diy solar

diy solar

IoT (cell) uplink for GroWatt SPF 5000ES

OK
7 installed
4 facing South aiming pretty much at the sun
3 facing East aiming er... on the way there

They are all putting out about 6A this morning.

Went thru 5 permutations of ground mount, hated every single one.

Now to get some data on thruput current when one module in series is shaded. Properly bypassed, I should be able to continue to get full current, but at a lower voltage. Yea yea... I said properly bypassed. In this way I hook up 400V worth, maybe I get 300V, maybe 350... but I should NOT be limited by the lowest current in the bunch

Eh hem
About to prove it or build it

IF I have 6 panels trying to push 6A and the 7th is weak, that one should just get stepped over. Not that hard to do. It might not add to the sum, but it cant block the sum. That is the whole point of bypass diodes.

-methods
 
4 panels facing south we're getting over 8 amps
Three panels facing East were getting 4 amps

I was fooled by the morning light!

Drug the three panels facing East to face South, one on a bracket two on buckets... Even the buckets are getting 7.5 amps

Was getting 1.2 kilowatts (meaning the weakest link was limiting the current)

Now I'm waiting for the machine to boot (sigh)

-methods
 
1641072581820.png

Looks about right. Seeing 5.8A of a 10A absmax. January 1st. If I moved 3 of the panels I am positive I could bump that 5.8 up to 6.8. That leaves us about 3A short of absolute optimal... and given that it is Jan1, leaning way over... I would say at this point it is "Good Enough"

Open circuit voltage is 350, that sags to 300, 7 panels, 5.8A.

4 panels fixed almost directly facing
3 panels flopping around, definitely not facing
Peak short circuit measured at over 8

Carry the one, divide by 2.... multiply by pi....
Yea
Running Series is problematic without a proper bypass.

OK
Proven to myself. Collecting 1.6KW and that will suffice for the time being.

-methods
 
Solar Meter says the following

4 panels around 700W/m^2
3 panels around 600W/m^2

So we know we have to take the worst>

Any of those, if I lean the meter up, gets over 1000W/m^2.

We have a hypothetical 3KW we are after, we are getting about 1.6KW. Meter says we are down by 400W due to our angle. That is 2/3 gain waiting for us. That is about 1KW... so...

If we lean ALL of the panels up just a little bit more our 1.6KW turns into 2.6KW and THAT... is quite acceptable... given all the other variables.

* Rotation angle is not perfect
* It is January 1st
* Dust on most of the panels
* Cracks and crust on all of the panels
* Worst-Case panel is inline (bottom of the pallet panel)

So... to the best of my Engineering ability, I would say that there are zero unknowns and the system is doing exactly what it should do. The pocket meter REALLY helps with this... because...

LOL

Your BACK will tell you all the panels are pointed DIRECTLY at the sun (lol) when in fact they all need to come up another foot. Yepperz... It is January. Not only shorter sun hours... but sun is far lower in the sky... so simply not going to hit straight on for my compromise angle.

So
From that
From this point forward, I will assume a save "Half output" in the winter months. I am sure that exactly what I have here would belt out another KW in June

Sweet
Satisfied

-methods
 
This video was produced for a different purpose, but it backs my statement about the importance of angle in the winter months


The absolute values dont even matter. What matters is the relative difference. The 3rd party measurement (Solar Power Meter) near perfectly matches the nameplate math and collected energy.

SO
Myth Dispelled. It is not just the "Solar Hours" in the winter, it is the different angle. I dont know why its so hard to find a graph of that on an image search, but it is measurable fact.

If you dial your system in June... its going to put out less in January. I say it is so.

So
That is why I have the Linear Actuators on order. They will arrive in 2 days.

-methods
 
I bet many folks are getting on the order of 30% output from their Solar Install*

Especially around here. . . where... when I take a drive the visuals are dismal. Panels aimed in all directions but toward the sun. :) Wasted space at some point. Just dead weight up on the roof, aging in the sun.

Time to make some progress toward lower cost, single degree of freedom, tracking.

