Page 1 of 2 12 LastLast
Results 1 to 10 of 14

Thread: Solar battery charging

  1. #1
    licensed to kill - FKA kev000
    Auto Apps:loading...
    tripzero's Avatar
    Join Date
    Aug 2006
    Location
    16.40618, 120.61106
    Posts
    2,494

    Solar battery charging

    This is a continuation of this thread in effort not to hijack the thread further.


    Background:

    I have an "always on" system that (depending on hw) consumes 2-10W). This system is powered by a kinetic hc-800 AWG wet cell battery. I've also got a 30W solar panel connected to this charge controller.

    I've measured the solar panel current at up to 800mA, but I notice the voltage readings hang out at below 12.6. Hooking the 10W system up, I drop the voltage to 12.3V (from 12.6V).

    I've logged the data for 4.5hrs at the link below:

    https://docs.google.com/spreadsheet/...3hKd1lvdTRQV2c

    Chart2 shows the battery voltage over time. As you can see, I don't ever drop below 12V and I seem to bounce around between 12.2V quite a bit. I poll for voltage from the dcdc-usb power supply every 30 seconds. I'm driving around as the voltage goes > 13V.

    Chart1 shows solar panel charge current. I'm missing quite a bit of data however because the logging stopped at some point (you can see from the time stamps in Sheet 1).

    Question:

    Why am I not going above 12.6?
    Can I still link my 2 batteries when the solar panels are producing n mA?
    Last edited by tripzero; 04-16-2012 at 04:35 PM.
    Former author of LinuxICE, nghost, nobdy.
    Current author of Automotive Message Broker (AMB).
    Works on Tizen IVI. Does not represent anyone or anything but himself.

  2. #2
    Raw Wave
    Join Date
    Nov 2009
    Posts
    2,118
    What's the (specified) float current of the HC800? (I couldn't find any spec).



    The SunGuard 4.5A controller has a standby current of 6mA.


    Certainly with a 10W load alone, a mere 800mA will cause a drop to the battery voltage and begin its discharge.


    And IMO, for a full-time 10W drain, 30W of panel is not enough.
    I usually at least quadruple the load power - unless perhaps good tracking and clean sun power is available, then trebling may be ok except near & between the darkening solstices.

  3. #3
    licensed to kill - FKA kev000
    Auto Apps:loading...
    tripzero's Avatar
    Join Date
    Aug 2006
    Location
    16.40618, 120.61106
    Posts
    2,494
    Quote Originally Posted by OldSpark View Post
    What's the (specified) float current of the HC800? (I couldn't find any spec).

    The SunGuard 4.5A controller has a standby current of 6mA.

    Certainly with a 10W load alone, a mere 800mA will cause a drop to the battery voltage and begin its discharge.

    And IMO, for a full-time 10W drain, 30W of panel is not enough.
    I usually at least quadruple the load power - unless perhaps good tracking and clean sun power is available, then trebling may be ok except near & between the darkening solstices.
    so with a 10W load, I won't see it above 12.6V even under the best solar conditions with that panel. With no load (except the 6mA draw from the solar controller), I should see it charging above 12.6V on a sunny day, correct?

    I suspect a lot of my problems will go away once I'm using 2.5-3W hardware.
    Former author of LinuxICE, nghost, nobdy.
    Current author of Automotive Message Broker (AMB).
    Works on Tizen IVI. Does not represent anyone or anything but himself.

  4. #4
    Raw Wave
    Join Date
    Nov 2009
    Posts
    2,118
    Quote Originally Posted by tripzero View Post
    so with a 10W load, I won't see it above 12.6V even under the best solar conditions with that panel.
    No, I'm talking about a full-time load, and using solar to replace that power.


    A 30W panel should put out 30W under good conditions. Many panels can put out more than their rating.


    However, some solar issues....
    Note that within [brackets] is extra definition or detail etc, and within [bold brackets] is incidental FYI or ramble. Both are optional (I should use a different font, or hypertext!).

    Misalignment drops output significantly. As I recall from some webbed research paper, a 2 means ~5% less; 5 is 20% less - DON'T take those figures as being correct, but it was something like that (I was surprised; it really did make tracking worthwhile!).
    [ That misalignment is from the normal - ie, panel surface perpendicular (90) to the sun, ie, pointed directly at the sun. ]

    A "12V" solar panel typically puts out 17-21V at full rating. With a linear regulator you lose a lot of that, ie, 30W out =~20 @ 12V.
    [ Viz: 30W @ 21V = 1.4A 1.4A @ 14V = 19.6W ~20W/ 1/3rd of the power is lost due to linear regulation (dropping 21V to 14V etc). ]
    IE, you can get 20W from your 30W panel, enough for the 10W load and hence should be above 13V etc assuming the battery is not drawing more than ~80mA (charging and then its fully charged float current refer).
    [ SMPS-type dc-dc converters overcome much of that, but for solar panels MPPTs are used - that's a Maximum Power Point Tracker which is not a "panel-sun tracker", but a dc-dc converter with a variable front end (input voltage) to extract the maximum POWER from the panel (it may reduce or increase its voltage to get the maximum current times voltage = power output).
    MPPTs should probably be the first step before buying another panel (for systems of say 50W and higher). ]

    For full time loads, at least double its power for solar panel rating for night & day alone. (Assumes a 12 hour day.)
    [ IE, the total Wattage used can only be recovered during daylight. That's an ~50% panel "duty cycle" - they sleep during dark & night. ]

    Then add whatever factors contribute to less than nominal/rated power, eg, sun not directly overhead (time of day and latitude), clouds, atmospheric contaminants, and of course solar eclipses (joke!). ]

    And remember the battery overhead - ie, its inefficiency [~30%(?)], and its float current.

    And some panels put out NIL power if partially shaded. (A shaded series cell blocks the other cells' output.)



    So my ROT/rough guesstimates for a 10W full-time load is x2 for day/night, & x2 for clouds & the weaker morning/evening sun.
    Hence 4x10W = 40W.
    Mind you, that assumes a bit of tracking - NOT being in a fixed position on a car even if it does align with the sun at some time.
    [ Hence why I think solar panels for car battery trickle charging are not worthwhile - they are too expensive for their size, and people don't realise what little power they get from them (especially behind windows!).
    Besides, with batteries having a typical 200mA to 1A float current, what's the point of a 5W panel (~500mA max)? It could all go to float, though it might charge a little when above the float current (ie, battery not full).
    However, as a means of preventing sulfation when the vehicle/battery is not being used (ie, not charged), it's a great idea, provided it has a regulator (since it's a long-term connection, and anything above 14.4V means gassing). ]


    But if the panel is in a fixed position and not tracking, or behind windows etc, or now and again shaded by trees, or rainy days... then 40W times what?


    { FYI - for my camping or "self sufficiency" purposes, I factor another x3 for a tracking system to allow for 2 days of non-sun.
    My fall-back is to charge off the car alternator which generally can easily supply whatever the battery accepts.
    [ I so often read how they can't because "alternators are made for to minimum cost". Well people, they have to handle lights and wipers... etc.
    For normal running, alternators should have quite a bit of reserve, not just "a bit more" than the "normal" vehicle needs (ie, without lights & wipers, though I'd suggest most are designed to reasonably cover lights, wipers, electric engine fan, heater & fan, and brake light in peak hour traffic - ie, near-idle speeds. ] }


    And just for your general FYI - the maximum sun energy is ~1kW per square meter at the Equator.
    Factor in solar panel efficiencies of say 25% to 40%, that's only 250W to 400W per square meter at the equator. I have seen some amazing claims of higher than 1kW/sq.m output (concentrator system excluded - ie, reflectors & lenses).
    So for a 10MW terrestrial solar power station, you can ask what "at least 1hec" (2.2 acres) of vegetation is to suffer - more like >5hec (10 acres) if photovoltaic panels, then x 2 for day/night, and x? for weaker sun, rain, shorter days.
    Just be thankful our domestic solar grids do NOT use batteries!!!


    Geez, another -101 lesson. This time my "homebrew" solar and alternator design opinions...


    Stop after the next line if not wanting the incidental FYI etc....
    But yes, the key with any solar (and battery!) design is to minimise the loads.
    [ That's why I stopped laughing at people that spent $1500 on 40L car fridges when I got one for $150 (a Peltier cooler).
    But instead of spending another $1,000 on panels and batteries (for a still limited reserve time), I spent that on an expensive fridge so I can now laugh at myself.
    The load has now dropped from ~6A-8A max to 2.5A max (usually 1.5A max).
    But the "duty cycle" is way less - an estimated average 3-4A down to probably under 0.5A - say 5 to 10 times less.
    And the bonus - I can now cool to -19V (-2F) even in ~45C/115F) heat compared to (up to) 20C/36F below ambient.
    Icecream in a tent in the outback, and cold beer for my girlfriend! We should both be happy. Should. ]

    Incidentally, my 100W "panel" is an array of 20W panels. I consider it quite likely I'd use one 20W panel for occasional vehicle battery charging, maybe 2 panels (or more?) if reasonably discharged etc. (That's if I have a "no engine running" policy, or a dingo steel some engine part since I'm probably engine-overhauling somewhere...)


    Not that it is relevant here, but on another forum is someone that - on the cheap - wants high camping power reserves. Admittedly the microwave and some other loads are now gone, but there is still a (24V) welder.
    Now batteries are not "power supplies", or rather, they are a very expensive source of power, and there comes a point where generators are the real solution - and are often far cheaper (with essentially infinite reserve time if you keep refueling).
    That expense is often not apparent at first - even if they only spend $hundreds on batteries. But come the time for battery replacement... (especially if all were paralleled and one bad battery has killed the others!). Yep, even I bought a fleaBay 2.5kW Chonda generator (admittedly for my gf's electric chainsaw whilst collecting roadside firewood, though I might rig its alternator to my vehicle instead for long-term use..).

    Yeah, sorry, more ramble. (For the big-picture of course!)
    Last edited by OldSpark; 04-17-2012 at 10:53 PM. Reason: fixed bolding; minor edits. Geez this reply is bad...

