Solar Install; the on-going saga

Well if your batteries are in good nick they'll only self-discharge at 5%-20% per month which is easy to monitor. ie. cycle and/or charge them once every 2-8 months (and check them more frequently in the Winter).
Most controllers have different algorithms for different battery chemistries (and usually they're set too low at that), but I think the TS was the cheapest programmable one I could find. It'd be difficult to find one with a charge completion voltage around a float set-point as it would be completely useless day to day. Perhaps if you were to partially shade the module with a blanket you could achieve a low tech solution.

You need to turn off the charge controller when changing the charge profile by the way. They don't like hot switching.

SmartGuage

There is a lot of hearsay in the industry that overcharge is bad for batteries...which it is, and most err on the side of caution to under-rate chargers. However undercharge is worse so if you plan on getting it wrong buy some de-ionised water and go too high.

SmartGuage wrote:

In general, batteries don't die of old age. Instead, they are murdered by being continually undercharged. i.e. not charged to 100%. Continually undercharging batteries is far more detrimental to their health than continually overcharging them.

Overcharging in this context refers to holding them at the acceptance voltage for longer than usual - it does not refer to charging at too high a voltage. Undercharging them means not reaching 100% charge state which could be the result of either too low an acceptance voltage or too short an acceptance time.

Obviously charging them exactly to 100% is preferable (with the occasional controlled overcharge in the case of wet cells) but this is rarely possible. If this is the case, err on the side of overcharging them rather than undercharging them. Continually overcharging by a certain amount might shorten a battery's life by say 20%. Continually undercharging the same battery by the same amount might shorten it by 90%.

tehehe...lead acid chemistry get's pretty complicated when you get down to it.

If anyone's charging batteries outside-ish on a non-temp compensating charger, I'm seeing 15.1v on a 14.4v set-point lately so it'd be pretty safe to charge sealed batteries on a flooded set-point, wouldn't worry about electrolyte loss, the gassing voltage point rises in the cold too.
I'm not suggesting trying that on AGM's or gel's though they can be finicky about these things and I never bothered to learn their requirements.

Sir Liamalot

Sir Liamalot wrote: »

Charging a battery at less than 10% DOD...is a very bad habit indeed

Just dawned on me; the same could be said for wiring the leisure battery relay from the D+ .
Depends on how you wire it I suppose, under-rated cable and a running 12v absorption fridge could offset this one.

moodrater

I think we should take a collection an buy liam a new battery

+ =

Winter's over folks!


35Ah harvested today in the Wicklow mountains, 'til the clouds came down about 16hr.
Module tilted 50° (ish), and "tracking" (rolling the van downhill south for midday and south, south west after.)

Lord William T. Kelvin

I may have found a solution to the final puzzle piece. Autonomy in the Winter/night-time. Thanks to Aidan_M_M for explaining to me how a gas valve solenoid is powered from a thermocouple, I've been plotting the next electrical endeavor (this may never happen but I still think it's worth a mention).
I've been looking at the Peltier generator with a wry eye on my wood turf burning stove.

Found this online;

http://www.tegmart.com/datasheets/DW-SM-45W.pdf

Which is a ludicrous price tag for a plate mass, thermal paste, a few TEG's, heat sink, two fans, a boost regulator and a few sockets.
One could DIY this simple device for easily half that, or perhaps for a fixed installation (risk of de-soldering the TEG), boost reg. to TS and sockets are already hardwired downstream. All I'd need is the TEG's and a temp. management system (self powered fan with heat sink).

Semi-similar to this except larger hot plate, and more TEGs with HT wires in conduit running to the reg & TS.

http://www.stovetopfan.co.uk/#/portfolio/copper-fan-c-small/


...I'm off to find a stove-top thermometer. :pac:


Few amendments while I'm at it;

