 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Electronics : Multifunction Picaxe Charger - need help
| Author | Message | ||||
| boelle Newbie  Joined: 02/01/2011 Location: DenmarkPosts: 13 |
Hi To be the nice man I am, I started my own thread. I got the idea of building my own charger when i saw gill's mod of gizmo's charger. It's close to what I need and then my brain started to spray out a few addons that could come in handy. My main problem is that I'm not very good a designing so I will need all the help i can get. "Why not just buy something off the shelf?" I can hear some say. Well first thing: most cheap battery chargers are simple models with slow/rapid charge option. They just stop charging, if you are not lucky to have the "boiler" model that just keeps charging. Next thing is that stuff like trickle charging and amphour tester would add lots to the price tag. And its more fun to build yourself. Functions I would like to have in the charger: *Charging of course, but with an option to have it adjusted to battery's current temp. This of course mean that a probe needs to be fixed to the battery. PWM should be used so the charge tappers off when reaching float charge. After that it should just trickle charge so it can be left connected without worries. here the display should just read voltage and amps that go in the battery. maybe state of charge as an % number. when at 100% it should trickle charge and beep every 1 min to say its done. *Amphour test. For this my idea was to charge the battery full then discharge it and count amphours finishing off with a charge. Only need to figure how big a load is needed for this number to come up so you can trust it. Display could start at 0 and count up amphours until it reaches the lovest healthy voltage for that battery. *Load test. Load with very high amps for like 10 secs and see if it drops below 9.7V from a full charge and if it jumps up at least 50% after load is removed. The display could just say passed / not passed plus maybe lowest voltage messured. *Desulfating. I allready found a good schematic for that here: http://home.comcast.net/~ddenhardt201263/desulfator/schemati c.gif The idea is to charge to float and then switch in this and let it run. Only need to monitor voltage and change to charge to keep it above 11V. For every X number of hours it should do an amphour test to see if amphours increase. If it does not for like 3 desulfating runs the routine should be stopped and display start amphours and how many amphours gained and resulting amphours. This is the ideas I have for now. Logging would be nice but I think that should be left out for now. For the hardware side my ideas are: One box that holds the display and picaxe etc... And another close to battery that holds dump loads etc to keep the heavy wires short. The boxes could be stacked and locked together or with a multiwire cable between to make the setup more flexiable. The part with dumploads need a good cooling fan. For the dumploads my plan is to use car main light bulps to make it simple, the heat generated are removed by the fans, one bulp on low beam for the amphour test and a lot more on high beam for the load test. Off for some thinkering... Upps, forgot to say that the base of the charger is a simple trafo from the boiler model i have... it can give 17Amps should be plenty |
||||
| kevindion Regular Member  Joined: 28/07/2010 Location: ChinaPosts: 59 |
i hate electronics. |
||||
Gizmo Admin Group  Joined: 05/06/2004 Location: AustraliaPosts: 5024 |
Hi boelle Big project you got going there, lots of functions. I thinks thats why kevindion said what he said, its going to be a lot of work to get it all working, but it could be done with enough effort and time. One tip I have, for the dump load, instead of car head lamps, use those 50watt 12v down lamps. They are much much cheaper. Another thing to conside with lamps as dump loads, is they have a high inrush current as they light up, maybe 3 to 4 times higher than their normal rated current. A 50 watt lamp can have a inrush current of 12 to 15 amps, before settling back to 4 amps after a few 100mS. http://www.thebackshed.com/windmill/articles/200wDumpLoad.as p Or you could look at making your own encapsulated resistors from stove elements like Pete did http://www.thebackshed.com/windmill/articles/BuildingEncapsu latedResistors.asp. It works well and is easier than you think. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
i have started to hack up your charger that gill mod'ed to better understand it would this be what i need to charge? http://billedeupload.dk/photo/2011-01/4c9ae2ab/charge%20part .JPG.html |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5024 |
Hi boelle I'm not sure what it is you're trying to do with that circuit. Could you tell us what you want it to do and maybe draw more detail. The transistors are not needed, a PicAxe can drive the mosfets directly, and not sure what the 18v zener is there for. I would suggest you try building a working circuit before attempting to modify it, sometimes thats the best way to learn how something works. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
Downwind Guru  Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
What frequency range do you intend to operate the mosfets at to achieve trickle charge and charge control etc. This will make a big difference to how you setup the driving of the mosfets. Test these functions before going to far with the design, as i think you will find problems. Tip.....Stock up on lots of mosfet spares. Pete. Sometimes it just works |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
firstly i'm trying to understand the curcuit, I found a mod by Gill and he added PWM to your charger so it could tapper off slowly. But as i'm not that good at designing i have to split it up in smaller blocks to understand it. |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
as for the trickle charge.. my main idea is a low freq so that the battery will get a few long burst's of charge to keep it topped but without boiling it. for normal charge i would start out with the fet's on all the time and slowly lower the freq as float is reached. |
||||
powerednut Senior Member  Joined: 09/12/2009 Location: AustraliaPosts: 221 |
