Help with solar tracker?
- Thread starter Depori
- Start date
By solar tracker, do you mean something that rotates solar panels to face the sun?
You can use light sensors such as LDRs or infrared sensors (the 2 leg ones, not the 3 leg ones) pointing in multiple directions to measure the light levels to determine where the sun is. The 18M2 has enough ADCs for this purpose. Then, a servo or stepper motor can rotate the solar panel stand.
You can use light sensors such as LDRs or infrared sensors (the 2 leg ones, not the 3 leg ones) pointing in multiple directions to measure the light levels to determine where the sun is. The 18M2 has enough ADCs for this purpose. Then, a servo or stepper motor can rotate the solar panel stand.
Last edited:
I have designed several of these and built as many, after much research and testing the assumption i have concluded to is it is best not to use light sensors and just track on the time of day with 2 or 3 programs to suit the changes throughout the year.
Outside of 1 axis tracking the percentage of gain for cost falls away rapidly, a single axis tracker can add 25-30% increase and a second axis might add 2-5% gain there after, when the sums are done its cheaper to add a extra panel than the second axis.
The sun travels a steady path of consistant travel so time is far good enough for tracking, where light sensors can cause hunting with cloud cover and also use more power than just a simple timed movement.
Im sure many will disagree, but i would like to see their logged data to back up any trials and testing they have done to prove me wrong.
Outside of 1 axis tracking the percentage of gain for cost falls away rapidly, a single axis tracker can add 25-30% increase and a second axis might add 2-5% gain there after, when the sums are done its cheaper to add a extra panel than the second axis.
The sun travels a steady path of consistant travel so time is far good enough for tracking, where light sensors can cause hunting with cloud cover and also use more power than just a simple timed movement.
Im sure many will disagree, but i would like to see their logged data to back up any trials and testing they have done to prove me wrong.
techElder
Well-known member
Finally! Some common sense!
I'm with you SAborn. The sun moves (ok, I know it doesn't actually move) very predictably with time. Put a timer control on to adjust from dawn until dusk. Then reset for the next dawn.
Get out and do a manual adjustment for the seasonal changes.
PS. Too many of us get enamored with the capabilities and technicalities of electronics and microprocessors to see through the weeds to a simple solution.
.
I'm with you SAborn. The sun moves (ok, I know it doesn't actually move) very predictably with time. Put a timer control on to adjust from dawn until dusk. Then reset for the next dawn.
Get out and do a manual adjustment for the seasonal changes.
PS. Too many of us get enamored with the capabilities and technicalities of electronics and microprocessors to see through the weeds to a simple solution.
.
I thought the same thing. Where I work they research solar panels and most have just single axis tracking.
One application that might need two axis tracking is for a small solar furnace based on a magnifying lens. That would make a cool science project. I wonder how hot you could get the focal point?
PS: I am off to see the Mythbuster's live show tonight, woot!
One application that might need two axis tracking is for a small solar furnace based on a magnifying lens. That would make a cool science project. I wonder how hot you could get the focal point?
PS: I am off to see the Mythbuster's live show tonight, woot!
Hey all
Thanks for quick answers to my question
But I mean a solar tracker in one direction from east to west,
with a scheme based on a piaxe, preferably one with a 18M2
program for that circuit.
On the Web I've seen a circuit for this purpose but with a arduno.
I can build a circuit. But not design
greeting
Thanks for quick answers to my question
But I mean a solar tracker in one direction from east to west,
with a scheme based on a piaxe, preferably one with a 18M2
program for that circuit.
On the Web I've seen a circuit for this purpose but with a arduno.
I can build a circuit. But not design
greeting
This one? http://www.instructables.com/id/Arduino-Solar-Tracker/step4/The-Code/
This one's code should be convertible to PICAXE code.
Use the design in the link above.
I'll go with the fellow sun belt posters too- that's assuming you mean photo voltaics (PV) rather than solar heated water. It would greatly help if you specify the scale of your setup too - Watts or KW?! Are you after attractive Feed In Tariff (FiT) benefits that may apply in Belgium? I'll assume you mean kW...
With PV prices around US$5 a Watt (& falling thanks to the Chinese...) it's increasingly more cost effective to just add extra panels,which can be firmly & discretely attached to suitable roof tops. In contrast to such "fit & forget" systems, tracking systems (although entrancing) may need ground level mounting, & rugged costly mechanicals that are prone to wind damage/frayed wiring/rust/insects & animals (& wayward humans!) etc.
A key daily issue can relate to the local solar resource anyway. Morning cloud/mist/tree & building shading/terrain etc may mean a mechanical tracker blindly follows virtually a solar ghost, while afternoon sunbeams (which extra panels could better exploit) may go begging. Air clarity,panel temperature & UV levels may be significant too. During heavy cloud,rain or snow your tracker will keep working of course & may wastefully take more energy to do so than you collect...
Seasonally it's worth stressing that low angle winter sunshine is usually far more valuable for off grid systems than in high season. Summer electrical demand may be lower (as fewer lights etc) in many regions too. I've an off grid mate here in NZ (latitude 41 South) who has his batteries "virtually boiling" in summer, yet in winter the then poorly slanted PV panels are gasping to even trickle charge. His dwelling's roof pitch (& his age!) is a tad daunting & doesn't suit seasonal adjustment. With rapidly falling PV costs it's however become tempting to just add a further bank of suitably winter slanted PVs onto a steep pitch shed roof. Stan.
EXTRA: Check this highly informative FiT slanted Australian site,which indicates single/dual axis tracking may increase output by ~20-30% for kW level systems. This however may NOT be cost effective (trackers may cost as much as the PVs!),& the 2010 calculations are now hazy due to lower PV prices & reduced FiT payments too. In contrast, mechanical tracking supply/installation & maintenance costs have probably increased.
The author sums things up by saying- For small 1.5kW installations the upfront cost of a tracker is high, and the payback period is long. In reality, investing the same amount on additional solar cells would yield higher generation volumes, more income, as well as greater environmental benefits..
With PV prices around US$5 a Watt (& falling thanks to the Chinese...) it's increasingly more cost effective to just add extra panels,which can be firmly & discretely attached to suitable roof tops. In contrast to such "fit & forget" systems, tracking systems (although entrancing) may need ground level mounting, & rugged costly mechanicals that are prone to wind damage/frayed wiring/rust/insects & animals (& wayward humans!) etc.
A key daily issue can relate to the local solar resource anyway. Morning cloud/mist/tree & building shading/terrain etc may mean a mechanical tracker blindly follows virtually a solar ghost, while afternoon sunbeams (which extra panels could better exploit) may go begging. Air clarity,panel temperature & UV levels may be significant too. During heavy cloud,rain or snow your tracker will keep working of course & may wastefully take more energy to do so than you collect...
Seasonally it's worth stressing that low angle winter sunshine is usually far more valuable for off grid systems than in high season. Summer electrical demand may be lower (as fewer lights etc) in many regions too. I've an off grid mate here in NZ (latitude 41 South) who has his batteries "virtually boiling" in summer, yet in winter the then poorly slanted PV panels are gasping to even trickle charge. His dwelling's roof pitch (& his age!) is a tad daunting & doesn't suit seasonal adjustment. With rapidly falling PV costs it's however become tempting to just add a further bank of suitably winter slanted PVs onto a steep pitch shed roof. Stan.
EXTRA: Check this highly informative FiT slanted Australian site,which indicates single/dual axis tracking may increase output by ~20-30% for kW level systems. This however may NOT be cost effective (trackers may cost as much as the PVs!),& the 2010 calculations are now hazy due to lower PV prices & reduced FiT payments too. In contrast, mechanical tracking supply/installation & maintenance costs have probably increased.
The author sums things up by saying- For small 1.5kW installations the upfront cost of a tracker is high, and the payback period is long. In reality, investing the same amount on additional solar cells would yield higher generation volumes, more income, as well as greater environmental benefits..
Last edited:
Hiya Depori,
When I was looking at the lego mindstorms toy some time back...
I came across a sun tracking device and although there was no source code the device was manufactured as follows....
It used TWO light sensors aligned next to each other (like the fingers on your hand)
And between the two light sensors was a piece of CARD...
And basically the key to the whole equation was the Card!
Because it cuts down the SUNLIGHT that arrives at the furthest Sensor and takinh light readings fromeach sensor then points you in the direction the sun is coming from.
I've attached a crude picture and hope it helps explain better.
DocZaf
When I was looking at the lego mindstorms toy some time back...
I came across a sun tracking device and although there was no source code the device was manufactured as follows....
It used TWO light sensors aligned next to each other (like the fingers on your hand)
And between the two light sensors was a piece of CARD...
And basically the key to the whole equation was the Card!
Because it cuts down the SUNLIGHT that arrives at the furthest Sensor and takinh light readings fromeach sensor then points you in the direction the sun is coming from.
I've attached a crude picture and hope it helps explain better.
DocZaf
premelec
Senior Member
some considerations
It's great how much speculation can occur from an inquiry - we don't know the size of the thing or it's shape or if it's thermal or electric output etc - i just read an article in Home Power magazine on a 14 foot diameter parabolic hot water unit - needed to be tracked very accurately [and in my opinion was not a needed device as very hot water not required - but it looks cool...] . As far as tracking with a clockwork and then resetting you are simply wasting power on rainy and cloudy [or volcanic ash] days so it's good to have some intelligence as to there being power there to harvest. In any case hurray for more sun power use!
BTW one of the "cutest" tracking setups I saw over 50 years ago - two rows of PVs with a shade between them vertical - and that _facing_ a cylindroparabolic mirror which faced the sun - they just rotated the assembly till equal current from each PV array was seen. It had mechanical balance putting the pivot pint where the mirror and PVs had equal torque [leverage?].
It's great how much speculation can occur from an inquiry - we don't know the size of the thing or it's shape or if it's thermal or electric output etc - i just read an article in Home Power magazine on a 14 foot diameter parabolic hot water unit - needed to be tracked very accurately [and in my opinion was not a needed device as very hot water not required - but it looks cool...] . As far as tracking with a clockwork and then resetting you are simply wasting power on rainy and cloudy [or volcanic ash] days so it's good to have some intelligence as to there being power there to harvest. In any case hurray for more sun power use!
BTW one of the "cutest" tracking setups I saw over 50 years ago - two rows of PVs with a shade between them vertical - and that _facing_ a cylindroparabolic mirror which faced the sun - they just rotated the assembly till equal current from each PV array was seen. It had mechanical balance putting the pivot pint where the mirror and PVs had equal torque [leverage?].
Last edited:
Why preferably an 18M2 ?
Given the cost of solar panels and the mechanics of a tracking system, the cost of a particular PICAXE seems insignificant.
This is the last controller i built for someone to run their solar tracker, and the OP looks to be asking for code more than the circuit, but as my system uses displays and keypad systems that are not an off the shelf item, there is no point in me posting code for this system.
The basics of it is a DS1307 RTC and a picaxe driving 2 relays for motor control, the system is user programable via the keypad to how often it advances and for what run time period, but on an average it advances about every 15 minutes to a preset maximum arc, at the set start up morning time it reverses back to the start limit switch and the process starts over for the day.
The whole assembly is mounted on a modified car diff, set in a vertical position, this gives a very strong assembly.
Additional gearboxes are used on the input to the diff to add about a 400:1 ratio, the motor used to drive the gear reduction is a small DC motor that has about a 20rpm output, the overall power consumption is next to nothing.
The basics of it is a DS1307 RTC and a picaxe driving 2 relays for motor control, the system is user programable via the keypad to how often it advances and for what run time period, but on an average it advances about every 15 minutes to a preset maximum arc, at the set start up morning time it reverses back to the start limit switch and the process starts over for the day.
The whole assembly is mounted on a modified car diff, set in a vertical position, this gives a very strong assembly.
Additional gearboxes are used on the input to the diff to add about a 400:1 ratio, the motor used to drive the gear reduction is a small DC motor that has about a 20rpm output, the overall power consumption is next to nothing.
I did not build the mechanical part for this tracker the owner did, although the idea was copied off a earlier one i had made.
The reason for using this method is for a smaller foot print on the ground, as rows have shading problems and need large spacings between rows.
The motor in the photos has been changed now to a small geared motor like you would buy at a hobby shop.
I had avoided linking to this as its a long thread of chatter and on a different forum.
http://ozelecforum.com/viewtopic.php?f=33&t=108&start=40
The reason for using this method is for a smaller foot print on the ground, as rows have shading problems and need large spacings between rows.
