Particle in the kitchen

Hi Particle Community,

I’d like to introduce [Stash][1], a small start-up centered around automating when food needs to be ordered using Particle devices.

The Setup

Currently we are prototyping with the Particle Core. When the lid closes on a Stash container, an update is sent to our cloud servers with information on how full/empty the container is.

From there we use that data to generate statistics that calculate how much of a particular food you will use in a given day, week, month, and year. After about a week of use we have enough data to accurately calculate when you will run out of food.

We hope to shortly couple it with Amazon Dash Replenishment Services. With that, a customer is able to have food re-ordered for them and sent to their doorstep before they run out. All without having to lift a finger.


Currently we are trying to see what market demand is. If you’d like to learn more about us just visit [][2]

Questions/Comments? Respond below :smile:


Interesting project, however it would be nice to get an actual level and not only the times the container is opened or closed.

Maybe an optical sensor can do or a capacitive sensor of sorts.

Also you have the issue of power… Maybe you can use the magnetic switches to turn on the power of the photon, take a reading each time the lid closes and send the statistics… AAAND! Have a solar cell with a battery… You might “MIGHT” have enough power for those transmissions with a days of indoor solar cell power…

Although maybe NOT :blush:

If this issue does not work, I would suggest attempt capacitive charging since I don’t see people charging all their containers…

Good luck! Nice project!

Honestly, I credit your creativity but to me it is in the realm of “because we can” but isn’t really all that practical. Plus it seems that there are a lot of unsolved problems, like power which @frlobo pointed out.

How does one describe the contents of the container to the Particle device in your example?

How about inserting/attaching RFID or scan tags to the containers and have the user brush the “nearly empty” container past a particle enabled scanner? (humans are pretty good sensors for that type of application (plus what if I want to replenish my Cheerios with Lucky Charms this time).

The enabled scanner can collect the list of things to buy and upload daily, weekly, etc.

If you can use a UPC compatible scanner, you may be able to use a public database to attach the food type and size to the container’s tag.

Great directional thinking, I will say. It just needs some fleshing out.

1 Like

@frlobo, I guess the open/close is only the trigger to do the measuring and communicating the new amount (which I guess is measured by the extra sensors inside the housing of the Core)

But true, open/close might not be the best trigger. What if you have the container in any non-vertical position or its contents not shaken down to have the container filled evenly when you close the lid?
Maybe contacts at the bottom of the container to detect “home-position” and two distance sensors might account for that - or go for pressure sensors to weigh the contents (but make sure nothing is stacked on top).

Thank you guys so much for the feedback! I’ll respond to some of the points you’ve all brought up :slight_smile:

@frlobo Sorry for not clarifying better, the container does return the actual level once the lid closes. We chose the lid as a trigger because it helped us with your second point which is power.

To power the device we are using a 2400 mAh rechargeable battery which currently allows it to last about a month and a half under the conditions of being used once a day everyday. We could increase this by tweaking the software / hardware though. Or like you said, adding a solar cell :smile:

@ScruffR The point you raised about the contents not being level is good, currently a user needs to give the contents a little shake to make them level. If they are skewed, then the reading will be whatever level the contents directly below the infrared sensor are. The issue with adding more infrared sensors is cost. But if not for cost, more sensors could certainly be used to increase accuracy like you stated.

@BulldogLowell Thanks for the feedback! To answer your question, when a person connects their container to our cloud services for the first time they can enter in a name for what the contents are. Currently we are only showing the landing page to the public to see if there is enough interest in this to make it a viable product.

Also, if we were to make an app it would have a bar-code scanner built into it which could scan something like a box of cereal and be used to selecting the contents of the container like you said. But for the moment, users need to manually enter in that information.

So far we haven’t had many people hit our landing page, and of the ones who did, not many left their email.

What is everyone’s thoughts on this as a viable product to market?

Thank you all once again, your opinions are highly valued!

1 Like

I’m all in for putting intelligence into everything - and thats ok when i built it myself.

But honestly for such a simple task like this the cost per item would be way to high. what’s your target price for this product? Using a photon +lipo+sensors+box something like ~ 50$ ?

That would definitly over the top IMHO regarding cost/use balance. And then you’d only have ONE box for ONE Product.

you could make up for that by addings things to the backend like automatic ordering via amazon or something like that - but still…

My wife is definately the best sensor when it comes to grocery (-;


It’s been a while since I have shamelessly self promoted :grinning:

Bluz would solve your battery problems. Put a gateway in the home and then cheaper, coin cell powered bluz boards on each container. You’d have the exact same functionality but potentially years of battery life at a much cheaper cost: