 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 : Charge controller help
| Author | Message | ||||
terrapinlogo Newbie  Joined: 26/10/2009 Location: New ZealandPosts: 21 |
I have built a diy charge controller from the schematic attached and it will not charge the battery. I have it connected to a 12v supply and when I short the header (instead of adding buttons) to force it to charge it only turns the dump led off and not the mosfet and it doesn't turn the charge led on. The blue and red wires go to the relay and the other two go to the battery. Any ideas? Thanks Damian 
If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
| Robb Senior Member  Joined: 01/08/2007 Location: AustraliaPosts: 221 |
What voltages do you get on pins 4, 5 & 6 of IC3? Do they change when pressing "buttons"? |
||||
| terrapinlogo Newbie Joined: 26/10/2009 Location: New ZealandPosts: 21 |
pin 4 is at 0v all the time. pin 5 is at 5.8 when there is no jumper and when the jumper is in charge position, it is at 1.95v when the jumper is in the dump position. pin 6 is at 1.96v with no jumper 5.88 with the charge jumper and 0v with the dump dumper. If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
| Robb Senior Member Joined: 01/08/2007 Location: AustraliaPosts: 221 |
Well if you can get a low on both pin 5 & 6 at once pin 4 should go high. More info about NOR gates |
||||
| terrapinlogo Newbie Joined: 26/10/2009 Location: New ZealandPosts: 21 |
After discovering I have a p channel irf9540 instead of the n channel irf540 i measured the voltages again and got 4.9v when the jumper was in the dump position, but the led still wont light and the mosfet doesn't turn the relay off/on. Would an irf840 work as I have heaps of them lying around. I know it's overkill but if it saves me spending another 8.50 then I'm happy. thanks Damian Edit: I have the led going! There was a direct short to ground. The lower variable resistor had a little piece of solder between its top pin and the led track. The mosfet still won't switch which has me scratching my head as it should still switch I thought, just in reverse. If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
| davef Guru  Joined: 14/05/2006 Location: New ZealandPosts: 499 |
For testing you could manually connect the 10K resistor on the gate of the FET to the regulated supply. This should turn the FET on. I can't read what voltage the regulator is. Do you know what the guaranteed turn-on voltage for IFR540 is? 5.8 volts should turn most FETs on, but check. Dave |
||||
| terrapinlogo Newbie Joined: 26/10/2009 Location: New ZealandPosts: 21 |
It's an 8v regulator. As for the FET I have no clue what half the specs in the data sheet are but Vgs is +\- 20v and Vgs(th) is 2-4V. Is one of them the value I am looking for as they are the only two apart form the 500v Vds which is just the maximum voltage is it not? Connecting the 10k on the irf9540 to the supply would turn it off as it is an p channel FET. If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
I dont see how you could expect to replace a N channel fet with a P channel fet and everything still work, without doing circuit changes to suit. Use the right channel fet and things might work as they are ment to. I see little purpose in asking why it wont work if you subsitute parts without knowing what you are replacing things with, or we dont know what is different than the schematic. 99.9% of the problems folkes have with building controllers and them not working is due to solder bridges, mainly with vero board as you have used. The other problem has been wrong parts used. It looks that you are lucky and have both these problems 
Give the bottom of the board a scrub up with some metho and an old tooth brush to remove any resin and gunk, then let it dry and check the circuit for solder bridges. You are also a game man not using IC sockets for the chips, i once done the same but learnt fast the cost of IC sockets was well worth the use of them. Pete. Sometimes it just works |
||||
Bub73 Senior Member Joined: 10/12/2009 Location: United StatesPosts: 116 |
I don't know about the irf540 but the irf510 takes 10 volts to turn on; so study the data sheets when making substitutions carefully. Also as Pete said above use the sockets for the ic's and save yourself grief and money. Bob |
||||
| terrapinlogo Newbie Joined: 26/10/2009 Location: New ZealandPosts: 21 |
Yes I know I cant substitute a p channel for an n channel. There is suppost to be an IRF540 on the board but I miss read the markings and put a IRF9540 in which is the p channel equivalent of the IRF540. I have a big pile of IRF840 n channel mosfets which i want to know if they will work as that will save me buying an IRF540. I will compare the datasheets and see what I can find. If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
