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 [2]   Go Down
  Print  
Author Topic: Measuring / Estimating DOD whilst discharging .. How?  (Read 7189 times)
Dyslexicbloke
Full Member
***
Offline Offline

Posts: 159


Blue sky thinking ...


« Reply #15 on: June 20, 2011, 09:52:23 PM »

You say you get 13.3v when using the system yet you also say it never sees over 12.3-12.5v.

Can you explain further?

Sorry typo .... should have read 11.3 which was the setpoint I initially had to auto strat the genny and charger.
since reading all this stuff I am being more coutious but I am still discharging to 11.5@ 6-8A before recharging.

I absolutly agree RE more PV .... Its a matter of affording some and is high on the priority list.

I dont get the point that the chart shows 100% DOD in all cases ....... What am I missing?
As you look at increesing final voltage for a given period the Ah delivered decreeses .... surly that is indicating a less deep discharge isnt it.

I take the point RE DOD and open circuit voltage (rested I assume) and I have found some data supporting what has been stated.

The bats have the following properties that may mittigate things somewhat .....
From manufacturer data:-
System ampere-hour range 104 to 4800 Ah to 1.75 VPC at 8-hour rate @ 25C (77F).
Terminals Solid copper insert.
Positive plate Patented lead calcium tin silver alloy.
Negative plate Lead calcium grid alloy.
20 years design life in float applications at 25C (77F)
1200 cycles to 80% DOD at 25C (77F)
Operating temperature Temperature excursions between -40C (-40F) to +50C (122F) allowed. (battery performance and life will be affected)
Self-discharge 0.5 to 1% per week maximum @ 25C (77F).
Float voltage 2.23 to 2.27 VPC.
Extended partial state of charge operation
Deep discharge recovery
Freezing tolerant

I have also attached some graphical data ... My pack is 400Ah nominal as far as I know.
I currently stop the generator charge at circa 15A which I believe is supposed to be aprox 80% full (see chart 2)
(The time and profile are about right given that my max current is limmited to 55A and the float is compensated for temp)

I am very interested to know what you folks make of this data although it is now clear that I will need to charge for longer.
The question is how short can I make it IE what current should I stop at?

I am also going to start logging the rested voltage after my overnigyht switch off along with the voltage and current I had immidiatly before switching off

Looking forward to comments
Al


* BatCHT1.png (91.33 KB, 658x422 - viewed 311 times.)
Logged

Off Grid - Big Caravan and huge enclosed gazzebo.
300W PV 12V system.
400Ah of AGM Absolyte GP cells. (Second hand)
600W Inverter (Maplin's finest :-) )
CHP in the works - Chinese Horisontal Diesel [S195 Generic - Kukje]
VAWT testbed flying - Back to that when its warmer I think.
Dyslexicbloke
Full Member
***
Offline Offline

Posts: 159


Blue sky thinking ...


« Reply #16 on: June 20, 2011, 10:01:07 PM »

More charts....


* BatCHT2.jpg (60.81 KB, 722x431 - viewed 346 times.)

* BatCHT3.jpg (60.84 KB, 682x442 - viewed 354 times.)

* BatCHT4.jpg (51 KB, 666x434 - viewed 347 times.)
Logged

Off Grid - Big Caravan and huge enclosed gazzebo.
300W PV 12V system.
400Ah of AGM Absolyte GP cells. (Second hand)
600W Inverter (Maplin's finest :-) )
CHP in the works - Chinese Horisontal Diesel [S195 Generic - Kukje]
VAWT testbed flying - Back to that when its warmer I think.
profp
Guest
« Reply #17 on: June 21, 2011, 10:49:58 AM »

Tonight with no heavy loads but a poor day of charging, it has gradually sunk to 24.6V and the Smartgauge reckons it's at 52% SoC. 

The charts I have estimate 24.64V to be 70% charged (which the smart gauge rates as 77%). Which I guess just goes to show no two banks are the same! (Assuming that's not a typo in your original post. FWIW 50% SoC is 24.12 on my curve at no load, obv. would be depressed further under load)
Logged
billi
Global Moderator
Hero Member
*****
Offline Offline

Posts: 9073



WWW
« Reply #18 on: June 21, 2011, 10:54:51 AM »

ambient (battery) temperature changes  figures  quite  a bit
Logged

1.6 kw and 2.4 kw   PV array  , Outback MX 60 and FM80 charge controller  ,24 volt 1600 AH Battery ,6 Kw Victron inverter charger, 1.1 kw high head hydro turbine as a back up generator , 5 kw woodburner, 36 solar tubes with 360 l water tank, 1.6 kw  windturbine
profp
Guest
« Reply #19 on: June 21, 2011, 11:27:13 AM »

ambient (battery) temperature changes  figures  quite  a bit
AFAIK Smartgauge doesn't compensate for temperature - I had a quick hunt (should be doing other things!) but couldn't find any curves to show standing voltage vs. temp for FLA.
The internal resistance increases as temp falls, depressing the voltage, but IIRC outtasight's batteries are indoors/in a centrally heated house, so his readings should be higher than mine all things being equal?

Dyslexicbloke - this article which google turned up may help with your estimates, as it includes voltage-load curves. http://www.scubaengineer.com/documents/lead_acid_battery_charging_graphs.pdf

I wonder if an arduino, programmed with the curves for a specific vendors batteries, and armed with a current shunt as well as ADC, and possibly a temp sensor, would do a comparable (or perhaps even better?) job than a smart gauge. It would lack the 'it just works' element of a SG, but the network interface & option of monitoring individual cells in the bank, is quite attractive... Another little project to add to the list...
« Last Edit: June 21, 2011, 11:40:29 AM by profp » Logged
Dyslexicbloke
Full Member
***
Offline Offline

Posts: 159


Blue sky thinking ...


« Reply #20 on: June 22, 2011, 04:35:58 PM »

Interesting artical thanks ...... It set the grey matter churning a bit.

I may be way off base here but I would apriciate some thoughts and opinions.

Given that I have data for open circuit voltage vs SOC and also charging current vs SOC (@flost) I should be able to extrapolate internal resistance vs SOC shouldnt I ?

Heres what I mean ... assume a consistant SOC for the following parameters
OCV is O, float voltage is F, charging current is I and the unknown internal resistance is R

The efective voltage producing a current into R is F - O right?
That makes R = (f - O) / I for any given SOC

Assuming that R for a spaciffic SOC is largly a product of electrolite resistance it should be consistant for charge and discharge so ...
The voltage drop due to R will be a product of -I * R (Because its discharging) and the terminal voltage will therfore be (-I * R)+O for any given state of charge.

Since R and O are both dependant on SOC that isnt easy to solve, at least not for me, If there are any mathmatisions out there who can tell me how to solve that, other than itterativly, I would apriciate it just to test the hypothasis.
Off to play with Exell ..... I will leave out temperature compensation for the time being, my head is already hurting.

Al
Logged

Off Grid - Big Caravan and huge enclosed gazzebo.
300W PV 12V system.
400Ah of AGM Absolyte GP cells. (Second hand)
600W Inverter (Maplin's finest :-) )
CHP in the works - Chinese Horisontal Diesel [S195 Generic - Kukje]
VAWT testbed flying - Back to that when its warmer I think.
EccentricAnomaly
Guest
« Reply #21 on: June 22, 2011, 06:35:56 PM »

Looking at the graphs in the HP article profp referenced I'm afraid I don't think it's a good idea to assume that the internal resistance is consistent.  E.g:, 50% SOC, charging at C/5 gives 13.4 V and C/10 13.2 V, a difference of .2 V whereas discharging C/5 11.53 V and C/10 12.01 V, difference 0.48 V so apparently over twice the internal resistance.

Similarly, looking within the individual graphs things don't look too linear, either. E.g., for discharge the currents chosen are C/100 = 0.01C, C/20 = 0.05C, C/10 = 0.1C, C/5 = 0.2C, and C/3 ~= 0.333C so the steps are 0.04C, 0.05C, 0.1C and 0.133C. Running down any of the SOC lines the gaps between the current lines really don't look proportional to those steps.
Logged
Dyslexicbloke
Full Member
***
Offline Offline

Posts: 159


Blue sky thinking ...


« Reply #22 on: June 22, 2011, 07:10:21 PM »

Ahh well ..... I just proved that I am no scientist and know / understand as little as I feered RE batteries.

I worked the figures, having spent ages extracting them from the charts, and found that my hypothosis is completly flawed.
Nevermind none linear .... Plotting internal resistance from charging current, having taken OCV into account, produces a nice looking curve which moves from nearly vertical to nearly flat in a consistant sweep.
The problem is it cant have anything to do with discharge voltage drop because it would produce the bigest drop at the maximum SOC ....  banghead

Sort of obvious now I see it, given that the charging current continues to fall with a constant float.

Back to the drawing board I think .....  surrender

Al
Logged

Off Grid - Big Caravan and huge enclosed gazzebo.
300W PV 12V system.
400Ah of AGM Absolyte GP cells. (Second hand)
600W Inverter (Maplin's finest :-) )
CHP in the works - Chinese Horisontal Diesel [S195 Generic - Kukje]
VAWT testbed flying - Back to that when its warmer I think.
Justme
Hero Member
*****
Offline Offline

Posts: 3575


« Reply #23 on: June 22, 2011, 07:36:17 PM »

A banks condition is also not so simple that you can just use what its done since its last full charge to work out whats what.