The motor in the photos has been changed now to a small geared motor like you would buy at a hobby shop.
I had avoided linking to this as its a long thread of chatter and on a different forum.
http://ozelecforum.com/viewtopic.php?f=33&t=108&start=40
As i had said before, its all works off a Real Time Clock (RTC) DS1307 and is all operated on time, the time is set by the user for when to start the day, how often to move the panels, how long to run the motor for (or how far this moves the panels per run time) and the end of the day time.
Its all very simple stuff with using a RTC and a picaxe, the only problem area i had was coding for the display i used and the keypad system used, if i had used a LCD it would have been simple, and if you dont mind making the time adjustments in code and reloading the chip the keypad is not needed, and keeps the program much simpler, as with the keypad you need to keep scanning the keypad for key presses and have code to respond to the key pressed and then store all functions into eeprom so the program is not lost on a power failure.
I had all the bits and had worked with them previous so it was not so bad but did make the program long and messy.
Also stated earlier the keypad and display used are not avalible and are based around a pic (16F88) driver programmed to function as a I2C device, so the code is of no use to you or anyone and not listed as without the purpose built items you cannot reproduce the circuit without major redesign and would be easier starting from scratch.
(also why i did not post the link earlier as seen no point without all the data.)
My original point was its easy to do tracking just based on using time and get very good results, and no need to use light sensors and other methods, that make programming harder and can cause other problem, when all is said and done the sun travels an arc over the length of the day and regardless of cloud, rain, snow, or if hell freezes over, to track the sun you need to also travel an arc across the course of the day.
Its all very simple stuff with using a RTC and a picaxe, the only problem area i had was coding for the display i used and the keypad system used, if i had used a LCD it would have been simple, and if you dont mind making the time adjustments in code and reloading the chip the keypad is not needed, and keeps the program much simpler, as with the keypad you need to keep scanning the keypad for key presses and have code to respond to the key pressed and then store all functions into eeprom so the program is not lost on a power failure.
I had all the bits and had worked with them previous so it was not so bad but did make the program long and messy.
Also stated earlier the keypad and display used are not avalible and are based around a pic (16F88) driver programmed to function as a I2C device, so the code is of no use to you or anyone and not listed as without the purpose built items you cannot reproduce the circuit without major redesign and would be easier starting from scratch.
(also why i did not post the link earlier as seen no point without all the data.)
My original point was its easy to do tracking just based on using time and get very good results, and no need to use light sensors and other methods, that make programming harder and can cause other problem, when all is said and done the sun travels an arc over the length of the day and regardless of cloud, rain, snow, or if hell freezes over, to track the sun you need to also travel an arc across the course of the day.
premelec
Senior Member
Hi SAborn - thanks for the pictures - I knew cars were good for something - now I'm wondering if golf carts have differentials that are smaller - anyhow I'm all for recycling and have considered bicycle parts in the past for small units [I quite driving 7+ years ago and bicycle exclusively so I'm well aquainted with with their mechanics - thinking a bunch of front forks with one panel per and chain and sprockets to couple to common motor drive ...] - We have big winds here on the edge of the Rocky Mountains so wind loading is a consideration.
Problem #1 is the large foot print it will take up due the spacing needed between panels to prevent shading.
Problem #2 See problem #1... you would need a mile of chain to link it all together.
Problem #3 Do the calculations on the wind loading over the square inches of panel area and i doubt your fork assemblys will take the sideways thrust (they are not designed for high loading in that direction)
Problem #4 Best not to use your own car for parts anyway, better to use someone elses car.
premelec
Senior Member
@SAborn - I was thinking pretty small arrays and panels were 50 watt... things have shifted since then I've got old VW so swinging half axles could add another degree of motion - floppy to feather in the wind... with old coil springs... with PVs so cheap it's rare anyone is tracking around here though we have many megawatts installed in the area. Expecting possible 80MPH winds tonight so I hope they're all tied down... Miles of old bike chain available... BTW I built a drive for a Mienel Cusp water heater using bike cable to open and close insulated reflector leaves around the tank - about 36 years ago - a clever but not wind resistant design... Some of these designs are more art than efficiency and I think that's good as it educates people inadvertently as they look at it artwise - dark stuff on roofs is not very exciting though it puts out the kilowatts...
