those 3 mppt controllers. how ? in parallel, is my 1st guess, if in series, voltage will (potentially) increase way over, resulting in battery explosion of world-cataclysmic proportions. we do not want that. yet.

Too complicated.

Connect all panels in parallel. Feed them to one boost mode charge controller. The 7210A should work well; a Genasun GVB will work as well with a little clipping. You will get, at most, about 150 watts at any given time while riding, so both should be able to handle the current.

I'd recommend against the hingey-swing-out-open-up thing. Your trailer will sound like a bag of broken springs being dragged behind you and won't last long. Unsuspended trailers take a lot of abuse, and anything remotely complex will have trouble.

 

[aum]

Too complicated.
Connect all panels in parallel ... You will get, at most, about 150 watts at any given time while riding, so both should be able to handle the current.
I'd recommend against the hingey-swing-out-open-up thing ...

thx for the reply

regarding "all ... in parallel" - i am a bit skeptical, because with each sunpower solar panel comes a small leaflet (safety & installation instruction), which hosts a small print, qoute: "... connecting panels in parallel .... additionally, if one of the panels were to go into reverse bias, for example by shading, damage to the by-pass diode in the junction box may occur. avoid shading either panels. sunpower does not recommend connecting these panels in parallel if shading will occur." end quote.
though i might be unnecessary paranoid - this text refers to charging the battery directly with sol panels, while i have a mppt controller in between sol panels and the battery. du no. "... if shading will occur." - of course it will.
beside, connecting all in parallel would give 3 x 6.3 a + 3 x 3.1 a, which sums to 18.9 + 9.3 = 28.2 a, way over 16 a, nominal value for grin t's solar current sense. if i am missing here something, i beg you to let me know.

regarding "... against the hingey-swing-out-open-up thing ..." - well, we'll see how this turns out, i actually did not, not even for a moment, thought this would be a bad idea. but, reading some 'e-trailer' threads, it seems that (flexible) solar panels should be mounted rigidly, that is, backed with rigid, sturdy area / plate, to minimize flexing. this on-the-road-bumpy-micro-flexing seems to result in micro-cracking of the solar cells (not panel itself), and consequently, with time, lowers the overall power output of panels. there is not much we can do against this, it falls under 'wearing out' label.

godspeed

aum
s love nia

 

[aum]

... one of them Green mpt-7210a mppt boost charge controller. It lasted about one summer ...

yea, i heard that this green little monster might come from few different china factories, consequently varying much in quality - some might work well for a long time, some others might not work, if at all, lol

i can only hope, if mine breaks, that it breaks somewhere near the big city, so i can hopefully get some (better) replacement, or electrician capable of repairing (and possibly upgrading ?) it.
grin technology recommends a (chinese) silver mppt model, as a cheaper alternative to genasun (48 v 8 a costs 158 $, plus import and transport fees, mpt-7210a costs 32 gbp, free shipping from china)

godspeed

aum
s love nia

 

[ZeroEm]

Maybe I should try to fix the one I have might be something small but my eyesight is not as it once was. Bought a Green buck DC-DC converter for my trike and it lasted as long 6 months.

 

[JackFlorey]

regarding "all ... in parallel" - i am a bit skeptical, because with each sunpower solar panel comes a small leaflet (safety & installation instruction), which hosts a small print, qoute: "... connecting panels in parallel .... additionally, if one of the panels were to go into reverse bias, for example by shading, damage to the by-pass diode in the junction box may occur. avoid shading either panels. sunpower does not recommend connecting these panels in parallel if shading will occur."

Yes, that's what those bypass diodes are for - to isolate sections of the panel that might see reverse bias.

However, in your case you're just not going to be generating that much power. You likely won't be able to generate enough reverse current to damage anything even under worst case conditions. In large installations the rule of thumb is to not parallel more than two strings. In smaller installations people regularly go to four or five panels.

If this worries you, use two Genasun controllers to avoid paralleling all the panels at once. I wouldn't worry about it though.

it seems that (flexible) solar panels should be mounted rigidly, that is, backed with rigid, sturdy area / plate, to minimize flexing.

Definitely. I am using 1/4" plywood. Coroplast can work if you reinforce it, and carbon fiber panels would be awesome - but super pricey. (And it's conductive, so watch for shorts.)

 

[aum]

... in your case you're just not going to be generating that much power.

aah, good to know this, thank you, you made my day

forward:
using beech plywood, for solar panels back-plates (beech is strong, the rest not so, but they are lighter), i get:
2.5 m2 (square meters) of solar panels area (3 m2 with top 110 w panel) x 0.006 (for 6 millimeters plate thickness (= 1/4 inch)) x 800 kg/m3 (kilograms per cubic meters) = 12 kg (26.5 lbs) of wood for solar panels back-plates - too much. we'll see about that, skip for now
or, perhaps reinforcing the solar panel longitudinally (parallel to travel (= wind) direction), ie with 2 pcs of rigid (ie carbon fiber) sticks, might do the trick

forward:
so far, battery and mppt boost charge controller are equipped with xt connectors (60 & 90), also 'chocolate' where solar panels wires will come, is mounted, and waiting for wires
q #1 - where should i put fuse(s)
option #1:


option #2:


please advise, and thank you

mounting of solar panels:


godspeed

aum
s love nia

 

[aum]

hello

i have
1) golden motor magic pie edge e-motor, 2 pcs
2) grin tech double shunt
3) grin tech cycle analyst 3.1

i need to connect all together:


a) is this wiring correct ?
b) do i need to connect the gnd & shunt-ve & shunt+ve (pin 2) to e-motor ground ?
c) just green wire (pin 3 - throttle out) to green wire (throttle) on e-motor ?

thank you for your help

aum
s love nia

 

[aum]

hi

i also need speed wire, connected to ca3.1 (wp), to measure the speeed
1 option is to draw / connect a hall sensor wire, directly from internal controller, in golden motor edge e-motor(s), but i have opened the e-wheel, took a look at the controller, and to draw an extra wire from the controller, to the outside of e-wheel, would require a massacre, which i strive to omit, for now. trying to find a better solution.

here is the situation, in image:


q: is it possible that (my edge) white wire (+5v – VSP5V) is a speed wire ? it says v-SP-5v, as SP for speed. can i connect this to cycle analyst 3.1 speed wire (yellow; ca3-wp) ?

if none knows the right answer, can i:
- wire everything up, incl. green (throttle) wire from my double shunt, to gm edge green (throttle) wire;
- then connect my voltmeter (multi-meter), to gm edge white & black wire (supposedly speed & ground);
- manually rotate the wheel, and see if i get any readings on voltmeter ?

what readings should i expect, if the white wire is indeed speed wire ? cca 5 v every 1 turn of the wheel ? that would i expect, similar as a hall sensor 1 wire (ie yellow) – 1 signal per 1 rotation of the wheel. i just do not know what signal.

i also have golden motor bluetooth module, which goes into 5-pin wire harness connector, and if i use smartphone 'smart meter' app (from golden motor), i can read the speed on my smartphone screen. therefore i concluded there has to be 1 wire for speed.

thank you for your help

aum
s love nia

 

[LewTwo]

Just happen to have this setting on my desk
I wasn't even sure I could get enough light on this for the picture.


The green wire is labeled "SP"
The white wire is labeled "SP+"
The red wire is labeled "CR"
... as an aside that hex socket head cap screw with the green paint is for the green phase wire.


The orange wire is the "TX" (Transmit) wire
The gray wire is the "RX" (Receive) wire ... it may not look grey in this picture (but it aint brown).
... as an aside that hex socket head cap screw with the yellow paint is for the yellow phase wire.

The black, blue and yellow wires are bit more of a challenge and all that silicon good does not help matters.

Best I can tell ...
The black wire is labeled "PA12"
The blue wire is labeled "BKL"
The yellow wire looks to be labeled "TA" (required 10X loop)


That rectangular white connector at the top right is for the internal hall sensors.

This controller is from a Magic Pie V4 but my understanding is that all GM V5 and V4 controllers use the same external (god awful) wiring harness. "SP" and "SP+" are used for 'speed control' (i.e. the throttle). There is NO external "speedometer" wire for the GM controler. As it can read the halls internally it does not need one. If you need a "speedometer" wire for the CA then you will need one that is based on a magnet attached to one of the wheel spokes.

 

[aum]

lewtwo, thx for the reply

yours is indeed (mp) v4
mine is (mp) edge v5

and, yes, all that silicone is a good, and me disturbing this internal zen protection, is a bad idea
i have put aside the project, just over the weekend, contemplated a little, decided to 'try' to squeeze 3 thin, extra wires, into the controller, 2 for thermistor, 1 from hall sensor, for speed
so, unscrew and put apart everything ... again ...
i also need to count the magnets in the rotor

best regards

aum
s love nia