battery issue 0007 - bad thermistor

[mmm2000]

Hello all,

I have purchased 2 Li-ion for my Neato XV about 2 years ago and they work like new even now, except that I get the 0007 error, after a full charge.

After I connected the robot to the Neato Control while charging I've noticed the BattTempCAvg[0] and BattTempCAvg[1] went up to about 40C around when batteries were full (then stopped charging) and the temperature for the 2 thermistors went down to about 23C, which is the temperature of my room.
Then BattTempCAvg[1] started to go down one degree at a time to about 16C, then it went up to 20C then back down and after about 5 minutes it dropped to -2C.

My guess is that the thermistor for the battery 1 is bad.

My questions:

1. Has anybody replaced the thermistor from a neato battery? These batteries are awesome and they were expensive. Trying to save a few bucks if I can.
2. Which side is the battery that gives the reading for BattTempCAvg[1] (the power button side or the other side)?
3. Is this a good replacement for the defective thermistor :

https://www.ebay.com/itm/5x-10k-Precisi ... 5111!US!-1 ?

The thermistor is specified as 10K ohms at 25 C temperature and the resistance falls as the temperature rises: 6K ohms at 37C.

Thanks all in advance.

[mmm2000]

After some investigating I was able to figure out that the BattTempCAvg[1] is for the battery that is on the opposite side of the power button.

The way I found out is by measuring voltage between the 2 yellow wires, with the batteries connected to the robot and with the robot on. The robot reads the temperature based on the voltage that the yellow wires circuit produces:

1.25v 12C
1.3v 13C
1.35v 14C
1.52v 19C
1.55v 20C
1.56v 21C
1.63v 23C

The temperature for the battery with the bad thermistor was going up and down slowly, every few seconds.
The temp for the good battery was a steady 23C (room temp) with the corresponding voltage of 1.63v

I also tried to warm up the batteries with a hair drier, but to my surprise the temperature/voltage did not change. I got them pretty warm. So what am I missing here? Isn't the thermistor inside the batteries?

Here are pics of the inside of the battery:

[vic7767]

The Lithium packs do not have a thermistor since those packs don't generate heat during charging. There is a thermistor simulation circuit within the Power Control Board in each pack. It would appear that you have a defective PCB.

[glnc222]

Many technical details of the thermistors and their role in charging NiMh batteries is included in the experimental LiFePo4 battery project report thread at
http://www.robotreviews.com/chat/viewtopic.php?p=124772#p124772
Operation and construction of a DIY adapter for Lithium batteries is included. Skip through irrelevant parts.
The software terminates charging in NiMh mode (set by unpublished USB command SetConfig BatteryType) when temperature rises faster than 1 degree centigrade per minute, easily simulated by an artificial jump of several degrees. The high side battery pack (with the 15v terminal) thermistor is monitored for charging. With a lithium battery adapted to the NiMh charger, the temperature readings shown over USB will not be meaningful as far as actual temperature goes.
The NiMh chemistry heats up when approaching full charge. Enclosed pack chargers use this temperature to detect full charge as the heating needs to be stopped. Open single cell chargers use a small drop in voltage which occurs when full, as they are open to dissipate the heat. Lithium cells do not heat up and are controlled by reaching a maximum voltage.

On the newer Rev64 XV system boards, the software can be switched to the Lithium charging algorithm and lithium replacements made without adapters to the NiMh charger, but they have been made only DIY. The older Rev113 boards do not respond to the SetConfig command. The software is shared with Vorwerk and configured by setup to match the particular hardware, with Vorwerk using lithium batteries in their VR100 version of XV Neato models.

While it was not practical to make a LiFePo4 battery with the required capacity at the time of the experiments, later model cells were found adequate for converting original Botvac models to lithium. The Botvac runs on 12v vs 15v in XV models, reducing the number of cells required. LiFePo4 is more durable than regular lithium, but at the expense of lower capacity in cells. They are used in some robot lawn mowers, with room for enough cells.