Last modified:
All Roombas have traditionally used slotted-optical-switches for their Bumper "sensors". However, for Scooba, the designers decided to shift to mechanical-switches for the same application. The purpose of this section is to reveal where the Bumper-Switches are located, what they look like, and how they get actuated. Two of these switch-mechanisms are used in the robot, but we will concentrate only on Scooba's right-side switch and actuation hardware.
A large portion of the robot's right, front quadrant is shown in Figure 1. It shoud be apparent that the Bumper-Shell and upper-chassis assemblies have been dismounted. We are interested in the black structures; the curved inner-bumper assembly (low in the image), and the polygonal 'arm' (which is an integral part of the bumper-casting) jutting inboard from it. The large white object occupying the middle to upper portion of the frame is the Controller's enclosure.
Figure 1. RHS Inner-Bumper and its Cam-Operated Switch-Actuator Arm
Moving in now, to get a closer view of the actuation hardware (the switch we want to see is hidden by the the arm) the arm is seen to be held down by small blocks of plastic, through which screws pass. We will call the part that is somewhat rectangular, the "Retainer/Cam-Follower", and the "D"-shaped piece, the "Retainer-Link". We have peeked ahead, and can tell that those will be fitting names! Of course the "arm" is free to horizontally slip around under those blocks.
Figure 2. Enlarged View of Actuator Arm and its Guide / Retainer Fittings
The retaining screws have now been removed, the two parts lifted off, and the arm & bumper have been shifted out of position a little -- to give us the scene shown in Figure 3. The Retainer/Cam-Follower (with its screw passing through it) is lying on its edge; and that view of it reveals the "cam-follower" portion, i.e., the short, cylindrical boss which projects out its underside. It is on the surface of this immovable cylinder that the hollow-cam rubs, as the Bumper is depressed.
Shifting your view to the 'arm' now, the cam (which is the irregular hole through the arm), which makes the bumper-end go through some complicated monkey-motions as it actuates the switch, is clearly seen in profile, and is seen to be an 'inverted' cam. IOW, the cam rides on a stationary cam-follower, in this inverted design. How long, do you suppose, it took the engineer to come up with that optimum, filleted-polygonal profile?
Figure 3. Dismounted Retainer and Guide Show Cam's Profile and Cylindrical Follower Under the Retainer-Bar
Also notice, in Figure 3, lower right, the switch-body has been uncovered, and is visible for the first time. But, you may as well look at Figure 4, where the switch is more clearly seen -- at least its top facets are well defined. It is the black, fat, tee-shaped block, pushed into a rectangular holder. A pair of orange wires come out of its back-side. On its front side, there is a spherical-radius, blister shape -- which is the um... 'switch-button', i.e., the item that the bumper pushes inward to register a bumper-contact. Note that the "blister" is over emphasized in the photo, since there is half of a drain-hole, beyond the button, but fairly well in alignment with it, which results in making the button appear to have twice its actual protrusion.
Figure 4. Plan-View Shows Rubber-Encased Momentary-Contact Switch
We also get a good view, the first complete view, of the "Retainer-Link". There is more to it than one could see at the assembly level. We are talking about that "L"-shaped part. After dismounting it, it was moved down and to the left, to place it just a little more than half an inch from its normal, mounted position. You can probably tell that the hole at its right-end mounts on the post where its screw was loosely inserted; while the opposite end of the link fits the other post -- where the cam-follower normally resides. This simple looking link performs a number of functions. For example, it:
Figure 5 gives us an oblique frontal view of the switch. It is a difficult photographic subject, because it is all black! It is also a flat black; which is brought on by totally encasing the switch in a black-rubber jacket! Water-proofing, we can presume.
Figure 5. Oblique-View from Front Shows Switch-Button and Cam-Plate-Arm that Presses It
This view is also the only one that shows a 'dropped-leg' (right-center, in the image) at the switch-end of the actuator-arm. That is the structure that presses the button, when needed!
Having finally pressed the button, we conclude discussing the RH Bumper Switch. There is no plan in place to repeat the discussion for the switch at the robot's left side. Readers will simply have to envision a similar arrangement of hardware items over there, and with some of the items -- like the actuator-arm, and the Retainer-Link -- will be of opposite hand.
In closing, there is one additional item that is related to bumper-switching that may be of interest. Roger has observed that Scooba momentarily stops its brush and pump motors while it responds to a Bumper press; a tactic to limit puddling, most likely.
If you wish to return to the main (Scooba Technical Comments) document, click Main
Otherwise you may return by pressing the 'Backspace' key.
This page is currently maintained by G. Plews