  5. #5
    North of the land of Hey Huns
    Auto Apps:loading...

    Join Date
    Jun 2004
    Location
    Westminster, MD
    Posts
    2,127
    FWIW, most solar charge controllers (including trip's) are not linear regulators, and the voltage drop across the PWMed mosfet is the only lost energy.
    "stop with the REINSTALLS, what do you think we got some lame-o installer!!!" - mitchjs
    RevFE
    My Shop

  6. #6
    Raw Wave
    Join Date
    Nov 2009
    Posts
    2,118
    Good point, but a PWM is not an SMPS.
    And I was considering it from the "lost power" POV, and PWM still has the same lost power - it just isn't given off as heat (because it doesn't conduct when OFF to "average" out the Vdc delivered.

    Only the "dc-dc converter" (meaning SMPS types) and MPPTs overcome that "voltage dropping" lost power.

  7. #7
    licensed to kill - FKA kev000
    Auto Apps:loading...
    tripzero's Avatar
    Join Date
    Aug 2006
    Location
    16.40618, 120.61106
    Posts
    2,494
    A few more details now that today has some sunshine...

    400-600mA is typical current from the controller to the battery in the back window of my car. I moved the panel outside and angled it towards the sun and I started getting 1610mA and the battery voltage went up to 12.9 (from 12.6). I think if I"m going to get anywhere, I need to somehow mount it outside...

    Also, apparently just moving the panel at an angle instead of lying flat while in the back window added about 200mA (700mA +).
    Former author of LinuxICE, nghost, nobdy.
    Current author of Automotive Message Broker (AMB).
    Works on Tizen IVI. Does not represent anyone or anything but himself.

  8. #8
    Variable Bitrate mayhembdm666's Avatar
    Join Date
    Mar 2008
    Location
    Adelaide, Australia
    Posts
    233
    Remember that solar under a standard window will not perform as well if at all at times.. Tinted or non tinted.

    I've also been curious about the idea lately of trying to use some solar panels as a replacement for external plastic strips/sections of the car.

    For example the side skirting and some cars having the large dark plastic sections between the tail lights.
    Last edited by mayhembdm666; 04-21-2012 at 07:44 PM.
    2004 Holden WL Caprice Auto GENIII
    Base System = Raspberry Pi
    Everything else is pending for now as switched from a Mini-ITX setup

  9. #9
    Raw Wave
    Join Date
    Nov 2009
    Posts
    2,118
    That's 3 to 4 times the output.
    And a good example of what difference the angle makes, and how much is lost thru glass etc - especially car windows. Unfortunately I only wrote "But if the panel is in a fixed position and not tracking, or behind windows etc, ...". You'd think I'd add a few words like "car windows can halve panel output". (Just keeping my replies short I guess. )


    So going from [email protected] to [email protected] (still only 20W, 2/3rds its rated "30W" output) tells me the load is somewhere between those 2 currents - probably closer to 1.5A.
    That's likely to be your battery's float current unless other loads are also connected, and unless the battery isn't fully charged in which case that voltage will increase over time.

    But at 12.9V, the battery is definitely charging, albeit slowly. And I'd guess not enough to replace lost charge on a permanent basis.


    Note that 1.6A @ 21V = 33W. IOW, if it is rated for 30W@21V, an MPPT could extract that extra power, minus whatever power the MPPT consumes.
    But a bigger panel may be cheaper, especially if a 30W rating still isn't enough for what you want (eg, my x4 rule).
    Last edited by OldSpark; 04-21-2012 at 08:41 PM. Reason: trivial edits

  10. #10
    licensed to kill - FKA kev000
    Auto Apps:loading...
    tripzero's Avatar
    Join Date
    Aug 2006
    Location
    16.40618, 120.61106
    Posts
    2,494
    Need to find a way to get the panel in the sun. I've determined the best way is either a roof rack mount, or to replace my sunroof with a custom fiberglass sunroof where I can mount the solar panel. Who is the resident fiberglass expert?
    Last edited by tripzero; 05-08-2012 at 01:24 AM.
    Former author of LinuxICE, nghost, nobdy.
    Current author of Automotive Message Broker (AMB).
    Works on Tizen IVI. Does not represent anyone or anything but himself.

Page 1 of 2 12 LastLast

Similar Threads

  1. Solar battery charging
    By tripzero in forum Power Supplies
    Replies: 15
    Last Post: 04-12-2010, 06:49 PM
  2. 2nd battery charging
    By richard888 in forum Newbie
    Replies: 2
    Last Post: 03-30-2009, 06:01 AM
  3. bad to run carputer while charging battery from AC?
    By jrtf83 in forum Power Supplies
    Replies: 2
    Last Post: 09-21-2006, 10:33 PM
  4. Battery / charging probs
    By RichardB in forum Power Supplies
    Replies: 6
    Last Post: 04-18-2006, 10:31 PM
  5. Second battery charging.
    By mindracing in forum Power Supplies
    Replies: 1
    Last Post: 04-18-2005, 04:50 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •