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 JEfromCanada » November 30th, 2013, 2:55 am

glnc222 wrote:Only now recall seeing LiOn batteries for Roomba's in the aftermarket years ago.


I bought an aftermarket LiON Roomba battery from someone years ago. As I recall, the entire batch of cells were defective, and he ended up getting many complaints/returns. I didn't try another one after that.
Current robots:
Roomba Discovery 4210 (R.I.P. - Freecycled)
Roomba 560 (retired)
Neato XV-11 SW version 3.1.17844, LDS version 2.6.15295, with Pet Brush and Vic's filter (sold)
Neato Botvac 80 SW version 1.1.97, LDS version 2.6.15295, Board SW version 22753
JEfromCanada
Robot Addict
 
Posts: 662
Joined: November 13th, 2007, 5:16 pm
Location: Canada

Re: Neato lithium ion battery revisited

Postby glnc222 » November 30th, 2013, 5:12 pm

Details of Roomba batteries are in the Roomba forum, where there is some experience with LiOn previously reported.
The only point here is that LiOn batteries (with proper accoutrements) have been shown to properly charge on machines supplied only with NiMh batteries, increasing prospects of success with Neato, and confirming TimTaler's results. Battery packagers are looking into what can be supplied for Neato and reports may be available soon.
http://www.robotreviews.com/chat/viewtopic.php?f=11&t=17491
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 8th, 2013, 9:24 pm

A couple loose ends in discussion of the battery balancing circuit:
PeteX posted could not get German materials to work in Neato, but the battery pack photos show no balancer circuit boards. Was he testing Taler's akku-balancer or something else? Original reading suggested he was only addressing the general idea of charging Lithium on Neato and not the specific balancer circuit; now am not sure -- which German discussions was he referencing? [edit] member PeteX joined the day of that post and has no other months later.

Developer TimTaler reports several months Neato use but does not specify which Neato model, which might be a Vorwerk with LiOn firmware, being in Germany. (Though VR-100's use a 4500mah battery I'm told, compared to Taler's 2600mah). No photos to clarify.
The balancer circuit is not advertised for vacuums on the dealer's German website (after applying Google's translator to it).

Any clarifications would be welcome and help promote use of this interesting product.

[edit]That only 2600mah hours batteries were used does suggest Neato instead of Vorwerk, which has a higher load and 4500mah batteries. Explicit confirmation would add to the impression.
Last edited by glnc222 on December 11th, 2013, 1:56 pm, edited 1 time in total.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 10th, 2013, 2:41 am

Issue of Discharge Curve
One dealer's graph of the discharge curves for Panasonic 3100mah LiOn batteries below shows about 3.25v at the end of a typical Neato 2ah run, where with NiMh Neato is at around 14.4v -- higher than the Lithium 14v or less. Neato dies completely at 14.2v -- shuts off totally.
So questions arise about application of a much smaller 2600mah Samsung battery to the Neato. Is the discharge curve so radically different? Or is the 50 minute run time less than the 2ah run observed, at least on old Neato's?
So at the 14.4v point in the graph (green line) only 1.5ah is used, 25 per cent less than maximum run observed,
which is 90 minutes on hard floors in the manual, or 67 minutes reduced. Newer Neato's with higher loads for filter and brush run less so that could fit with the 50 some minutes reported I suppose with this 3100mah battery. But with the same curves a 2600mah? I don't have all the technical data on the Samsung. Just some "known unknowns" as they say.
Attachments
orbtronic.jpg
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby mfortuna » December 10th, 2013, 8:03 am

I'm surprised the cells were discharged down to 2.8V. Normally you don't want to go below 3.0V or even a little higher just to treat the cells nicer.
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
Shark Ion 750
User avatar
mfortuna
Robot Master
 
Posts: 5838
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Neato lithium ion battery revisited

Postby glnc222 » December 10th, 2013, 1:29 pm

Full battery property analysis display -- not an application, which would leave a margin for durability.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 10th, 2013, 7:23 pm

battery properties data -- by brand

Comprehensive LiOn battery characteristics with copious charts, links to discharge curves at
http://lygte-info.dk/review/batteries2012/Common18650Summary%20UK.html
and
http://lygte-info.dk/info/Batteries18650-2011%20UK.html

Some of these show 2400mah batteries with adequate voltage at 2ah, different from the Panasonic previous post.
[edit] more detailed data on Panasonic 2900 similar 3100 of prior post in the above at http://lygte-info.dk/review/batteries2012/Orbtronic%2018650PD%202900mAh%20%28Black%29%20UK.html
(to read the more densely lined graphs, the magnify ctrl+ key in firefox is helpful)

Effect of protection circuits, omitted from the Akku-Balancer, needs clarification.
[edit] see http://www.lygte-info.dk/info/battery%20protection%20UK.html in the above. A "pressure valve" used to disconnect, contrary to a dissipating transistor. Fine details of the PCB's shown.

Trustfire's 2400 looks interesting. Seems well recommended.
Main use of these cylindrical cells appears to be flashlights, popular with campers.

Much higher capacities available in a fatter 26850 size. Might still fit, not sure.
Last edited by glnc222 on December 11th, 2013, 2:42 am, edited 1 time in total.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby Gordon » December 10th, 2013, 10:19 pm

Nice batch of charge & discharge data you have rounded up!
glnc222 wrote:...Effect of protection circuits, omitted from the Akku-Balancer, needs clarification.
Taler may pop in and comment about this but I think he indirectly stated that a primary goal of developing his balancer circuit was to avoid the errors produced by certain robots' charging systems when charging current is interrupted by something not resident in the robot's control system.
... A "pressure valve" used to disconnect, contrary to a dissipating transistor. ...
Maybe I am missing something, but I think those are apples & oranges topics, for the reasons: I believe when a protective pressure valve breaks, it also breaks the cell's continuity; whereas, (and I assume that "dissipating transistor" is in reference to Taler's balancer) when the big FET switch closes in Taler's circuit the cell begins to discharge through the circuit's 11 ohms resistance. The power FET's ON resistance should be found to be much less than 11 ohms.

Those are, IMHO, totally different processes, stimulated by different forces.

I might also say that although Taler's balancer is momentarily tossing away charge, the cell's reduced SoC will tend to be recovered while overall battery voltage is less than the system's full-charge voltage (unless some other sensed function, such as battery temperature, halts the charging process before voltage causes the halt).
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 » December 10th, 2013, 11:48 pm

Apology was not clear protection referenced as addition to the balancer, not in it. I was concerned transistor switches to disconnect in the cell package would have a junction voltage drop (and a lot of heat), maybe explaining how the protected Panasonic cell voltage looked too low for Neato compared to bare Samsung. Not very familiar with power transistors.
Protection trip levels don't seem restrictive, so maybe they can work with the balancer -- Taler didn't say -- or show the schematic -- despite the PCB layout (what do you put on it?).

Since the Lithium cells seem mostly charged in separate chargers with parallel circuits, Taler's means of charging in series might not be so common (except in laptop batteries). Handy addition.

Whether the akku-balancer is the best or only option will take a few months as old batteries wear out and suppliers research more. By then maybe the data I found can help someone else try it first...
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 11th, 2013, 3:04 pm

comparison to NiMh

Looks flatter to me than LiOn. The 2/3 discharge point where Neato would end run has much higher voltage than the same size Lithium (12 cells NiMh, 4 LiOn), if I've read the graph correctly, bit difficult. You have to figure what size Lithium is needed for equivalent performance voltage wise, not just capacity.
Attachments
NiMhCurve.jpg
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 11th, 2013, 5:24 pm

Samsung Discharge Curve
This looks usable at 3000mah, so how smaller 2600mah Taler used works is a puzzle -- maybe less than maximum run time (50 vs 90 minutes). But there are many different models of these cells, labeled A, B, C... etc.

Caution visual slopes can mislead, have to be interpreted numerically -- graphs on different scales.

From "Samsung 3000 mAh 18650 Testing"
http://www.candlepowerforums.com/vb/showthread.php?275159-Samsung-3000-mAh-18650-Testing
samsungcurve.jpg


[edit]IC Protection Incompatible with balancer?
TimTaler posted:
Neato errors only with low/High Voltage cut off, like 18650cells with internal IC protection.
If you charge a 18650 with internal circuit above 8.4V it turns off (cell goes to high impedance) the same at low Cut off voltage --> Neato Message Battery error/Fail ...

A bit terse but seems to say the usual protection circuits trigger a Neato error from interrupting charging when some limit is reached. But Neato has no dedicated charging hardware, just a cpu running all sorts of stuff which could crash, and what state is the charging in then? Lithium carries far more warnings for this than Nickel, so I want the protection. Or the quote is misinterpreted?

[edit] voltages again -- reference notes cells which could charge to 4.35v get only to 4.2 from protection cut offs -- highest line in some graphs -- for a different source of discrepancy between protected and unprotected discharge curves.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby TimTaler » December 12th, 2013, 3:15 am

removed
Last edited by TimTaler on December 12th, 2013, 5:22 am, edited 1 time in total.
TimTaler
Robot Groupie
 
Posts: 88
Joined: August 6th, 2010, 5:42 am

Re: Neato lithium ion battery revisited

Postby Gordon » December 12th, 2013, 4:01 am

{deleted msg after source msg got removed}
Last edited by Gordon on December 12th, 2013, 12:22 pm, edited 1 time in total.
Gordon
Robot Master
 
Posts: 4304
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: Neato lithium ion battery revisited

Postby mfortuna » December 12th, 2013, 11:48 am

When I first measured a LIPO cell on my RC discharger I was surprised to see how pronounced the discharge curve was for a 7.4V 3800mAH pack. One of the perceived advantages of the LIPOS was the "flat" discharge curve.

I think the primary advantage of Lithium based cells over NIMHs are

Lighter weight for a given capacity.
Longer life span.
Ability to deliver high amounts of current.

If you look the the NIMH discharge curve at 5A (2.5C) the battery has dropped to min voltage at about 1.5 Ahr which is a capacity reduction of 25%.

The Samsung graph doesn't have a 2.5C rate (7.5A) but at 2C (6A) it is still above minimum voltage at 2.75 Ahr which is a capacity reduction of 8.3%.
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
Shark Ion 750
User avatar
mfortuna
Robot Master
 
Posts: 5838
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Neato lithium ion battery revisited

Postby glnc222 » December 13th, 2013, 2:42 am

protection voltages
neato cut off voltages (end of charge to) it always stops at 8.2-8.4V

i.e. 4.2v max per cell four pack. For one example, Panasonic 3100mah spec:
Protection: Over-charge (Activated @ 4.35V) ...

So Neato should never trip protection normally and the balancer never see that feature except when warranted.
(discharge protection 2.5v, way below Neato's return above 3.6v or dead at 3.5v). Specs might vary with cell model, not checked.
[edit] discharge current protection limits can be too low for Neato on some cells. needs above 3A.

[edit]Lithium Battery Cost
The reference posted has two bar charts: capacity down to 3.6v, around Neato's run end, and current ratings. For the full 2ah run time and 3A load -- more than the Samsung 2600 50 minutes -- a few listings all $60-80 for four, for not much saving. The lifetime (not run time) isn't touted anywhere longer than Nickel. Unless there are other lifetime factors (high discharged Nickel worse?), the extra balancing electronics cost may be what dictated using NiMh. Substantial savings on smaller, unprotected batteries. The maximum run time on carpet is 2/3 the 90 minutes on hard floors, so full capacity needed there. Maybe some smaller hard floor venues can benefit.

[correction] the German magazine mentioned is shown at the ELV dealer site and is their newsletter. A more extensive PDF article in German was found http://www.elv-downloads.de/service/manuals/LBA1/66446_LBA1_km.pdf I will try to translate with Google in a later post. Appears to be a commercial product not intended for Neato per se, just an application of it.
Last edited by glnc222 on December 14th, 2013, 2:15 am, edited 1 time in total.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 13th, 2013, 7:54 pm

Larger Battery Needed
Closer inspection of the discharge charts show LiOn capacity usable down to 3.6v decreases with current load.

At 2-3A 18650 size batteries cannot supply more than half Neato's run time 2ah capacity -- in line with the 50 minute run time reported. Lot of the capacity is at voltages too low for Neato.
So either higher capacity 8-cell parallel packs are needed, which do not appear supported by the akku-balancer for series packs (and inherently more complex), or a larger capacity 26650 size (corrected from 28650 mentioned before).

candidate
A number of 26650 are listed in 4000mah. The Chinese UltraFire brand nominally 5000mah but tests at 4000mah. [correction]UnProtected. Detailed Ultrafire test data at http://budgetlightforum.com/node/20791
and http://lygte-info.dk/review/batteries2012/UltraFire%20UF26650%205000mAh%20%28Flame%29%20UK.html may indicate the class generally. Neato will use about half the capacity, just to get the voltage range needed. Some 3500's seen not examined.

Size 68.9mm x 26.7mm compared to wrapped Neato pack 70mmL x 32mmW x 57mmH (2 high LiOn 53.4mm).
Fit will depend on dimensions of the akku-balancer, unavailable, just looks like it might go in, to the side of the cells. Thickness needs to be within around 7mm and some perfboard things made with IC's here fit that; would have to test (no pictures of the original installation in a Neato). [correction] board size listed as
23 x 16.5 mm (additional load resistor not fitted) 30 x 16.5 mm (additional load resistor fitted)

[edit] concerned with text found later post board doc: "Due to the possible temperature the Balancer must not be stuck directly on the (plastic) Housing of the Batteries." No photos of his Neato installation.

Price widely varies by source from $5-10, and wholesale advertised. So ambiguous, or adventurous....
They claim enormous durability, number of cycles, but seem to be into exaggerating specs. Others claim 500+, still better than usual Neato. Using only half the capacity, though, might be favorable to longer life.
Last edited by glnc222 on December 14th, 2013, 2:55 pm, edited 3 times in total.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 14th, 2013, 1:19 am

correction -- protected 26650
The cheaper Ultrafire is unprotected. Protected versions include Trustfire
http://www.amazon.com/Trustfire-26650-5000mah-Lithium-Rechargeable/dp/B008YAPRUQ#productDetails
Data at http://lygte-info.dk/review/batteries2012/TrustFire%20TF26650%205000mAh%20%28Flame%29%20UK.html
(two versions are noted -- an extra "manafont")
Maybe a bit less capacity on the curve at 3.6v than unprotected Ultrafire.
Protected versions appear to run $10-15, so that feature adds to the cost around $20 on four. It has to be done on each cell in case a link in the series chain is shorted out raising voltage on a cell. It's in a house, not your model plane in the Elysium Fields.
Still talking $40 replacements (keeping old balancers) instead of $80 NiMh replacement packs.

[edit]am a bit concerned protection voltage level could be 4.2 instead of 4.35, and might get tripped by Neato charging. Specs omit fine details for our weird applications.
Last edited by glnc222 on December 14th, 2013, 3:35 am, edited 1 time in total.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 14th, 2013, 3:30 am

Documentation for Akku-Balancer LBA-1
TimTaler developer (document may be by a journalist or tech writer, unknown)
[edit] May have confused Taler's developing the Neato kit using the balancer with that circuit -- could be someone else's product years ago.
Parts previously posted from ELV's newsletter magazine shown at the dealer site; this reference shown in google searches.
Note the product is several years old, so obviously not Neato related per se. Thanks to Mr. Taler for bringing it to our attention.
Orders to dealer: http://www.elv.de/lithium-polymer-lipo-akku-balancer-lba-1-komplettbausatz.html
[edit] appears discontinued at ELV Dec. 2013-- see later post.

Google Machine Translation from German PDF article http://www.elv-downloads.de/service/manuals/LBA1/66446_LBA1_km.pdf
Machine translation may be limited in accuracy or completeness. Not further edited. Some gaps where text could not be copied by Windows into the translator form. [edit]forgot to feed only complete sentences to the AI across column ends.
Not for general readers -- might have specific technical points buried in the poor translation for engineers.
---------------------
April 2006
Lithium-Polymer (LiPo)-
Akku-Balancer
akkuschematic.jpg

akkuIC.jpg

akkuload.jpg

AkkuParts.jpg

[added]
akkudata.jpg

The advantage of lithium polymer batteries is particularly high energy density at the same
time low volume and weight. A disadvantage is that these batteries are very sensitive
are against overcharge, so that for a long life series-cells only
should be loaded with appropriate protective circuits
While NiCd and NiMH batteries very
are robust and wrong treatment
do not take immediate evil, it is already
with a slight overload of
LiPo cells to injury. Even with
a deep discharge these cells
quickly destroyed.
If you want to own battery packs confectioning
tioning , it is essential to ensure ,
that all cells connected in series , the
have the same voltage and the
first interconnecting the same La -
have state of charge . If this is ignored ,
it comes already with the first charging
to irreversible damage.
With ready-made battery packs
are, in principle according to
Cells selected for use . As long as
the packs are new, all are in series
cells connected to the same voltage
General
Lithium batteries , especially the modern
tions lithium polymer cells have already
a large part of the market for portable
Devices conquered. But even in model -
area , this battery type is due to
the weight and size advantage always
more by . But we must not
remember that the use of incorrect
Treatment is fraught with danger .
The manufacturer's charging instructions are
must be observed , and for loading
this battery - type only appropriately concentrator -
pierte chargers are used .
Lithium Polymer (LiPo ) -
Battery Balancer
tion level is loaded and everything is (still)
in order. In the course of time it is
but the drifting apart of the cells ,
for which various causes responsible
can be Lich .
No matter whether it is assembled to itself
Battery packs or whether the cells
already interconnected at the factory
are drifting apart can without
corresponding non- protective circuit
be prevented. One cause may
For example, different warming
Be cells in operation. At high discharge
flow , the middle cells of a
Battery packs often warmer than the outer ,
which in turn variations in the
Capacity available to episode
has , and during the loading process they are
Cells the "full " state not at the same
Ships reach the outer cells .
In addition to temperature are capacity
tolerances , which is becoming more or less
-
size for the drifting apart . after
a plurality of charge-discharge cycles increases the
Voltage difference between the inputs
individual cells on and on. In extreme
case can lead to overload cause
Cells swell and burst or that
there is even a fire.
Consider the charging process based
a 3- cell battery packs , and we
assume in our example that the
Cells are already drifted apart
or different
-----
Used in the consumer electronics
Lithium battery packs are fundamentally
always equipped with protection circuits.
Figure 1 shows the example of a
Laptop battery pack.
If it is in the field of modeling
or in other applications with self-
assembled LiPo battery packs already
less than 100 charge cycles to education
come cases , it is usually due to lack of
Protection circuits.
ELV Balancer circuit
The records ELV Balancer circuit
by outstanding fea -
times out . Depending on whether the balancer
remain firmly attached to the cell
or is externally associated with a
Charger is used, are
different in the ELV circuit
Component versions available.
Especially if the circuit continuously with
should remain connected to the cell , is
it is natural to every gram of weight and
correspondingly small in dimensions.
The balancer has the energy of
the cell can no longer be taken
can convert into heat. Now it is
a law of nature that small on a
Area , only a limited energy
amount can be converted into heat.
To these conflicting requirements
to do justice to allow the board a

With up to 350 mA current can Balancer
the reaction in the heat with SMD Wi-
resistances occur while a
additional power resistor to
1.1 A balancer current options. the
Power resistor is on a Pla -
housed routines section, the simple
can be cut off when this
Resistor should not be fitted .
In operation, it is important to note that
the balancer in the field of resistors
R 3 to R 5 and in the area of the power
resistor R 6 can be pretty hot.
In particular, with stocked
Power resistor (R 6)
must reflect actual installation
in a shockproof ,
non-combustible housing
carried out . For fixing
communicates with a M3 screw
a mounting hole next to
the power resistor
available.[retranslated important point]Due to the
possible temperature must the
Balancer not directly on the (plastic)
Housing of the Batteries be stuck .

