6 VOLT 1Ah SLA Charging problem

lbenson

Senior Member
As I see it, your NPN mosfet would go between Q1 and the return to 0V, parallel to the 0V return from your device attached to H1 (and the battery). There may be reasons why this wouldn't work with the L317, but I am too inexperienced/untutored to know about them. More expert advice welcome.

How does D2 work?. Why isn't battery 0V connected to wall wart 0V?

How are you sensing when the battery needs to be charged, or are you assuming it will just trickle charge?
 
Last edited:

AllyCat

Senior Member
Hi,

First you need to delete (short-circuit) D2 because it prevents the battery from charging.

The FET needs a resistor from its gate to the supply rail. With an N-channel, the source will be a few volts below the supply rail (by the gate threshold voltage) which may be ok with 12 volts input, but nogood with a 9 volt rail (so a P-channel would be needed). In either case connect the gate also through R5 to the collector of an NPN transistor, with emitter to earth and base via a resistor to the PICaxe control pin.

But a FET is not needed. You don't show the value of R4, but I'd guess a few kohms. Just take a resistor of about 5 times R4 from the base to the PICaxe control pin. If the pin is low, the charging current would be increased slightly (increase R3 if necessary) but if it is set to float (an input) then the current will be unaffected. When the PICaxe pin goes high it will turn on the transistor and reduce the LM317 output, so that it no longer charges the battery.

Cheers, Alan.
 

ZOR

Senior Member
Many thanks Ibenson, Alan. I have removed D2, silly mistake. Alan, you say an FET is not required, that's good. So I have changed my circuit to hopefully suit. Sorry about component values missing, just getting my head around using Diptrace.

PIRBOXPOWER.jpg

Hope okay.
R1 - 240 ohm
R2 - 2.2k pot
R3 - 5.6 ohm
R4 - 100 ohm
R5 - Not selected yet to avoid future bang

I will use 9 volt rail now for sure. Again why can I not delete diagram below showing 12 volt rail.
 

Attachments

techElder

Well-known member
ZOR, on the forum page at the top, go to Settings. On the left near the bottom is Attachments.

Check the box on the right for the item to delete.

Scroll to the bottom, on the right, is Delete Selected.
 

AllyCat

Senior Member
Hi,

R4 is unnecessarily low, I'd use 1k, 2k2 or 4k7 with R5 at 4k7, 10k or 22k respectively. However, C.3 on a 14M2 (and Leg4 on most M2s) is "input only" so you can't use that particular pin to switch off the regulator*.

* Well, strictly you might, by increasing R4 to about 6k8 (R5 low value or not needed), and activating the "Weak PullUp Resistor" inside the PICaxe. That actually has an advantage that you can't accidentally activate a "low" output (which increases the limiting current by about 20%). Generally, the WPU seems to be about 32k, but the data sheet specifies a very wide range, so it's not guaranteed to "always" work.

Cheers, Alan.
 
Last edited:

ZOR

Senior Member
Thanks Alan. R4 value was taken from the earlier reference circuit. Sorry C.3 won't be used, was not thinking when using schematic drawing. I will try 1k and 4k7 respectively and try when I get back. Thanks again.

Thanks Texasclodhopper, that's solved the mystery. I saw it dissapear when I used delete after selection, but came back.
 

ZOR

Senior Member
Thanks Alan, made R4 1k, and with a 3.6k makes the 2N2222 switch off and charge current is zero.

I must try and reduce the battery voltage so I can see what maximum current could be drawn. At the moment the battery is up to 6.30 volts and reducing charging current at present 10.7ma. I need to find a value for when the battery is fully charged. No experience with SLA batteries, can I choose a regular maximum battery voltage or would it reduce life.

So far very pleased with your help and results.
 
Last edited:

lbenson

Senior Member
So far in reading about LM317 SLA chargers, it appears that part of the feedback to the "ADJUST" pin on the LM317 is intended to cause the output voltage to decrease as the battery voltage rises to the desired level, so that the output falls to the trickle state when the battery is fully charged. If this is the case, is there any need to be able to turn off the LM317 using the picaxe?
 