The Tri-Star soft starts just fine, I never noticed because I was driving, watching it in mains mode it's on the ball. I have mentioned elsewhere if your charger finishes too quickly then it's probably not doing it's job adequately. The TS spent over 40 hours charging after 3 weeks away from a hook-up during the Winter and the batteries were worked well hard in the cold for the duration, about 30 hours of that was driving home the last 400mA.
Fridge fan resistor is actually 1.8Ω (not 180Ω nor 100kΩ, which is scraping a pass but probably needs a companion when I switch to OLA) it's semi-self-regulating in that when temp comp. is high volts, low temp the fridge is generally cool and vice versa high temp + hot fridge = low volts.
13.8v Fan max tolerance
14.7v theoretical max temp. compensation with active fans on 14.4v set-point
14.7v - 13.8v = 0.9v required drop
V=IR (where continuous current is 500mA)
V/I = R
0.9/.5 = 1.8Ω (cable losses not included)
Original calculation was for rated (inrush) 720mA draw so that's why I'm so close to the margins. Working fine now I heard them engage twice about two weeks ago
The TM is a fantastic Ah counter & accurate split bank voltmeter, in this installation (grounding of auto electrics bypassing shunt make large inaccuracies) and given Peukert (1:1 setting on meter) it's a rubbish battery % monitor after day 4 without a full charge. It's saying 44% now and I understand it to be ~90%. It also resets to 100% at about 97% or lower depending on how the generators are behaving, I think the later discrepancy can be overcome with some tinkering in the advanced settings however. I believe it's set this way to make people feel better about their charging system when the difference is negligible and people get upset when they never see 100% (which realistically only ever happens a handful of times somewhere between charge cycles 50-100)

Lord William T. Kelvin wrote:

"If you can not measure it, you can not improve it.".

Is it worth the investment? 1 kWh (70Ah) at the alternator = ~€1 (diesel) + maintenance.

Typical engine efficiency of 40%, a belt efficiency of 98% and an alternator efficiency of 55%, this leads to an overall energy conversion efficiency of only 21%.

Happy to entertain ideas on a compact Stirling engine too

moodrater

Sir Liamalot wrote: »

I may have found a solution to the final puzzle piece. Autonomy in the Winter/night-time. Thanks to Aidan_M_M for explaining to me how a gas valve solenoid is powered from a thermocouple, I've been plotting the next electrical endeavor (this may never happen but I still think it's worth a mention).
I've been looking at the Peltier generator with a wry eye on my wood turf burning stove.

Found this online;

http://www.tegmart.com/datasheets/DW-SM-45W.pdf

Which is a ludicrous price tag for a plate mass, thermal paste, a few TEG's, heat sink, two fans, a boost regulator and a few sockets.
One could DIY this simple device for easily half that, or perhaps for a fixed installation (risk of de-soldering the TEG), boost reg. to TS and sockets are already hardwired downstream. All I'd need is the TEG's and a temp. management system (self powered fan with heat sink).

Semi-similar to this except larger hot plate, and more TEGs with HT wires in conduit running to the reg & TS.

http://www.stovetopfan.co.uk/#/portfolio/copper-fan-c-small/


...I'm off to find a stove-top thermometer. :pac:


Few amendments while I'm at it;

The Tri-Star soft starts just fine, I never noticed because I was driving, watching it in mains mode it's on the ball. I have mentioned elsewhere if your charger finishes too quickly then it's probably not doing it's job adequately. The TS spent over 40 hours charging after 3 weeks away from a hook-up during the Winter and the batteries were worked well hard in the cold for the duration, about 30 hours of that was driving home the last 400mA.
Fridge fan resistor is actually 1.8Ω (not 180Ω nor 100kΩ, which is scraping a pass but probably needs a companion when I switch to OLA) it's semi-self-regulating in that when temp comp. is high volts, low temp the fridge is generally cool and vice versa high temp + hot fridge = low volts.
13.8v Fan max tolerance
14.7v theoretical max temp. compensation with active fans on 14.4v set-point
14.7v - 13.8v = 0.9v required drop
V=IR (where continuous current is 500mA)
V/I = R
0.9/.5 = 1.8Ω (cable losses not included)
Original calculation was for rated (inrush) 720mA draw so that's why I'm so close to the margins. Working fine now I heard them engage twice about two weeks ago
The TM is a fantastic Ah counter & accurate split bank voltmeter, in this installation (grounding of auto electrics bypassing shunt make large inaccuracies) and given Peukert (1:1 setting on meter) it's a rubbish battery % monitor after day 4 without a full charge. It's saying 44% now and I understand it to be ~90%. It also resets to 100% at about 97% or lower depending on how the generators are behaving, I think the later discrepancy can be overcome with some tinkering in the advanced settings however. I believe it's set this way to make people feel better about their charging system when the difference is negligible and people get upset when they never see 100% (which realistically only ever happens a handful of times somewhere between charge cycles 50-100)




Is it worth the investment? 1 kWh (70Ah) at the alternator = ~€1 (diesel) + maintenance.

Typical engine efficiency of 40%, a belt efficiency of 98% and an alternator efficiency of 55%, this leads to an overall energy conversion efficiency of only 21%.

Happy to entertain ideas on a compact Stirling engine too

"The output above is under 840 °F (449 °C) heat source surface temperature"

If you get your stove top up to 449C you'll at best be sleeping outside at worst you'll kill yourself. You can be sure they fed those fans with cold air too to skew the figures upward.

TEG efficiency is at the very most 8% probably more like 5% so to get 45W out the unit would be drawing 500w of heat from the stove top which will drop the stove top temperature and hence the output.

We heat 120m2 with 4kw stove and it doesn't get over 180! Which according to their graph would generate a whopping 0 watts. If I was to draw another 500w from the stove top the temperature would be even lower.

Real TEG modules are very expensive most of the ones on ebay are tec modules stamped teg by our enterprising chinese friends even the genuine tecgmodules on ebay are only good to about 200-230C.

Wasn't thinking of buying a pre-built unit for a second. Just threw it out there as a "we have the technology" example. Wanna see a real rip off?

These were more what I had in mind if I can find a less expensive source because ideally I'll be running them in parallel-series and they add up to a pretty penny quite fast.

My burner is a wee 2.5kw which I'm dubious will get up to temp but worth a punt on a thermometer at least.
So let's conservatively say 800w @ 3% efficiency for 8 hours = 192W or 14Ah, factor in Puekert reduced discharge curve and that's near enough the lighting on demand.

Pretty expensive electrons alright, even before the hidden losses. But at the end of the day a quick calc. (real-life solar figures) put's it only about €1.50 per watt more than the solar rig which for portability is worth it. I know it's a much lower output but there's been hot embers in my stove pretty much since October. I reckon done right it could beat a modest wind turbine for the sake of consistent operation and discretion.

I'd imagine the heat I'm drawing off the stove-top doesn't leave the space, if anything the convection may distribute it more evenly and at a lower operating temp.

oh no! my fancy calibration labels are melting! :pac:

Voltage figures not accurate. Guesstimate of the above module spec with a ~30° temp. offset to the cold side. Any ideas the peak cooling differential of a convection assisted heat sink?
The wood burner is elevated so colder air off the ground isn't a problem.

Haven't committed yet. I'm still sourcing and comparing the benefits against the cost effectiveness of larger batteries, albeit with a bias towards an extra generator and an end to a stand-alone regenerative electrical system rather than a depletive one. Genuine TEG's with datasheets are well hard source...almost like nobody thinks they're viable :rolleyes:.
Anyone got any recommendations?

moodrater

Sir Liamalot wrote: »

oh no! my fancy calibration labels are melting! :pac:

Voltage figures not accurate. Guesstimate of the above module spec with a ~30° temp. offset to the cold side. Any ideas the peak cooling differential of a convection assisted heat sink?
The wood burner is elevated so colder air off the ground isn't a problem.

Haven't committed yet. I'm still sourcing and comparing the benefits against the cost effectiveness of larger batteries, albeit with a bias towards an extra generator and an end to a stand-alone regenerative electrical system rather than a depletive one. Genuine TEG's with datasheets are well hard source...almost like nobody thinks they're viable :rolleyes:.
Anyone got any recommendations?

If you're pulling a few hundred watts of heat it will be much colder than that. I used tecs for my stove fan but stood them off the stove with an inverted aluminium heatsink to reduce the temp to a safe range. CPU heatsinks will have a thermal resistance spec thats good to about 150C where the heatpipes stop working.

Page 6 here shows how to calculate the net delta T from the gross delta T adn teh thermal resistance of the teg and thermal resistance of the heatsink. As far as I can remember reduce the amps drawn and the TEG thermal resistance and delta t falls. So its not a simple thing.

http://www.micropelt.com/downloads/thermo_generator_package.pdf

[edit] by the way I gave up on the stove fan and put a brushless fan at ground level to blow cold air and accelerate convection its far far more effective at distributing the heat than trying to blow hot air around at waist heigh the inlaws made a clothes airer with 12v fans that does the same thing and dries the clothes at the same time[/edit]

Meh, less heat = more fuel. :rolleyes:
I was hitting 200°C boiling 2 litres of water.

Yeah I realised that, was hoping for an easy answer.
Down about 150°C now once the space is up to temp...let's sweep that under the carpet as a Springtime misnomer
Plough on, I know this is a probably a bad idea, much harder melt batteries, determined to pursue it now until I can quantify how bad

moodrater

Sir Liamalot wrote: »

Meh, less heat = more fuel. :rolleyes:
I was hitting 200°C boiling 2 litres of water.

Yeah I realised that, was hoping for an easy answer.
Down about 150°C now once the space is up to temp...let's sweep that under the carpet as a Springtime misnomer
Plough on, I know this is a probably a bad idea, much harder melt batteries, determined to pursue it now until I can quantify how bad

A small turbine with wavin blades should knock out 40-100watts a lot of the time. The magenets for this one would be only €30 or so from china: http://www.bright-planet.org.uk/

Knockin' wind turbines together is way easier than thermodynamics.
I'm gathering gubbins for a 3kW fixed installation. Not so keen on the little ones. Economy of scale and all that.

I don't think they're very suitable for me in terms of camper use though.
Problems are;

• Not stealth (honestly yes! A smoking chimney is more stealth, "At least it's inside" they'll say!)
• Inconsistent.
• Doesn't work low down or in cities; too much turbulence. (Edit: except perhaps a VAWT might)
• Won't compliment solar as well as TEG's (in theory).
• Noise.
• Marauding s***bags late at night.

Alternator is winning that round I'm afraid.

I have thought about it though and may come back to it, reckon the 12v fridge element would make a perfect dump load although I'm not entirely sure how the food would feel about running on gas and sporadic electric.

The themoelectric generator died on the drawing board.

Optimised for 300w heat tranfer @ 180°C figures in at €43.50 per watt and 35dB fan Noise.

Expecting a nominal generation of 10watts after self-powered fan and DC boost reg but before cable and charger losses.

The only reports I could find of anyone making this work is CLICKY LINKY.
He made a 10 TEG thermopile, water blocked to DHW and a 4.5kW burner.
He reddened and warped his burner, bolstered it with fire bricks and air blasted the vents to get space heating requirements.

Direct youtube link

Low tech. science


Anyone throwing away an old printer?

Tristar Input tolerances:

• Min. voltage to operate 9V
• Max. solar voltage (Voc) 125V

Dump Load? Diversion control?
TS-45 heat sink! :pac:
It's a 45A controller with, over-voltage & over-current protection! :cool:

Stepper motors are a bittova chocolate teapot. My initial guesstimate of the youtube link was highly overinflated. More likely it's a 0.25watt.
Most of the output goes to heating the inductive coils (Coil resistance 57.5Ω per phase). I managed an underwhelming 1watt AC with phases in series at full tilt on a drill with a guinea-pig.
By the time I have that rectified, boosted and transmitted all it'll do is reduce the solar output with contact resistance of the new connections to get the power down from the roof. Cogging is too high to consider multiple steppers wired in series geared off a cam, there just won't be enough torque to cut in.

Research on VAWTs demonstrates that they are prone to failure due to blade fatigue as one blade is always acting against another and although they cut-in at lower speeds and aren't concerned about wind direction, one can expect ⅓ the output of a comparable HAWT.

Looks like I'll have to wind my own coils then :rolleyes:

Aidan_M_M

Sir Liamalot wrote: »

Stepper motors are a bittova chocolate teapot. My initial guesstimate of the youtube link was highly overinflated. More likely it's a 0.25watt.
Most of the output goes to heating the inductive coils (Coil resistance 57.5Ω per phase). I managed an underwhelming 1watt AC with phases in series at full tilt on a drill with a guinea-pig.
By the time I have that rectified, boosted and transmitted all it'll do is reduce the solar output with contact resistance of the new connections to get the power down from the roof. Cogging is too high to consider multiple steppers wired in series geared off a cam, there just won't be enough torque to cut in.

Research on VAWTs demonstrates that they are prone to failure due to blade fatigue as one blade is always acting against another and although they cut-in at lower speeds and aren't concerned about wind direction, one can expect ⅓ the output of a comparable HAWT.

Looks like I'll have to wind my own coils then :rolleyes:

Purely playing Devils advocate here, we've met , chatted and shook hands so you know I'm not out to randomly knock you Liam, but after nearly a year chasing the best way to harvest solar energy, and countless hours perfecting it, is it really , honestly worthwhile ?
I only ask as I can be a perfectionist doing my own stuff too, but sometimes the marginally less effective off the shelf solution works out handier in the long run. And damn near as good.

killalanerr

Aidan_M_M wrote: »

Purely playing Devils advocate here, we've met , chatted and shook hands so you know I'm not out to randomly knock you Liam, but after nearly a year chasing the best way to harvest solar energy, and countless hours perfecting it, is it really , honestly worthwhile ?
I only ask as I can be a perfectionist doing my own stuff too, but sometimes the marginally less effective off the shelf solution works out handier in the long run. And damn near as good.

I think its more a labour of love at this stage

You're completely right Aidan. System has been working as well as could be hoped for, since the beginning of this month I've had no need of a hook up. Portable electrons in the Winter are proving the greatest challenge. Which could probably be catered for least expensively with another 80w of solar and 350Ah of batteries instead of the 220Ah I have spec-ed and yet to commit to. No need really with my twin elecsols producing an astounding ~100Ah between them.

Realistically this is what I ought to be looking at and as you say won't cost a whole lot more than my time, effort and components.


As regards worthwhile, yes definitely, what price do you put on independence? Up until recent contrivances ie. TEG's onwards the returns on alterations were substantial. Might suit most to stop after solar especially if you have the roof space for more panels. Even with a viable wind turbine it's a lot of setting up, down-time between moving and low mounted turbulence issues.
MPPT is an option too but I'd need to respec the module for a higher voltage one and MPPT is really pointless on small systems compared to the price of more modules.
However to my stubborn mind the camper is just a small-scale test bed for greater endeavours so I'm happy to finance my own learning curve rather than plug and play.
Another important consideration is that generation during the night goes much further as it offsets the lighting & boardsie demands and generation on demand is not subject to lead acid charge reluctance.

Meanwhile weather, PWM subsidised charging and a high-side alternator regulator (14.6v) seems to have recovered my starter battery. It's sitting 0.2v higher than it did a few month's ago.

That all I need to keep things ticking over year round is another 10Wh gene should say it all really

What's that? Electric bicycle dynamos you say....:pac:

Down the economy of scale road I'm currently designing an off-grid workshop too that the van plugs into, except counter to tradition the van will be charging the workshop. The workshop will have 440Ah storage and maybe ~1.5kW DIY wind and solar, the van will provide another 220Ah and it's generators. By this approach I'm hoping to reduce the discharge curve enough to increase Peukert compensated storage by ~100%

I might have to build one of these too

It may "go miles" to redeem my mad notions from my better half when the batteries are too low for a washing machine

Sir Liamalot

Sir Liamalot wrote: »

That all I need to keep things ticking over year round is another 10Wh gene should say it all really

Perhaps I'm even there already if the batteries weren't self-discharging.
They're getting replaced before the Summer Hols.

moodrater

Sir Liamalot wrote: »

Found myself a suitable motor dynamo at last

Here it was hiding all this time



Complete with a blade hub vice (chuck) , gearbox, and diode bridge
Here I was expecting to have to run an excitation field and low and behold the screws started sticking to the permanent magnets as I was taking them out :cool:.

I may have to compromise on the blade diameter, it'll take at least a 1.5m span to turn this.



18v motor, coil resistance < 5Ω, seen here powering a 12v 5w bulb @ ~200 RPM

Problem is its got brushes and they're usually about 3mm long and designed for a life of about 5 hours. Cheap ones use plastic bearings and gears too and they're nor designed for any lateral load. Also you lose 30-40% in the gearbox/windings. A mini version of the piggot style design is the way to go.

Haha busted post deletion...I was thinking I've had enough failures on the workbench, I'm gonna just wait til I have it finished and working before I upload another. Not that I don't appreciate the input.
This one has permanent magnets, no brushes...had thought of that and the excitation malarky too and was going to do it anyway, free stuff is free stuff afterall.
It's not going to see the use for the bearings to fail or if they do at least I'll know if it's worthwhile by then (can't get the chuck off at the moment spent an hour on it with an impact screwdriver). Gears are metal. It'll only be in operation 16 - 24 hours a week for 3-4 months. If it proves practical then I'll consider spending on it.

Yeah the gear losses are a concern, I'm contemplating overcoming them with brute force and ignorance by increasing the wingspan and/or extra blades. I can lose the gearbox and get 30% to 40% electronic losses instead on a boost reg. :rolleyes: {....or keep searching for a suitable motor...:p}


I know, a Piggott turbine would be best I'm just too lazy to wind coils and all that GRP neodynium magnet craic for the sake of a stick it up take it down every night jobbie. I'll probably use his blade design, haven't decided whether I'll go birch ply or whatever's to hand....probably birch come to think of it for the sake of the windscreen.

20watts no bother at 600RPM, >25V OC @ 1000RPM.

[Edit: the Piggott turbines don't scale as well as you'd think, it's not as simple as dividing everything by two. He's tried to make the book as accessible as possible by cutting down on the maths and formulae and opted instead for providing multiple size models. Things like furling and torque over swept area, amount of coils you can fit on a reduced hub size and what they'll produce, blade integrity etc. are more complex than first impressions of his guide. I'm coming around to the notion that a blade diameter less than 1m won't work and his smallest is 1.2m so still practical, I was set on a < 0.75m diameter when I last discarded the idea.

Interesting design here for a dynohub. You can get 3W hubs since that was made]

moodrater

ahhrghh was typing a reply and hit ctrl r by accident.

That motor is a rare beast indeed, even good makes like makita, hitachi, dewalt are mostly brushes expect for the impact and highend drills. Would be a nice rc motor.

You shouldn't need to do any calculations there are loads of variations on the internet that are built, tested, mistakes already made.

A 1.5m blade on that certainly won't be safe.

My 14.4v hitachi has permanent magnets too.

I think I'll run a cam through some bearings onto the chuck rather than mounting the hub straight to the chuck to reduce the lateral strain. My junk horde is in the wrong country at the moment though so options are kinda limited.

And now I'm thinking PVC blades from wavin pipe. :eek:
I'll start with a smaller diameter and see how it goes.

I can't run it through the Tristar in solar reg. mode. It'll just open circuit when the batteries are full and then overspin. Not sure I need a regulator considering the batteries will be low if it's worth setting up. Fridge will be too powerful a dump load and counterproductive...a little dichroic or two will probably do with a wee PWM diverter or something....I was thinking of automatic electric breaking but I don't think it'd do much other than melt.

moodrater

Sir Liamalot wrote: »

My 14.4v hitachi has permanent magnets too.

I think I'll run a cam through some bearings onto the chuck rather than mounting the hub straight to the chuck to reduce the lateral strain. My junk horde is in the wrong country at the moment though so options are kinda limited.

And now I'm thinking PVC blades from wavin pipe. :eek:
I'll start with a smaller diameter and see how it goes.

I can't run it through the Tristar in solar reg. mode. It'll just open circuit when the batteries are full and then overspin. Not sure I need a regulator considering the batteries will be low if it's worth setting up. Fridge will be too powerful a dump load and counterproductive...a little dichroic or two will probably do with a wee PWM diverter or something....I was thinking of automatic electric breaking but I don't think it'd do much other than melt.

Hard to get small length of the heavy stuff, you could ask the council who takes their offcuts for reycling. The wavin biaxial water main stuff would be ideal, if it works out you'd want to paint it. Theres a simple opamp circuit floating around for dump load it uses high voltage jfet opamps think its tl084 probably lf347 or tl074 would be fine too if you have one in scrap.

I've 3m of pipe, the orange stuff not the recycled junk. Should do me for a few hubs. I'll be inspecting them every time I pack it so should be able to spot fatigue early.
Op amps make my brain hurt but its a good solution, anything that's latching won't last.
Paint...eh yeah...function over beauty me.