It is like $42 in brackets, a $60 Linear Actuator, and $30 in electronics. I will now prove that is fact... by doing it, showing results, and posting a BOM.

Yea
If someone else installed your solar, non-starter. If you installed your solar... very do-able.

-methods
 
When I am solving broken windows OS issues, I almost always find the command line answer out of India. I cannot understand what they're saying, but I watched the screenshots.

...

After scouring the internet for hours... The only tests I found, directly correlating tilt angle to Short Circuit current, we're coming out of India. There were one or two ladies in the woods as well.

...

In this video he is at about my angle... Getting 4 amps and change. He then tilts up and gets about 7 amps and change.

That is what I see here, but a bit more exaggerated... Likely due to our coastal conditions, the time of year, and of course my cracked panels.


I'm trying to quantify how much the cracking affects sensitivity to angle. I do not have a point of reference! I do not have a known good panel to compare to!

... It is known that some panels are more sensitive to angle. It makes sense to do this, depending on how they will be installed, where they will be installed, ...

If I'm putting a panel on a tracker, then I want to maximize power Dead on. If I'm mounting a panel fixed, I want to scavenge more from the sides.

... This is what I'm trying to understand

It is almost impossible to understand something without a point of reference. Without a known good.

How do I quantify the effects of the cracked glass, when I cannot compare to something without cracked glass

The scientist dilemma. I will find a way.

I am nearly positive that coatings will help alleviate some of this angle sensitivity***

-methods
 
Now that I have 2 units operating

* GroWatt 5000 ES
* Solar City 5000W

Now we have reason to get in there and forward the data via cellular to the cloud

(I.E. almost back on topic)

The growwarr is standalone in the bush, must be cellular. The SolarCity is within range of WiFi

Both fire out on a serial line...

* Lay circuits to read rs232, CAN, UART, USB
* Relay to Cell or Wifi

I will include an standard FTP implementation


-methods
 
Both the cellular and Wi-Fi antennas will have to be remote! The boxes are metal Faraday cages.

Antennas can be a flat panel sticker or a standard aerial

Power
I'm probably going to use an isolated 240 volt to 5 volt converter. You can't reliably pull power off the UUT!

No cooling will be required, no heat sinking.

The box is going to have to be highly insulated electrically. Wires are going to have to be very abrasion resistant. Attachments are going to have to be locking and true to vibration.

I don't want to bring any wires out of the box... So the electronics are going to have to live in the box

A small hole may need to be drilled*. Great attention to where and how... To maintain the Faraday cage

....

Both will want to radiate via BLE for direct phone interaction, OTA update, and "fiddling"

Yes yes... I will get around to laying out a PCB and open sourcing it. Good designs come from someone who actually works with a product!

Notice that I didn't come out of the gate and just make something. I first got into the hobby for a month to figure out what the difficulties are.

What you're going to want to do while you're experimenting. What you want to do while you're troubleshooting. What I don't know yet is how to interact with it long-term

I presume I just need to fire to a weblet. Nobody wants to come through 5,000 Excel spreadsheets.

A few accumulators, some reset buttons, some sliders to graph a range, some graphing

Yea - I know how to make something that's rad.

... Will want to be able to turn the system on and off from the phone!

-methods
 
Requirements

In order to justify this pilot production run the Data logger is going to need to serve two purposes.

* Provide wireless access
* Actuate Sun tracking

I'll throw together some simple low side switches that can either directly control linear actuators that are 12 volts to 72V, or can control an isolated SSR for AC.

We will have to have a small Cable bundle that is UV and abrasion resistant. It must be totally and completely electrically isolated!

This will go out to the photo sensor*

You achieve total and complete electrical isolation by using digital isolators along with isolated DC to DCs.

... Ingress and abrasion and UV
These won't be much of a problem.

I may power the system off of its own small solar panel... But we need to think about this. Linear actuators take quite a bit of energy to run

At least the linear actuators that can move 600 lb around :)

So we need to think long and hard about what that source of energy will be. A small battery, high C-rate, with its own tiny solar panel... It's probably the answer

Lifepo4 pack, so we don't have any fire starters

... If we do this we can get away from tapping power off of the inverter. It's a far better idea, keeps us away from any DC or AC high voltage...

