Rebuilding a Roomba Battery for Less

Inside the Roomba and Scooba and more, Cool mods, Repair and Upgrades - including the all new iRobot Create Kit. Let's void that warranty baby!

Re: Rebuilding a Roomba Battery for Less

Postby a1robotrepair » May 19th, 2016, 12:59 pm

See previous post. Here is a picture with the guts removed. The other 4 batteries, also ICRs, are used as spacers, extra fire power when one of the LGs starts self heating and vents fire. I didn't even mention they aren't 3Ah, oh, now I did.

image.jpeg
:text-banplz:
There is also no thermal protection.
LG 18650 B4 LGABB41865
0303J307A9. Do I have the date code correct? 2010, Day 303, Lot J30, Line 7, Lot serial A9
Last edited by a1robotrepair on June 7th, 2016, 6:45 am, edited 2 times in total.
Roses are red Violets are blue Poems are hard Flowers and Chocolate
http://www.daycounter.com/Calculators/
a1robotrepair
Robot Master
 
Posts: 758
Joined: October 14th, 2013, 8:40 am

Re: Rebuilding a Roomba Battery for Less

Postby a1robotrepair » May 21st, 2016, 5:45 pm

Someone, Please help me respond to this email from the ID 10 Ts.
Hello,
thanks for your reply. we are sorry to hear that you have problem in our battery, our battery defective rate is low. we have to tell you that, we are very concerned about the safety problem, we have protection circuit board in order to protect the battery from high heat and over charging. If the the charging over 5A, the fuse will burn. If the temperature reach 80℃ or more, the machine will stop to run. For the balance circuit , we use LG or Samsung cells, their voltage are the same, and have stable voltage, due to limited space, so we have not use the balance circuit. it dose not burn. For our battery for Irobot, the running time in the machine is similar in original irobot battery. thanks for your time to let us explain the details. we find that, you have some researches on the battery, we also willing to discuss the technology issue further about the battery with you if you have any question.
Best regards
Roses are red Violets are blue Poems are hard Flowers and Chocolate
http://www.daycounter.com/Calculators/
a1robotrepair
Robot Master
 
Posts: 758
Joined: October 14th, 2013, 8:40 am

Re: Rebuilding a Roomba Battery for Less

Postby vic7767 » May 21st, 2016, 6:53 pm

It would seem that if their statement includes "no balancing circuit is active" then I would not purchase those packs for use in the Roomba or any other similar robot.

Don't waste your time generating a response.
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 14290
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Rebuilding a Roomba Battery for Less

Postby a1robotrepair » May 21st, 2016, 7:27 pm

Hi Vick!
Hope all is well with you and yours.
I'm actually hoping one of the RR bots picks this up and sends it to Amazon, eBay wouldn't care. I also hope RR members read this and don't buy these. I wanted them for the case and balance/protection/thermistor circuit but I ended up with just a case of... I designed a simple thermistor circuit that works great in the R3 but lacks what I want to do in the 700, 800 series, charge below 600mA, above 500mAh to 4.2V/cell then trickle charge to 50mA/cell and stop charge. Back to the drawiging board.

From this photo I'm assuming the iRobot Li battery is INR 2Ah.
image.jpeg
image.jpeg (102.13 KiB) Viewed 2688 times
Roses are red Violets are blue Poems are hard Flowers and Chocolate
http://www.daycounter.com/Calculators/
a1robotrepair
Robot Master
 
Posts: 758
Joined: October 14th, 2013, 8:40 am

Re: Rebuilding a Roomba Battery for Less

Postby Kelad » July 3rd, 2016, 3:55 am

Hello All
I purchased a 780 with battery installed and who knows how long it had been without a charge. I did an initial charge via side port for days (battery light flashing fast inidicating 16 hr refresh cycle running). I then used my IMAX B6AC to cycle it several times to see the discharge value. Cant remember if i cycled down to 10.8 or 12 V. Anyone the discharge value was approx 2100mAh for all 4 cycles. I was thinking 'you beauty'. Anyway, i install this battery in several roombas and as soon as main wheels start it dies with 'please charge roomba'. I followed Vic's website and took a 35W/12V downlight. The current is only 1.9A (should be 3A), however the voltage dropped to 6-7V immediately. I then remove the load and the voltage jumps back to 16.5V. Can someone give technical explaination as to what is happening? I assume battery has an issue, but how can the IMAX say all good, but in practice its no good. Is issue the battery can not supply the required current for roomba ? :thanks:

p.s attached charging log for once i initate side port charge after 'please charge roomba' message. Log is strange as it starts at 18V and reduces.
Attachments
R32_B34 0107.txt
(251.72 KiB) Downloaded 37 times
Kelad
Robot Groupie
 
Posts: 67
Joined: February 19th, 2015, 4:50 am
Location: Australia

Re: Rebuilding a Roomba Battery for Less

Postby mfortuna » July 3rd, 2016, 7:41 am

What discharge rate did you use on the IMAX. I noticed my RC charger that is capable of charging a 12 cell pack only discharges at a max rate of 0.5A.
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
User avatar
mfortuna
Robot Master
 
Posts: 5751
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Rebuilding a Roomba Battery for Less

Postby Kelad » July 3rd, 2016, 7:55 am

mfortuna wrote:What discharge rate did you use on the IMAX.


0.4A. So my battery has good capacity discharging at 400mA, but not at 2A?
Kelad
Robot Groupie
 
Posts: 67
Joined: February 19th, 2015, 4:50 am
Location: Australia

Re: Rebuilding a Roomba Battery for Less

Postby vic7767 » July 3rd, 2016, 9:32 am

It is very possible that there are defective cells within your pack.
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 14290
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Rebuilding a Roomba Battery for Less

Postby useRoo400 » August 10th, 2016, 11:00 pm

Hi Roo hackers. I've got 3 battery shells for Roomba 400 model for sale cheap on eBay. Sub-C batts are removed, but all other wiring, connectors, thermistors, case screws, etc. are included for someone who wants to rebuild with NiMh or Li-Ion cells. I hope these will be of use to someone!

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... 1643525057

In my 2 Roo 400's I use a 3500 mAh NiMh, now almost 3 yrs old, and a Li-Ion "R4" from Lithium Power Inc, allegedly 4500 mAh, that's 13 mo. old. The promised increased run time with the lithium batt really hasn't been observed. Both run to 'yellow' display in about 45-50 minutes on bare concrete, 40" on carpet. The one clear advantage the li-ion batt has is much quicker charging and staying much cooler than the NiMh when charging.

Since both batt types and the internal charging components really dislike overheating, I've been charging them in the Roomba using a timer on the chargers: on for 15 min, off for 330 min -- 4 charge periods every 24 hours. In 15 min. the batts and Roomba case overlying the charge circuit only get slightly warm to the touch. And 330" is enough for everything to cool to room temp. I only run them once a week, so this has worked fine, and I hope will prolong life of the batts and the Roo internals!
useRoo400
 
Posts: 5
Joined: August 10th, 2016, 10:26 pm

Re: Rebuilding a Roomba Battery for Less

Postby vic7767 » August 10th, 2016, 11:24 pm

You can leave your Lithium Power 4400 mAh pack inside your Roomba and let it charge normally without having to be concerned with cell heat. The R4 pack has built in power management circuits to control charging, discharging and most important, cell balancing. If the pack is not left installed so that the float voltage can be applied to top off pack voltage then your battery pack will become unbalanced and provide poor performance.