I've got old VW so swinging half axles could add another degree of motion - floppy to feather in the wind... with old coil springs... with PVs so cheap it's rare anyone is tracking around here though we have many megawatts installed in the area. Expecting possible 80MPH winds tonight so I hope they're all tied down... Miles of old bike chain available... BTW I built a drive for a Mienel Cusp water heater using bike cable to open and close insulated reflector leaves around the tank - about 36 years ago - a clever but not wind resistant design... Some of these designs are more art than efficiency and I think that's good as it educates people inadvertently as they look at it artwise - dark stuff on roofs is not very exciting though it puts out the kilowatts...While this mainstream school-fed logic may apply to most - I say "Balderdash".
If you are going to think like that, why even use solar electrical at all? The grid is cheaper. On that note, why eat, intravenous is cheaper and less effort.
My first solar tracker electronics were made solely from a discarded UPS. In those you will find voltage comparators, DPDT relays, and MOSFETS from an H-bridge.
Then I got my hands of a couple of discarded mountain bikes (preferably older steel framed jobs)
The last item was a drive motor from walking treadmill.
With the right gear ratio, and a PWM to slow the motor... just about any prime mover would work.
My total cost was less than 20 bucks.
If you into it, and want it, make it happen - don't let silly old-school economist thinking dampen your spirit -- cause let's face it, what do economists actually know about economical living? Nothing.
Fair comment, but perhaps get back to us after a good storm or ice up. Since my earlier comments PV prices have further fallen, & panel costs below US$2 a Watt are now common down here in the sunny South Seas. Reliable & robust mechanical systems,especially those with such boring items as galvanised & s/steel fittings,predictably remain at much the same "OMG-how much!" prices...
Last edited:
Gee Stan you need to shop around, I looked at 250w panels a few days ago for $0.50 a watt (Chinese made) and settled on some high quality panels for $0.80 per watt (German made).
I do agree with tracking systems if one has the yard space, but disagree with push bike parts being sturdy enough, then most trackers i have built take 1-2kw of panels and in no way would i want them to end belly up in a storm.
Although Stan is correct with todays cost of panels its hardly worth building a tracker for a 20-25% return, other than a little workshop therapy and the ability to use a Picaxe in the system.
Nice find- what is your source? Is this nominal rating justified however?
FWIW -aside from MPPT issues, my rule of thumb is to take ~2/3 of advertised power levels as more like the true bright sun output. I typically find most mono PVs actually better their stated current output here in bright sky & UV rich NZ conditions, but once loaded the delivered open circuit voltage drops seriously. Hence claims of (say) 19V at 10A (thus nominally 195 Watts) are usually more like 13.8V @ 10A (thus 138W) in working conditions.
premelec
Senior Member
@manuka - that's what the MPPT is for - kind of DC impedance matching for varying load & temperature etc... probably a better value for the money than tracking... lots of microinverters seem to be appearing [one per panel and grid tieable]. A recent comment from the editors of HomePower magazine indicated up to 40% overall gain in far latitudes - however wind loading and complexity still work against it for medium size installations. Whatever works!
Last edited:
I have been involved with testing of some Micro Inverters, and would argue the opposite of a 40% gain.
The testing even involved a picaxe that read several I2C sensors, there was a 16 bit ADC per cell monitoring the DC voltage and a 2 x I2C temp sensors per cell monitoring panel temp and inverter temp, and a single I2C temp sensor monitoring ambient air temp.
All the inverters AC wattage was also recorded, everything logged to a PC every 10 seconds.
In full sun on average the micro inverters worked reasonably well and on an equal to string inverters also being monitored as a comparison.
Now where the big difference showed, was in less than ideal sun conditions, as when the panel voltage dipped due to cloud cover of shading etc, the micro inverters backed off quickly compared to the string inverters.
Then with a little worst conditions of grey skies (nice English day conditions) and high cloud cover the micro inverters would often shutdown due to low DC panel voltage, (voltage below operating voltage) then when the sun increased the micro inverters would wake up again, go through the connection procedure including the anti island delay of 1 minute, (all up around 2 mins) before connecting to the grid again, only to then shut down due to lower light conditions again, repeating to the point they would spend more time off line than connected to the grid.