Within just have to isolate the comms to the box

Also
We don't even have to wire to the box if we're very clever. You can shoot ble a few feet even through a farad a cage

If at all possible we want to not touch the qualified electronics, we don't want to drill any holes in it, we don't want to do anything to sacrifice it's qualification

... So now with these requirements I'm leaning more towards an external standalone system

Two stages of relay
* Relay from the box to the outside local
* Relay from there to the cloud

... Yeah it'll cost an extra 20 bucks, who cares about 20 bucks. You'll make it up and energy savings in a week if we get this right.

I'm at 25 cents a kilowatt right now! Believe it or I'll show you my bill. I'm going to break a thousand bucks this month

-methods
 
Yes
Stand alone system which is agnostic to the inverter used and does not compromise any sign offs.

To accomplish that means NO ELECTRICAL CONNECTION.

Easy enough.
So what is MVP inside the 3rd party box?

* Nothing Ideally! So hack the protocol for listen-only
* A small, short range, ble dongle

There are rules around taking two "qualified" radiating devices and putting them in the same place at the same time. Both qualifications can move forward, if you achieve the required spacing.

But
Since it is all a racket anyway, and we don't seek to profit... We can create an "off road only" edition, which in this ethos - we call "Off Grid"

Moving right along

-methods
 

It was between building out the BMS and implementing the active tracking. I am going to go ahead and build out the tracker first.

The big question with these cheap linear actuators is the side-load point of Yield. I don't know that yet, so we are going to need to destroy one.

-methods
 
Biggest concerns

* Ingress failure modes
* Side-load catastrophic failure modes

I'm far more worried about ingress than sideloading. Anything you can do to shield the unit from a constant rolling stream of water will help.

I don't care if some guy on Amazon claims IP68 or IP54 or whatever. Usually they're just copying and pasting jargon. Sometimes they're actually rated... But those tests are there withstand ingress for like 15 minutes!

Surface tension alone will get you through nearly all IP testing!

This is something people don't know.... An IP test only runs for like 15 minutes or an hour or two. On that time line just some grease will get you through.

Over the period of a year?

Maybe not so much. Maybe when that tiny little gasket embrittles from all the UV and thermal pounding... It starts letting a little water in.

So
We're trying to get away from Grease for sealing water. Something like silicone is preferred. The problem with grease is that it picks up particulates. Those particulates then act like sandpaper.

... I often use a PTFE (Teflon) infused lubricant that cures dry. This would be like the stuff you use on guns. You can buy a big yellow can at Walmart and it works beautifully.

That will reduce the friction and drag. You have to check its compatibility with your gasket...

But basically you can make a line on line gasket, get rid of the stiction and friction, then rely on surface tension.

That's better than grease. Notice the units don't come with any grease on them, like cheap steel from China does

BTW
If you ever get anything from China That's wrapped in wax paper, smothered in cosmoline, oiled heavily... Etc. You need to keep that part oiled for life or strip it and paint it.

This is why we really prefer aluminum... But watch out! Lower grade aluminum will oxidize horrifically fast. That anodizing they put on things will come up in a matter of 3 weeks in direct sunlight.

All you have to do is throw it up on the roof and wait. Your answers will come but no work.

* Parts Pedigree
* Proper Seals
* Long Term Failure Modes

-methods
 
* 1pcs angle iron along the bottom rail
* 1pcs heavy angle along the top rail
* 1pcs link between the actuator base and bottom rail

I think that will cover MVP. It can't fall over side to side and it can't actually far enough to flip without 50 mph wind.

Provisions for staking, but also a platform that can be sandbagged.

Sand and bags are cheap*
Since we're building something mobile it's important! I can pour concrete with the best of them, but that doesn't help you when you figure out your angle is off by 10° from optimal... Are the tree shadows encroach.

Nope
I really believe that anything good needs to be modular and mobile.

Ok
Going to survey the site soon. It's a hillside... So it's going to be a real hard space to work in.

Working on flat hard packed ground is a luxury! As soon as you're trudging bushes, climbing Sandy hillside, or knee deep in bug-mud you will remember!

* Access for the build
* Access for maintenance

. . .

-methods
 
Many lessons learned on my path into solar!

Here are some of my most recent leanings on the way to hacking into the Firmware.

* Wind blew two of my 460W panels off of an elevated structure. They were both already shattered. One seems about the same, the other was *almost* punctured but still electrically connected. Threw them back up, 6 panels put out 300V and I need to get back for some short circuit current testing and MPPT load testing.


* Had my first bad(ish) interaction with Signature Solar. I bought right before the Holidays (Xmass Eve,... remember... I am a TEST ENGINEER) lol. Took them forever to ship my goods, so... they are having growing pains. My second 5000ES arrived with a battery buss bar nut rattling around in the main compartment. That was fun.

I bought the combo kit (with Auto Inverter) and this time it shipped with a different version. This one showed up with undersized and rattling loose mounting fasteners. One was rolling around, three were backed out, all were missing washers. Fine - but - two boxes of loose screws???

I dont mind Loose Screws*


* Signature Solar has not been answering the phone, ... and when they answered Emails it was very weird. They did not "reply" with quoted text, but replied out of context in a new thread and did not answer my question. Eh hem, bad news. Good Communication is paramount for business.

I have been shaking down outfits like this for a LONG TIME. Folks will learn what is important and what is not. For me... Loose Screws and sparse customer support is not a problem at all. BUT... when it comes time to start dealing with FAILURES... they will figure out RIGHT QUICK when methods starts ripping into them. I am a fierce Engineer with 20 years experience and I set a pretty low bar for "Works". So far what they have sent me "works" - so we are on good terms until things start breaking.

SO
Putting them on Warm Standby until the failures start occurring. I dont like slow shipping or poor communication, so for the next setup I will probably use a different source. I still need to buy 1 more 5000ES to make my 10KW setup (15KW inverters to support 10KW continuous draw).



* Got into that Solar City 5KW inverter for a while. I have it wired in on some 8AWG with 7pcs of 460W panels. Due to my angles I was only able to pull a couple KW, but... it works pretty damn turnkey. I was very unhappy with the MPPT algorithm they are running. It is complete trash at anything but full sunlight. I absolutely guarantee that at low light levels the box is under-performing. Clearly on purpose, ... so... perhaps for some reason that I do not yet understand (or maybe thats why they went out of business).



* Scratching my head hard on why the bypass diodes in the panels dont allow full current to flow when 1 of 7 modules is shaded. I cant for the life of me figure out why 7 modules pushing 9A would drop to 3A when one module shades out. What should happen, is the array voltage should drop by 50V and short circuit past the shaded module. Then, the remaining 6 should keep pushing 9A.

I need to do the MPPT math on that, and see if it is just my GroWatt that is doing it.

Example
50V * 6 modules = 300V
10A * 300V = 3KW
50V * 5 modules = 250V
10A * 250V = 2.5KW

These are round numbers only, for relative comparison. So it we completely drop a module out that is 500W off the stack. Now how does that play out when we start to MPPT it?

All the modules will open circuit 50V
All the modules MPPT to around 42V
All the modules will push 9 - 10.5A short circuit
4 of the modules will short circuit 8A as placed
2 of the modules will short circuit 6A as placed

So what should I see in this example? What is the Maximum Power?

42V * 6A * 6modules = ~1500W
42V * 8A * 4modules = ~1344W

OH - DOH
It is higher power to run all of the modules at the lowest modules current while keeping its Voltage UP, as opposed to shorting across that module - IN THIS CASE - but as the array gets longer, the current should outweigh the voltage loss as the ... er....

Power = Volts x Amps

If a Panel shorts out to conduct current, that is a loss of 50 in the above equation... to gain 2...

