Stuck Wheel Problem and Fix

[aziesemer2]

I bought a Neato XV-21 "used - like new" from Amazon a few days ago during a business trip to the US. The robot arrived in its original box and appeared never used but after charging it for 1-2 hours, to my surprise, the robot didn't worked as expected. It kept moving in circles, back and forth... so I guess I got a defective robot which was returned to Amazon.

But in my second attempt, it cleaned all the apartment and returned to the base accordingly. It's insane the amount of dust it collected from the carpet - hotel carpet, not mine

After successfully test it a few more times, I decided not to return it to Amazon but bring it with me to Brazil where I live.

Unfortunately, after get home, it started to present the stuck wheel problem much more often (I got this message on the display). So I decided to inspect it closer and figured out that one of the wheels was much harder to turn by hand than the other one. Because the warranty is not valid in Brazil and I'm not supposed to return to the US soon, I decided to fix it by myself.

I removed both wheels and pulled both drive motors off. Than I realised that the problem was in one of the motors that wasn't spinning smooth enough. I contacted Neato's support and asked them if they could send me a new motor by mail just to hear that they don't send parts and I should return the robot so they can fix it... I searched in eBay and could find it for $20 + shipping. Before doing that, I decided to plug the motor (+ wheel) in a 5-9v power adaptor and run it for a few hours to smooth the mechanics inside... During this time it became obvious that something was wrong with the motor since it constantly changed its speed between fast and slow periods. Setting the power adaptor to 9V appeared to work better to fix this issue, but I had to pay attention to the temperature and let it "rest" in 5V or unplugged when it got too high (it's not an exact science, sorry). After some hours the motor finally appeared stable, for good or bad, I had no idea yet...

It worked!!!! After putting everything back in place the robot never got stuck again and both wheels are now equally "soft".

Hope my post can help you guys having the same problem...

Difficult level: easy if you have the right tools and a minimum knowledge of electronics.

Detailed instructions:
1 - Turn off the robot
2 - Remove the 4 screws from the wheels cover and open it
3 - Pull the wheel off, removing the spring from the botton and disconnecting it from the main board. (special attention to the left spring so it won't touch the main board)
4 - Get a power adaptor and connect it to the red and black wires on the plug of the motor: http://www.hizook.com/files/users/3/Nea ... _wheel.jpg. The 2 adjacent lateral wires on the right of this picture. In my case, it was easier to connect the wires to the back of the plug, instead the front. You can use an alligator and connect direct to the wires too.
5 - Run it for a few hours as described above, adjusting between 5-9V (I used an 1A max adaptor). Don't forget to reverse the polarity too to spin it in both directions.
6 - Put everything back in place and you're done!

[vic7767]

Was shaft alignment checked and was any lubrication attempted?

[aziesemer2]

Was shaft alignment checked and was any lubrication attempted?

As long as I could see, the alignment was OK. Didn't try any lubrication inside the motor since I thought could make it worse.
The principle I used was the same that happens with new cars that need to run a few thousand miles to achieve it's optimum performance.

[glnc222]

So the wheel motors aren't digitally speed controlled with pulse modulated signals like the brush and vacuum? And don't run on 15V like the vacuum? It seems to have speed control in operation.

[aziesemer2]

So the wheel motors aren't digitally speed controlled with pulse modulated signals like the brush and vacuum? And don't run on 15V like the vacuum? It seems to have speed control in operation.

Yes, they are digitally controlled but I'm not sure about the 15V power source. I read somewhere it was just 5V. In fact, the wheel speed when I run it on 5V is very similar to the speed Neato actually works. When I increase it to 9V it runs much faster than that. I definitely wouldn't run it as far as 15V!

[glnc222]

How does it run at all without the pulsed rpm signal? The vacuum power line is on without turning at all, comes on under USB just going into testmode. Aren't there four wires on the motor?

[aziesemer2]

How does it run at all without the pulsed rpm signal? The vacuum power line is on without turning at all, comes on under USB just going into testmode. Aren't there four wires on the motor?

It's a standard DC motor with a RPM reader in it's back. You don't need to codify the pulse to make it work because it's not a step motor or something like.
The RPM reader is similar to the one you put in your bike. That's the reason for the extra wires.

[vic7767]

They are 12 volt DC motors that will run with dc power applied to the (+), (-) terminals and are also controlled using a PWM signal via the microcontroller inside the Neato. H-bridge control in concert with PWM leads and tachometer (rotation reports) run both drive wheel motors.

[neato]

[glnc222]

Nice to learn about H-bridges.
Looks like Neato reports stuck wheels when the motor can't come up to demanded speed according to tachometer.

Might use USB just to run the wheels conveniently -- if can set fast enough. Manual at website shows SetMotor operates wheels -- set unit upside down, or on blocks (if needed a command to ignore microswitches under the wheels sensing springs extended all the way as when off the floor -- or just tape the wheel mounting pushed in).
Wheel assembly pics http://random-workshop.blogspot.com/201 ... embly.html. H-bridge http://en.wikipedia.org/wiki/H_bridge. Fancier fan has circuits inside the casing http://www.robotreviews.com/chat/viewto ... or#p106626.
Nominal 12v gearmotor I used for side brush worked fine on 15v, and when pulsed even less likely over-drive coils; large built-in margin for reliability. Fan is labeled nominally 12v but runs on 15-- at 90 per cent capacity though, probably just for reasonable discharge rate, insufficient battery. Have to check what Neato uses on wheels. Neato voltage drops a volt or more as battery discharges over each run, so starting at 15 possible.

[aziesemer2]

Yes, it would be much more convenient doing this procedure without need to open the robot. Plus, Neato could fix this problem without ask their customers to send it back... it would be much more cost effective since it seems there are many people around with this problem.

Or a custom firmware update...

The only problem of running it directly from the robot is that I'm not sure if it have enough power to run the defective motor. It may not be a good idea...

[glnc222]

Did you have to pull the motor out of a gear train? Maybe should not run with all the gears under USB without disassembly. The speed control though, might solve the heating problem. But then a fast speed is needed to get enough smoothing... You can use 12v portable (or not) car batteries, too.

With the main load of the vacuum off, battery should give a lot more run time on USB. I think motor commands are disabled under USB when the charger is attached (so you don't trip over the cable -- product liability dragons; and new models don't have the charging cable anymore).

Thanks for cross-reference posting.

[aziesemer2]

Did you have to pull the motor out of a gear train? Maybe should not run with all the gears under USB without disassembly. The speed control though, might solve the heating problem. But then a fast speed is needed to get enough smoothing... You can use 12v portable (or not) car batteries, too.

With the main load of the vacuum off, battery should give a lot more run time on USB. I think motor commands are disabled under USB when the charger is attached (so you don't trip over the cable -- product liability dragons; and new models don't have the charging cable anymore).

Thanks for cross-reference posting.

I didn't pull it out because I decided to run it as similar as I could get from the real world usage. Not sure if it makes any difference, though, probably not.
You can use mobile phone chargers, 9V batteries or whatever can generate anything from 7-9V. More than that I wouldn't advise since the stucked motor probably generates more heat than a normal one.

[vic7767]

It is best to remove the motor and dis-assemble it in order to locate what is causing the binding. That is not a normal condition and a break-in run is not necessary with new robotic vacuums. aziesemer2 was just lucky that the motor windings didn't short out and burn up the motor from the high amount of current utilized to drive the wheel.

[aziesemer2]

It is best to remove the motor and dis-assemble it in order to locate what is causing the binding. That is not a normal condition and a break-in run is not necessary with new robotic vacuums. aziesemer2 was just lucky that the motor windings didn't short out and burn up the motor from the high amount of current utilized to drive the wheel.

I don't think it was just luck. Anyway, disassemble the motor is a very difficult and risky task. I wouldn't recommend doing that, but buying another one ($20-30).
I believe the problem was some dust inside the motor, misalignment or oversized component and the break-in fixed it.

[glnc222]

Roomba motors I had could not be disassembled except for attached gear cases in the motor housing; the rest was sealed by soldered boards related to brushes. Are the Neato motors more accessible? Fun taking things apart...if they can be reassembled.

[vic7767]

The Neato drive wheels can be dis-assembled much easier than the early R2 4XX, 4XXX model Roombas. The 500, 600, and 700 Roomba modular wheels can also be dis-assembled with the proper tools. Of course all dis-assembly can be difficult if the correct tools are not available.

[prjimm01]

great stuff but its easier for me just to return it and not endanger the warranty

[vic7767]

great stuff but its easier for me just to return it and not endanger the warranty

Very good point. The many posts I make on any of the rr forums are not intended to have owners violate any specific warranties that are in place on the robots they are having difficulty with. There are owners that have robots that are either beyond the warranty period or just curious as to what resolutions might be posted that they could perform on their own robots if they decide to do so.

[SyRenity]

I searched in eBay and could find it for $20 + shipping.

Will XV-11 motors work on XV-21?