ZOR

Senior Member
Thanks Ibenson. If that were the case would make life even better. I found discepencies between my cheap multimeters readings, however my charging is well below 100ma, and it does fall right down as the battery is charged. Leaving it in this state would always leave the battery good in the event of an untimely power cut. Where did you read this? Thanks again
 

lbenson

Senior Member
A full explanation seems hard to come by. I read this site regarding charging 12V SLAs with the LM317: http://www.circuitstoday.com/battery-charger-circuit-using-lm317

About three quarters of the way down in the comments section, the author says, "... without battery set the output voltage at 13.8volt with pot for 12 volt battery for normal charge. set the voltage at 14.5 volt for boost charge. as the terminal voltage trying to reach the set voltage the charging current will fall down automatically. this setting will suit only 12volt lead acid batteries."

I interpret that to mean that the rising battery voltage causes the LM317 output voltage to drop to the trickle level. More info would be good.
 

ZOR

Senior Member
Thanks again. As you say it needs more information. Certainly don't want 400ma charging current. I think I will stay as is because the battery seems to be good at holding it's voltage when not connected and will be that way most of it's life. So I will play safe. Just looking into using ADC on voltages higher than 5 volts. Regards
 

ZOR

Senior Member
A question on recharging battery. It's easy to tell when the battery voltage has dropped enough to recharge, however when charging, the battery voltage is clouded with the voltage coming from the recharging circuit. Although creeping up in voltage my question is should I look at the charging current falling to a value or wait for the final battery voltage to be x.

I am finding that if I run the battery down a bit, and recharge, the voltage goes up to the original voltage quite quickly but then drops rapidly again when charging taken off (with a 10ma load on the battery) So maybe I am not charging it fully looking at a false voltage value? So maybe I should be looking at when the current drops to a value. Monitoring current?
 
Last edited:

ZOR

Senior Member
Thanks for the consideration. I will be reading your threads. Will be good for future reference. Regards
 

hippy

Technical Support
Staff member
If you are using a typical constant voltage, current limited, charging system; when you first connect the battery it will have high voltage, high current, as it reaches charged the voltage stays the same but current drops.

You therefore have to monitor current drawn to see when charged.

If the charger drops the voltage to deliver a floating charge voltage, the charger itself is monitoring the current and you can use the lower voltage to tell it is fully charged and now on trickle charge.

Charging voltages have to be higher than battery voltage to cause current flow into it. Battery voltage will then drop to its actual voltage when the charger is disconnected. If the charger disconnects after trickle charging then you could use that voltage drop.

Measuring current works in all cases, measuring voltage depends on the charger used. You may have measure additional things to determine if the battery has actually been charged and it's not just mains power for the charger having gone off.
 

ZOR

Senior Member
Thanks hippy. I think as I have the circuit with the LM317 earlier, producing a fixed voltage reccomended and current limitation, I will leave charging circuit on all the time unless an easy way to disconnect when current falls below X. Another issue in my head is the battery is 6 volts, but measured new at 6.47 volts, so what is true maximum charged voltage or does it peak when no more current is drawn.
 

ZOR

Senior Member
Think I've twigged it. I can measure (think) the voltage accoss R3 with ADC as drops when current reduces. Only problem is voltage maybe too low for Picaxe ADC to read? < 0.1 volts. Can I do any multiplication to bring in line with something to monitor? Also am I on right track Y/N?

PIRBOXPOWER.jpg

R1 - 240 ohm
R2 - 2.2k pot
R3 - 5.6 ohm
R4 - 100 ohm
 

hippy

Technical Support
Staff member
Another issue in my head is the battery is 6 volts, but measured new at 6.47 volts, so what is true maximum charged voltage or does it peak when no more current is drawn.
The battery voltage will peak when fully charged, and this should be when the current has reduced to a minimum.

A battery is only nominally 6V or 12V, is constructed from cells which add up to make the final output voltage. I don't know the exact values and it depends on the battery itself but one can expect 13.8V from a 12V SLA, 6.75V from 6V, perhaps higher, when just charged.
 

ZOR

Senior Member
Thanks hippy, I will just let it carry on charging and see the point where the battery stops drawing current or reaches a minimum. Then I will hopefully be clearer on setting its limits.
 

ZOR

Senior Member
Another string to my bow. Just received this item.
http://www.ebay.co.uk/itm/201310245111
Has anyone used it. No information as to which preset does what, 1 for variable voltage/set, 1 for variable current/set, but nothing about the 3rd preset.

I set my required voltage out, then put my meter in 10amp current mode accross the output but unit got very hot and I could not adjust current reading on my meter but the leds on the unit went down to one green. My variable power supply got very unhappy???

EDIT: Don't worry not going to use them. Had 3, cannot control output current, gets dangerously hot. Duff Ebay Chinese product
 
Last edited:

hippy

Technical Support
Staff member
I do not think you can measure current by putting the meter across the output. The meter is ideally a zero-ohm load in current measuring mode so you are just shorting the output, and it's perhaps no surprise things got hot. You need to measure current into a load.
 

ZOR

Senior Member
Thanks hippy. I thought it was suicidal to do, but it says

Use the multimeter in 10A current scale to measure output short-circuit current, and adjust the current potentiometer to make sure the output current to the expected charging current value
Think one of my meters went to heaven!

Going to stick with LM317 circuit.
 

ZOR

Senior Member
Many thanks for the information/help.

Unfortunately I don't speak French but I think it still has the instruction:

Utiliser le multimètre en 10A échelle courante pour mesurer la production de court-circuit, et régler le potentiomètre de courant pour vous assurer que le courant de sortie à la valeur attendue courant de charge
 

ZOR

Senior Member
Thanks Flenser, interesting. I hoped he would go through current setting method just as confirmation to it being wrongly described in page. However from that video it looks like there is a PCB track in series with the output that might/could be used internally to see current. Maybe the three boards I have are fake one's. Just waiting for Ebay supplier in China to comment.
 

Flenser

Senior Member
Zor,

If the current limit pot just happens to be at it's max value than the procedure described on the ebay listing seems a bit suspect to me as it would mean you could pass the max current for the chip (listed as 4A). These LM2596 converter boards seem to have a reputation of the advertised 3A rated current being very optimistic for the chip with no heatsink and I would expect it to get very hot if the short-circuit current was 3A or higher.

Have you tried turning the current limit pot all the way counter-clockwise to set the lowest current limit before doing the step of connecting the multimeter in 10A current scale to measure output short-circuit current?
 

ZOR

Senior Member
Hello Flenser, yes I did turn all pots counter clockwise first. I have just read about another similar device where a user was explaining to another how to set that boards current out. And again you short the output with a meter set at 10amps and then set your desired current after previously setting voltage output. The Ebay seller is refunding my money, doesn't want returned. I am waiting for another device to arrive to try, but have more faith in the LM317. rEGARDS
 

besupreme

New Member
For a regulator with unknown internal circuit, I try it first with a load resistor. Any 12V/5W..21W lamp will do. Some of these cheap modules have a load limit that you can adjust from "explode mode" to almost zero, and voltage limit typically between 1.2V and very, very much :) Resistors cost money in china, too, and board space.

I found that charging NiCd or lead-acid batteries with fixed voltage works much better with a temperature coefficient. I built a power tool charger sensing cell temperature, which is described here strippenstrolch-stammtisch: Temperaturkompensierter NiCd-Akkulader für 18V power tools in german. I intend to build a similar charger for 12V lead-acid batteries for my car, although I may rather use a specialized chip like the UC2906, which has temperature compensation for lead acid batteries built in.
 

ZOR

Senior Member
Thanks for warnings and suggestions. I think the boards I tried first had strips of copper track in series with output and the circuit looked at the voltage accross it. However it took out my meter, and I should have put in a series resistor. Never mind, no explosions!
I am going to stay with the LM317 so it's a known item.
 
Top