Yea - ok - what I am seeing actually makes sense

PAUSE
(I woke up at 3AM again, horrible nightmares)

ONWARD

* We broke ground on the power run to the 32 module array. This is the setup where we are taking 32 of the broken 460W panels and mounting them all together in some kind of Series Parallel. At this exact moment in time my plan is to pull 6pcs of 10AWG + ground thru the 2" pipes. I figure with this... I can configure anything that makes sense. I will have 3pcs of GroWatt down in the shack, so I will need to run 3 arrays. The GroWatt wont over-draw 10AWG, so... should be fine.

1/3 of the trench is cut. Still need to do the upper and lower trenches. I am going to break it up into 3 runs of about 100 feet. I am figuring something like... 10A continuous on the 10AWG so I do not anticipate a bunch of line loss. To accomplish this, I MUST run the three arrays in Series, ...

So back to Shading

* Array Math
Target voltage is 420V unloaded. 420V / 50V = 8.4 = 8 or 9 panels. Call it 8 to keep things symmetrical. 8 panels loaded down to 42V pushing 9A will come in around 378W * 8 = 3KW. Three arrays of 3KW should get me 9KW, which is close enough to optimal.

BUT
To even DREAM of landing that, I have to... have to... Have To... HAVE TO... aim the panels perfectly at the sun***

MUST AIM DIRECTLY AT SUN

So
I am calling my first install of 6 panels elevated a TOTAL FAILURE and I am moving over to Pole Mount. Poles allow both rotation and elevation adjustment - so - I can start by hand adjusting and collecting data, then I can get my Linear Actuators on there. The actuator system will come in around $100 (Two actuators, controller, sensor, brackets, cabling, housing.. all DIY) so we double that, figure $200 each for 5 systems and ... a lot of labor.

So
That will come together - but - what I am stuck on right now is the damn Pole Mount costs. FLIPPING EXPENSIVE - and... 8 panels are HUGE and the wind will catch them and they are HEAVY... 60lbs x 8 = 480lbs... add to that the mounting, so figure 20% at least, maybe far more... we are going to have to call it 600lbs.

I need to put a 600lb kite up on a pole (OMFG) and wiggle it around with cheapo linear actuators (sigh). 3 large poles, each holding 8 panels.

That is what I am currently trying to Riddle Solve


I am of course reverse engineering everything I can find.


Just harvesting ideas and things that WONT work - lol - like that Amazon system. Lol, bend that thing over double.

- Need large poles, 3 of them, must be round so I can rotate
- Need to build 1 proof of concept

ONWARD

Largest Risks so far
* WIND

My biggest mistake so far was underestimating the power of the wind.

-methods
 
Rough Numbers

Right now if I drive a bargain and handle all of the labor myself I should be able to get about 10KW for $10K consisting of
* New panels
* Pole Mounts
* Inverter

Add to that all the conduit and wiring... and endless labor... so... really more like $12k in materials I am guessing. Labor is free, we have 5 guys including an Electrical Engineer, 2 guys with Tractors, guys who can trench, etc... so... Labor is the big missing part that covers most of my savings.

I have to come in at HALF to make all of this worth my time... and... I have to perform comparably. That aint likely to happen, so i am going to complete the entire job, for no pay, with no materials left over - JUST FOR THE EXPERIENCE>

So Far
* 30pcs 460W panels cost me about $2.5k (10 at $100, 20 at $75)
* 10KW of inverter + Auto Transformers cost me about $2k
* Box of random bits and experiments, about another $1k

So after I pull away some extras for experiments and R&D, I am in about $5k already. I am at the ready to extract near 10KW from the sky and utilize that. I have a pile of pull-down batteries sufficient to buffer the loads (another story all together). If you told me SHTF right at this exact moment, I could take what I have in my possession, and produce NEAR 10KW for 6hrs a day out of thin air (on good sun days).

...

At a Minimum
I need to slide in UNDER $10k including all of the wire, conduit, mounts, install, etc. Turn-Key finished, producing honest power, for under $10k, in a way that is reconfigurable, upcycleable, etc... using Pull-Down gear...

So Mounting
Getting Killed on Mounting


To hit my power numbers I have to have the ability to articulate the panels on two axis. I have a real power draw need... not some hypothetical. I have to run a 7HP pump for 6hrs a day. That is a BIG LOAD... that you have to have some headroom on. I figure I need 10KW on tap to assure that 7HP works as desired.