Its real condition will be affected by its use over a longer period of time. Using a small snap shot of V wont help much.
Logged

Navitron solar thermal system
30 x 58mm panel 259L TS
1200watts solar 120vdc
FX80 Solar controller
2 x Victron Multiplus II 48/5000/70
Cerbo GX & GX 50 touch
BMV 700
6kva genny
48v 1000ah
Grid Possibly coming soon
EccentricAnomaly
Guest
« Reply #24 on: June 22, 2011, 09:16:56 PM »

Justme, are you talking about the effects of surface charge or something longer term?
Logged
Justme
Hero Member
*****
Offline Offline

Posts: 3575


« Reply #25 on: June 22, 2011, 09:37:25 PM »

Longer term. Like how you treated the bank last week will affect its capacity this week.
Logged

Navitron solar thermal system
30 x 58mm panel 259L TS
1200watts solar 120vdc
FX80 Solar controller
2 x Victron Multiplus II 48/5000/70
Cerbo GX & GX 50 touch
BMV 700
6kva genny
48v 1000ah
Grid Possibly coming soon
Dyslexicbloke
Full Member
***
Offline Offline

Posts: 159


Blue sky thinking ...


« Reply #26 on: June 23, 2011, 06:03:15 PM »

There is a huge amount of impirical test data in this collection of documents ...
http://www.tpscrail.com/products/gnb/PDF/Absolyte.pdf/Absolyte_engineering_papers.pdf

Its field and lab test results on Absolyte cells, lead,copper,antimoney plates, but I suspect it is true of most simmilay chemistryes.

There are interesting conclusions about what constitutes abuse and exactly what effect it will have over time.
Most of the studies were aimed at signaling or telecoms applications where periodic deep discharge was expected.

One of the studies looks at practical ways to predict SOC whilst maintaining a load although it dosnt actually recomend a circuit or equipment to do the job it dose explain how several different parameters can be utilised and what conclusions can be drawn from them.

An interesting note is that, as mentioned above, what you did in previous cycles will affect capacity and how that manifests its self this cycle.
The conclusion being that although manufacturers data may be the best way to predict SOC for ideal cells in the real world test data and observed parameters, although theoretically less accurate, will more oftern than not give a better indication of SOC for practical purposes.

When I have read all this 3 or 4 more times, which may take a while, I may actually know a little more about what I am attempting to do.

The final point is that all this data makes me wonder if a Smartguage would even work for these cells because they do seem to differ considderably from the norm ... perhaps I will send mail and ask!

Al
« Last Edit: June 23, 2011, 06:15:39 PM by Dyslexicbloke » Logged

Off Grid - Big Caravan and huge enclosed gazzebo.
300W PV 12V system.
400Ah of AGM Absolyte GP cells. (Second hand)
600W Inverter (Maplin's finest :-) )
CHP in the works - Chinese Horisontal Diesel [S195 Generic - Kukje]
VAWT testbed flying - Back to that when its warmer I think.
Outtasight
Hero Member
*****
Offline Offline

Posts: 886



WWW
« Reply #27 on: June 29, 2011, 11:13:18 PM »

A banks condition is also not so simple that you can just use what its done since its last full charge to work out whats what.

Its real condition will be affected by its use over a longer period of time. Using a small snap shot of V wont help much.

I'd second that.  If I shallow discharge and top up to full consistently, the pack holds more charge than if I deep discharge or have a run of days with deepening partial charges.  After such "abuse", it can take a run of good days to get the pack back into something like rude health.  A single good charge to "full" (pack stops absorbing charge) doesn't actually restore full capacity (and the Smartgauge seems to recognise this).  I can usually tell how much of a liberty I can take with it by looking at the power curves from the previous couple of days.  If it showed solid charging with several hours of unbroken floating then it holds more reserve power than if it has been very fragmented charging due to "sunny intervals" weather.
Logged

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

Posts: 9073



WWW
« Reply #28 on: June 29, 2011, 11:30:08 PM »

Sure  battery monitors should have their place , sure mine adjusthimself  and 100% full  does not mean full in my head
If battery is charged fast with  lots of Amps  100% state is reached too fast  , but charging goes on and monitor resets  after ongoing charge to 100 % again
So an ongoing charge and an ongoing 100% charge reset here at my home , sometimes i get 50 ah more ore even 100 ah more pumped in the battery and monitor resets to 100 %

More important is to have enough charge  Grin

Billi
« Last Edit: June 29, 2011, 11:35:09 PM by billi » Logged

1.6 kw and 2.4 kw   PV array  , Outback MX 60 and FM80 charge controller  ,24 volt 1600 AH Battery ,6 Kw Victron inverter charger, 1.1 kw high head hydro turbine as a back up generator , 5 kw woodburner, 36 solar tubes with 360 l water tank, 1.6 kw  windturbine
Pages: 1 [2]   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!