9v 1200mAh Lithium Battery too good to be true?

JohnR2010

Does anyone have experience with this 1200mAh 9v lithium-thionyl chloride battery or one like it? http://www.batteryspace.com/primary-lithium-battery-9v-1200mah-10-8wh-50ma-rate-compatible-with-energizer-l522-9v-ultralife---ul-listed-un-38-3-passed-ndgr.aspx
It looks like this is going to be a perfect fit for my project as long as I keep my peek power demand under 100mA. It seems like the specs are too good to be true (1,200mAh) for a 9v battery?? Tye typical 9v Lithium specs I have seen are more around 800 mAh.

Tracy Allen

There is specsmanship going on there. Huh,1200mA capacity at 0.0mA discharge rate?? The characteristic curve is more telling...

The 1200mAh figure applies at a discharge rate of 0.5mA. At 10mA it is down to 850mAh and only 300mAh at 50mA.

The Energizer L522 is also listed at batteryspace.com. It has a much higher current capability and delivers about 750mAh at 25mA drain. The two batteries are comparable at that level. I bet the L522 too would perform near 1200mAh at 0.5mA drain. The Energizer data sheet does not cover that possibility. Can they say for sure? Maybe there are tradeoffs in the chemistry between life capacity and high current capacity.

In any case, add a humungous capacitor in parallel with the battery to supply the surge loads.

I buy LiPoly batteries in quantity from that company, BatterySpace and have no complaints. Their warehouse is only mile from my shop so I can pick things up will-call.

JohnR2010

I noticed the 0.0mA statement as well, that is what made me question the specs. I feel better now that you have found a comparison to the Energizer L522 and pointed out the performance in the Characteristic Curve chart. My circuit (The TDS ZibBee Server you have been helping me on) now pulls an average of 0.100 mA. That average is based on an hourly peek to ~20mA for about 1.5 seconds.

What would you consider a humungous capacitor? I currently have the 220uF cap we discussed back in the power supply tread http://forums.parallax.com/showthread.php/160483-Powersupply-design-with-Microchips-LDO-MCP1701A?p=1321547&viewfull=1#post1321547. I can go bigger. If I can get close to 1000mAh out of the battery I should get over a year run time.

Tracy Allen

I figured it was for the TDS project. Maybe you addressed this before, but why can't you use a small LiPoly 3.7V battery and an energy harvesting scheme to keep it charged? That's another option in addition to the one of AA cells + boost circuit.

The 9V battery is convenient certainly with the enclosures available with battery compartment molded in. My guess is that the battery would give you pretty long life under the condition you describe. It has a long time to recover between each pulse of 20mA current. I think you'll just have to try it and see.

For the capacitor, the defining equation is Q=CV, charge equals capacitance times voltage. Or the derivative, I = C dV/dt. Current in relation to rate of change of voltage. It will take a pretty big capacitor.

Suppose you want a capacitor initially charged to 9V to supply all the 0.02A current during that 1.5S period, forget the battery. The final voltage should be no less than 3.5V so that your regulator can still supply 3.3V to the circuit. So a change of voltage from 9V to 3.5V is a drop of 5.5V, and the rate of change of voltage then during the 1.5 seconds has to be less than 5.5/1.5 = 3.67 V/s. Divide that into the current and you have 0.02A / 3.67V = 0.00545F = 5450 µF. Big?

Now the battery, Another important piece of information is the internal resistance of the battery. You can read from the characteristic curve that with 50mA current the battery terminal voltage is 8V instead of 9V, thus, a 1 volt drop through the internal resistance and 1V/0.05A = 20Ω. That resistance increases as the battery ages and you can be sure it will also increase substantially at low temperatures.

The fact is though that even if you have a 5450µF capacitor, the battery will supply all of the current as soon as the voltage across the capacitor drops to 8.6V. The capacitor doesn't help much. To transfer most of the short term workload to the capacitor, you'd have to add a resistor in series with the battery, in effect increase its internal resistance. Say 1000Ω, and let the capacitor recharge through that resistance at a peak of 5.5mA during the 1 hour interim. All in all, I'd say forget that approach and give it a try with just the battery. The 220µ capacitor is meant only to supply micro or milliseconds of transient current demand.

JohnR2010

Thanks Tracy for the explanation I should have looked up the equation. Your walk through was a great help I would have probable missed the need for the 1K resistor between the battery and large cap.

I did consider a rechargeable LiPoly early on but I dismissed it since I didn't have a way to charge it. I have another project (Solar Powered ZigBee router) that is based on a 6,600mAh LiPoly battery pack and would love to use a similar one here. I'm curious what did you have in mind for an energy harvesting scheme? The TDS sensor will be inside most likely mounted in the basement. The sensor's probe tees into a 1/4" water line like the one that runs to a refrigerator's ice maker or a reverse osmosis system.

I'm trying to keep the unit small and portable, the long term goal is to have a platform I can base other sensors on that is why I favor an off the shelf battery and long service life.

kwinn

In this application with that battery you may want to look at one of the constant current led drivers to convert the 9V 0.5mA current to a lower voltage/higher current that charges a large capacitor (supercap?) or battery.

Tracy Allen

The usual energy sources are light, heat, vibration, flow, etc. and there are specialized chips to capture miniscule amounts and pump it up to charge a battery to run a wireless node. However, there may not be easy options in your basement on a 1/4" feed line. One year on a battery is certainly the simplest-change it at the same time as you do the smoke detectors.

At average current 0.1mA, that comes to 876mAh in one year.t

TDR tee into a 1/4" line? That sounds coaxial.

JohnR2010

Yep I think I have a good chance to make a year as things are now. I'm going to turn my focus back to the code I want to make this rock solid reliable and with ZigBee that means I have to take into account a lot of possible network changes.

Atlas Scientific is going to release a new TDS circuit soon and I plan on incorporating it into my board. When I do that I may go the voltage boost route and see if I like that better. Thanks Tracy!!

Here is a pic of two of the probes I'm using from HM Digital. They each have four wires two for the electrode and two for the thermistor.