navitron
 
Renewable Energy and Sustainability Forum
Welcome, Guest. Please login or register.

Login with username, password and session length
News: Anyone wishing to register as a new member on the forum is strongly recommended to use a "proper" email address. Following continuous spam/hack attempts on the forum, "disposable" email addresses like Gmail, Yahoo and Hotmail tend to be viewed with suspicion, and the application rejected if there is any doubt whatsoever
 
Recent Articles: Navitron Partners With Solax to Help Create A More Sustainable Future | Navitron Calls for Increased Carbon Footprint Reduction In Light of Earth Overshoot Day | A plea from The David School - Issue 18
   Home   Help Search Login Register  
Pages: 1 ... 14 15 16 17 [18] 19 20 21 22   Go Down
  Print  
Author Topic: Cobbled together PV in W.Sussex  (Read 152092 times)
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #255 on: March 09, 2012, 12:35:19 AM »

More catch-up posting on the blog today...

Today's episode is about bottom balancing the pack with a bunch of log data graphs for the discharge curves and then the charge curves.

Nicely shows the radically different behaviour of lithium-ion cells from lead acid ones.

The fact that the terminal Voltage hardly changes at all during charge and discharge (until you get to the saturation / desaturation points) makes it hard to use Voltage as a good indicator of state of charge.  EV people use Ah or kWh counters, which actually do work on lithium cells.  Lead acid cells have massive problems with Ah counters due to the Peukert factor (apparent capacity is inversely related to discharge current) and drift due to self-discharge and a generally lossy charge efficiency.  Lithium cells exhibit 95% charge efficiency, even at charge/discharge rates over 1C.

I'm seeing significant improvements in usable energy usage due to this high charge efficiency alone.

Lead acid cells waste over 20% of all the input energy through charge inefficiency.  This is due to two main effects: heating due to internal resistance and undesirable chemical reactions (gassing and gas recombination in gel and AGM cells).  Lithium cells do not gas in normal use and have VERY low internal resistance, leading to very high charge efficiency - almost as good as capacitors.

Having batteries that are 15-20% more charge efficient is like having a 15-20% bigger PV array or 15-20% longer charging time on a Winters day.  You get more usable energy at night for the same energy input during the day.

So although expensive, lithium cells actually have a bigger payback than just longer cycle life and affinity for partial charge compared to lead acid.
Logged

http://solarbodge.blogspot.com/
3.58kWp & 800Ah LiFeYPO4 off-grid(ish). See 'Cobbled together PV in W.Sussex' (in "Show Us Yours")
biff
Global Moderator
Hero Member
*****
Offline Offline

Posts: 12274


An unpaid Navitron volunteer who lives off-grid.


« Reply #256 on: March 09, 2012, 09:22:45 AM »

Good stuff Outta,
               This kind of information is invaluable to anyone thinking of going down this route.It looks very promising and by all accounts must be the way forward.
      Biff
Logged

An unpaid Navitron volunteer,who has been living off-grid,powered by wind and solar,each year better than the last one.
powerbynature
Newbie
*
Offline Offline

Posts: 10


WWW
« Reply #257 on: March 09, 2012, 01:42:59 PM »

some great information on this thread.  As I've been going down the same route you have managed to give me some ideas although the wife is not happy, on the flip side though my son things it's all great as he gets to help (help as in he's 2 and a half) but they are never to young to learn Smiley
Logged

http://www.powerbynature.co.uk - reviews and information
dickster
Sr. Member
****
Offline Offline

Posts: 271


« Reply #258 on: March 09, 2012, 02:30:56 PM »

This is all very exciting, perhaps a good alternative to lead acid, but as a non educated layman, it all seems very complicated. What are the chances of said layman running a set up like yours without the in depth knowledge that you have?
Logged
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #259 on: March 09, 2012, 04:32:26 PM »

All... Thanks for the interest and encouragement.

@dickster:  The charging of lead acid batteries is no less fraught with danger.  In fact more so, especially with AGM ones that can go into thermal run-away quite easily and burst into flames or explode.  In comparison, although the lithium cells require more close monitoring and automation to ensure they don't turn into expensive bricks, they are actually now safer if abused (at least the ones I'm using are).

But it has to be said that this is an experimental power system I have here.  I doubt there are more than a couple of lithium off grid systems in Europe in residential use at the moment.  Products I've heard about are still in pre-production or are just toys (I saw a portable solution that used a 20Ah 12V battery and a briefcase solar panel at EcoBuild last year).  I'm an electronic engineer and seasoned bodger, so it's ok... "Honest, guv' ".

The "lay person" shouldn't be playing with anything more dangerous than AA batteries in their walkman.  Certainly shouldn't be "playing" with a 3kW PV array and a battery bank capable of delivering sustained short circuit currents of 8,000 Amperes...  They definitely shouldn't be hacking the internals of a 230V AC mains inverter that is connected to the kids Playstation!

Having said all that, my wife manages to "operate" the system just fine with nearly no knowledge about how it works.  That's because she doesn't have to "do" anything to it... Just be aware that the house is running on a limited amount of battery power and so be careful not to waste it (leaving lights, TV, laptop on when not needed).  Basic training for all "lay people".

If we don't learn to manage our energy use a bit more intelligently, it won't just be my lights that go out when the energy reserves have been squandered carelessly.
Logged

http://solarbodge.blogspot.com/
3.58kWp & 800Ah LiFeYPO4 off-grid(ish). See 'Cobbled together PV in W.Sussex' (in "Show Us Yours")
StBarnabas
Hero Member
*****
Offline Offline

Posts: 2221


St Barnabas Chapel (2009)


« Reply #260 on: March 10, 2012, 09:50:46 AM »

Outta
keep up the excellent work. This thread remains a true gem.
StB
Logged


Gestis Censere. 40x47mm DHW with TDC3. 3kW ASHP, 9kW GSHP, 3kW Navitron PV with Platinum 3100S GTI, 6.5kW WBS, 5 chickens. FMY 2009.
biff
Global Moderator
Hero Member
*****
Offline Offline

Posts: 12274


An unpaid Navitron volunteer who lives off-grid.


« Reply #261 on: March 10, 2012, 11:41:44 AM »

Exactly St/B
                This thread really is a true gem with the most modern up to date information available.
                                             Brilliant stuff Outta!!
                                                               Biff
Logged

An unpaid Navitron volunteer,who has been living off-grid,powered by wind and solar,each year better than the last one.
wookey
Hero Member
*****
Offline Offline

Posts: 3221


WWW
« Reply #262 on: March 10, 2012, 10:31:10 PM »

OK 'idiot' is a bit harsh, but he's certainly a man with a huge ego, and ultimately I found other people to be doing a better job of presenting impartial engineering data without excessive bluster. This was the thread:  http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=15247
Read it yourself and see what you think. You clearly understand this stuff so I'll let you get on with it.

You are quite right that the series-only case is a lot simpler and is less likely to have problems. My sets of 3 that have got poorly are almost certainly due to one weak cell in those sets, which after a while ends up hurting all 3. I would love to have some time to take it to bits and check out exactly where I am at, but extension-building has put all my proper electronics-geeking on hold for the last 18 months or so. To put things in perspective my pack is still working fine and shows signs of being able to do so for a long time yet. And if I have to replace 3 or 6 cells, to get it 'good as new' at some point, that's not too big a deal. I can even buy them in the UK now for 14 each.
Logged

Wookey
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #263 on: March 10, 2012, 10:54:23 PM »

Thanks guys.

Probably now it will go quiet for a while on the thread as the system is in and appears to be working, so it's just a case of letting it run and see what happens.

No news is good news, as they say.

The one area where I'm thinking of change is in how to increase the charge capacity.  The 60A charger is now routinely maxing out and so there's a couple of options... Change the controller for a more powerful one (maybe 100A, but Morningstar don't make such a controller) or split the main array into top and bottom.

The 1,540Wp roof array is just at the maximum 60A battery current limit (in full sunlight and cool weather).  That lot terminates in the upper junction box on the wall with a 35mm2 feeder cable down to the lower junction box that piles the wall, patio and garage PV on top, before going into the house.  The lower arrays are at differing angles and suffer various amounts of shading but add up to another 915Wp (another 35A).

I'm quite surprised the controller hasn't complained at all, with 57% over-current input.  I've seen the posts about a couple of TS-MPPT60 controllers letting go of magic smoke, but that seems to have been an issue with high Voltage handling, rather than over current.  I'm running quite low Voltage but huge current.

If I split the array and use a 60A controller and a smaller 45A controller, I can get the full 95A potential (the MSView log graph was conspicuous by the battery Amps haircut the curve took on, relative to the solar input curve) but it means more wiring, more switchgear, more fusing, higher night time parasitic load and more networking to capture the data from a 3rd controller.

Alternatives do exist... There's the Outback FM80 controller that can do 80A (still not enough for the whole array) or the Midnite Classic controller that can just squeak 94A output.  But then both completely break my logging and AC load management solution, which is built on the Morningstar MSView MODBUS software.  The Midnite Classic does speak MODBUS but the MSView logger doesn't know the Midnite Classic commands or variables...  I'd have to write my own MODBUS client Shocked

The previous lead acid battery needed over-current protection in the form of an AC dump load.  The new battery can easily charge at speeds of up to 1C (400A) with no supervision and speeds of up to 3C (1,200A) with temperature monitoring and compensation.  In short, it will absorb almost any amount of power that you could conceivably throw at it, up to a 31kWp PV array.
Logged

http://solarbodge.blogspot.com/
3.58kWp & 800Ah LiFeYPO4 off-grid(ish). See 'Cobbled together PV in W.Sussex' (in "Show Us Yours")
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #264 on: March 16, 2012, 03:49:08 PM »

Got some graphs up on the blog from the CellLog8s of the pack in operation over a number of days and ranging from quite empty to full.

All the cells seem to be tracking each other quite well and I'm keeping a daily log of the cell differential when making my usual meter readings at midnight.

The CellLog8s has a useful display that tells you the highest and lowest Voltage cell in the pack and the difference in millivolts between them.  You can even program the alarm to sound if the cell differential value exceeds a value, so that you can spot a cell going bad in the pack before it self destructs at the high or low end (under charge or discharge) and causes the rest of the pack to get damaged too.

So far it's reading just a steady 6 to 8mV difference under the usual light load at midnight.  The traces confirm this, showing no serious deviations under high charge rates (up to 76A) or high discharge rates (up to maybe 70A).

More of the house is going off grid now.  I ran a new solar spur up to the airing cupboard and now run the central heating gubbins off grid.  Despite the boiler being in the kitchen and the heating controller being in the kitchen, the whole system was powered from a socket in the airing cupboard upstairs (where the circulating pump is).  So it's easy to unplug it from the grid socket and plug it into the solar one.

Now, with only really a couple of pesky wall-warts for a couple of things in far flung reaches of the house, everything bar the instant water heating shower and the cooker are off grid.  As I write, the wireless energy meter for the utility is reading zero Watts Grin.  In truth, there's some power being still consumed but it's now below measurement levels.

And maybe I don't even  need to bother with the wall warts... It's not just the wireless meter that is reading zero.  The actual utility meter also has a minimum power threshold of recording, below which the LED that usually blinks to indicate the passing of Watt-hours just lights up continuously to indicate that it has stopped recording. These meters have an "anti-creep" mode to prevent spurious billing of energy when the amount is so small that it could just be a tolerance error in the meter itself - say, reading 5W falsely when the consumer unit was actually turned off completely.

So a house can consume maybe 5-10W of energy and pay nothing for it, as the meter will stop.  The meter itself consumes 2W of energy just doing it's thing and we wouldn't want to pay for that.

The weather has been quite kind over the last 3 weeks, but not wall to wall sun.  I've not had to put the house on-grid since installing these new batteries (apart from the maintenance day when recharging after bottom balancing them).  I wonder how long I can run the house without reverting things back to the grid.

E.On don't believe my self-submitted meter readings any more.  I sent some in for the latest bill and they sent a guy out a few days later to re-read them.

Unlike grid tied solar generating stations that still import lots of energy at night (even if they are a net exporter over 24 hours), an off grid station just negates importation and so my import meter is gradually slowing to a stop.
« Last Edit: March 16, 2012, 03:53:51 PM by Outtasight » Logged

http://solarbodge.blogspot.com/
3.58kWp & 800Ah LiFeYPO4 off-grid(ish). See 'Cobbled together PV in W.Sussex' (in "Show Us Yours")
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #265 on: March 20, 2012, 06:43:07 PM »

Well, according to my calendar, it's been just over a month since I installed my lithium ion battery bank.

It's still working.  I'm still alive.  The sofa is still... not a pile of ash.

Almost on the first month anniversary I did have the battery run out on me in the night, after a run of dull days.  But I propped it up with some help from a hastily bodged together reserve battery (one of my lead acid 12V batteries).  It was just enough to see me through the wee hours of Sunday morning and then luckily Sunday was sunny, so it's back to business as usual now.

I've spotted another bug in the Morningstar Tristar MPPT-60 firmware.  Nothing serious.  The internal logged data on the controller is supposed to record the Pout Max value (the maximum power delivered to the battery in a day) but the logged figures are off by a factor of 10 (maybe they just forgot the decimal point!).  It logs my peak ouput power as 15,000W... I wish  Grin

While I was at it, I pestered them on when they are going to release a 100A version of the controller.  The lady replied, "I'll forward it to the tech department and we'll get back to you."  Yah... Right.

I pointed out that Outback and Midnite solar both have 80-100A controllers out in the market, but I'd prefer to stick with Morningstar (if only so I don't have to reprogram my load controller).

The Tristar will handle 3.2kW of PV, but only with a 48V battery.  I have 24V everything now, so I'm not going to change to suit them.  Besides, the only reason most people go to 48V is because you couldn't easily get a lead acid battery to do more than 3kW load at 24V.  You have to keep the current down to get the rated capacity.  C/10 is about as good as you can get.  That meant a 1000Ah battery could do 100A at 24V for a 2.4kW load.

Going to 48V meant you could have a 1000Ah bank and still only do 100A but get 4.8kW load on it.

