XG11 battery voltage
 

The manual doesn't give a minimum voltage to allow the battery to go down to. As its a LiFe battery I get worried when it goes to 6.3V. Should I be does anyone know. Is it true that LiFe batteries tend to drop very suddenly if they go too low?

There appears to be a typo in the manual. The illustration on page 91 shows the low voltage alarm is set at 9.0 volts while the text says 6.0. The alarm on mine was set at 9.0 and the use time with that setting is acceptable. If your transmitter is the same as mine, you can’t set the alarm voltage lower than the 9.0 volts!
　
Allan

Thanks for the reply Allen. I'm not sure we have the same radio but what you said prompted me to check out the manual and my tx setting page is on 82. It says low voltage is set for 6.0V. But my real question is, is it safe to fly with radio battery as low as 6V or should I set this to a higher value. Can the transmitter operate safely at 6V?

You are correct about the transmitters. I’m afraid I replied before giving adequate thought to your post. My transmitter is a XG11MV, the module version. It uses a three-cell battery and not a two-cell as used in the non modular XG11.
　
There is not much info on the type of LiFe battery used in our transmitters. From what I found on the web, it is not exactly the same as the A123 cells which are LiFe ion. I don’t think they can be discharged as deeply or at as high of rate. Yet the discharge curves through the normal operational range are similar.
　
The best piece of information I’ve found so far is on rightbattery.com. They ran a test on the 2000mAh Robitronic LiFe battery, http://rightbattery.com/190-2s-6-6v-...battery-tests/ Those curves are similar to information I have on the A123 cells. Assuming what we are using are similar to the Robitronic, I wouldn’t be reluctant to setting the low voltage alarm for the XG11 at 6.0 volts. You could up it 0.1volts if it gives you more piece of mind. Setting it at 6.3 volts would waste a lot of its capacity.
　
I’ve got my alarm set as low as it gets, 9.0 volts. I haven’t noticed any precipitous voltage drop when approaching that value. I suppose the thing to do the next time the transmitter needs charging is to leave the transmitter on when the alarm sounds and see how long it lasts before the discharge knee occurs.
　
Allan

Hi Allan, I contacted the JR help desk and they advised that one can go as low as 6.0v safely but as the LiFe batteries drop off dramatically after that so the advice is to set the alarm to 6.1 or 6.2 to be safe. Thanks for your help.

My thoughts are that those numbers are too low. According to the curve charts, there is very little charge left after 3.2 v per cell so that is only a .1 volt difference from the nominal voltage across most of the capacity. It is then important to get the warning as soon as the curve is reached because of it being so sharp with such little capacity remaining. A better number for two cells is 6.4v.


If one examines the LiFe curves, it is important to pick the curve line for the proper current draw. I think we should be looking at the lightest loading line and while 6.2 might be the right number for heavier loading, it is too low for the light loading of our transmitters.

As an aside, I've used RX telemetry warnings of 6.5v for more than two years and never had a false warning from brown out on any flight system I've got, which includes up to 35cc planes and consuming half of a battery capacity. If running digital servos on a 100cc plane, a different loading curve line might be needed and a lower setting to avoid brown out false warnings... YMMV.

Hey guys,

I just picked up an XG11 and have been familiarizing myself with it; programming feels and looks similar to my 9503 so that is good.

The problem I have with the LiFe transmitter battery though is that it challenges everything I have learned to date about the A123 chemistry; that being, you don't measure battery discharge state with a voltage reading! The discharge curves that AA5BY has posted are naturally correct...I think that is the document from the NoBS Batteries website if I'm not mistaken.

I have A123's on all my airborne packs at this point and through reading and personal experience I manage the discharge status of those packs primarily on mah used as provided by trial with my FMA Multi charger data. I find the NoBs site information very authoritative and have studied it pretty good. It is clear to me that the discharge curves for this chemistry are just too flat for voltage to have much predictive value on state of charge. Looking at any of the lines in the graph the voltage is flat...really flat. That is great battery performance for our RC application, but you simply don't know where you are on state of discharge by looking at voltage. At 6.2 you could be 5 minutes off the charger or 50 minutes off the charger! This is why the FMA chargers use a "gas tank" type methodology. On a new airplane, I fly a couple flights and recharge so I can measure how many mah on average I use per flight...then three, four flights etc. until I get a really good average per flight. At that point you can take whatever safety margin percentage of total effective mah capacity you are comfortable with and then recharge after whatever number of flights has you using that capacity based on the average per flight. Everything I know at this point tells me that is the correct way to do it.

Then, a radio turns my process and knowledge upside down by using a LiFe battery for power and displaying voltage as an indicator of battery status. It doesn't make sense to me at all.

In my opinion, a transmitter fuel gauge reading that displayed the mah used is the best way to do this and my guess is it can be correlated to a time algorithm if mah can't be measured directly. I'm thinking the current draw on a transmitter is pretty constant regardless of what you are doing, unlike an airborne pack where number of servo's, flying style and other variables play into the battery discharge rate.

I'm not sure what I am going to do yet but I have a thought to charge the battery on my FMA Multi to measure mah used and compare that to transmitter data on elapsed time the power switch was "on" to try and figure something out myself or at least study this some more with my own data. Perhaps this kind of thing can be addressed in software updates but I don't know too much about that yet.

What are your thoughts on my comments guys? You can call me crazy but you are going to have to take a number; not to mention you won't be the first!

Thanks,
Tom

The chart is for the AI23 ANR26650M1A cell that uses a metal can container. NOBS had used this battery in its receiver packs. To get a better idea what the discharge characteristic for the XG transmitter pack is, one should do a discharge curve of voltage versus time.
　
I did this for my XG11MV. It took almost six hours. I would post the results except I have no idea how applicable it would be for the XG transmitters that use a two-cell pack. The MV has a three-cell, 2100mAh pack. The nominal cutoff voltage for it is 9.0 volts and the transmitter functions down to 8.5 volts.
　
One observation is that the voltage doesn’t fall off the end of the end of the world as it passes beyond the knee. The discharge rate is much greater but still linear. In fact the factory setting of 9.0 volts is about half way between the knee and the final 8.5 volt cutoff point. Twelve minutes of run time remains between 9.0 and 8.5 volts. This is more than sufficient warning time for power application. It is a little short for a sailplane.
　
One final caveat, the results I write about are for a pack that had no more than four charge/discharge cycles. I expect the pack will deteriorate over time. The warning point will have to be adjusted.
　
Allan

Tom... I'm in agreement with your management scheme. I do not rely on battery voltage to determine LiFe state of charge or capacity.... but (there always is one isn't there), that doesn't mean voltage measurement has no application with LiFe batteries. As a secondary safety measure, I think voltage is of value but only for real time voltage via telemetry. It has no value to pre-determining capacity for a flight as traditionally done with other battery technologies, but if the warning is given at the initial point of the curve, then there is likely time to immediately land.

aworrest and aa5by,

Thank you for responding and I can see those points of view. I made a really stupid move years ago and lost a beautiful Extra 300 due to an undercharged receiver pack (NiMh). It was totally my fault and I'll spare you the details other than to say the crash, in retrospect, was simply spectacular! I had an OS 1.08 in it that I had to dig out of the ground with a shovel and the biggest part of the airframe left intact was the rudder; even blew the receiver apart! I can kind of laugh about it now but at the time I was totally disgusted with myself.

Anyway, that experience has left me all goofy with batteries...I'm probably overreacting, but it is what it is.

As to telemetry, I pick that technology up on my new XG11 and I will enjoy that for sure. I'm sure you guys know that on A123's there really isn't much voltage sag under load either. So between the super flat discharge curve and the lack of voltage sag under load I suppose the telemetry voltage is good to have but you just need to be extra careful in paying attention.

Thanks again for your comments.

Tom