Solar Install; the on-going saga

moodrater

Sir Liamalot wrote: »

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.

Paint for the uv not beauty, orange stuff is for underground it has zero uv stabilisation, its pretty brittle to begin with and gets extremely brittle in jig time when exposed to uv. Grey is overground and uv stabilised but still too thin for the size blade you're taking about. Remember the tips even at moderate windspeed the tips can be going over 100kph.

Unless you're taking about orange polyethylene gas main, you could swing the van around on that stuff But tis probably a bit flexible, although the centrifugal force might straighten it out under rotation,

Prototype turbine is ready for field tests.
I'm planning mark 2 & 3 already. :pac:
Turns out it is a brushed motor. So is my Hitachi. The magnets are on the stator.
Brushes were hidden...bit longer than 3mm though :P ...maybe 7mm :rolleyes: Non serviceable, manufactured obsolescence tripe. Ah I might be able to bodge new ones in there if needs be.
I don't expect it to perform very well to be honest, cogging is heavy, blades are inferior, undersized tail, no external bearings, motor has gears...I do spend a lot of time up mountains where wind is abundant though! If it cuts-in below 5m/s I'll be happy. It can live dismembered under my solar module, til I make something better.

Here's a teaser.

Rough estimate seeing as it's not been Proven in the field yet ; €0.26 per watt.
Handles 800 RPM pretty well on a bench test.
Still need a diode (to stop the battery powering the motor), and a mounting pole. If it performs ok, I'll even paint it. Diversion controller isn't happening until Mk 2 at the earliest.

PVC Blade templates here.

Oh and I seriously don't recommend choosing a diamond coated angle grinder blade for a hub...I've a fine collection of broken and blunted drill bits, alas as needs must.

Fries-With-That

Interesting use of a drill. I can't wait to see your results as I have a few old cheap as chips drills here.

Why not use 5 blades and have them smaller, catch more wind and less flexing.

Because it's a diamond coated blade hub with pre-drilled holes in all the wrong places and I didn't fancy fighting it anymore. I don't want to spend on this before I get some results. It's a bit annoying buying "suitable" materials when I have a shed full of free materials back in Éire. I had originally intended a 5 blade hub, it's easy to swap in any case as the hub detaches from chuck quite easily. Mk-2 is going to be a dynohub maybe with 5 blades.
I reckon the Mk-1 will be very hard to start, perhaps that's because I crushed the stator a little in a vice...or maybe it's just the iron in the windings.

The more blades you have the easier it'll start but the lower the max RPM (I'm sure you understand this already Fries...just saying for anyone else interested). I think the Mk-1 is going to be my high-wind turbine.

You're right I ought to hurry up and test it already.
I pulled a diode off a car radio...looks like a 5A, google-proof code on it, can't find a listing...eh...it'll work until it breaks :pac:

Production date Dec. 2013 within 11 days between them
235Ah @ C20
290Ah @ C100 (nominal) - de-rating (10%) for ambient 15°C: 260Ah -> 1600Wh to 50% DOD

1kg each heavier than the "equivalent" Trojans and a month fresher. Also have dual battery posts including standard automotive.

All this for about €40 less than I paid for my previous lemons.

113mm² battery inter-connector compared to standard domestic AC 13A
mmmm copper...

12.6v outta the box...no point sticking a hydrometer in there yet.
Where's me slow and nasty chargey do-fer...

Not bad on a new battery...don't trust this contraption a second though...soon as my back is turned it'll get up to mischief...wait until you see the "float" drift.
I'll get back to yee in 40 hours when it's done. :rolleyes:

Fun filled weekend ahead: [you know the tune] "We're all goin' ona...electron holiday"
Test new batteries, refit and rewire for series config., program custom profile into the TS, pull some data.
Reprogram the TM and set to advanced user profile, edit the charge 100% reset stipulations, set equalisation reminders, set charge efficiency to 85%, pull some data.
Upgrade the fridge fan advanced and highly technical protection systemᵀᴹ.
Test the discharge curve.
Might even get a chance to fly my new windy. Hub dynamo is arriving Monday.
Also have a cheap and nasty "100W" Chinese boost reg, IP 0 on the way over land...see how long that lasts between a pulsing DC (ish) generator and a battery (capacitor across input (any recommendations in Farads?) and shottky diode on output).