Going to lithium means you can have a 26V, 400Ah bank and do 1C continuous load for 400A at 26v = a stonking 10.4kW load... Enough for an instant water heating shower with change to run a toaster.

So the makers of solar chargers are going to have to up their game, because much more power is available at lower Voltages than before.

Hell, if I was to start again, I might even consider using a 24V nominal PV array (like I do now to keep the wiring cost down) but use a 1000Ah 13V (4 cell) lithium bank.  The fewer and bigger the cells, the easier it is to manage them.  You'd just need a MPPT charge controller that can handle 250A at 13V...  tumble
Logged

http://solarbodge.blogspot.com/
3.58kWp & 800Ah LiFeYPO4 off-grid(ish). See 'Cobbled together PV in W.Sussex' (in "Show Us Yours")
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #266 on: March 23, 2012, 09:28:20 AM »

The guys at Morningstar sent me a reply.  One of the techies said that the reason the Tristars only go up to 60A is because of the relationship between the switching devices they use and the size of the heat sinks required.  The heat generated increasing with the square of the current.  So increasing the current with the same type of switching device from 60A to 100A requires a lot more cooling.

The Tristar MPPT60 already has a quite impressive bit of finnage on the outside and on my system that is pegged at the 60A limit for a couple of hours a day now, it gets pretty warm.  Still cool enough to touch but definitely very warm.

Midnite and Outback both use fans to cool their controllers and so can increase the current the controller handles but Morningstar are sticking to their design principle of only using passive cooling.  I kinda agree.  When I went to the Earthship near Brighton (why does a house with some passive heating, rainwater harvesting and PV have to have such a hippy name?  It's counter-productive to getting the technology adopted by "Dave from Croydon"), I saw an Outback controller hard at work with the fan going and full of dust.

He didn't rule out them building a more powerful controller, but they have nothing in the pipeline.

A pity as there are a lot of advances in MOSFET and especially IGBT devices that are getting ever lower "on" resistance and faster slew rates that reduce the big heating effects when switching in PWM situations.  These are all being driven by improvements needed in EV motor controllers, where you have a 190V battery and a 200kW motor and a PWM speed controller that needs to deliver over 1,000A at 190V.

So it should be getting easier to do 100A at 24V with the same size heat sink.   Granted the EV controllers have fans and even water cooling, but they are handling 100 times as much power.

No word yet from their software guys yet on the spurious power reading in the logs.  They asked me for a copy of the log file from the controllers memory and went off to check out what might be wrong.

Meanwhile, on the blog... I've been experimenting with eeking up the final charge Voltage a bit on the lithium battery bank.
« Last Edit: March 23, 2012, 09:32:38 AM by Outtasight » Logged

http://solarbodge.blogspot.com/
3.58kWp & 800Ah LiFeYPO4 off-grid(ish). See 'Cobbled together PV in W.Sussex' (in "Show Us Yours")
marktime
Sr. Member
****
Offline Offline

Posts: 278


« Reply #267 on: June 11, 2012, 07:01:13 PM »

You're on the telly Smiley
Logged
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #268 on: July 24, 2012, 12:16:03 AM »

Ha ha... Yeah as well as doing the Britain in a Day film, BBC South Today did a 2 minute feature on my solar system as a trailer for the film on the morning news.  I haven't seen it because our digital telly is from the London region and it only went out on BBC South.  But some random woman at a lavender farm we were visiting a couple of weeks ago remembered me... She was giving me funny look and I was wondering if I'd had a wardrobe malfunction when she suddenly blurted out "You're that solar guy from the news, aren't you?". 

That aside, things are going well at the ranch.  156 cycles and counting with no problems.  After tinkering with the charge settings over a few weeks, I settled back on the original settings, with a short 30 minute absorption charge at 3.50Vpc (28.0V for the pack).

Did about 12.5kWh today with 4.1kWh in the water tank via the immersion heater.
Logged

http://solarbodge.blogspot.com/
3.58kWp & 800Ah LiFeYPO4 off-grid(ish). See 'Cobbled together PV in W.Sussex' (in "Show Us Yours")
StBarnabas
Hero Member
*****
Offline Offline

Posts: 2221


St Barnabas Chapel (2009)


« Reply #269 on: July 24, 2012, 06:42:46 PM »

Great stuff Outta. Sadly you did not break into the Northumberland Region TV region! Great things are going well and your batteries are surviving the strain!
Sean
Logged


Gestis Censere. 40x47mm DHW with TDC3. 3kW ASHP, 9kW GSHP, 3kW Navitron PV with Platinum 3100S GTI, 6.5kW WBS, 5 chickens. FMY 2009.
Pages: 1 ... 14 15 16 17 [18] 19 20 21 22   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.20 | SMF © 2013, Simple Machines
SMFAds for Free Forums
Simple Audio Video Embedder
Valid XHTML 1.0! Valid CSS!