Even a new battery won't hold charge

The very latest news and updates for the iRobot Roomba robotic vacuum, the Scooba Robotic Washer and the Dirt Dog workshop sweeper. All discussion and troubleshooting questions go here.

Even a new battery won't hold charge

Postby Arctander » April 5th, 2009, 2:08 pm

Hi everyone, I searched this forum for some answers, and only seemed to find myself a bit more confused, so I figured I'd register and just fire it out there since there seem to be a lot of you with a ton of knowledge about all this. I realy did try to find an existing topic, and found some similar, but none quite seemed to answer my problem since it's a new battery.

Here's the story - just under two years ago I got the pet roomba - I think technically it's the: iRobot Roomba 4105 (at least that's what it looks like.) It ran fine for a year, everyone was happy. It put up with the "perfect storm" of hair - my fiancee's long black hair and her dog's endless piles of short blonde. But after a year the battery just gave out. I ordered a new battery, it ran fine for about 6 months. I've always been very good about cleaning the bugger and taking care of it (mostly talking nicely to it when no one else was around.) All in all I was a happy customer.

However after the 2nd battery made it only 6 months until it started dying sooner and sooner - and more often in random places, usually not making it back to the charge base. Over the next two months it got worse, dying faster and faster, til it got to the point where it'd make it just 15 seconds before dying. I read the forums, I tried the "hold power in for 30 seconds" then charge battery for 72 hours. That didn't fix it.

So, I got a new battery this week. Charged it up for a full day, it ran the roomba for about 15 mins, then just died on the floor - didn't make it to the base. I charged it again overnight, and it ran for 15 seconds. I'm not sure what to do. It's a new battery. If it were some diagnostic problem, why would it have made it for 15 mins on the first charge? I'm charging it w/o the base now in case there's something wrong there - but does anyone have any particular suggestions?

Thanks in advance for any suggestions...
Arctander
 
Posts: 6
Joined: April 5th, 2009, 1:58 pm

Re: Even a new battery won't hold charge

Postby vic7767 » April 5th, 2009, 3:19 pm

You might consider that the internal charging circuit of the Roomba may have failed. The reason the Roomba ran at all was due to the residual charge within the new battery. Leave the battery charging overnight in the Roomba without the home base. Remove the battery and measure the voltage level after the overnight charge, it should indicate a level of 16 to 17 volts if the charging system is working. You might also have just a power supply issue, verify that the power supply is delivering 22 volts at the barrel end. The inside of the barrel is positive, the outside is negative.
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 11442
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Even a new battery won't hold charge

Postby Gordon » April 5th, 2009, 4:00 pm

After you have started the charging that vic7767 suggested, you may acquire a certain degree of confidence that a reasonable charging current is coursing through that battery by flipping over the robot after 30 to 60 mins of charging so you may lay a hand on the battery-case. You should sense that case temperature to be much warmer than your body temperature. If it remains cool, and if you have confirmed the charging power supply to be OK, your robot very likely has an electronic failure that will require some outside help to correct.
Gordon
Robot Master
 
Posts: 4081
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: Even a new battery won't hold charge

Postby Arctander » April 5th, 2009, 6:45 pm

Thanks for the replies. I cleaned the bedevil out of it earlier today and now it's plugged directly in (no home base) I'll leave it charging for the night and check it tomorrow.

vic7767, how does one measure the voltage level / supply?

Gordon, the battery has been charging for a couple hours and is cool to the touch. I assume that means the issue is that it simply is not actually charging it at all - correct?
Arctander
 
Posts: 6
Joined: April 5th, 2009, 1:58 pm

Re: Even a new battery won't hold charge

Postby vic7767 » April 5th, 2009, 7:10 pm

Arctander wrote:vic7767, how does one measure the voltage level / supply?


I've included a pic showing how to measure the PSU output. The same meter is used to measure the battery voltage, you just need to remove the battery and make electrical contact with the two terminals on the battery. The terminal farthest away from the battery is positive.
Attachments
PSU.JPG
Roomba and Neato Mods, come visit: http://www.vic7767.com/

http://www.Robot-Doc.com/
User avatar
vic7767
Robot Master
 
Posts: 11442
Joined: January 14th, 2006, 8:31 pm
Location: Haughton Louisiana - USA

Re: Even a new battery won't hold charge

Postby Gordon » April 5th, 2009, 8:58 pm

Arctander wrote:...Gordon, the battery has been charging for a couple hours and is cool to the touch. I assume that means the issue is that it simply is not actually charging it at all - correct?
Correct.

Probably, that battery is chargeable. But, prior to tasking a Roomba to charge it, you would like to know that its terminal voltage is 12Vdc, or greater.

Also, checking the PSU's voltage, as vic7767 outlined, at its pigtail-plug is a good idea, but is insufficient to prove the PSU can deliver current. A second test, using only a light bulb, will draw current from the 22Vdc source and give moderate confidence that the PSU can do work when tasked. I just connected a household type 100-watt (a.k.a., "95 watts") candelabra-based lamp across my Fast Charger's power-plug, and it drew 0.33A (and made the lamp glow). The PSU's voltage held at 22Vdc while that load was applied.

If your PSU passes voltage and current checks, the charging fault is inside Roomba, and will require repair to, or replacement of the main circuit board assembly.

If you need a picture showing the current-check arrangement, I can probably make something similar to what vic7767 did for the volt-meter check.
Gordon
Robot Master
 
Posts: 4081
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: Even a new battery won't hold charge

Postby Arctander » April 6th, 2009, 7:19 am

I guess I need to a radio shack to get whatever that device is you are using in the picture... I assume it something called a "voltage checker"?

Like this? http://www.amazon.com/Wilmar-Battery-Alternator-Voltage-Checker/dp/B0002KO1HQ

Thanks a bunch for the feedback vic7767 & Gordon.
Last edited by Arctander on April 6th, 2009, 8:22 am, edited 1 time in total.
Arctander
 
Posts: 6
Joined: April 5th, 2009, 1:58 pm

Re: Even a new battery won't hold charge

Postby mfortuna » April 6th, 2009, 7:43 am

Look for a Digital Voltmeter (DVM).
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: 5021
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Even a new battery won't hold charge

Postby Arctander » April 13th, 2009, 1:37 pm

Well, it looks like the PSU is fine.
Image
Unfortunately, plugging the PSU directly into the roomba doesn't give it a charge (it lights up and pulses like it would charge it, but it only lasts for a few mins.)

I assume this means that the chip inside the roomba is fried and needs to be replaced then - right? Is that something I can do myself? I'm not afraid of opening the roomba up and mucking around inside of it - as it's already out of warranty and it's worthless as it is now...
Arctander
 
Posts: 6
Joined: April 5th, 2009, 1:58 pm

Re: Even a new battery won't hold charge

Postby mfortuna » April 13th, 2009, 4:33 pm

People have changed both U2 and U4 and very often it fixes the issue. You can buy the transistors from protech robotics, mouser, digikey etc. The part number is STN3PF06. Search the forums and you will find more information on changing them.

Even though these parts are surface mount, they are relatively easy to replace with average soldering skills.
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: 5021
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Even a new battery won't hold charge

Postby Arctander » April 13th, 2009, 5:16 pm

Thanks for the help mfortuna.

How hard is it for below average and well below average soldering skills?

I don't think I've soldered anything since, well... 1980 (gah that makes me sound old...)
Arctander
 
Posts: 6
Joined: April 5th, 2009, 1:58 pm

Re: Even a new battery won't hold charge

Postby mfortuna » April 14th, 2009, 7:55 am

You have to be careful not to damage the delicate pads. What I do is clip the smaller leads off, then heat the large tab and remove the MOSFET. Then I remove the remnants of the small leads. Just heating them with an iron and flicking them off works great. To install the new MOSFET I first line up the pins and heat one of the small ones untill the solder remaining on the pad flows. Then I solder the other small pin and add a little solder to it. This is followed by the large tab and then back to the original pin. Not much solder is required since a fair amount remains on the pads. I just try to freshen up the connection.
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: 5021
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Even a new battery won't hold charge

Postby Arctander » April 14th, 2009, 6:23 pm

Thanks for the follow up mfortuna. I'm heading out of town for a wedding but when I get back I'm going to do my darnedest to figure out what all that means and not break my roomba while trying to repair it :D

Is there a site that has a diagram showing where all those different parts are on the chips? Or are they all just labeled once I take it apart and look closely? As it stands now I don't have any idea what exactly the U2, 4 or the MOSFET are.
I was able to find the STN3PF06 when I googled, so I have at least a clue.

In the end, I'm going to be working to replace the failed part(s) on the main circuit board, right? And if I blow that, I should be able to just order one (like this one link) and be good.

Right?

I really appreciate all the help and sorry for being such a noob.
Arctander
 
Posts: 6
Joined: April 5th, 2009, 1:58 pm

Re: Even a new battery won't hold charge

Postby mfortuna » April 14th, 2009, 11:24 pm

If you search these forums you should find pictures of U2 and U4. They are labeled and are on opposite sides of the board.

No worries, we were all noobs at one point!
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: 5021
Joined: February 5th, 2006, 10:35 am
Location: NH

Re: Even a new battery won't hold charge

Postby wwallender » June 16th, 2009, 9:26 am

hopefully this will help
your battery should be out putting 16-17 VDC
my charger is out putting 23VDC (charger works great) if your roomba battery keeps dieing check the voltege with a meter if it still is above 12 VDC your battery is fine if its below your battery is shot
just for the heck of it (it just may help) measure the voltage output to your dust bin fan i couldn`t get a picture
but it should be 16-17 VDC
107_0102.JPG
1800 MAH Nicd battery out putting 17 VDC
(13 days old) 10-20 cycles

107_0100.JPG
2600 Mah battery outputting 16 VDC
about 3 months old 20-40 cycles

107_0114.JPG
charger out putting 23 VDC (note i have 2 chargers and they both output 23 VDC
wwallender
 
Posts: 3
Joined: June 16th, 2009, 8:55 am

Re: Even a new battery won't hold charge

Postby Gordon » June 16th, 2009, 11:44 am

wwallender wrote:...your battery should be out putting 16-17 VDC...
You should qualify that level of battery voltage by saying something like: "...outputting 16-17 VDC right after charging.".
my charger is out putting 23VDC ...
You are misleading yourself and possibly others by using a meter that rounds its reading to the nearest volt! iRbt's PSUs are voltage regulated and generally output a voltage the well matches what's printed on the case label. If you had noticed vic7767's pictured measurement of 22.5Vdc, upstream in this thread, you might have saved publishing incomplete data.
... if your roomba battery keeps dieing check the voltege with a meter if it still is above 12 VDC your battery is fine...
That is not necessarily true. Member glo69 harped at us about finding batteries with terminal volts of ~14 that would not support a Roomba (4XXX) load. Here is a picture that shows the effect:
Image
The story that goes with the image is here:
viewtopic.php?t=9845
Gordon
Robot Master
 
Posts: 4081
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA


Return to iRobot Roomba and Scooba Chat

Who is online

Users browsing this forum: Lacheby, TechGuy, vic7767 and 157 guests