Hello folks. At this time of date, i have been working weeks on off trying to make my first serious pcb. It is a 3 cell li-ion charger, which when i get the charging part to work, will have more features. But it does not work...

I am using the consonance CN3763 charging IC with the typical application circuit and calculated value parts on 15 v, but when i test the completed circuit, only the charging LED is on but it does not provide any current for the battery. I have tried 3 times soldering, and testing, because i thought maybe i had fried some of the components but no, same thing happens. No current. I also tested for shorts, right grids, and component orientation off course.

Therefor I believe I must have messed up making the circuit. Though i can not find an error. Any help, comments, suggestions will be so appreciated. I have provided schematic below and components list. Any help or feedback will be appreciated

Thank you - Ejnar

 

Hi, your circuit seems to be OK as far as I can see. The only difference seems to be the RX resistor between FB & BAT pins imissing which is present in the circuits in the datasheet.
Maybe there is an issue with your PCB? Are you not getting any voltage on the battery connector pin 1 at all? What do you get on pin 10 of U2, it should be going low to switch on M1.

 

Just to be sure: are you testing with the battery connected?

 

Even if it works, it is not going to work. You need cell balancing for cells in series and this circuit does not do that.

I would be a little worried about a part with a datasheet that claims to be a 3-cell charger and includes this statement,

l Complete Charge Controller for single cell Lithium-ion Battery

 

Even if it works, it is not going to work. You need cell balancing for cells in series and this circuit does not do that.

I would be a little worried about a part with a datasheet that claims to be a 3-cell charger and includes this statement,

They sell 2, 3, and 4-cell versions. The last digit in the model number is the number of cells. The only difference, as far as I know, is the charge voltage. These chips work fine, the only thing is no balancing which is not the best thing but keeps costs down and only becomes a problem occasionally or when the pack ages a lot.

What you spotted was probably due to clumsy reuse from another product that, I think, doesn’t have the boost converter (to go from 5V to something that can handle charging 2+ cells). Definitely the wrong way to reuse documents, but common everywhere. If you are going to do things that way then you can’t afford to skimp on the proofreading.

 

Hi, your circuit seems to be OK as far as I can see. The only difference seems to be the RX resistor between FB & BAT pins imissing which is present in the circuits in the datasheet.
Maybe there is an issue with your PCB? Are you not getting any voltage on the battery connector pin 1 at all? What do you get on pin 10 of U2, it should be going low to switch on M1.

Yes, the RX resistor was not neccessary for my needs according to the datasheet way down. I tested the circuit pcb with the battery attached, therefor all i cant see, is that the board is not outputting any current to the battery pack. I will check the U2 pin 10 as you said and write a follow-up

 

Just to be sure: are you testing with the battery connected?

Yes, I have only tested with the battery connected as it would not output otherwiser according to the datasheet

 

Hi, your circuit seems to be OK as far as I can see. The only difference seems to be the RX resistor between FB & BAT pins imissing which is present in the circuits in the datasheet.
Maybe there is an issue with your PCB? Are you not getting any voltage on the battery connector pin 1 at all? What do you get on pin 10 of U2, it should be going low to switch on M1.

I think you might be on to something with that mosfet signal. Or else i am testing wrong. It should be pulled low, when the battery is connected, and therefor should short when i connect my multimeter between that and the ground grid, right? What the board is doing is shorting, when i connect pin 10 to the power grid, which I guess is because it pulls up instead of down, that is a problem. Please correct me if I am wrong

 

They sell 2, 3, and 4-cell versions. The last digit in the model number is the number of cells. The only difference, as far as I know, is the charge voltage. These chips work fine, the only thing is no balancing which is not the best thing but keeps costs down and only becomes a problem occasionally or when the pack ages a lot.

Is this another way of saying that using this chip to charge series-connected batteries will only occasionally cause them to burst into flames and/or explode and burn your house down?

 

I think you might be on to something with that mosfet signal. Or else i am testing wrong. It should be pulled low, when the battery is connected, and therefor should short when i connect my multimeter between that and the ground grid, right? What the board is doing is shorting, when i connect pin 10 to the power grid, which I guess is because it pulls up instead of down, that is a problem. Please correct me if I am wrong

Not sure how you are checking/testing pin 10 of the chipp when you say 'it should short when you connect my multimeter between that and the ground grid'. It sound like you have your meter set on resistance reading and looking to see a short, I dont think thats the right way to go about this.
The chip is described as PWM type which I would expect would be pulsing the gate of M1 into conduction, you would need a scope to see the signal there. Also, there may be a good reason for the chip not charging as expected such as a faulty battery or a faulty cell within the battery so, you need to be careful how you go about investigating this. I would expect if the chip decides the conditions are not right for a charge, then it will not by pulling pin 10 up. Have you checked and rechecked your PCB and connections?