Where the string inverters would drop back in the same conditions to a trickle of power to the grid, and instantly respond to increased light levels, they also maintained connection with the grid for several minutes after the DC voltage fell below operational voltage, The string inverters remained grid connected down to around 30-40v with an operating voltage of about 150+v.
The other factor that showed was to mount an inverter directly under the panels is not the best location on hot still days, the increased heat from the panel to inverter with still or low air movement, decreased output efficiency, compared to string inverters located in ambient air temps.
Advantages are any sparky ( mains licensed electrician) can install the panels and no requirement for a solar installer accreditation, and the system can be modular and panels added as money permits without the need to purchase multiple panels and a suitable string inverter all in a single designed package.
My conclusion ....... i would go the string inverter with series panels any day over a micro inverter system.
Premelec: I'm naturally up with MPPT features- my point was that actual bright sun field testing usually shows that no combination of voltage & current from the panel under solar,charge or load variation would deliver the stated power output! Marketing hype means it's the easiest thing "under the sun" to just slap a big number on the PV, which is often much too optimistic. Failing field testing a quick measure of usable PV surface area can be helpful too. At 1000-1200 W/sq.m & 10-15% efficiency then a 1 sq.m monoX PV will probably be good for just 100-150 Watts peak power output.
SAborn: Very interesting - thanks for the heads up on this. But let's have details of those 50 - 80 cent a Watt 250W panels please! Grid parity is easily feasible at these attractive prices,as is wired rather than piped water heating -classic solar water heating outfits may be in for a real fight! (I'm reminded of this prediction in fact) Stan.
SAborn: Very interesting - thanks for the heads up on this. But let's have details of those 50 - 80 cent a Watt 250W panels please! Grid parity is easily feasible at these attractive prices,as is wired rather than piped water heating -classic solar water heating outfits may be in for a real fight! (I'm reminded of this prediction in fact) Stan.
Last edited:
What can i tell you other than it was trade pricing given to me from someone i know in the solar bussiness. (goes to show the price mark up some add)
One link that popped up on a different forum recently is this one
http://www.ebay.com.au/itm/Key-Solar-Poly-120W-Solar-Panel-Home-Power-Generator-/271155635129?pt=AU_Solar&hash=item3f2222a3b9#ht_3035wt_1160
Although only 120w panels but still around $0.70 per watt.
The person who purchased thes panels commented they were not supplied with flying leads and MC4 connectors as most panels of today are fitted with, or the bonus leads as listed in the sale.
Great -but I'd be very wary of that bargain 120 Watter's claims & durability! Although more costly,their larger 200W polyX offering has the sort of product details & features that reassure. Of course a mere 12V at higher surrents is an issue, as further costly copper cabling and heavy duty connectors will be needed.
I'm actually in the market for a folding PV system ( c/w leads etc) in the 100-150W range for field work, so am interested in the likes of this. Anyone from Oz. over this way care to bring one along!? Stan.
I'm actually in the market for a folding PV system ( c/w leads etc) in the 100-150W range for field work, so am interested in the likes of this. Anyone from Oz. over this way care to bring one along!? Stan.
techElder
Well-known member
Gentlemen, I have been unable to find a reference that someone made to a PICAXE project to monitor and charge control a battery within a battery group with a solar recharger. I believe these batteries were destined for an automotive application. There was a networked (I2C?) set of PICAXE boards with overcharge loads on them to keep from overcharging the batteries. Each battery had a board, but all the boards were connected to this network.
The link I've lost goes to this person's web site about the project, but I have been unable to filter through all the search references and find it.
Anyone know what I'm trying to remember?
The link I've lost goes to this person's web site about the project, but I have been unable to filter through all the search references and find it.
Anyone know what I'm trying to remember?
The PV panels I've tested here in Colorado, USA, have all met (exceeded) manufacturer's specs for full sunlight days. But we are above 6,000 feet elevation and have very clear skies compared to places like NZ or GB. When I set panels up at 9,000 elevation in winter, things really start cranking. Low temperature, low humidity and high altitude work wonders.
The answer to the question of whether or not to use a tracker is that it depends on your goals and finances.
If you want to learn more about robotic mechanisms, have a lot of materials lying about waiting for a project to utilize them in, and you're short on money and long on time, a home-built tracker is the way to go.
If you are a wealthy retiree with a 40 foot motor-home who wants as much PV power as he can get, a commercial PV tracker is right for you. If you are asked to develop a cost-effective PV installation for powering a sewage treatment plant, (like Boulder, CO has,) then an array of fixed panels is right for you.
I'd personally opt for using some ingenuity, (and using up some of the electro-mechanical junk I have in storage) to build a home-brew tracker just so I could sit back and watch it work. More satisfaction, and more power.
The answer to the question of whether or not to use a tracker is that it depends on your goals and finances.
If you want to learn more about robotic mechanisms, have a lot of materials lying about waiting for a project to utilize them in, and you're short on money and long on time, a home-built tracker is the way to go.
If you are a wealthy retiree with a 40 foot motor-home who wants as much PV power as he can get, a commercial PV tracker is right for you. If you are asked to develop a cost-effective PV installation for powering a sewage treatment plant, (like Boulder, CO has,) then an array of fixed panels is right for you.
I'd personally opt for using some ingenuity, (and using up some of the electro-mechanical junk I have in storage) to build a home-brew tracker just so I could sit back and watch it work. More satisfaction, and more power.
Last edited:
premelec
Senior Member
Micro-inverter performance
@SAborn - aside from installation and heat issues was it your finding the micro-inverters were just deficient in efficiency over usual power solar insolation conditions? [there are also cases where partial shading issues come in play - forget those as well].
Did you try more than one brand? Could you say what brand you tested? Thanks
PS The 40% figure I mentioned seeing about higher number latitudes was for tracking vs no tracking - nothing to do with micro-inverters
@SAborn - aside from installation and heat issues was it your finding the micro-inverters were just deficient in efficiency over usual power solar insolation conditions? [there are also cases where partial shading issues come in play - forget those as well].
Did you try more than one brand? Could you say what brand
PS The 40% figure I mentioned seeing about higher number latitudes was for tracking vs no tracking - nothing to do with micro-inverters
Last edited:
The tests were done using a small single solar cell as a light reference meter to monitor light intensity to give a base line for all other panels output.
The tests were done for a solar panel manufacture using micro inverters with their panels, several micro inverters were tested as supplied from the panel manufacture, i dont wish to quote what inverter brands was tested as this was done on behalf of the manufacture.
I can say the Chinese made micro inverters worked best from the several tested, the German made product was the worst and the USA brand was a tiny bit better than the German brand, others were in the middle.
The big problem with all is the very narrow operating voltage window from a single panel for the micro inverter, as i quoted previous under good light conditions they all worked well, but as we all know we dont live in an ideal world with perfect conditions everyday of clear skies and bright sunlight, its the not so perfect days the product marketers dont tell you about, or forget to show in their product test results.
For my location its about a 50/50% of good solar days to not so great solar days for a 12 month window, so to have the panels off line for 50% or more for the 1/2 year of non perfect days is not a option i would choose.
Jeremy Harris
Senior Member
I'm pretty sure this is Peter Perkins BMS project, linked to above (his username here is retepsnikrep). It used a small 08M board on each cell, networked via a daisy chain serial link to a master control board and display.
techElder
Well-known member
My apologies to eclectic for not delving deep enough into the Peter Perkins site when he proffered it and many thanks to Jeremy for reinforcing the validity of that posted link.
What I hadn't recognized was that Perkins has "sold" his SolarVan site, so a lot of the links no longer work. The key link to the PICAXE hardware is here (http://www.batteryvehiclesociety.org.uk/forums/viewtopic.php?t=1245). A link to his original SolarVan site is at the Wayback Machine archive.
Thanks so much for allowing me quit kicking myself for not saving these sites a long time ago!
What I hadn't recognized was that Perkins has "sold" his SolarVan site, so a lot of the links no longer work. The key link to the PICAXE hardware is here (http://www.batteryvehiclesociety.org.uk/forums/viewtopic.php?t=1245). A link to his original SolarVan site is at the Wayback Machine archive.
Thanks so much for allowing me quit kicking myself for not saving these sites a long time ago!