iRobot Scooba: experience and solutions to common problems of a washing robot cleaner

    iRobot Scooba (models 5800, 380 and analogs look almost identical)

    When I first heard about the robot vacuum cleaner, I immediately wanted to buy one to try it in the “field conditions”. I read about them somewhere in 2010, and after a while I managed to fulfill my little dream. Then I randomly met the manager of the iRobot product distribution company, and I was offered to pick up a Scooba 5800 stand at a good discount.

    Without thinking twice, I agreed, brought the novelty home, tested it and ... the robot completely met my expectations. He washed the floors just perfectly, in any case, much better than I do, even when I try very hard. Under the cut - a few more memories, experience with the robot and the solution of several common “iron” problems that are common among all Scooba, which lead to the inoperability of the device.

    A few more memories and evaluation of the work of Scooba

    The robot that I bought successfully worked for several years. Then it seemed just amazing to me that such a complex device works without any problems, despite daily contact with water, which is simply fatal for any electronics.

    He worked exceptionally well - he washed almost everything, leaving behind a perfectly clean floor. I forgot what mopping is. He washed everything even in the common corridor of my house, which the tenants did not wash, because the corridor was common, and the cleaning lady from the public utilities could not reach him because of the installed doors. In general, perennial dirt and dust (of course, there were not directly pieces of dirt, otherwise I would not mock the device) was washed in just one robot cleaning session lasting about an hour.

    Yes, the robot does not touch the corners, leaving small dry areas a few centimeters in size. But compared to the usual quality of manual floor cleaning, it’s just space. It turns out especially good if you add a little cleaning agent (you can buy a specialized one from iRobot, or you can add a couple of drops of the usual one, the main thing is not to foam too much, otherwise the robot will not be able to work normally).

    Scooba 5800 shortcomings and honorable death from water

    During this time, a couple of Scooba flaws came to light. The first is the plastic lock-handle of the robot tank. This handle locks the tank. And it is completely plastic, despite the fact that the force when pressing the handle should be sufficient to open the lock and the tank can be removed.

    That same pen. By the way, her 3D model for a 3D printer is on the network on Thingiverse.

    I understand everything, but this pen fails in a year or two (for someone, in general, for several months). Perhaps this is deliberate obsolescence, although, rather (judging by the reliability of the robot itself and its consumables) there is simply an oversight of the developers.

    The second drawback is that the robot battery, Ni-Mh, by 4.5 Ah * relatively quickly lost capacity. Yes, instead of a lithium battery with a robot, a battery with a “memory” and relatively quick degradation was supplied. I had to buy a battery, and then I went somewhere for a few weeks.

    Upon return, it turned out that the robot does not want to continue its career. He started, crawled on the floor a little and almost immediately gave a sound signal about the problem. There was no time to figure it out, the device seemed very complicated (it's a robot!), So I just left it alone (but did not sell it).

    IRobot Scooba Vacuum Turbine

    A year later, finding time to figure it out, I realized what his problem was. It turned out that the impeller of the washing system stalled, which drove the flow of water and air through the robot. She simply withered and could not start. I disassembled the robot, slightly shifted the blades of the turbine (it really dried out a little), assembled it, and it all worked again. But the joy was short-lived - I redeemed the robot in water, in its entirety, and here its subtle nature could not stand it anymore.

    As far as I remember, this year happened in 2013.

    New - well forgotten old

    Over the past few years, I moved to another country and changed several apartments. All this time there was no time for robots, although lately I recalled my Scooba, my floors, and I wanted to delegate washing duties to the robot.

    But now, for some reason, I felt sorry for the money, because even a used Scooba model in normal condition costs about 150 euros (you can find it cheaper, but no one guarantees the operability of the device). In the end, I found a compromise with my own greed and I decided to try my luck by buying another Scooba (in addition to the existing one) inoperative to fix it. The idea was to use parts from one robot to solve problems with another.

    Nevertheless, over the past years, a lot of materials have been accumulated on the Network about the repair of these and other robots, and I have gained more experience in fixing them. And I ordered the Scooba 380, which was sold for a ridiculous 35 euros, with a battery, a brush and filters.

    Characteristics of the 380th model
    • Type: Robot
    • Sensors: Optical
    • Cleaning: Wet
    • Dust collector: cyclone filter
    • Power controller: no
    • Cleaning area limiter: virtual wall
    • Modes of movement: in a spiral, zigzag, along the walls
    • Cleaning Modes: Local Cleaning
    • Installation on the charger: manual
    • Area processed on a single battery charge: 80 m2

    In addition, I asked relatives to send me an old Scooba 5800, which was still gathering dust in an empty apartment in another country.

    We assemble one of two robots

    Models 5800 and 380 are almost completely identical, only the “brains” differ slightly, the 380 robot, as a more modern one, cleans up a little better, plus it treats the battery more carefully. Therefore, it was calculated that if something had been broken in the purchased robot, then it would be possible to replace it with spare parts from 5800.

    The new tank had a broken water tank - there was a large crack on the side that let water through. It could be sealed, but the look is not the same, so I took the tank from the 5800 model.

    In addition, the purchased Scooba also had a problematic turbine. If I hadn’t dealt with the first robot, I probably would have lowered my hands and put aside the second robot to get dusty, frustrated by the stupid purchase. But I already knew what the problem was, plus I knew how to solve it without dismantling the robot.

    It is necessary to fold a dense fabric in several layers and lay it on the floor. Then take Scooba, put it on its side, where the turbine is located (if you look at the robot from above, this is the left side), raise it by 2-3 cm and let it fall onto the fabric. The device itself is strong, it does not threaten such a fall, but the turbine shifts under the influence of vibration and then, after adding water to the system, it starts to work. If this method does not help, you can still return the robot to its previous position, and in the same way raise it by 2-3 centimeters, then release it so that the robot falls onto the fabric.

    Everything turned out, the turbine worked. I pressed the clear button and saw ...

    Dance of Death Scooba

    This is the name for circular irregular movements of the device, which, after a couple of minutes of activity, stops and signals a problem with the brush (which is usually in perfect order).

    I turned off the robot, upset, and began to search the Internet for information on what can be done. After reading a couple of hours, I tried randomly turning on the robot again, and it worked by itself. The batteries lasted about 10 minutes, but it worked fine! For two days I was happy with clean floors, and then one fine morning, filling the robot tank with water and running it for cleaning, I saw the same death dance (it is also called Scooba Circle Dance).

    At one of the American forums devoted to robots, it turned out that there could be two reasons. The first is the failure of the IR sensor of the robot wheel module. It ceases to give a normal signal and the wheel spins much faster than usual, which leads to circular movements, because the second wheel works at normal speed. The robot seems to be going crazy, because the computing unit simply does not understand what is happening.

    Scooba Wheel Module

    The second reason is the failure of some elements on the motherboard, which is being treated, but for this you need to be a good electronic affairs master. An ordinary person, even with the experience of soldering, will not cope - it is not so easy to find a problem element. You need to know where to look.


    Hoping for luck, I decided to simply replace the wheel module and then check how everything works. At one of the forums I met another solution - just solder one of the module wires, so that it goes directly from the cathode of the LED to the connector.

    You could try, of course, but I didn’t want to disassemble the robot, remove the wheel module, solder it, assemble it and then find out that the problem has not gone away. We all have little time, and the robot vacuum cleaner is still not the AK-47, so that it is constantly disassembled and assembled. A lot of plastic parts that become dilapidated over time. And if a robot can withstand one disassembly-assembly, there is no guarantee that it will survive the second.

    By the way, already during the first disassembly procedure, it turned out that one of the plastic wheel mounts had flown off - it just broke off. Fortunately, the module can normally hold with one mount. If two broke off at once, then the robots could be thrown away. Glue will not help.

    Below is the process of disassembling the robot and removing the wheel module. The video turned out so-so, but gives a general idea of ​​the structure of the robot.

    And here are photos of the disassembled robot and wheel module with the above-mentioned IR diode. Apparently, it is clean (some users of the forum from which I got information helped clean the item). So the problem is not in the dirt, but in the degradation of the element (90%, that is so).

    And a few more photos of the insides of the robot:

    And, finally, an example of the correct operation of the robot after assembly:

    Solving Another Problem - Batteries

    So, I had two batteries. The first is the one who arrived with an expat robot from my former apartment. The second is the one that came with the robot bought. Both were almost dead.

    I tried to leave the batteries on charge for a long time (night), but there was practically no hope of their recovery. After all, this is Ni-Mh, these elements deteriorate very quickly. If you consider that my first robot stood idle for about 4 years, then this battery could just be waved.

    I thought and was wrong! And twice - both batteries managed to revive. At first I tried using the smart charger IMAX B6. He honestly charged the batteries, but they lasted for 10-15 minutes of the robot.

    So, both batteries were restored in a rather unexpected way. First, I installed one and then a second battery in the first robot, which was not charging. This is done in order to quickly discharge the battery. Even when the robot is off, there is some kind of energy consumption (the board is not completely dead). After waiting for the voltage drop across the terminals from 16 to 7 volts, I connected one of the batteries to the second working robot. Then he put the robot on charge.

    The LED blinked faster than usual, which meant the activation of a special 16-hour cycle for charging new batteries, provided by the developers of iRobot. Also, this mode starts if the battery is discharged to 7 V. Leaving the battery to charge for 16 hours, I achieved its almost complete recovery. The robot turned on in the morning worked for about an hour. And, as it turned out, it was a battery from my first 5800 robot, which stood idle for about 4 years! They wrote on all forums that just a few months is certain death for the Scooba battery, but look at that.

    The second battery did not help. But I still tried to charge it, I tried to discharge it again. The robot did not enter the 16-hour cycle, but after a couple of days I once again left her for the night. In the morning I tried it, and it turned out that now it now also holds about an hour of battery life. As a result, I had two batteries that work almost as much as new ones. The price of the new one is about 40 euros, the same amount I paid for the second robot. I did not have to buy batteries, which I am extremely happy about. Well, if they fail, then I will try to make a Li-ion battery from new cans of used laptop battery. I hope in the near future I do not need to do this.

    And now what?

    The robot has been washing floors in the entire apartment for about a week, during this time I did not notice any glitches. Perhaps it will be enough for a while, something else will fail. But I have spare parts - the first robot, model 5800! So I look to the future without fear, in the near future I will not have to wash the floor. In any case, I hope so.

    Also popular now: