Roomba Discovery starts then stops

[bem684]

I just inherited a Roomba Discovery from a friend when he was moving to a new apartment. I saw him use it dozens of times last year but it suddenly stopped working a few months ago and he was too lazy to find out why. I took it, figuring that it was probably just clogged and dirty. After removing what must have been years worth of hair and dirt from every conceivable part of the vaccuum, it still doesn't work. I gave it a charge (3+ hours), and even ran some of the diagnostic tests I found at http://mysite.verizon.net/gsplews/Diagn ... .v.0.5.htm. It seems like the battery is charging, since all the lights worked and I had no battery issue during the tests.

What happens is as follows. When I turn it on, the Power, Spot, Clean, and Max lights are all solid green. I hit the clean button, it plays a little song with 4 ascending notes, moves forward a few inches, then stops with a new 4-note song (lower notes than the first song). The power button flashes red, the status light is a solid orange(maybe thats what people are calling amber?), and this new 4-note song keeps playing every 30 seconds or so.

I don't have a user manual since I got it second hand and I've never personally used one of these so I don't even know what's normal in terms of beeps and light patterns. Anyone have an idea as to what might be wrong? All the wheels spin, the brushes move, the side brush moves. I think I've reached the end of my competency in the matter. I am at the mercy of your genius, anonymous internet gurus. Thank you.

-Ben

[vic7767]

All indications are that even though the Roomba has charged its battery it does not have the ability to provide the operating current required to run the motors. You may suspect that the battery is shot.

[Gordon]

You may d/l a Discovery O-M from Support at iRobot.com.

[joanne0012]

I have this same situation with my 4290: "When I turn it on, the Power, Spot, Clean, and Max lights are all solid green. I hit the clean button, it plays a little song with 4 ascending notes, moves forward a few inches, then stops with a new 4-note song (lower notes than the first song). The power button flashes red, the status light is a solid orange(maybe thats what people are calling amber?), and this new 4-note song keeps playing every 30 seconds or so."

BUT if I then remove the battery, reinstall it, and press Power, then it again shows all 4 lights in green. Eventually (after a couple of minutes) the 3 lights (for Spot, Clean, and Max) turn off and the Power light is green and blinking. I've done the battery reset and overnight recharge routine.

Can this 4-note distress song be something other than a battery problem? The fact that it will run for a few inches on the green lights and THEN the Power and Status lights turn red (and will do this repeatedly if I remove and reinstall the battery) seems to indicate that the problem is some sort of blockage rather than a tired battery. This is a household with long-haired people and cats so it's possible that there's something twisted up that I haven't found.

[vic7767]

4 beeps is an indication that one of the drive wheels has issues. You might do a thorough observation of the drive wheels for some type of obstruction.

[Linux.Is.Skynet]

I have this same problem BUT I've ran the diagnostics and everything works fine. When it's not in diagnostics mode it starts forward and dies with red ring of death and an evil tune. But I can have it in diagnostic mode running for more than 30 mins (I stopped there) without any signs of weakness.

[Linux.Is.Skynet]

Well I re-read the first post and he's ran fine through diagnostic mode as well. So I just simply can't be the battery ....

[mfortuna]

Make sure the front wheel turns OK and it passes the appropriate test. That is an easy one to skip.

[vic7767]

The Roomba factory tests don't require that the battery be at the same level needed to actually run a cleaning mission. I would still suspect that the battery cannot provide the operating current needed for normal cleaning missions.

[Linux.Is.Skynet]

I noticed that the Home Base doesn't light up when I hook the fast charger up to it. Not a single light comes on even when I try the diagnostic test. Could the charger have something to do with this. When the fast charger is hooked up to it , it passes the trickle-charging current levels and voltage verification. I'm about to give up and buy a new one.. Bleh

[mfortuna]

Did you try charging the roomba by plugging directly into it?

[Linux.Is.Skynet]

Yes, I've tried that. It takes about eight hours from a fully drained battery to a "charged" one. I have another question. I've heard that OSMO by-passes some sensors and embedded software, but I'm not really sure. Because before I charged it the last time I drained the battery running in diagnostic mode (Running brushes, vacuum and drive wheels) for two hours. Maybe there's something screwy with it's mobo. I've heard of that happening. This is my first Roomba (4120) so maybe another (newer) one is due anyways.

[mfortuna]

I'm not sure how you run the diags but if you run them with the robot off the ground the tests that runs the drive motors and main brushes is in a no-load condition. A roomba running on the floor will draw more current.

As far as I read in the manual, flashing red + 4 beeps means a drive issue. Flashing red + a 4 note song means a dead battery.

[Gordon]

...before I charged it the last time I drained the battery running in diagnostic mode (Running brushes, vacuum and drive wheels) for two hours. Maybe there's something screwy with it's mobo. ...

There is no evidence that this is a mobo fault; however, I can point to a faulty Roomba operator as being a major threat to the hardware!

This is a perfect example supporting iRobot's policy to not honor a valid warranty after determining the operator had run the robot's factory tests!

You, L.I.S., have ignored the user warning given in the test procedure, where it says:

...a Roomba in Diagnostics Mode is operating without the normal safety features that stop the motors when Disco is either lifted off the floor, or a stout article becomes wedged in a running mechanism. Not only must due care be exercised when using hands to reorient Roomba while its motors are powered, but if anything halts a running motor, for more than a second or two, there is great probability that the motor or its related electro-mechanical parts could be ruined! ...

...and, specifically the phrase "operating without the normal safety features" is the critical portion. I will grant you that the warning's wording stresses operator safety, and mechanisms' safety (including their electronic drivers' safety), and says nothing about battery safety. So that warning is weak in that it is not idiot proof.

By operating Roomba for two hours (possibly locked in Test-12) it is highly likely that you drained so much charge from the battery that its terminal voltage fell far below the 12-volts that Roomba will ordinarily halt a mission. If your Roomba stopped with it Power button lit RED, then Roomba's low-voltage sensor was still operational (and the battery had a safe load shut-off), however, if it stopped with all LEDs dark it is likely that battery voltage fell below five volts (expected to be a battery-killer)!

Once the load had been removed from it, the battery voltage may have climbed a few volts as chemical equilibrium was restored, and the influence of internal-resistance removed. Assuming a no-load terminal voltage of 8Vdc, the mean cell voltage would have been on the order of 8/12 = <0.7V>. But, if one or more weak cells are in the pack, they could have shown voltages near zero level. This guesswork is merely saying: You may have ruined that battery. If not that, then you shortened whatever life it may have had remaining.

OTOH, that battery was already under suspicion as being a contributor to your Roomba's fault symptoms. I'm a little surprised that it ran Roomba (in T-12) for as long as it did. Two features may have assisted that longer than expected run-time: 1) (primary) Wheel motors were using much reduced power -- compared to normal mission speed, and 2) (minimal) Additional charge was drawn out of the battery by blowing past the normal 12-volt limit.

IMHO, further troubleshooting of your system should be preceded by confirmation that you have a battery capable of operating Roomba in clean mode. That can be accomplished by load testing the (apparently) charged battery. If interested, do an "Advanced search" using this query string "+battery +load +test", then search those results for "lamp", and visually scan through the twenty-some posts to get an idea of what battery load-testing involves.

[Linux.Is.Skynet]

Actually this is not an example of the dangers of using the diagnostic tests. All of the errors I noted happened well before I ran any of the diagnostics. The home base never worked before I received the robot,the robot didn't work before I received it and it still doesn't work at no less degree than when I received it. I posted to ask questions about my Roomba, not to be singled out as an example to the dangers of voiding my warranty or wrecking my robot by running diagnostics. I'm well aware of my own safety as not to put my hands down inside of it while it's running. I see little point to your post. My roomba is doing the same thing it did before I ran diagnostics. I like the Roomba but I've heard alot of complaints about battery failure and mobo failure in less than a year of ownership of the second generation models. So dude before you throw iRobot's equivalent of the EULA at people, please read through the other forums and maybe you'd see that there are many other people with the same problem as me.

[Gordon]

Actually this is not an example of the dangers of using the diagnostic tests.

This is a great example of the danger to the hardware while running tests on a machine that has its normal protections suppressed!

All of the errors I noted happened well before I ran any of the diagnostics.

I have not lost track of that. Did you notice my recommendation about battery testing?

The home base never worked before I received the robot,... and it still doesn't work...

You must put that Home Base out of sight until we get the robot, PSU and battery working together. Don't even think about the HB until that happens.

... I see little point to your post.

We had been running the built-in tests for a couple years before iRobot changed the warranty rules without warning. We were mystified by the Company's action. Eventually I began to see how mechanisms, drive wheels in particular, could suffer damage by allowing the robot to wander around on the floor bumping willy-nilly into objects while having no capability to do normal Clean-mode evasions. Your act of running for hours in a test-mode was the first ever reported, and allowing the robot to fully discharge its battery is an excellent example of electrical abuse of the diagnostic warning.

My roomba is doing the same thing it did before I ran diagnostics. I like the Roomba but I've heard alot of complaints about battery failure and mobo failure in less than a year of ownership of the second generation models. So ... read through the other forums and maybe you'd see that there are many other people with the same problem as me.

I am well aware of the quantity of those problems, and their nature. They are so much the same, but the owners come in with a mind set, thinking their battery just can't be the problem, after all it measures 15-volts! I've long wanted to be able to write a general procedure that owners like you could use to get from a non-working system to a working system, but the weak point in that process is always "what level of understanding does the user have about electricity, and how well equipped is s/he to do necessary tests". If people are ill equipped, one approach is to substitute known good sections into the system to see what fixes the problem. Another approach is to ship the system to a repair facility, where all will be taken care of for a fee.

Where do you see yourself in that fix it loop? If you are DIY type, have some electrical test gear on hand, then I am willing to work with you to check out your battery -- as a first step.

[sjamison]

I have 3 faulty Discoverys (given to me to repair by 3 different people). Two of them have the same symptoms at described here. New batteries were purchased (according to one of the people who gave them to me) and still the problems persist. I have cleaned them out with compressed air and alcohol swabs and still no change. any ideas for some high level trouble shooting.

[vic7767]

Begin by picking just one Roomba to begin troubleshooting and put the others away for now. You can go to my website www.vic7767.com to the "How Do I" section and perform a load test on your battery to see if it is up to par. Then even if the battery is not top notch you can still perform the Built in Tests to help identify what sensors may be at fault.