We will see
I never fail, it just sometimes takes me an iteration or two to hit target. In the end I will have gone from knowing nothing about solar to knowing how to set up a 10KW (then 20KW) Solar Farm, and... we can get on to hacking that firmware. :)

Mounting

I am on the round face of a hill with full sunlight. Soil is 200 million year old sand. The face of the hill wraps around enough that I will have to tip over at least 2/3 of the panels. Very few of them will naturally face directly at the winter sun path. Summer? Wont matter, gets POUNDED by sun. But Winter? DOH -> Much harder to collect energy than I was lead to believe.

eh hem

So
* Laying them on the ground wont work
* Propping them on simple 2/4 platforms could work, but would really suck for seasonal adjustment
* I am not building any more large, elevated, monolithic structures. They simply dont collect energy thru the winter unless perfect

I have to run something like 24 - 30 huge panels, so... fence posts and individual mounts are pretty much out of the question. It would be way too fiddly.

I have eliminated a pole based ground mount, as it is no different then building it out of scrap lumber

Poles
Big ass Poles...
I was aiming at 4 panels a pole for a reasonable 420lb load scenario... that would cause me to have to set up at least 6 stations. The price on hardware is outrageous right now, so I have to build from all pull-down and scrap materials. Whatever I can find... is what we will use. I have a large truck, tractors, etc.... so I can pick up a good deal if I can find it. Some kind of beams or old telephone poles or timbers from an old cabin

Something
But.. I need to get one of those planted in the ground and start working thru the logistics of getting a rack elevated. Super difficult work on a steep slope of sand. SUPER difficult, almost impossible to do "Safe" - so - strictly DIY

Bah
Other options

* 24 separate, stand alone, knock-together ground mounts built out of scrap
* 12 pairings, standing tall, knock-together from what I can find - two panels each
* 6 stations, 4 panels each, call it 420lbs minimum... that is as big and heavy as I want to go

Could cut into the hill a bunch, but that wont hold together well. Just end up having to do a bunch of retaining wall. Nope... I think ground mount is really out of the question for anything other than the initial bringup testing.

Pole Mount

Wont Work

Wont Work

We can see that most of them are nothing but user supplied pole stock, some U-Bolts, and some brackets cut out of Angle Iron. No magic there, just paying for someone else doing the thinking for you. Price of pole stock varies WILDLY and you really have to call around to see what a pallet of them will cost you picked up.

Wont Work

Doing something like that on a standard Fence Post could come in under budget, but 4" steel pipe is relatively expensive when purchased in small quantities. Building one would be fun, but multiplying that by 30.. nope. Does not scale.

1642949686804.png
I think the above is as simple as it gets. The bracketry on the sides is trivial to knock together from scrap. The vertical is 4" and the horizontal is 3". I bet it rocks and flaps around like crazy!!!!

Sigh

Rotation would be trivial, but leaning would be bugger. 5 clamps with 2-4 fasteners each... I would of course leave them loose and fix it all in place with the Linear Actuator. Weight rests mostly on the pole, with only the difference weight and dynamic wind weight out on the end. Balanced***

1642949853371.png

This one above must be a Typo looking at all their other prices. Saying $1.24 for the top bracket.

Yea
See? They want $90 for this one

Knock that together out of scrap in 2hrs.

Sigh
No progress

Plan is about to switch over from Lofty to Emergency... about to make some compromises to GTFJD.

-methods
 
SIMULATED EMERGENCY

Ok, here we go

* Its GO TIME
* We have a few thousand bucks
* We can only source local
* I can work metal, I have a welder, chop saw, angle grinder, mag drill

I do not have the time to ENGINEER anything that can hold 4 or 8 panels. Too much damn risk, too hard to assemble... I watched a few videos of guys pulling up EPIC arrays and they were using Crane Trucks and major rental pull-along hardware. I could RENT some stuff... and I know a guy with a Crane... but the location and reality just dont add up.

Very remote
Very soft soil
Very steep

About as bad as it gets, short of having to fell a bunch of trees.... so... What now?

...
I KNOW that if I go and try and cut out a bunch of flats to get the angles right on a ground mount system I am going to erode the mountain to such a degree that I will spend double shoring up. Cant do it, wont work. I have to set poles... and set them very deep or distribute them.

Poles
Ok - how about 24 of them (barf)


* How high do they need to be?
* How deep do they need to set?
* What is available?

Fence Posts are as common as dirt. I could literally pick them up off the side of the road here and there. The "End Poles" are the ones I am after, something like 2 3/8". It is possible that a 1 5/8" pole could handle a single 60lb panel, but... it wouldnt feel right. I need more diameter to work things out how I want to.

But - lets not eliminate that

* 6' lengths of 1 5/8" for $12 right now

* 6' lengths of 2 3/8" for $21 20 miles away, get them today

So we are looking at $300 or $600 for just the posts to cobble together 24 fiddle-ass, individual mounts. Add to that

* At least 2 U-bolts if we want to rotate but not lean
* At least 4 U-bolts and more pole + bits if we want to lean in two quadrants

Ok
Well... The reason I wanted to do them in groups of 4 or 8 was to better leverage the linear actuators. In singles, cant justify tracking... so there will not be tracking. Manual aiming of the panels, 4 times a year.

IT IS GO TIME

I can absorb that labor for now.

* Poles must be mounted in such a way that they can be Up-cycled once replaced
* Poles only need to be round to allow rotation, so - NOT A REQUIREMENT - U-Bolt can rotate on a 4x4 just fine
* Poles probably need to sink 2-3 feet into this sand, it is soft stuff. 2' would be sketch, 3' would hold.

Ok, so what does the top look like?
What do I have on hand?

* I have a box of brackets and such, but I have no problem drilling, fastening, tapping DIRECTLY into the panel frames.
* The panels come with big thick frames, so why not utilize that
* These big-ass panels need MINIMUM 4 points of attachment to hold together, no way will 2 points of attachment hold them

MAYBE if you held them on the middle of the short ends (long ways) you could swing 2 points of attachment... maybe... but better assume you want to grab the panels at about 1/3 and 2/3 along the LONG SIDE of the panels. That is sure to work and will distribute the load out better. 15lbs static on 4 points is better than 30lbs static on 2 points. Dynamic WILL be pretty gnarly, so we are going to have to use #8 or bigger hardware.... or double up on everything.

Steel
I am thinking Steel for this project. Aluminum angle is expensive and does not weld as easy. Steel man... some kind of pull-down scrap that is in lengths too short to be of use for a big project. Steel.....

Sigh
What if we Hang the panels?

* Pole rises up 8 feet, turns over like a street lamp
* Bracket sort of like a Universal Joint hangs down
* Hang the solar panel, from the long side?

Ever see that anywhere? Meh, would flap in the wind... but just a tiny bracket at the bottom would hold it steady

-methods
 
We will work from only what the HomeDeblow offers

Aluminum angle in 1/16 goes floppy pretty quick, especially at only 1"
It is reasonably priced tho

2" Steel angle in 1/8" - $22

Simple is, is simple does. No need for a bunch of holes, I can cut holes where I need them.

Slotted

I need to go handle some samples and see how the 14G deflects. I know the 18G wont cut it, just twists. ... Gonna mock this up quick and see what works and does not work.

REQUIREMENTS
* 1 single pole, made of common stock fencing materials, diameter is Buyer Beware
* 1 panel mounted to that pole via 4 points of attachment
* Ability to rotate and lean, so two axis
* Horizontal or Vertical Mount, but must stick to standard low cost fence posts, considering soil (so horizontal)
* Must be one-man adjustable, so weight has to be caught on a collet and greased
* No loose fasteners!!! To adjust must not mean to re-align with holes, that is too ghetto. Must adjust using U-Bolts continuous.
* PAUSE

We need two hinging mechanisms, and hopefully those will also act as mounts. U-Bolts, I do not believe... can be beat for this.

(Wife woke up, must go be human)
EMERGENCY SIMULATION PAUSED

-methods
 
Back
Top