Crown recommend a charge termination of 14.5v and weekly equalisation :eek:!
Right, much as I respect their products this is not for an industrial golf cart mains charger and the very fact they put a wholly mechanical wind operated water pump (high torque, low rev. turbine) on the top right front cover of their brochure above all the devices that do actually use their batteries, but from mains, gives me every confidence to set the max absorption to 14.7v and monthly equalisation.
Max permissible charge voltage including temp. comp. : 16.2v
Check the water levels in a month or two and if they're low I'll reduce the set-point.

This charger's only rated for 14Ah - 120Ah batteries by the way

Mod complete...it was getting grumpy.
:pac:

moodrater

Sir Liamalot wrote: »

Production date Dec. 2013 within 11 days between them
235Ah @ C20
290Ah @ C100 (nominal) - de-rating (10%) for ambient 15°C: 260Ah -> 1600Wh to 50% DOD

1kg each heavier than the "equivalent" Trojans and a month fresher. Also have dual battery posts including standard automotive.

All this for about €40 less than I paid for my previous lemons.

Battery porn :pac:

Drift isn't as bad as I've previously seen it on older batteries. It's hit >15v on occasion.




~700mA Float :eek: ...back in yer box!!:mad:

End of absorption 2.2A :rolleyes:
...ya get what you pay for.

20 squid sez the Tristar could squeeze another 15Ah in there with less boiling.

I didn't get as far as I'd hoped today



Batteries are in.

Heavy bar stewards, suppose handles would be too much to ask for...better off I guess...there's loops for a sling...if they had handles someone might foolishly try to lift them. Specific gravity of 1.275 on all 6 cells.

Reprogrammed the TriMetric, added an absorption time to the 100% charged criteria, so now it's amps (less than 2% of C20), volts (>14.6v) and hold for 1h30m before it resets.
Looked at the data and it's not much...I mustov reset battery lifetime discharge a while back, other than that it's just self-updating info on the last 5 cycles.

I emailed Ralph the founder & owner of Bogart about the inaccuracies I was experiencing in the Winter on the TM and he admitted it's never going to be accurate after a week without a full charge and then cheekily suggested that the onus is on me to maintain my batteries properly and charge them to 100% every week anyway. Nice guy all in all, I'm gonna stick with the monitor, he's right, but the meter's still out. Problem is they all will be, his is least so.
Thinking about the SmartGuage, unless I go for the whole SmartBank package I can't see how it's going to be any better with the engine battery dropping in an out of circuit seemingly at random to the processor.
After some more correspondence I learned I had actually wired the TM properly, the charge to the engine battery has never been a factor because it's on the load side of the shunt. I'm gonna do some clamp-meter testing on the alternator just to verify the TM (yes I know shunts are more accurate than induction meters...anyone got a meter that can take >20A through the prongs? :P).

So programming the TriStar is a pain in the hoop. Moringstar Corp. "World Leading Manufacturer of Solar Controllers" thought USB was a passing phase so I have the splendid alternative options of communicating with the controller through RJ-11:rolleyes: or serial

...12v serial none-the-less. My USB to serial was female and the TS is female. So I had to neuter it with a hack saw, the driver disk was a half-size cd...computer says no!...eventually got it to recognise the TS and then it just wouldn't play at all. Apparently if you buy 5 adapters you might get lucky and one might work, or find a machine from the 90's.
Wouldn't talk to the router either or straight through. Can you make a rj-11 to rj-45 crossover cable
I've ordered a male FTDI USB adapter

Right this has me stumped...I'm just wondering what to do with my retired lemons. I've discharge tested them consistently to understand the pair together are 55% DOD at this stage, but the thick plottens.
The eldest (4 years, 2 months): most abused and the one I was about to retire this time last year is now the strongest.
I separated the pair and the voltage rose when the other fell (newer one is 3 years 3 months) I just did some hydrometer tests and the eldest is only 0.005 off on one cell from a brand new (Elescol) battery.
The younger is pretty healthy in SG terms..only 0.025 difference between highest and lowest cells with the centre 4 pretty much on spec.
I didn't take Paddy's suggestion of contaminating the electrolyte with epsom salt way back in case anyone's wondering.

