I work in a restaurant which has a standard fountain soda machine. You place a cup on the angled grating, it rests against a metal bar which dispenses the drink. Pretty basic.
Here’s what has baffled me and my co-workers (including managers): it always stops automatically when the cup is full. Always.
You see, we had assumed it was based on weight, and stopped when the cup reached a certain pre-set weight. But the problem is that we have different size cups (such as kids’ cups), and they too fill until full and then stop. You can even fit a pitcher in there, and it will fill until full and stop.
There are no motion sensors or anything as far as I can tell - so how the heck does it know when to stop?
What does it do when you’re pouring something very fizzy, like root beer? Does it fill until the suds reach the top, or does it let the suds overflow and fill until the liquid reaches the top? Or is it really smart, and waits for the suds to subside, and then finish filling?
The last one, actually. It fills until the suds reach the top. The suds subside, and as long as you leave it there it will fill it a little more at that point.
WAG: perhaps the metal bar the cup is resting against is connected to a circuit which checks for capacitance. That way, when a liquid (or even just the foam of the soda) comes into contact, the capacitance changes and the machine stops. This would also be coupled to a physical switch which doesn’t dispense soda unless the bar is depressed. Those two combined would produce the effect you’re looking for.
What happens if you put other liquid on the bar while it’s filling?
Sounds like either an ultrasonic or optical fill level sensor. Try cupping your hands around the top of the cup as it fills and see if you can fool it.
Now that I think about it, there must be a small amount of overflow that I don’t actually see, and a sensor along the bottom under the grating that catches that and stops the pouring. Although it’s interesting that that would work even with the suds. Am I correct?
Possible - but the problem with that is that those bars get all kinds of liquid and stickiness on them all day long… although it does malfunction every once in a while - could this be a cause?
Yeah, I don’t know how my theory works given all the other crud that winds up on the bars, but I don’t think your other suggestion would work:
Once the overflow stops, how does the machine distinguish the full cup from someone having placed a new, empty cup against the bar? If it’s able to the start-and-stop thing with root beer, then it can’t be an issue of releasing the bar first…
Wow, I thought this was an issue of us being a bunch of young dumb blue-collar workers. Didn’t know even Dopers would be confounded. I’ll have to try Q.E.D.'s suggestion.
It seems to me that it’d be more complicated than that, even. That’d tell it the height of the surface, but a pitcher and a kiddie cup are going to have significantly different heights. It’d also have to have some way of knowing how high the container is.
I’ve a hunch that it would only work correctly with a handful of specific container sizes, and that when it was installed, some long-ago manager, or maybe a guy from the pop company, had to calibrate it to the cups and pitchers that you use. If that’s the case, it would only have to be able to distinguish between those specific containers, and then once it knows that it’s filling a 64-ounce pitcher (say), it’d just pour exactly 64 ounces.
Chronos, your idea couldn’t work as simply as you describe it, because it knows to stop when the foam reaches the top, but then to top up more when the foam subsides. Also, consider **Rigmarole’s ** comment about being able to fill pitchers: it doesn’t sound like he’s talking about a specific cup size.
I’m interested in **MilTan’s ** idea of capacitance. A bit of crud on the bar isn’t going to alter capacitance as much as being in contact with a whole cup of liquid.
Rigmarole, what happens if you touch something metal (a knife maybe) to the bar when a cup is half filled?
Probably two optical/ultrasonic sensors. One to measure the height of the cup and another to measure the fluid level in the cup. The machine shuts off when those equal each other.
The patents for these things all generally seem to be aimed at having the cup tipped a bit, so there will be a known location for an overflow to happen. When the cup begins to overflow, the arm senses the overflow and shuts off the soda. Earlier designs had the overflow sensing done at the cup’s rim, but sanitation concerns with re-fills led to the sensor moving down a bit so it wouldn’t be touching the cup where a person’s lips would have been.
Earlier designs were time based - they’d have “small” “medium” and “large” buttons, as well as a “flow” or “top off” button. These worked on the concept that a given size cup would take a given amount of time to fill, so they’d set the timer for something just under that to allow for variances in how much ice is in the cup, and the operator would then add a bit if needed with the flow button.
One of my ex’s was an assistant manager at a movie theater. I was too completely baffled by the automatic stopping and asked her. She looked at me like I was a complete moron and explained it to me. The cup is on a ramp. When the soda reaches the top of the cup the rim pushes too far on the lever and the flow stops. The flow only happens when the lever is in an in-between position, push it too far and it will stop. You put the empty cup on the thing and the rim pushes on the lever and starts the flow, when the level reaches the rim, the rim pushes farther and the flow stops. :smack: Brilliant, no? And I’m the one who went to college…
OK, I tested it tonight and it indeed seems to sense liquid on the arm - I tried splashing some water on the metal arm while it was filling a glass and it stopped then and there.
The only thing that still seems curious to me is that all the liquid stickiness that accumulates there doesn’t regularly throw it off. But like I mentioned, it does occasionally malfunction and I suspect this may very well be the reason. Next time people complain about it not working I can appear the genius by simply wiping off the arm and voila! I see all kinds of promotions in my near future…
Heh, I asked a friend to check a machine yesterday (I’m in Tokyo, so I can’t do it myself) and that was indeed a design I described without any liquid sensors. The arm only fills when in an intermediate position. Must be two competing designs?
I think this topic deserves more experiments, thought, and research.
By “push farther” do you mean that the weight of the cup + liquid when full being tilted on the ramp pushes more weight against the bar than the spring loaded maximum set on the machine which pushes the bar to the off position and turns off the flow? :dubious: Because I can’t seem to figure out how the same machine would distinguish between a 4 oz. kiddie cup and a 64 oz. pitcher. Surely the pitcher puts as much pressure against the bar as a full kiddie cup when it is only partially filled.
This also seems :dubious: to me. If you put a cup on there to fill and it fills up to the top but with lots of foam and shuts off because the foam runs down onto the sensor, then reason has it that it will not start again until the sensor has dried off. At that point the fluid would flow again and the cycle would repeat. If you leave a cup there, will it fill-stop-dry-fill-stop-dry-fill-stop-dry to infinity?
Also, surely you can fill one cup right after another with no drying period in between. How could that be with the sensor you describe. How would the sensor know if the wetness was from the current drink or the one that was filled 8 seconds ago?
I was thinking the same thing… Consider if the bar is a long lever, hinged at the top. I imagine that a kiddie cup pushing at a very low part of the bar would have a better mechanical advantage over the pitcher that is pushing at the top of the bar, closer to the pivot, thus leveling the playing field. Both the little cup and huge pitcher would cause the same torque to be exerted on the sensor.
It sounds fidgety and tricky to tune, but if there is a tolerance of an inch or so of soda, then perhaps the lever principle might do the trick.
I’ll have to try Q.E.D.'s suggestion.