Neato BotVac 75/80/D85 "...push switch Down" (RESOLVED)

[Deleted_Permanently]

Hi,

For now I have replaced mostly wheel motors, front bumper buttons and brush motors on neatos. But now I have 4 robots with mysterious problems and I'm unable to find solution. I have to rescue these 4 botvacs, because mehanical and visual side of these robots are excellent. Problematic models: 75, 80 and two D85.

All of those robots have same problem: I turn robot ON, place it to DOCK, it starts to charge, after about 5 to 15 minutes it displays message: Please open my dirt bin & push switch Down to ON. If I turn robot OFF and ON again, it may charge battery to 100%. But sometimes it need another restart. If battery is full, robot will work around 1 hour to 1.5 hour. After fully cleaned my apartment robot returns to DOCK, charges battery to full and everything seems to be OK. Something happens in between night and morning, in the morning same message again and all restarting process starts again.

Neato Control shows "BatteryFailure 1" if this problem occurs.

For months I have browsed the internet and looked possible solutions, but nothing works. Before anyone thinks "have you tested new battery?", yes, I have, all four of robots have new batteries inside.

Maybe MAX17047 has something to do with this, but so far every problem on internet with this IC has been on temperature side.

What I have done so far:
1. Tested and cleaned main switch - seems to be OK (occiloscope and laboratory PSU stress tested)
2. Solder shorted main switch to ON position
3. Replaced all electrolytic capacitors
4. Resoldered every possible THT component pad
5. Replaced battery connector and pins with new MOLEX ones
6. Soldered wires from battery to mainboard directly
7. Used Neato Control software to see if thermo sensor is OK
8. Tested DOCK with ocsilloscope, when charging voltage seems to be OK

I hope that someone can help me to find solution for that. It would be very sad to scrap these.

[CodaCM]

I will look into this further when I have a chance if you get no more replies. I hope you can get these Neatos working.

[Deleted_Permanently]

Have done some test.
When i restart robot and battery temperature is above ~35°C, then it seem to charge to full without any error and without any need to restart. But when robot stays longer in to dock (like over night), problem returns.

Robot with fully charged battery will run/clean over a hour without any problem.

[EDIT]
It seems, that not only temperature must be right, but battery voltage must also be right, to start charging. Right now battery voltage is 14.3V and temperature is 41°C. Charging process does not start, neato shows empty battery icon. No charging, no error (about 20 minutes), dock voltage fluctuates between 12V...18V.
It took 25...30 minutes to start charging. Charging completed within ~10 minutes, end voltage 15.6V and temperature 48°C. Now shows 100% and no error.

[EDIT]
After some tests, it occurred that charging starts if battery temperature is below 39.6°C.

So far what's happened:
1. Neato won't charge battery. Battery at room temperature and about 14V.
2. Did some restarts, that forced littlebit juice into battery and rised temperature and voltage.
3. At some point it charged battery to full, temperature ~48°C.
4. After that I did restart neato to see what will happen next.
5. Neato did delay charging (and display empty battery icon) until battery temperature dropped below 39.6°C.
6. Neato started charging. Battery temperature rised to ~42°C, voltage rised to 15.3V and then that error occurred (...switch Down).
7. Charging ended, error on display. Battery voltage dropped to 14.4V and temperature rised to 45.5°C after 20 minutes.

[Deleted_Permanently]

After hours and hours of testing I found anomaly that produces this error. When battery is almost full and you switch robot from slide switch to "ON" and after a while error occurs (robot in dock). MAX chip measures voltage/temp rise delta to determine battery status - I have measured, that with all my four robot/battery combo, temperature does not have time to rise enough fast with voltage rise, and then MAX decides that battery is bad. This is not realistic that all four new batteries are bad, but I think that all four MAX chips are bad - these are old robots. So I decided to test this theory. I added 10R 2W resistor, 15V 1.3W zener and 0.5A fuse in series and soldered to battery +/- pads. Zener was placed aside battery NTC temperature sensor. This will produce small amount extra heat if battery is near full (over 15V) and still charging. And it worked with all four robots. No error anymore.

[EDIT]
After this hack was applied, first two clean cycles battery lasted only 30...40 minutes. Third time already over an hour. So it seems MAX learning new battery charging curve. So far all robots behave same. Charging is normal and no errors.

[Deleted_Permanently]

Robots have worked 5 days without any problems. Batteries did last over an hour. I can't say exactly how much over an hour because when robots finished their job, the battery icon showed more than half of the battery left.

This "benchmark" is based on my D5 with an old 5000mAh Li-Ion battery. D5 gets job done with an hour (+/- 10 minutes depending day).

Everything seems to be OK with this hack. But I can't guarantee that this will work with all robots with that "...push switch Down..." error.

[Deleted_Permanently]

UPDATE! Everything OK with batteries and robots. So far no problems.