| Robb Senior Member Joined: 01/08/2007 Location: AustraliaPosts: 221 |
I just grabbed and IRF840 and stuck a 12volt 21 watt break lamp in as a load. It seems to saturate at about 6-7 volts. Don't forget your only driving a relay. If its not that bigger relay why not just put in a small NPN transistor and remove the 100k to ground. An IRF 840 is like using a truck to carry a feather if your relay has a 100mA coil current  |
||||
| terrapinlogo Newbie Joined: 26/10/2009 Location: New ZealandPosts: 21 |
I have just clipped an IRF840 in with croc clips and it works. Just going to solder it in now and see if it works. Thanks for all your help. Damian If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
> Vgs(th) is 2-4V. This is what you want. > Connecting the 10k on the irf9540 to the supply would turn it off as it is an p > channel FET. Correct, connecting the 10K to supply would keep it off as well as connecting it to ground. The gate would need to be negative w.r.t. ground to turn a P channel FET on. I was referring to the IRF540, which I now realise you never had  |
||||
| terrapinlogo Newbie Joined: 26/10/2009 Location: New ZealandPosts: 21 |
An easy mistake to make with all the part numbers being similar. It was going for about 5 mins but then I turned it off and on again and it doesnt work properly now. Something to do with the variable resistor connected to pin 5 of the comparator. You turn it to on end and you get 6.2-8v and the other and you get 0-2v depending on which end you start at. Those voltages cover 90% of the range depending on the end you start turning from. Which means I cant get the 5.95v required at pin 5 to set it up properly. It is probably a short somewhere bu I'll be damned if I can find it. Damian Edit: Fixed it. I had totally overlooked a track cut. Pin 5 of the comparator was shorted to pin 7 of the comparator and pin 6 of the nor gate. Hopefully this time it stays going. If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
| Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
Ahrrr a construction error, it would put you in the same league as the rest of us. We all make mistakes, as you have found finding it is the challenging part. Good to see you didnt give up and isolated your problems. Better still its great you posted your mistake, as often many of us give lots of thought to the problem and never get to know what the problem was, and if it was ever solved. Glade to see a positive result and sure you have learnt lots along the way. 
Good result. Pete. Sometimes it just works |
||||
| terrapinlogo Newbie Joined: 26/10/2009 Location: New ZealandPosts: 21 |
I have the board all going now but the turbine is strong enough that it pushes the voltage that the charge controller sees up to 15v when the battery is still at 12v. This is causing the relay to click on and off and sound like an angry blow fly. Any ideas on how I could add some sort of time delay so that the relay can only change state every minute or so. I'm thinking some sort of capacitor and resistor network but I have no idea on how to hook it up. If it doesn't work hit it with a bigger hammer. If it still doesn't work you probably shouldn't of hit it. |
||||
| Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
What this normally means when you have relay chatter is the hysteresis is not wide enough. To fix this you need to lower the cut back in voltage on the trimpot For example the cutout (high) set to 14.0 volts and the cutin (low) set to 12.8 volts. What this will do is make the dump stay on once tripped for longer until some power is used from the battery and then it will let the controller reset to charging. At present as the voltage goes up the relay trips then the voltage falls a fraction and the relay switches off and the volts go up and the relay switches on again in a cycle...this will kill the relay REAL FAST....NOT GOOD!!!! Just lower the cutin setting (LOW) till you get the relay to hold in, the battey voltage will start to fall back to resting voltage and your low point needs to be a fraction below resting voltage or very close to resting voltage. Pete. Sometimes it just works |
||||
| tytower Newbie Joined: 19/11/2017 Location: AustraliaPosts: 21 |
Just a small quote from Loius the Iphone /Ipod repairer on UTube A P channel needs the gate to go low to conduct (remember p tail is below line) An N channel needs the gate to go high. When its open current flows from Source to the Drain. |
||||
| tytower Newbie Joined: 19/11/2017 Location: AustraliaPosts: 21 |
deleted Edited 2020-02-24 07:44 by tytower |
||||
 Print this page |
