Neato lithium ion battery revisited

News and information about the Neato XV-11 Robotic Vacuum. All discussion and troubleshooting questions go here.

Re: Neato lithium ion battery revisited

Postby glnc222 » August 23rd, 2015, 11:33 pm

Commercial Neato Lithium NASA Test
Crush tests by NASA for airplane safety cover Boston Power / LithiumPower/ AnewPow 4400mah LiNmC batteries sold for use in Neato XV models.
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20130000056.pdf
These do not catch fire when internally shorted, so appear safe for vacuum use in houses.
Thank the different lithium chemistry used. Also mentions long battery life.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » September 7th, 2015, 1:49 pm

LiNMC cell
Safer LiNMC cell with same chemistry as LithiumPowerInc/AnewPow mentioned at German Roboter-Forum Panasonic CGR 18650 CH 2250 mah (German dealer https://www.akkuteile.de/lithium-ionen-akkus/18650/panasonic-cgr-18650-ch/a-100602/)
Use eight (2-parallel) with USB SetConfig BatteryType 3 Vorwerk lithium charging.
Problem I see with some higher capacity 18650's is maximum charging rate only 1.65A compared to Neato's 2 amp charger.
Panasonic battery guide http://na.industrial.panasonic.com/sites/default/pidseu/files/downloads/files/sfc-2014_interactive_20_11_14_0.pdf
Does not show maximum charging rates. Sometimes in dealer listings.
Some cells with protection circuit caps are too long, 68mm for the 65mm Neato compartment. With only two cells per compartment, might fit at an angle. But the charging rate is an issue with only two. Maybe four would fit in the six-cell compartment, don't know.

[edit] CGR 18650CH data sheet http://www.bto.pl/pdf/04444/cgr18650ch.pdf
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby Lewiy » September 9th, 2015, 3:28 am

I have assembled battery as written here -- viewtopic.php?f=20&t=16982&start=320#p132117. 4S Samsung ICR18650-26F + Protection Board in one battery compartment and in the other - nothing. SetConfig Batterytype 3. DC-DC converter in the charging unit was required to limit the charging current and to ensure full charge. It works fine . The duration of cleaning (hard floor and carpet fifty-fifty) 57-58 minutes before full discharge.
Lewiy
Robot Master
 
Posts: 143
Joined: September 27th, 2014, 3:33 am

Re: Neato lithium ion battery revisited

Postby glnc222 » September 9th, 2015, 11:54 pm

If a current limiting device is placed between the Neato charger and the battery, it might confuse the constant-current 2 amp charging circuit (standard constant-current NiMh charging method). The Neato charger would raise its voltage to the maximum which has been found to damage the system board. Some component is not rated for the 24v maximum dock power. Have you tested all this?
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby Lewiy » September 10th, 2015, 1:59 am

I have. Experiments have shown that 4S Li-Ion batteries are not charged enough by current 1.85 A (CC). The charge lasts about half an hour, the battery capacity is about 1000 mAh. LM2596 converter allows to upload 2 times more capacity by descending current while the battery voltage increases. This is not the classic CC-CV patern, but the result is suitable. The converter connected between the DC-block and protection block using the DC-connectors. It can be readily removed when the need to return to the original configuration for use NIMH batteries. The electrical circuits of the robot is not supplied voltage and current that exceeds a predetermined value. How damaging can be caused?
I admit that there is no need to limit the current when using 4S2P battery configuration. In the pair of accumulators charging current is divided by 2 and less than 1 A per cell.
In the experiments, the battery capacity is a calculated value and can differ slightly from the actual.
Attachments
Charge 4S Li-Ion in XV21 without current limiting.jpg
Charge 4S Li-Ion in XV21 with LM2596 current limiting.jpg
IMG_20150910_080156.jpg
Last edited by Lewiy on September 10th, 2015, 4:38 am, edited 3 times in total.
Lewiy
Robot Master
 
Posts: 143
Joined: September 27th, 2014, 3:33 am

Re: Neato lithium ion battery revisited

Postby glnc222 » September 10th, 2015, 2:15 am

In my experiment with LiFePo4 battery packs a mistake in my custom thermistor control circuits prevented the charger from terminating properly. The charging voltage rose beyond 21v and a system board was ruined.
I do not have the resources to make component level repairs of the system board. I do not know what part was damaged. F_Robot in Germany has used a lower voltage power brick in the dock (actually a bench power supply) to prevent accidents.
The cells used by Hausman in Germany was 4S Samung without any buck converter.
You should do a data logging experiment to monitor the charger voltage during lithium charging with the converter.
I used an inexpensive PIC Malaysian device but the Neato Control Program has some logging feature I have not figured out.
I see you have logging equipment. I cannot read the graph titles well.
They could be added manually with Irfanview.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby Lewiy » September 10th, 2015, 3:15 am

I updated photos with increased signatures on the axes.
XV21 configured for use with Li-Ion battery stops the charge on reaching 16.6 V or if the battery temperature more than 39 degrees Celsius. This is the first line of protection. DC-DC converter is configured to ceiling voltage of 17.2 V. This is the second line of protection. Finally, thermal fuse is installed inside the battery package to break the chain if something goes wrong.
Lewiy
Robot Master
 
Posts: 143
Joined: September 27th, 2014, 3:33 am

Re: Neato lithium ion battery revisited

Postby glnc222 » September 10th, 2015, 12:14 pm

We need the voltage on the input to the converter (the voltage in the Neato system board) not the output to the battery. The system board output voltage can be read over USB.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » September 11th, 2015, 2:21 am

OK, I see the two charging voltages, input and output of the converter-limiter, on the graph.
It appears the current regulator in the Neato charger is being fooled somehow. Without the limiter, the current is regulated to around 2 amps. When the limiter is inserted the current is lower; the current regulator in the neato is not working, not raising voltage. So what is the Neato charger regulator doing? Is their some difference because the battery type has been set to Lithium instead of NiMH? Both use a Constant Current charging phase.
If you just inserted a resistor between the battery and the charger you would expect the charger voltage to rise under control of a current regulator in the Neato.
The limiter device is not having that effect.
So I do not understand how the Neato charger and limiter work. Both involve some high frequency components as I understand it. It would help to understand what principle is at work so one could expect a good result in all Neato systems.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » September 11th, 2015, 12:15 pm

Perhaps the Neato charger has a voltage limiting function for Battery Type 3 lithium in addition to constant current regulation, a maximum voltage limit for lithium battery protection. Instead of simply shutting off the charger and declaring an error condition, the voltage is simply clamped to the maximum allowed while the current regulator continues to operate. Then the additional current limiting device between the charger and the battery can further reduce the current without the charger output voltage rising beyond the maximum allowed for lithium batteries.
To test this hypothesis the difference between operation with battery type set to lithium and NiMh would need to be observed. To prevent damage to the system board found when charger voltage exceeds 21V, action is needed to protect the system board during the process. For the LiFePo4 experiments I made a protection circuit which disconnects the dock power supply from the Neato system board when voltage exceeds a set level (circuit shown earlier in this thread). Manual intervention to stop charging can also be used.

[edit] Having the charger voltage remain at 17v while charging current continues poses questions as to how the lithium type 3 algorithm terminates charging. It would first be assumed the charger simply terminates when voltage reaches the maximum 16.8v used by the battery. Perhaps instead the charger looks at the slight drop in charging current when the maximum voltage is reached and lithium batteries start limiting current. NiMh chargers use a similar method based on very small current drops at the end of charging, though the thermistor is also used instead on larger packs to limit the end of charging heat generation.

current regulator links
Lewiy supplied example of the current limiting device buck voltage converter for driving LED's
http://www.ebay.com/itm/400440723717?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT
An additional trimmer adjustment is included for current regulation besides the usual voltage setting.
A youtube video is available on buck voltage converters http://www.youtube.com/watch?v=qXEZNNKWC9s
and various tutorial articles can be found on the web, Wikipedia etc.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby Lewiy » September 11th, 2015, 2:58 pm

Neato does not regulate the charging current in the circumstances discussed. The controller simply opens the MOSFET on the charging circuit and passes a current to the battery. Thus, the charging station determines the charge current. The controller monitors the voltage on the battery during charging and closes the MOSFET when the voltage reaches 16.6 V. Then the battery level is 100%. The second charging circuit, which is controled by other MOSFET and limits the current to 0.4A at NIMH charging algorithm, is not used after switching to Li-ion mode.
The charge current of about 2 A is suitable for 4S2P battery as it is in VR100 (any alterations to the charging station is not required). I was forced to apply a current limiter in the charging unit because the current of 2A was redundant for 4S1P battery.
Lewiy
Robot Master
 
Posts: 143
Joined: September 27th, 2014, 3:33 am

Re: Neato lithium ion battery revisited

Postby glnc222 » September 11th, 2015, 3:54 pm

How does current flow back through the limiter board to power the Neato when the battery is discharging?
[edit] photo shows the limiter board with a heat sink in the dock, not in the battery pack. Moreover, it is powered directly by the PSU brick, not the charging plates governed by a circuit in the dock. So the battery powers Neato normally. This is similar to a German user replacing the dock PSU with a lower voltage bench power supply to prevent some over-voltage conditions in battery experiments.

Another voltage converter video https://www.youtube.com/watch?v=NyTze5HU4nI
Last edited by glnc222 on September 11th, 2015, 6:11 pm, edited 1 time in total.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby Gordon » September 11th, 2015, 5:19 pm

glnc222 wrote:How does current flow back through the limiter board to power the Neato when the battery is discharging?
First, I must admit to not being up to date on all that has been going on in this thread.
2nd, what I next offer only applies to the rev64 main board.

In short, I would say not to expect Neato's operating current take the path you mention. By referring to a cleaned up version of devilmike's pencil schematic, brought into the forum by Lewiy in 2014, it is clear that the two violet colored paths for battery current required to operate Neato on a cleaning mission are most likely.
rev64ChargingCtrlrSchemat.PNG

I like the path through MOSFET Q8, lower right, because Q11 (gate driver) provides the system with single-command capability to power OFF everything driven through Q8.
Gordon
Robot Master
 
Posts: 4304
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: Neato lithium ion battery revisited

Postby Lewiy » September 13th, 2015, 10:57 am

Gordon wrote:
glnc222 wrote: the two violet colored paths for battery current required to operate Neato on a cleaning mission are most likely

Not sure what we're talking about the same. Lost in translation . Here we are talking about charging the battery, not the battery load. Two charging paths:
high-rating U8/61 -- Q15 -- Q14 (unlimited current, for NIMH and Li-Ion)
low-rating U8/62 -- Q13 -- Q12 (limited current, for NIMH only)
Lewiy
Robot Master
 
Posts: 143
Joined: September 27th, 2014, 3:33 am

Re: Neato lithium ion battery revisited

Postby Gordon » September 13th, 2015, 12:22 pm

Lewiy, I'm sorry for any confusion. My post was in response to glnc222's question about the battery discharge path, repeated here:
glnc222 wrote:How does current flow back through the limiter board to power the Neato when the battery is discharging?
I thought by posting the battery charging schematic it would be clear that the battery has no current path off the main board and to the charging power supply; so it was I that responded to glnc222 about the diagram:
Gordon wrote:... the two violet colored paths for battery current required to operate Neato on a cleaning mission are most likely.


I hope confusion has disappeared.
----------------------------------------
Lewiy wrote:
Gordon wrote:
glnc222 wrote: the two violet colored paths for battery current required to operate Neato on a cleaning mission are most likely

Not sure what we're talking about the same. Lost in translation . Here we are talking about charging the battery, not the battery load. Two charging paths:
high-rating U8/61 -- Q15 -- Q14 (unlimited current, for NIMH and Li-Ion)
low-rating U8/62 -- Q13 -- Q12 (limited current, for NIMH only)
Gordon
Robot Master
 
Posts: 4304
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: Neato lithium ion battery revisited

Postby glnc222 » September 13th, 2015, 12:32 pm

There has been a confusion here. My comments refer to the path of current within the buck voltage converter and current limiter, and not within the Neato system board. There is no issue regarding paths within the Neato system board. I originally assumed this component was added to the battery pack and was in between the battery and the Neato. Actually it is between the dock PSU and the Neato. The brevity with which these things get described and documented can produce such misdirection.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » September 16th, 2015, 11:47 pm

Panasonic NCR18650B LIthium Ion Cell
Interesting lithium ion cell Panasonic NCR18650B, available with tabs on ebay in the U.S. and Europe.
Cheaper without tabs. Soldering lithium may not be as wise as with NIMh cells, though the Germans did it -- quickly...
so get the tabs (and still do it quickly).
Capacity 3400mah, maximum charge rate 1C (3.4 amps), discharge 2C (6.4 amps), seems to meet Neato requirements. Only four cells needed (so only a four cell protection PCB).
The price $12 or more may not save over regular Neato batteries, but might last twice as long.
The safer LinMC and LiNM chemistries have lower capacities (and prices) and require eight cells.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby crateofbeer » September 17th, 2015, 3:17 am

Might also be worth looking into Konion cells since they don't NEED balancing and are generally very safe.
A while ago I upgraded an electric scooter with them and it's been working great for over a year now.
You might be able to get 3 rows of 4 cells in the neato but not sure.

From the net:
One of the beneficial quirks of these cells is that the factory claims that they don’t need to have the parallel strings balance-charged. In theory, the entire pack can be bulk-charged, and each individual cell will stop taking a charge when it’s full, even though the charger will continue charging up the other cells…the ones that need just a little but more to be full.
crateofbeer
 
Posts: 14
Joined: September 17th, 2015, 3:01 am

Re: Neato lithium ion battery revisited

Postby glnc222 » September 17th, 2015, 12:14 pm

Sony Konion cells are LiNm previously mentioned. One of the more inherently safe alternatives to lithium ion.

Some Neato packs using balancer heating for the standard NiMh charger were made by German owners. (The standard Neato battery charges with control by temperature; the Vorwerk lithium batteries charge by voltage only. The unpublished SetConfig usb command changes the procedure.)

There is a 4000mah 26650 cell but with less than adequate 4 amp discharge rate, requiring eight 186650 instead.
There is a whole class of "high drain" cells with higher charge/discharge rates but lower capacity.
This division applies to NiMh cells as well as lithium. No such thing as just a battery anymore.
Neato needs cell types with the same maximum voltage as lithium ion or NiMh, for the power and charging system design. LiFePo4 has a lower voltage and needs a longer series, besides being too low in capacity to fit.
The safer types of lithium could be soldered instead of spot welded, less fire hazard. Only local dealers provide spot welded tab services to simplify constructing packs.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » October 25th, 2015, 10:23 pm

Protection Boards

I looked for possible protection boards of use with lithium cells in Neato and found one 4 cell version rated 6 amps
on ebay http://www.ebay.com/itm/4-cells-6A-Li-ion-Lithium-Battery-18650-Charger-Protection-Board-14-8V-16-8V-4S-/331630880247?hash=item4d36bda5f7:g:FssAAOSwjVVV0bHw
Seems about 66mm long, tight fit but probably can be used, maybe at an angle. (18650 cell size = 18mm x 65.0mm; made to fit across the tops)
All the two cell boards appear to have currents way too small for Neato's 2 amp charging rate and observed running loads around 3 amps with some peaks observed around 4.
[edit]Except the 6 amp 2 cell AliExpress board mentioned by Lewiy http://www.aliexpress.com/item/Protection-Circuit-Module-PCM-for-2S-7-4V-7-2V-Li-ion-Li-Po-Battery-C/32308955059.html?spm=2114.31010208.0.332
There are 4 cell boards rated 4 amps but that seemed maybe too low. Any opinions welcome.
I did not look at boards with additional balancing features. Those seem to be mostly oriented towards inclusion in chargers more than battery packs.
A four cell board would require either a single series of four cells all in one compartment, or extra wires between compartments when using eight cells.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

PreviousNext

Return to Neato Robotics

Who is online

Users browsing this forum: calebleemcd, moradi2, vic7767 and 362 guests

cron