 

Hello folks. At this time of date, i have been working weeks on off trying to make my first serious pcb. It is a 3 cell li-ion charger, which when i get the charging part to work, will have more features. But it does not work...

I am using the consonance CN3763 charging IC with the typical application circuit and calculated value parts on 15 v, but when i test the completed circuit, only the charging LED is on but it does not provide any current for the battery. I have tried 3 times soldering, and testing, because i thought maybe i had fried some of the components but no, same thing happens. No current. I also tested for shorts, right grids, and component orientation off course.

Therefor I believe I must have messed up making the circuit. Though i can not find an error. Any help, comments, suggestions will be so appreciated. I have provided schematic below and components list. Any help or feedback will be appreciated

Thank you - Ejnar

View attachment 293861

Hello there,

First, read and then re-read post #9.
You should never, ever, ever, ever, try to charge three (or even two) Li-ion cells in series without some sort of charge balancing mechanism.
I wanted to stress that, but there is one exception if you are willing to take the risk. That is, if the cells themselves have built a built in battery management system, which is a small pc board with some mosfets and other stuff. It's still a bit of a risk because you are now depending on the BMS of each cell to protect it during charging and charging happens to be one of the most stressful times for these cells and is a time when most cells end up with a problem.

There is also the question of the charging current. The higher the charging current, the more the design depends on each BMS or your external balancing mechanism. For lower currents it should be safer, but it is hard to put a number on the threashold.

I almost hate to see people with little experience build battery charges for Li-ion cells, but i know we all like to do it at some point. It's a challenge, and can be rewarding when it works right. You just have to take extra precautions or else you could end up being very, very unhappy.

I'll take a look at your circuit drawing better a little later too, but do you happen to know what the function of RCS is (0.2 Ohms)?
I am just wondering if it is being used strictly as a current sense for the output or if it also plays a part in some state variable feedback mechanism as i was wondering if anyone did that yet in these things. I'll be able to take a closer look later.

 

Hello there,

First, read and then re-read post #9.
You should never, ever, ever, ever, try to charge three (or even two) Li-ion cells in series without some sort of charge balancing mechanism.
I wanted to stress that, but there is one exception if you are willing to take the risk. That is, if the cells themselves have built a built in battery management system, which is a small pc board with some mosfets and other stuff. It's still a bit of a risk because you are now depending on the BMS of each cell to protect it during charging and charging happens to be one of the most stressful times for these cells and is a time when most cells end up with a problem.

There is also the question of the charging current. The higher the charging current, the more the design depends on each BMS or your external balancing mechanism. For lower currents it should be safer, but it is hard to put a number on the threashold.

I almost hate to see people with little experience build battery charges for Li-ion cells, but i know we all like to do it at some point. It's a challenge, and can be rewarding when it works right. You just have to take extra precautions or else you could end up being very, very unhappy.

I'll take a look at your circuit drawing better a little later too, but do you happen to know what the function of RCS is (0.2 Ohms)?
I am just wondering if it is being used strictly as a current sense for the output or if it also plays a part in some state variable feedback mechanism as i was wondering if anyone did that yet in these things. I'll be able to take a closer look later.

Oh okay. Actually this circuit is ment to be placed in some products i land to launch, therefor balance charging might be quite necessary. Regarding the RCS, i did only find in the datasheet, that it was used to set the charge current, which i tuned to 600mA.
But a question of pure curiosity, what excatly bad can happen to the batteries without balance charging? will it explode, or just degrade the batteries? I am currently looking through the datasheet, if something is wrong with my mosfet, as it seems to not pull low

 

it was used to set the charge current, which i tuned to 600mah.

A nit, but current is in amperes (or mA).
mAh would be a battery capacity, thus 600 mAh means the battery is rated (typically) to deliver 60mA for 10 hours.

 

Not sure how you are checking/testing pin 10 of the chipp when you say 'it should short when you connect my multimeter between that and the ground grid'. It sound like you have your meter set on resistance reading and looking to see a short, I dont think thats the right way to go about this.
The chip is described as PWM type which I would expect would be pulsing the gate of M1 into conduction, you would need a scope to see the signal there. Also, there may be a good reason for the chip not charging as expected such as a faulty battery or a faulty cell within the battery so, you need to be careful how you go about investigating this. I would expect if the chip decides the conditions are not right for a charge, then it will not by pulling pin 10 up. Have you checked and rechecked your PCB and connections?

Yeah, i do not have a scope yet to test that. But it is weird, because, without power to the circuit, there is a resistance from pin 10 to both power and groung grid, but with the power on, There is no resistance between pin 10 and power grid(it is shorting) but there is no connection what so ever between pin 10 and ground grid. If it was doing PWM i would assume to see at least some connection. Therefore I think you would be right, that the the chip is turning it off for some reason. Would it be possible to simulate in other way the presens of the batteris without the batteries? And yes, i have tried rechecking every connection on my pcb

 

A nit, but current is in amperes (or mA).
mAh would be a battery capacity, thus 600 mAh means the battery is rated (typically) to deliver 60mA for 10 hours.

Yes, ofcourse. I changed the reply. What a mistake, think my head is spinning a bit from all this datasheet digging

 

Also, I get a output voltage of around 2.5 volts without the battries connected. And according to the datasheet the indicating LED should begin blinking when the batteries are not present, which it does not. So it seems, that the problem is not contained within the batteries

 

Oh okay. Actually this circuit is ment to be placed in some products i land to launch, therefor balance charging might be quite necessary. Regarding the RCS, i did only find in the datasheet, that it was used to set the charge current, which i tuned to 600mA.
But a question of pure curiosity, what excatly bad can happen to the batteries without balance charging? will it explode, or just degrade the batteries? I am currently looking through the datasheet, if something is wrong with my mosfet, as it seems to not pull low

Hi,

Yes, unfortunately it can catch fire or explode and burn and burn and burn and cause huge problems. New York just went through this a couple weeks ago and are now looking into putting in place some laws to force quality assurance for Li-ion batteries. Some fire departments don't bother to try to put out the flames, they wait for it to burn out.
You can see how important this is.

Oh also, some batteries do not have a built in BMS so they would be really bad for a charger that has no charge balancing built into the charger itself. That means if the customer is free to use their own batteries, they may choose the wrong kind and wreak havoc.
There are battery charger chips that do have charge balancing for these cells maybe you can look for a different kind. I know i sure as hell would if it was going into production.

Best of luck to you.

 

Yes, ofcourse. I changed the reply. What a mistake, think my head is spinning a bit from all this datasheet digging

Hi again,

Sometimes you have to step away and relax and do something else, or even sleep on it. When you go back to it, you will have a fresh mind and can usually find things that were not right much faster.

Another trick is to look away, then look at it again and quickly look away again. Do this several times. That helps sometimes too because your brain tends to work harder when it has to be fast.

 

Hi again,

Sometimes you have to step away and relax and do something else, or even sleep on it. When you go back to it, you will have a fresh mind and can usually find things that were not right much faster.

Another trick is to look away, then look at it again and quickly look away again. Do this several times. That helps sometimes too because your brain tends to work harder when it has to be fast.

Yeah, thank you very much. It always just is hard to start up troubleshooting, as you do not know where to begin, but I will follow your advice. Although according to your sayings, it may be best to create a whole new charger with charge balance... The batteries is going to be contained and out of reach for consumers, so if I insure to buy these BMS built-in batteries, do you think I will be fine with this circuit? (if i get it to work)

 

Yeah, thank you very much. It always just is hard to start up troubleshooting, as you do not know where to begin, but I will follow your advice. Although according to your sayings, it may be best to create a whole new charger with charge balance... The batteries is going to be contained and out of reach for consumers, so if I insure to buy these BMS built-in batteries, do you think I will be fine with this circuit? (if i get it to work)

Hi,

Well my preference is to have both built in BMS for the batteries AND for the circuit. That way you have two levels of protection. For a consumer circuit i would definitely do this.
Since your design incorporates built in batteries, you could also include some temperature sensing, which is not really that hard to implement. You could cut of the current if any of the cell temperatures reaches higher than some reasonable limit. This is the way professional fast chargers do it even when the batteries are not built in usually. Measuring the cell temperature is probably one of the best methods for sensing a problem. If the cell temperature goes too high, there is something definitely wrong even if the circuit does not know exactly what it is electrically.