You might consider performing the built in tests on your R2 Roomba and use the SCI port to determine the current used by the motors (drive, side brush, dust bin) to help find the reason for poor battery run times.
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 14290
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Rebuilding a Roomba Battery for Less

Postby useRoo400 » August 11th, 2016, 11:01 am

Thanks vic7767. You're correct, over-heating of batt or Roomba charger has never been a problem with the lithium R4. Definitely has been with the NiMh batt. I leave both chargers wired to same timer tho, for convenience, and alternate batts between the 2 400's. Since there are several days between running the vacs, I've been removing them rather than leave chargers on intermittent cycling all the time -- probably wouldn't hurt to leave them in place and intermittently trickle. Without constant/intermittent trickle charging they do run down a lot if left in place. Before running again, I put on the intermittent charging for at least 24 hours, so both end up in green and spend a while there.

I don't have the hardware (serial wire) or software to do the diagnostic tests. A friend here does have it but I've never been motivated to go that far since both work more or less OK. Both my 400's give nearly identical (low?) run times with both batts, consistently, so I've not actually suspected it was due to control system or motor faults. Maybe I'm wrong.
useRoo400
 
Posts: 5
Joined: August 10th, 2016, 10:26 pm

Re: Rebuilding a Roomba Battery for Less

Postby robocleaner » August 11th, 2016, 1:12 pm

a1robotrepair wrote:Here is a picture with the guts removed...

That's REALLY shocking!

To not provide a proper balancing circuit "due to limited space" when they're using extra/duff cells to fill the void is a ridiculous argument!

I've seen less scrupulous manufacturers sometimes just let undersized cells rattle around, or (a slightly brighter idea) use foam padding to at least pack the case, and in one case even a lump of scrap metaI used to compensate for the lack of weight... but to use actual cells?

I find it very strange that they're actually using decent (comparatively expensive) LG cells to execute this dangerous deception.
2 x Karcher RC3000 (2005 & 2012), Trilobite ZA1 (2005), Navibot 8855 (2010)
User avatar
robocleaner
Robot Addict
 
Posts: 672
Joined: May 3rd, 2011, 9:39 am
Location: Hertfordshire England

Re: Rebuilding a Roomba Battery for Less

Postby a1robotrepair » August 11th, 2016, 2:15 pm

The "good" cells are used ICR18650s. By the way, I have the date code correct, used, most dangerous type of lithium cell from day 303 of 2010. Total junk that's dangerous to use. Eventually the cells will become unbalanced, one will overcharge and could result in temperature runway up to 600degC. So sorry you house burn. We change name 2 on month. No us sell it.
Roses are red Violets are blue Poems are hard Flowers and Chocolate
http://www.daycounter.com/Calculators/
a1robotrepair
Robot Master
 
Posts: 758
Joined: October 14th, 2013, 8:40 am

Re: Rebuilding a Roomba Battery for Less

Postby robocleaner » August 11th, 2016, 2:47 pm

a1robotrepair wrote:The "good" cells are used ICR18650s... from day 303 of 2010... eventually the cells will become unbalanced

If the cells were already used and are being repurposed, there's a high probability that they already are unbalanced! Shocking.
2 x Karcher RC3000 (2005 & 2012), Trilobite ZA1 (2005), Navibot 8855 (2010)
User avatar
robocleaner
Robot Addict
 
Posts: 672
Joined: May 3rd, 2011, 9:39 am
Location: Hertfordshire England

Re: Rebuilding a Roomba Battery for Less

Postby a1robotrepair » August 11th, 2016, 3:21 pm

We Need a Bogus Battery Sticky! I could put quite a few in it from both Amazon and EBay. There are iRobot NiMHs out there too. They peal the stickers off iRobot batteries and put them on after markets or print their own. I'm sure it's happening with all major brand goods. Anyone who has helped set up a plant half way around the globe can tell you that while their plant is being built, another, identical plant is being constructed 3 kilometers away.
Next time I build a li battery I'll take pictures and draw some prints. I make them outside with safety glasses on. I've never had one fail or short, safety first.
Roses are red Violets are blue Poems are hard Flowers and Chocolate
http://www.daycounter.com/Calculators/
a1robotrepair
Robot Master
 
Posts: 758
Joined: October 14th, 2013, 8:40 am

Re: Rebuilding a Roomba Battery for Less

Postby vic7767 » August 11th, 2016, 5:31 pm

a1robotrepair wrote:Next time I build a li battery I'll take pictures and draw some prints.


Some of us would really like info like that and if possible a source for cells and control boards.
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 14290
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Rebuilding a Roomba Battery for Less

Postby robocleaner » August 21st, 2016, 4:29 am

Continuing on this topic of stacking 18650 cells without proper cell balancing or thermal protection (although slightly off-topic)...

Having just bought an e-bike (a motor-assisted normal pedal cycle) I was kinda curious to find out what was powering this thing...

The picture is not my own, but it's near-identical to the battery fitted to my new e-bicycle...

It claims to have some kind of battery management and there is some circuitry evident here, but the wiring suggests that at best only groups of 18650 cells are being controlled/balanced rather than individual cells. This thing is 36-volts, 10Ah (it charges from a 2-pin 42v 2A SM charger)... not the most powerful battery of this e-bike type available by any means, but it's kinda scary having this potential fireball strapped to a bicycle frame between your legs!

stacked samsung cells.jpg
2 x Karcher RC3000 (2005 & 2012), Trilobite ZA1 (2005), Navibot 8855 (2010)
User avatar
robocleaner
Robot Addict
 
Posts: 672
Joined: May 3rd, 2011, 9:39 am
Location: Hertfordshire England

Re: Rebuilding a Roomba Battery for Less

Postby mfortuna » August 21st, 2016, 9:30 am

This is the result of a 2S2C going up in flames!

I guess they caught a pro racer cheating with a similar bike setup. Said she took the wrong bike by mistake...
Attachments
100_5042 (1).jpg
Mike
Reds x 3, Dirt Dog, Disco (now a parts bot), Create, Scooba 350, and Security Dawg
Evolution Mint
Neato XV-11
User avatar
mfortuna
Robot Master
 
Posts: 5751
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Rebuilding a Roomba Battery for Less

Postby Atlantis » August 28th, 2016, 5:55 am

robocleaner wrote: but the wiring suggests that at best only groups of 18650 cells are being controlled/balanced rather than individual cells

How big are the clusters? If you have cells in parallel you don't monitor individual cells - there is no point because they have the same voltage etc - you treat the whole block like it is one big cell.
What you need to monitor is every "stage" in series. 36V=10blocks=10monitoring wires to the BMS
Atlantis
 
Posts: 7
Joined: August 16th, 2016, 7:15 pm

Re: Rebuilding a Roomba Battery for Less

Postby robocleaner » August 28th, 2016, 5:55 pm

Atlantis wrote:How big are the clusters? If you have cells in parallel you don't monitor individual cells - there is no point because they have the same voltage etc - you treat the whole block like it is one big cell.
What you need to monitor is every "stage" in series. 36V=10blocks=10monitoring wires to the BMS

I have no idea how big the clusters are, but judging from the photo and known capacity, your guess is probably right. My battery is new, under warranty, and $400... I'm not breaking it open. But however you configure it, unless extra/unseen cabling is running through the core, I can't see even the 10 wires you suggest for basic cluster balancing never mind thermal protection.

I traced the Chinese manufacturer, whose product detail claims "Large overcharge tolerance, self balance simplifying battery management system, high temperature performance". That smacks to me of the same cavalier, overtly optimistic attitude that a1robotrepair's supplier has also shown above... i.e. to hell with the extra expense of proper protection... it'll probably be okay.

I'll be charging this in my garden shed and not my home just in case!
2 x Karcher RC3000 (2005 & 2012), Trilobite ZA1 (2005), Navibot 8855 (2010)
User avatar
robocleaner
Robot Addict
 
Posts: 672
Joined: May 3rd, 2011, 9:39 am
Location: Hertfordshire England

PreviousNext

Return to Robotic Hacking

Who is online

Users browsing this forum: Shanew and 376 guests

cron