----
Balancer not directly on the (plastic)
Housing of the Batteries be stuck .
It is also advantageous that the ELV
Balancer does not need any adjustment .
On the contrary, the adjustment is carried out at

with trim pots , in which case quite
Doubts about the long-term stability con-
are introduced .
Another important requirement is a
extremely low quiescent current , since the cell
finally not in storage by Ba -
lancer discharged or even deeply discharged who -
the must . The quiescent current of the ELV - balancer
is at < 3V about 1μA and
at voltages > 3.5 V less than 100μA.
Compared to many other balancers
an extremely low value.
operation
As will be the so-called
Balancers overloading of individual
Lithium cells prevented? The radio -
tion example is quite simple. The Balancer
is basically nothing else
as an electronic load resistance
upon reaching the end of charge voltage
is connected in parallel to the cell and
a substantial part of the charging current
receives . The cells of a battery pack
are thus kept " in balance "
and thus the most important requirement when
Fulfilled charging. The lifetime of
Lithium cells can therefore with balancers
be significantly increased.
Since each cell or each parallel ge -
switched cell group a balancer
must receive , it is fixed at the
Battery balancers mounted on small
Dimensions and light weight of .
Finally, you do not want the benefits
LiPo cells through the circuit-
tions lose .
As seen from the specs ersicht -
Lich, the dimensions of the ELV
LiPo balancer in the manner with the cell
associated variant only 23 x 16.5 mm
---graphics
and the weight is 1.6 g almost ver-
negligible.
The actual operation is a-
times. Once the final charge voltage
reaches 4.2V, takes the balancer
its work and indicates this with
an LED on the board. below the
Final charge voltage, the Balancer
Basically nothing to do.
$ E?? 9? = HOOHQVSDQQXQJ? ALH? W? RI?
Balancer from the recorded charge current
no longer or no longer completely in the
Cell. In addition to protection of the battery pack
Balancer also allow an assessment
treatment of the corresponding cell in comparison to
the other cells in a pack. Since the
Charging the cell with the lowest
Capacity first the final charge voltage
reached, the LED is lit to
erst Apart from the first charging
a newly assembled battery packs
is it, this is the weakest
Cell in the pack. In well-selected battery
Packs come the balancer only briefly to
Deployment.
----
During the charging process is to be noted that
the final charge voltage of the battery charger
and the use of the balancer very closely
lying together . Operating at active
Balancers can the charger of course
not distinguish whether the current from
Balancer or absorbed by the battery
will . This may the end of charging
later recognized.
circuit
The circuit of the LiPo balancer
shown in Figure 2 . The main construction
element , the IC LM3420A -4 , 2 is of
National Semiconductor , which is
essentially a voltage
surveillance module is . the
Special is the high accuracy
of 0.5%. The internal structure of this construction
Einstein's is shown in figure 3, where
an addition to the precise voltage reference
OP , a transistor and two resistors
are integrated.
As can be seen in the diagram holds
the circuit complexity in
Limits. Particularly important
tig is a very small
Current consumption , so that
in the stored state, the
Cells not by the Ba -
lancer loaded appreciably
or deep discharged.
----
This is ensured with T 4, T 5 and external
Components circuit constructed ,
although the power consumption of IC 1 only
80μÅ.
As long as the cell voltage is below
RI ? FD ? ? ELV ? ? 9 ? OLHJW ? EH ¿ QGHW ? VLFK ? 7 ?
in the off state , which IC does not include 1
Voltage is supplied , and thus the cell
also not be charged. Only at a cell
voltage above about 3.5 V is T 5 via D 2
and the base voltage divider R 8, R 11
through controlled and thereby simultaneously
T 4 is set in the conductive state. about
the low-impedance protection resistor R 1
then enters the cell voltage directly
the monitoring module IC 1
When reaching the end of charge voltage
of 4.2 V supplies IC 5 at the output ( pin 5)
an output voltage at over R 10
the basis of the discharge transistor T 3 given -
ben is . Once T 3 by controls , loaded
the series circuit consisting of the
Collector resistors R 3 to R 6, and
the collector- emitter path of the transistor
tor , the cell. The implementation of the energy
takes place in heat depending on equipment with
the SMD resistors R 3 to R 5, or
in addition to the power resistor
R 6 Further, a part of the energy
via the transistor T 3 (collector -emitter
distance ) is converted into heat . Once the
Balancer is actively working on R 7 shows the
powered light emitting diode D 1 to operate on .
With the optional resistor to be fitted
was R 12 , the circuit for lithium
to - ion batteries with 4.1V final charge -
Voltage can be adjusted.
---
follow . The subsequently performed con -
control of the function is simply a
stabilized adjustable laboratory power supply
possible. To this end, the current limiter
the laboratory power supply to about 1.5 A and the
Output voltage is set to 3.0V .
The balancer is and with ST 1 at the positive pole
with ST 2 to the negative terminal of the power supply
to connect . The power consumption of
Balancer can now only about 1
μ
A amount .
The output voltage is slow to
approximately 3.7 V highly regulated. The LED D 1 may
not light up and the current consumption
me of the circuit should be 50 now
μ
A to
100
μ
A amount .
The output voltage will continue to he -
increased. From 4.2 V (± 0.5% ) has the luminous
diode D 1 is lighting and a load current to
be measured , which rises to about 350 mA,
only when the load resistor R 3 to
R 5 are fitted . If in addition the
Power resistor R 6 is incorporated , be -
carrying the balancer power about 1.1 A.
If the first test was satisfied
running , adjust the port can be connected to
to be monitored cell take place . da-
at ST 1 is the positive pole and ST 2
to the negative terminal of the associated LiPo
Cell connected . As already Spannungsdif -
ences in the mV range play a role
tion is recommended that no common
longer connecting cables, for several
To use balancer.
Note that the balancer on
End of charging quite heat up
the can . This is especially true ,
when the cells already differ widely
have drifted and in addition the power
is resistant R6 stocked.
In particular, with tipped performance
resistance must necessarily in the installation
a case be made that the An-
requirements of a fire enclosure
corresponds . Because without power resistor
on the bottom side no connection
---
wires or solder joints are present,
can Balacer with a tem-
peraturstabilen adhesive directly on tem-

(eg, metal). Figure 5 shows
a 2-cell lithium-polymer battery
retrofitted with balancers. must here
addition to all other cable connections
having a temperature-stable adhesive
be secured.
---
Last edited by glnc222 on December 26th, 2013, 11:54 pm, edited 2 times in total.
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby glnc222 » December 15th, 2013, 3:08 pm

separate protection circuits

Some protection circuits are on separate PCB boards instead of packaged inside cells. A single board handles multiple cells. While not exactly what is needed for Neato, it shows the possibilities. e.g. Tenergy 3-cell PCB/PCM-32006 at All-Battery.com

http://www.all-battery.com/protectioncircuitmodulepcbfor111vli-ionbatterypack3cellswith65alimit-pcb111v65a.aspx
data sheet http://www.all-battery.com/datasheet/32006%20%2811.1V_5-6.5A_3_SLi-ion%29.pdf
If it could work on two-cell packs and put in series might combine with the balancer -- unknowns. The in-cell protection is certainly more compact. Maybe a premium has to be paid just to get that fit.

Re-using protection and balancing circuitry reduces the cost of replacement batteries to around $40 for 26650 cells, about half the current NiMh 12-cell pack cost. A dozen bare cells for a 3800 NiMh Neato pack, essentially same as the factory 3200, can go as low as $48 -- without any boards (just thermistor and polyfuse in the pack, repackaged). Supposedly the LiOn version would last longer. Lasting twice as long would cost the same as two NiMh given cost of the boards. Looks like a subtantially greater advantage has to be shown. Maybe costs would be lower when made in quantity.

[edit]RC Model Batteries
Much more expensive and more capable. There are different ways to make batteries with no single optimum for all applications. One supplier Thunder Power says
When charging or storing your Lithium Polymer batteries it is also recommended to use a fire retardant bag.

Guess they're not protected. Innovative solution to the fire hazard -- a bag!
glnc222
Robot Master
 
Posts: 4871
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Neato lithium ion battery revisited

Postby mfortuna » December 15th, 2013, 9:17 pm

A LIPO sack is required for ROAR (the leading RC race organization) events. I have seen firsthand the damage a LIPO pack can cause. A friend of mine almost lost his house.

For RC applications the user is the brains.
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
Shark Ion 750
User avatar
mfortuna
Robot Master
 
Posts: 5838
Joined: February 5th, 2006, 10:35 am
Location: NH

PreviousNext

Return to Neato Robotics

Who is online

Users browsing this forum: vic7767, z_thompsonpa and 326 guests