The datasheets for the mosfets you select should give you information on how fast they can switch, which will give you your frequency limits. Make sure you select logic level mosfets (I think the ones gizmo used are), that way you wont have to worry about driving them with 12 Volts. You can drive the mosfets directly from your PWM output pin on the microcontroller. Just connect the gate pins to a 1k resistor and then to your PWM pin. If you want to just run it flat out, in a boost phase for example, then turn off the PWM and then drive the pin high as a normal output. Gizmo's charger can be used almost as is with PWM - or at least the first version can. I'm going to talk about this schematic, which is his first version. The far right pair of mosfets are responsible for connecting the charging power source to the batteries. In this case all you would have to do to have PWM control of the charging is have the PICAXE's PWM output pin connected to the lower optocoupler instead of Pin 15 (which may be the PWM output - i'm not familiar with PICAXEs). You would then have to write the code to make the microcontroller do PWM output. You can simplify a lot if your using a mains powered charger as your source instead of the wind turbine - you don't need the dumpload at all, just drive the mosfets low and it will look to mains charger like you've disconnected the battery. Most chargers handle that fine. The wind turbines need the dump load because they will probably overspeed if they do not have a load. This causes physical damage to the turbine. Solar panels and mains chargers shouldn't require the dump load. *Edit* damn i'm an idjiot. You need the dump load to do the load test etc.. As a side note - does anybody know why Gizmo used optocouplers to isolate the mosfets from the PICAXE? I'm not sure its necessary, but I've yet to build one thats dealing with a wind turbine... |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
thanks a lot for that, it makes the picture a bit more clear to me.... but yes i need the dump load but connected via a relay on its own so the pic can control it seperate. This is a long project for me but i need to understand the basics first, if i started out by experimenting blind it would tip me over and i have a bad habbit of just ditching projects that does that. would be a waste of money. But when i get it the big plan is to make a veroboard/stripboard layout and release the whole thing for people to copy as much they care. |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
i think the opto's in gizmo's design are there for extra security just in case something goes relly bad in the fet's. for the dump load my plan is an external case that holds current sensor and still i tend to opt for simple light bulps to do it... a bit inrush current do not matter that much since they will mainly be used to put a heavy load for 10 secs... or a few bulps to discharge. should they fail they are easy to replace.. any petrol/gas station will have spares 
The plan is to have two heavy copper bus bars in the buttom and the bulps connected here via relay's so that the load can be choosen by activating the right amount of relays... it's simple and dirty but should be 99% foolproof.. hmm... is there not a law from murphy that says its impossible to make things fool proof? 
at the end of the bars a pair of heavy battery clips like those on jumper cable's the charger and board with picaxe will be in a case much like the first one to they can be stacked or keept a bit apart, makes the setup a bit more flexible. connection between the 2 will be either a D-SUB 9 or 25... and maybe anderson connector for the charge wires... not sure yet but having it in the back of my head i have redone the charge part here, please tell me if there are any faults in my drawing: http://billedeupload.dk/photo/2011-01/08636e5d/charge%20part -1.JPG.html |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
for current sensing my plan is to use ACS758ECB-200B-PSS-T from allegro... seems to handle enough current and does not require any external components other than a cap. |
||||
| powerednut Senior Member Joined: 09/12/2009 Location: AustraliaPosts: 221 |
Its closer - you need a resistor (10k-15k or so should be okay, i think) between the source pin and the gate pin on your mosfets, and it would be a good idea to use a seperate 1k resistor from each mosfet's gate pin, rather than the single one you've got drawn. The source to gate resistor will stop the mosfet's from operating until you pull the pin high with your picaxe. I don't think you'd need the relay for your dump load - not unless your planning on loading the battery up to a couple of hundred amps. Mosfets will be cheaper than the relay. |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
hehe... the dump to discharge does not need to be via a relay... but the load to test if the battery can keep up with the load that happens when cranking the engine need to be 150 amps... i read a lot of good info here: http://www.civicforums.com/forums/109-7th-gen-do-yourself-ar ticles/307199-classic-diy-alternator-battery-faq.html that page did not sound like black magic but very reasonable if you have played arround with batterys a bit |
||||
| powerednut Senior Member Joined: 09/12/2009 Location: AustraliaPosts: 221 |
btw from Gizmo's advice in my charger thread the 1k resistors to the micro are a mistake - they'll slow the switching on the mosfets (really bad for PWM). He suggests 100ohms or less. |
||||
| boelle Newbie Joined: 02/01/2011 Location: DenmarkPosts: 13 |
me sitting an thinking here... i guess the 4 pos swith could be removed and place and extra START button next to the menu and up/down ones.... would make it a bit more simple i guess, could be wrong but done with schematic for the charing part, desulfating done almost as it's a copy from a old magazine... so think i will move on to the stuff close to picaxe |
||||
| powerednut Senior Member Joined: 09/12/2009 Location: AustraliaPosts: 221 |
Yup, you could do that - the 4 pos switch doesn't seem control anything directly, other than power on/off to the microcontroller. If you want to handle the other functions you want through a menu option that should work fine. Probably a good idea given how many functions you want to add to the controller. |
||||
 Print this page |