Any ideas? Grid corrosion? Buckled plates? They self-discharge about 0.2v per week with a ~100mA (C1000) constant load.

Initial tests of the TM are a consistent 0.5A lower than my clamp meter on solar and alternator modes. 2% error on the clamp meter according to Fluke (if it's dead centre in the clamp), and efficiency set to 85% on TM...% error doesn't make any sense either though because it's consistent from 0 to 10A....hrmmm need more data.

moodrater

Sir Liamalot wrote: »

Right this has me stumped...I'm just wondering what to do with my retired lemons. I've discharge tested them consistently to understand the pair together are 55% DOD at this stage, but the thick plottens.
The eldest (4 years, 2 months): most abused and the one I was about to retire this time last year is now the strongest.
I separated the pair and the voltage rose when the other fell (newer one is 3 years 3 months) I just did some hydrometer tests and the eldest is only 0.005 off on one cell from a brand new (Elescol) battery.
The younger is pretty healthy in SG terms..only 0.025 difference between highest and lowest cells with the centre 4 pretty much on spec.

I didn't take Paddy's suggestion of contaminating the electrolyte with epsom salt way back in case anyone's wondering.

Any ideas? Grid corrosion? Buckled plates? They self-discharge about 0.2v per week with a ~100mA (C1000) constant load.

Initial tests of the TM are a consistent 0.5A lower than my clamp meter on solar and alternator modes. 2% error on the clamp meter according to Fluke (if it's dead centre in the clamp), and efficiency set to 85% on TM...% doesn't make any sense either though because it's consistent from 0 to 10A....hrmmm need more data.

Normal wear and tear could account for it shedding of the positive, granulation of the negative, sulfation, impurities etc. hard to know without an autopsy. Just bring them to the scrap yard and put the money to good use. I wonder are the outside cells of the newer one lower due to vibration/movement via the terminals cracking plates or dislodging paste.

gctest50

Sir Liamalot wrote: »

I didn't take Paddy's suggestion of contaminating the electrolyte with epsom salt way back in case anyone's wondering. ...........

EDTA should be better than epsom salts afaik

ethylenediaminetetra-acetic acid tetrasodium salt <--- need to make sure that is the right one


.

Can't sulphation be read from the electroylte density though?
Sell them :eek:...they're only halfway through their life! They'll be worth just as much to the scrapper in 5 years when they're proper dead. :pac:

The outer cells being off is a common symptom it's more to do with proximity to the charger/load and lower internal resistance than anything else.

moodrater

Sir Liamalot wrote: »

Can't sulphation could be read from the electroylte density though?
Sell them :eek:...they're only halfway through their life! They'll be worth just as much to the scarper in 5 years when they're proper dead. :pac:

The outer cells being off is a common symptom it's more to do with proximity to the charger/load and lower internal resistance than anything else.

My sheds full have had to adopt a new criteria for keeping stuff - would i buy it off someone else its much more effective than will I actually use this to which the answer is always yes. Yer other half must be a saint.

I suppose you could run your shed lights off them with yer wind turbine.

I have a way of spreading the junk around, I just leave a little in everybody's shed.

You've got the entirely wrong approach Mr. M, soon as one shed is full I use the contents of the full shed to build another shed :pac:

Ok so this works it's an FTDI do-fer.



...Well except it doesn't fit...




Now it does!

Absorption termination voltage & rate of temp comp.




