RPS Errors

[sageman]

what im saying is that if the base can be tipped back enough to get bad alighnment so it buzzes it can be tipped back enough so the contack is on/off messing up the charging circuit info on the battery letting it go too far down......

[Racer686]

I've started getting RPS errors lately, after 3 months of uneventful cleaning by my Neato. I have been able to stop the errors by blowing compressed air into the turret. Dirt or hair must be getting in and throwing this off. Everytime I've had the error, it's cleared by blowing the air (happend 3 times so far).

[sageman]

ahh so dust is getting onto the sensors in the turret

[Fraggboy]

Interesting. Wonder if a piece of clear tape over both holes would work? I would think that the tape would scatter the beam and would produce ill results.

I haven't had an RPS error (yet), but if I do I'm going to do that first thing. Thanks for sharing that information.

[djos]

I believe the RPS errors are FW not Hardware related as on a couple of runs that I have manually initiated, after overnight charging, the XV-11 has got halfway thru the 2nd room (doesn't seem to matter which end of the house it starts from) and then has the RPS error. If I press "ok" it then starts back up and vacuums the rest of house with no problems at all.

Now it it completely random as to where it fails with an RPS error, most of the time it does the majority of the house before failing with an RPS error.

I have reported this to Neato.

For anyone having the climbing dock problem and or buzzing dock there is a very simple solution: 3m picture hanger strips



I placed 1 at each end of the dock as low down as possible and 2 (1 above the other) right in the center of the dock as low down as possible and My Neato never tries to climb it's dock any more. These are a little like Velcro (but not) and easily removable if you need to move the dock for whatever reason.

The dock climbing and buzzing is imo caused by the lack of height on some skirting boards allowing the Dock to rock backwards as the Neato contacts it, the 3m picture hanger strips prevent this and keep the Neato Dock firmly upright.

[johnnyh]

Just a followup to the RPS issue. As I've mentioned in the past, i just started getting the dreaded RPS error after owning the bot for just over a month. I called cust support and was told to turn off bot and plug it in directly for several hours and that should clear the error. I did this and the RPS error still remained. I tried again, leaving it plugged in overnight, and still the RPS error. (I didn't have much faith that this would fix the problem anyway).

Now for step 2. As Racer686 mentioned, I used dust-off to blow out the turret (though no visually noticeable dust/dirt). I then ran the unit and it did an entire cleaning cycle with no RPS errors. It's currently in suspended cleaning mode as it recharges.

This makes some sense, and my theory is that there is an optical encoder at base of turret (possibly a photo encoder at edge of bottom of rotating component, with a square notch pattern machined into the bottom of the rotating ring). This allows the unit to determine the angular position of the rangefinder as it rotates. I can see this encoder getting dust in it, thereby preventing the unit from getting accurate angular positioning info about the position of the rangefinder.

If this is the problem then I'm okay with the periodic spray of compressed air around the turret area as part of periodic maint. of the unit.

In the very least I think Neato should change the wording on the display when it determines that the optical encoder needs dusting out (if, of course, this is the problem)

Once Anita finishes her second cycle I'll post how she fares...

John Hanzl
Author, Out of Hell's Kitchen
http://www.outofhellskitchen.com

[johnnyh]

Quick follow-up -

The bot came out after her recharge to finish the cleaning cycle and once again - no RPS errors.

So - RPS errors with every time the unit ran for the past five consecutive runs (including after the power off / plug-in full recharge process), but after blowing out the turret with dust-off the bot as run just fine...

John Hanzl
Author, Out of Hell's Kitchen
http://www.outofhellskitchen.com

[neato]

Interesting. Wonder if a piece of clear tape over both holes would work? I would think that the tape would scatter the beam and would produce ill results.

I haven't had an RPS error (yet), but if I do I'm going to do that first thing. Thanks for sharing that information.

Placing tape or other covering over the RPS will distort the optical path and that will make the calibration invalid. The Neato would drive around blindly.

As far as low batteries causing RPS errors or clearing them by charging up the Neato using the external plug, those are old wives tales. They assume the RPS is run by an unregulated power connection to the battery.

If the RPS error immediately returns after clearing it you can try the system shutdown on the support menu. Then put the Neato on the base to charge and set the clock.

[sageman]

oh ok out with one of my therys

but i did sugest blowing on the laser turret a while ago a week or so as far as i can renember

[djos]

I got a can of compressed air yesterday arvo and cleaned out the turret so im curious to see if our Neato completes its mission and finds its way home without any dramas today.

[djos]

FIX CONFIRMED: I tried cleaning the laser turret with a can of compressed yesterday and our Neato is now perfectly vacuuming our house again including an extra room I added in today as a torture test (requires a mid mission charge to complete it).

[cool8man]

I'm glad you guys figured out a solution for the RPS error, as it was a deal breaker for me on deciding to purchase a Neato. However most people will not know to clean out the sensors with compressed air and will just assume the Neato is broken, they will return it and give the company bad reviews. Neato needs to fix the problem so that it does not happen at all.

[james_mac]

I was having RPS errors as well. Neato actually replaced my robot after their suggested reset and direct charging didn't work. I guess I found this suggested fix a little too late. I can say that their support is very friendly and helpful.

[djos]

Neato were going to replace mine as well but I was confident there was nothing physically wrong with it - I thought it was a FW bug but dust build up is nice and easy to solve.

I think Neato need to redesign the RPS turret with a clear cover over both holes or some aerodynamic trick to use air flow to clean the optics while it's spinning.

[johnnyh]

I think Neato need to redesign the RPS turret with a clear cover over both holes or some aerodynamic trick to use air flow to clean the optics while it's spinning.

One small point - and I could very well be wrong - but I'd make sure to spray compressed air all around the turret, not just where the laser lens and CMOS sensor holes are, because it's my suspicion that the thing that needs dusting is down in the base of the turret and not the laser / receiver at all...

John Hanzl
Author, action / adventure thriller
Out of Hell's Kitchen
www.outofhellskitchen.com

[djos]

You can very clearly see the laser and its sensor inside the 2 holes so I'll have to disagree.

[johnnyh]

You can very clearly see the laser and its sensor inside the 2 holes so I'll have to disagree.

Quite correct - you can see the laser and the sensor, however the turret has to have some kind of position detecting system to ensure the robot always knows exactly where the laser is pointing as it rotates (photo interruptor, hall effect sensor, whatever) and I suspect that it's this sensor that is getting dusty, and that sensor is probably located out of sight down below the base of the turret (but still 'reachable' by dust). I've built devices that use this type of optical encoder when i was a broadcast engineer, and I've experienced issues with dust build-up.

Of course it could also be dust on the CMOS sensor itself - but just make sure try to dust off all around the turret to reach the out-of-sight sensor I'm referring to...

John Hanzl
Author, action / adventure thriller
Out of Hells Kitchen

[djos]

Fair enuf, I only cleaned the 2 holes and the results where immediate - i assumed that the spinning turret has a simple optical "reader" similar to what you used to find in optical-ball mice to read it's position.

[johnnyh]

One other possible area of trouble for getting dust on it and causing RPS errors (and subsequently getting fixed by blowing out the turret area with dust-off) would be the slip-rings and brushes that probably exist in the shaft of the turret. These are required to make electrical connections from the robot motherboard to the spinning part of the turret. See the following image by Fraggboy that shows the four wires that come up out of the hub of the turret.



These would be attached to the slip-rings on the shaft, and the wires from the robot's motherboard would be attached to the corresponding brushes. This allows for the electrical connection while the turret spins.

Anyway, if that connection got dusty then the signal could become unreliable and therefore cause RPS errors.

Just a thought...

John Hanzl
Author, action / adventure thriller
Out of Hells Kitchen

[johnnyh]

I've found a link to some great images of the turret taken apart, down to where you can see the optical encoder mounted to the underside of the rotating component (pic #9) as well as the fixed plastic tabs mounted to the turret enclosure the the encoder rotates through (pic #16). This encoder is what I suspect crates RPS errors when it gets dusty (prob not only source, but a good culprit indeed).

The slip rings are also suspect. Interestingly, the original prototype of the LDS (known as the Revo) used a very short wireless transceiver to communicate between main unit and LDS. No doubt they went to slip rings due to cost, or perhaps RF noise interference...

http://random-workshop.blogspot.com/201 ... rther.html

John

John Hanzl
Author of action /adventure thriller
Out of Hell's Kitchen