Having trouble with LM2665

Thread Starter

Andrew216

Joined Jun 24, 2015
29
Hey Everybody,
I have designed a circuit that will turn on LEDs using a CapSense sensor. When I use my power supply with the LM2665 in the circuit everything works fine. When I use batteries and the LM2665 in the circuit the LEDs will lock on or constantly reset after my fade_on I have set in my firmware. I have checked with the scope and there is a noticeable difference in the signals between batteries and power supply. The power supply has more noise (probably because of the inductor??) and the batteries are smooth but there is a sag on the rising edge. Im using 2 CR2430 batteries in parallel and using the doubler for some over head voltage for my LEDs. Is there a better voltage converter out there that wont have this effect when used with batteries? OR maybe I have made a mistake in my design. I have attached a schematic any help is greatly appreciated.
 

Attachments

ronv

Joined Nov 12, 2008
3,770
Hey Everybody,
I have designed a circuit that will turn on LEDs using a CapSense sensor. When I use my power supply with the LM2665 in the circuit everything works fine. When I use batteries and the LM2665 in the circuit the LEDs will lock on or constantly reset after my fade_on I have set in my firmware. I have checked with the scope and there is a noticeable difference in the signals between batteries and power supply. The power supply has more noise (probably because of the inductor??) and the batteries are smooth but there is a sag on the rising edge. Im using 2 CR2430 batteries in parallel and using the doubler for some over head voltage for my LEDs. Is there a better voltage converter out there that wont have this effect when used with batteries? OR maybe I have made a mistake in my design. I have attached a schematic any help is greatly appreciated.
It looks like everything is right on the edge.
The batteries have a pretty high internal resistance so they probably are sagging quite a bit when hit with the 80 ma needed for the doubler. You might think about using a small rechargeable Lipo battery. This sag puts you pretty close to the minimum operating voltage of the doubler. But if all that works, the voltage drop of the doubler is about .5 volts at 40 ma so if the sage is to say 2.6 the output might be around 4.7 volts. The constant current driver needs 1.8 volts of overhead so it may not work to well either.
Something like this might do it.
http://www.all-battery.com/3.7V500mAhLIPORechargeableBattery-34063.aspx?utm_source=GoogleShopping&utm_medium=GDF&gdftrk=gdfV26767_a_7c354_a_7c922_a_7c34063&gclid=CL_D1-fX7coCFQiUaQodmuIFlg
 

Thread Starter

Andrew216

Joined Jun 24, 2015
29
Thank you for the reply! Unfortunately I have a space issue where coin cells are the only option. The CR2430's are as thick as I can go with the most mAH without going over the height restrictions.
 

ronv

Joined Nov 12, 2008
3,770
They make really thin ones.
What you might try is a fairly large cap across the battery. Or a 3rd coin cell in parallel (just to see if that is the problem)
 

Picbuster

Joined Dec 2, 2013
987
Hey Everybody,
I have designed a circuit that will turn on LEDs using a CapSense sensor. When I use my power supply with the LM2665 in the circuit everything works fine. When I use batteries and the LM2665 in the circuit the LEDs will lock on or constantly reset after my fade_on I have set in my firmware. I have checked with the scope and there is a noticeable difference in the signals between batteries and power supply. The power supply has more noise (probably because of the inductor??) and the batteries are smooth but there is a sag on the rising edge. Im using 2 CR2430 batteries in parallel and using the doubler for some over head voltage for my LEDs. Is there a better voltage converter out there that wont have this effect when used with batteries? OR maybe I have made a mistake in my design. I have attached a schematic any help is greatly appreciated.
Looking at the cr2430 spec's I see a std load of 0,2mA/cell. Your circuit is doing far more than that.
Suppliers define de std load as function of Vbatt_intern and voltage drop over internal resistance of that cell.
( eq @ 0,2mA a voltagedrop of xx mV is acceptable. xx suppliers spec's)
Advice; measure the used current and redefine your batteries.
 
Top