18.5volt/3.5A Laptop charger using LM3478_ Help needed

[kuttan]

Dear All,
I have designed a boost regulator(DC to DC converter) for my laptop using LM3478. My source is a 12volt/30ah Lead acid battery.

With no load the circuit is working perfect. However when the circuit is loaded(when connected to the laptop) its not delivering enough current. I tried connecting with both acer and thoshiba latop. Both cases its not delivering more than 500mA and its not charging the battery at all.

I have used National Work bench onlline design software for design assistance and followed the same circuit based on that. Also gone thorugh the datasheet of LM3478 and couldnt make out where the problem is.

If you have encountered the same issue, Please suggest me the options to tackle this issue.

Thanks to all for your assistance in advance,

Regards,
Kuttan

 

[davenn]

Dear All,
I have designed a boost regulator(DC to DC converter) for my laptop using LM3478. My source is a 12volt/30ah Lead acid battery.

With no load the circuit is working perfect. However when the circuit is loaded(when connected to the laptop) its not delivering enough current. I tried connecting with both acer and thoshiba latop. Both cases its not delivering more than 500mA and its not charging the battery at all.

I have used National Work bench onlline design software for design assistance and followed the same circuit based on that. Also gone thorugh the datasheet of LM3478 and couldnt make out where the problem is.
If you have encountered the same issue, Please suggest me the options to tackle this issue.
Thanks to all for your assistance in advance,
Regards,
Kuttan

ok it would help if we could see the actual cct diagram you used. Are many variations in the data sheet for various output voltages and current capabilities
otherwiise it would be impossible to see what you may have done wrong

cheers
Dave

 

[kuttan]

Hi Dave,

Thank you for your respose. Please find the attached ckt diagram.

Thanks and regards,
Kuttan

 

[(*steve*)]

Do you have the specs on that inductor?

My first bet is that it is saturating.

What voltage do you see across the output when you connect it to the laptop?

edit: a rough calculation is that the peak current in the inductor is around 12.5A

 

[kuttan]

Hi Steve,
Thank you for your reply.
The inductor specification is 2.2uH/19Amps. The output voltage is droping from 18.5 to 14 volt while the laptop is booting up and working. The current is variying from 100ma to 700mA. That time the laptop indication is 'power unplugged'. when the laptop is off and if we connect the power, its showing 18.5 volt consistently, current around 500mA and the laptop also showing battery charging indication.

As you have stated, it may be due to saturation. Here one more thing i suspect is, for testing i made a single sided PCB with hand. So there a few jumpers due to that. Also one jumper there in the feedback path as well as the inductor path. Is that also creating issues?

Thanks for your time,
Regards,
Kuttan

Do you have the specs on that inductor?

My first bet is that it is saturating.

What voltage do you see across the output when you connect it to the laptop?

edit: a rough calculation is that the peak current in the inductor is around 12.5A

 

[kuttan]

Hi All,

Expecting some expert opinion on the problem i have been facing on boost regulator circuit....

Thanks for your time,
Regards,
Kuttan

Hi Steve,
Thank you for your reply.
The inductor specification is 2.2uH/19Amps. The output voltage is droping from 18.5 to 14 volt while the laptop is booting up and working. The current is variying from 100ma to 700mA. That time the laptop indication is 'power unplugged'. when the laptop is off and if we connect the power, its showing 18.5 volt consistently, current around 500mA and the laptop also showing battery charging indication.

As you have stated, it may be due to saturation. Here one more thing i suspect is, for testing i made a single sided PCB with hand. So there a few jumpers due to that. Also one jumper there in the feedback path as well as the inductor path. Is that also creating issues?

Thanks for your time,
Regards,
Kuttan

 

[davenn]

Hi Kuttan

ok looking at your cct and comparing it to some of the layouts in the datasheet, there are some variations that maybe causing some of your problems.

firstly, I assume that you were taking the output from the correct place ?? as your cct doesnt actually show where you were obtaining it from.
it should be from the junction of the diode D1 and the 47uF capacitor.

now speaking of that 47uF cap I would say it is substantially too low in value
it should be at least 100uF and preferably double that. One cct shows 2x 100uF
another shows 3 x 100uF so both those examples show that you are at least 50uF low and maybe 150uF low
That output cap stores the charge (current) between the switching cycles of the MOSFET. So if you dont have enough capacitance there then you are going to suffer a signif voltage drop under load..... Sound familiar ?

Next thing the example ccts also an inductor value somewhat higher than what you are using. I would suggest that it should be up ~ 10uH rather than the 2.2uH that you are currently using

have a play with those 2 things and see if it makes the difference for you

Dave

 

[kuttan]

Hi Dave,

Thanks you for your reponse.
As you have mentioned, the ouput is taken from the junction of D1 and Cout only.
Presently Cout is 3x47uF. I will further increase the value and observe the variation. Also inductor value, i will try according to your suggestion and will update the result.

Thanks for your time,
Regards,
Kuttan

 

[NickS]

The main thing I would suspect for low current is the sense resistor. It is a very small value with very large currents flowing through it. So taking the voltage reading off it anywhere other than right at the terminal could easily be raising the impedance thus prematurely current limiting. Layout matters here.

Also I am not getting the same schematic as you by using the same IO specified. Did you alter the default frequency? Also what did you optimize for eff/size/cost?