Float Settings




Equalisation and HVD




Summary

EDIT: [ I've been adjusting the values since those, after rechecking the battery requirements: Temp comp is now set at 0.018V/°C (3mV per cell), equalisation 2.5hours max and low battery trip is 1 hour (extends the absorption by X time if battery falls below Y threshold)]

Live Data (module is shaded)

This can be setup up for remote logging in and uploading data online. The MPPT Tristars have ethernet ports not rj-11.


Did yee notice my (near) 6 month electron consumption is 16kWh (minus charging losses)?
I reset the counters so I lost the logging before I could upload it, It's a 3 month graphed log of battery voltage daily min/max, battery temperature daily min/max, alarms, faults, Ah generated daily and daily time in each charge stage.


TM is a consistent 0.5A higher on consumption figures vs my clamp meter...ok I'll admit I need a shunted DMM. I need an electronics meter anyways.

New 3W 6V Dynohub.

Chucked the over-voltage protection in the parts box.
This is a single phase alternator so it's AC output, I need a bridge rectifier on this later...then whip it up-to about 20V on a boost reg. through another diode with a few caps to be polite and send it on it's merry way to electron city.
I'm still debating making a dump load controller...eh I'll probably get around to it eventually.
I'm also debating if I should bother with some sort of furling system...

Blade Template

I reckon this'll knock the socks off that Rutland 504 I posted earlier. I did cheat though and increased the swept area. It's an 850mm diameter (Mk 1 is a 1.2m).

Coil resistance 2.6Ω
Bench tested very roughly 10W output at about 9v @ 400RPM (I had to drive it with a belt because the wires are on the axle. So very rough estimate indeed)
It'll happily light a 12v 5w bulb turning by hand.
V. low cogging too.

Mounted Blades

I quite like this design, I was going to go for 5 blades initially but the pre-drilled holes in the hub were a multiple of 3 so I just went with that...much easier divide a circle in 6 with a compass anyways. I'll see how noisey it gets, apparently even bladed turbines are the most noisey.
The beauty of this is I can just remove the blades and re-balance the hub for a 3-blade or 2-blade turbine.

Front Mounting




Rear Mounting

The biggest problem now is finding a suitable way to mount this. The Dynohub would be much better suited to a VAWT because the wiring is coming from the axle. So supporting the entire hub and blades by a single side of the axle is creating an awful lot of leverage....perhaps the answer is staring at me...hrmmm I might sacrifice that vice alright...

I went for a stainless steel hub this time...much easier

moodrater

Sir Liamalot wrote: »

Ok so this works it's an FTDI do-fer.



...Well except it doesn't fit...




Now it does!

Looks expensive for an ftdi cable, mine was 'borrowed' years ago. I use my old vagcom kkl cable now when I need one €3.50 on ebay, just resurrected a gigabit router with it in fact. I remember when the first ftdi chip came out cost me something like €27 + postage for the chip.

Yeah pricey enough...so is buying 4 more wrong ones.
I got lucky and found a "reconditioned" one.

moodrater

Sir Liamalot wrote: »

Yeah pricey enough...so is buying 4 more wrong ones.
I got lucky and found a "reconditioned" one.

Ah how bad. What did you end up with the first few times fake pl2303? Even the genuine ones were rubbish.

Yup. Prolific junk.

...and now I've just realised that I mounted the blades the wrong way on the hub.
I was all like direction of travel?! Ah sher it'll be grand it's only AC, that's probably to stop the wheel untapping. Actually it's a precession issue with the alternator gubbins.
Need to cut 6 new blades now..

gctest50

Sir Liamalot wrote: »

Need to cut 6 new blades now..

Put them on the other side of the hub

Thought of that; wiring issues unless I want to make it a downwind machine.

moodrater

Sir Liamalot wrote: »

Thought of that; wiring issues unless I want to make it a downwind machine.

Downwind has its advantages with flexible blade in high winds they won't hit the tower.

Indeed but the tower then causes turbulence and it makes braking that much more difficult. Works well if you're an engineer. I'll just turn the template backwards and get the jigsaw out again I think.