Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

in need of a buck boost converter

Status
Not open for further replies.

neilxavier

New Member
we are designing a battery charger from a solar panel. we have a power path battery charger which requires an input voltage of about 5V.

in this case, we need a buck boost converter with min Vin of maybe 0.3-0.8 and max Vin of maybe 6V (since that is the highest voltage our flexi solar panel can generate).

Maybe you have encountered a buck boost with these specs. Again here it is:
Vin (min) = 0.3- 0.8
Vin (max) = 6V
Vout = any range that would include 5V.

Your help will be very much appreciated. thanks in advance.
 
Your Vin min is very small. I think you can not find parts that will go below 2 volts. Some of the LED boost converters will go to 0.9 volts. Your panel will output very little power down there!
Try TI.com and national.com they have good tools to help design PWMs. Try using the LED boost converters.
 
Can you tolerate a simple current limiter when the input is above 5V? That would make a buck configuration unnecessary.

Where did my previous reply go?
 
Can you tolerate a simple current limiter when the input is above 5V? That would make a buck configuration unnecessary.

Where did my previous reply go?


Hi Mneray,
I undeleted that post and found it was a double post so i deleted it again. when i first looked it did say you delete it.

Cheers Bryan
 
Yeah I saw the other thread too, double thread posting is annoying.

Neil, what is the actual spec for your solar panel? ie the max power spec which should be @ a particular voltage.

With most panels the max unloaded voltage is about 1.3 times the max power voltage, so your "6v" panel probably makes good power at about 4.5v.

Maybe you could just connect the panel to the charger through a small schottky diode like a 1n5819. If you post better specs on the charger's required input voltage etc (and the panel specs!) then maybe people could help you without needing to guess. :(
 
You're doing it wrong.
Look up "maximum power point tracking". The thing is, if you draw too little current you get no power, 0.7v/cell * 0mA=0W. If you draw too much current- say a sunlight level produces 100mA if you short circuit a current meter across it. But 100mA at 0v is 0W! Look at the IV curve of a solar cell. You'll find the maximum power at around 0.5V/cell IIRC, which might be 90mA for that sunlight level. But the sunlight level changes so you have to keep sensing that and adjusting the load current used based on the cell voltage vs current to get that maximum power point.

If a 6v panel is dragged down to 0.3v, you've failed to find the max power point and even IF the converter works at 0.3v it's producing no significant power at that voltage. The power is greater if it reduces the current drawn off the panel so the voltage rises back up. That actually INCREASES the battery charge current!

The charger needs to be able to sense both the sunlight level and the charge state of the batt. Note that on NiMH, the charging algorithm relies on a transient dV/dT feature under constant current to detect full charge. It is not possible to provide constant current as the sunlight level changes. That creates what we in the engineering world call "a problem".
 
You can relate the max power point to a voltage, the assumption being it will always be in full sunlight at the appropriate angle to the sun. But who has perfect cloudless/shadowless sun with a tracking module the entire day?

It is because most people do not have the perfect sun, or tracking modules, that an MPPT charger is beneficial.
 
You can relate the max power point to a voltage, the assumption being it will always be in full sunlight at the appropriate angle to the sun.
Well, 0.5V is "sort of" the MPP. The thing is that the dI/dV slope is pretty flat around there and it's unclear what current you should be drawing just by looking at the voltage. The exact MPP voltage does change with sunlight levels due to I^R losses in the grid and other factors.

MPP also changes with temp. A LOT. Higher temps lower the MPP voltage quite a bit. And solar cells get very hot in the sun.

But no DC/DC converter which ISN'T designed as an MPP is going to regulate for input voltage anyways.

In short, to do this job properly, you need a PIC reading the current and input voltage and driving a PWM for a buck or boost converter (don't try to do a buck/boost unless you really need to- and you probably don't need to, because the panel voltage at MPP is almost constant). It will periodically vary the duty cycle by small amounts and see if it produces more or less power as a result, that's how it constantly hunts for the MPP.

If this is more than a "trickle charge" for the batts, then it will need to sense batt charge state too.

What cheap products do (well, just about EVERY consumer solar charger) is just add more cells so the MPP panel voltage is always greater than the batt voltage. The batt will just drag the voltage down to whatever the batt voltage is and the current is pretty much going to be equal to the maximum short circuit current. This does not "perfectly" utilize the panel but it's still cost effective.
 
Last edited:
In short, to do this job properly, you need a PIC reading the current and input voltage and driving a PWM for a buck or boost converter (don't try to do a buck/boost unless you really need to- and you probably don't need to, because the panel voltage at MPP is almost constant). It will periodically vary the duty cycle by small amounts and see if it produces more or less power as a result, that's how it constantly hunts for the MPP.

I heard an easier way, when you're making a charger, is just monitor the output current. Since you pretty much just want peak current into the battery, it's synonymous with peak power.
 
Correct. With a battery charger, peak output current=peak power. But you still need a power point tracker to make it work "best". And the end-of-charge sensing to prevent overcharge is still an issue.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top