I am doing a Solar powered power bank as my dissertation project. It's based on LT3652 charge control, a Lithium-Ion battery (4500mAh) and 2 boost converters (3.3-5V and 3.3-12V).

I am using a 21W, PV panel with a maximum voltage of 7.68V. I need 4.2V and 2A output to charge the battery.
In calculations I did set minimum voltage in, as 6V and float voltage as 4.2 and followed the datasheet.

My design is based on Sparkfun Sunny Buddy MPPT. I changed the inductor (SDE0805A-100M) and diodes (RBR5LAM30ATFTR) and also sense resistor.
Unfortunately I'm not getting what I need out. Pictures speak better than words.
1.My schematic.
2.The actual prototype board
3.The open circuit voltage out. (7V in, 4.2V out)
4.Voltage out with 1kohm load.(6V in, 1.5V out . If i change the input voltage any higher and lower than 6V output voltage drops!)
5.Current out with 1kohm load.(6V in, 220mA out It goes up to 240mA at 7.6V in).

I appreciate your thoughts and advise as my deadline is reaching and i have been all over the place but i couldn't find the fault. as you can see I have changed the calculated 3.2uH inductor with a 10uH one but it didn't make much of a difference!

Ps. Sorry for the long post, I just wanted to give you as much as information as i can.

 

"5.Current out with 1kohm load.(6V in, 220mA out It goes up to 240mA at 7.6V in)."

No. 220 mA into 1k? No.

 

"5.Current out with 1kohm load.(6V in, 220mA out It goes up to 240mA at 7.6V in)."

No. 220 mA into 1k? No.

What you mean?

 

You say you have a load of 1000 ohms and an output current of 220 mA. Ohm's law says 220 volts.

 

I think i have made an error. Load is 4.7ohm.

 

You say you have a load of 1000 ohms and an output current of 220 mA. Ohm's law says 220 volts.

You are right. I swapped around R4 and R3. how ever I still get 1.4v out over a 4.7ohm load. This means the output current is around 300mA which is not even near to 2A I designed it for.
Any idea where can I improve on to get the desired 2A out?

 

First, remember that this circuit does MPP "tracking" by preventing the input voltage from being pulled down too low. If you use a bench power supply that is voltage regulated and current limited, the peak power point is right at the threshold of current limiting and the "knee" of the voltage-current curve is very sharp. You can soften the knee somewhat by adding some series resistance.

Problems with switchers are very often due to noise.

The first thing I would do is solder the inductor directly to the PCB. Those parallel long lead wire will radiate a lot of noise that can easily interfere with normal circuit operation. If you need a long lead so you can get an oscilloscope current probe in place, keep the lead as short as you possibly can and make the other lead "zero length" if possible. You can also reduce radiation by twisting the leads together.

The input and output leads should also be twisted together in pairs to reduce inductance and minimize radiation. Right now you have a circuit that will likely produce horrendous ringing at Vin because of the ceramic capacitor resonating with the inductance of the input wire. It would not surprise me to see volts of ringing. I would add a bulk capacitor in parallel with the 10 µF ceramic cap. I can't tell from the photo what type C1 is, other than ceramic, but if it is a high dielectric constant type it may have very high voltage coefficient of capacitance and hence be much less than the nominal 10 µF. I regard the high-K capacitors to be pretty much useless - you'll get better performance with lower nominal capacitance in the same physical size with lower K dielectric material, though at slightly higher cost.

The layout of the current sense resistor and the output filter capacitors is not good. The sense track between the ceramic and tantalum forces the inductor discharge current to go a long way around, Part of the current does have a short path to the ceramic cap, so it may not be too bad. The sense point has significant inductance and resistance in series with it due to the longish narrow track between the end of the sense resistor and the capacitors.