Data gathering to be sold
Let’s talk about the definition of a smart device. Its a device that has sensors, a processor and can connect via a wireless connection to other devices to send data. That being said any vape that connects to an app is a smart device.
For the features you proposed its not necessary to have a wireless connection to a network.
That would require some kind of optical sensor and a control algorithm (for example like a PID controller) no network or data transmission required.
Thats just a given thing if the set temp is not to high and the temp control is precise enough.
That could be simple done by a puff counter just tell the user to clean the device after a certain amount of puffs and let the user confirm or ignore the cleaning info on the display.
And honestly if you use any vape device several times a day it should be cleaned daily in my opinion.
Technically it makes sense to give the user a choice if they want to get maximum battery lifetime or maximum charge capacity and maybe something in the middle. Also an automated algorithmic solution for improvement would not require a network connection.
Thats already happening for example in the Venty, which has a sensor for the draw speed. Also no network required for that.
The only use case i see (and i don´t think its a good one) would be to somehow check the how much is left in the cartridge to order new ones from the dispensary before you run out. But that would require that you always want the same stuff or you still need to tell what to order next.
So i think the only real use case is the collection of data.
I suppose one more “smart” feature is to authenticate the device as not fake. I’m not sure this is an issue.
Everything you described can be accomplished without “AI”. The term for this type of stuff is “control theory”.
As old-fart and Jonas said, you can’t control what you can’t sense. Sensing clouds and using that to control temperature might be interesting, but would be a whole new strategy.
The reality is that almost no vape out there has as close of control of the temperature as the manufacturers imply. Once you start drawing, the amount of power that would have to be delivered for the heater to not drop in temp is in the hundreds of watts. And the responsiveness of the temperature sensor is limited as well, there is always a delay. A ball vape is a good example of how this concern can be sidestepped: it just has enough heat capacity to deliver this power from stored heat alone. You also need a relatively extreme physical layout of the heater to have very responsive temperature control, if that is something you desire. Look at the tinymight and frolic for example.
Draw sensors could be potentially interesting as well, and are also used in the Tafee Bowle, which has the most sensors I am aware of in a vape. Still, no AI. Maybe it’s a “smart vape” though.
I think anyone that jumps to, we’re going to make vapes better by connecting them with a smartphone, AI, etc, doesn’t really understand the physical constraints of thermodynamics, heat transfer, and control theory that are the ultimate limitation on performance.
I like to add that the development of a good product should in my opinion always start with a problem, not with a solution.
5 Likes
True to a point but completely counter to innovation where it takes a product that has already addressed a problem and finds a way to make it better.
Our vapes are going to end up like our cars. Short life (plastics) and ready to brick at the most inopportune time. Focus on safety and develop devices that allow ease in the vape process.
1 Like