Z axis calibration idea

[Aztecphoenix]

While it's cool that this is everything we're looking for together in a package... It's also $79 for something that might take at most $30 to make as a DIY project.

Well I'm just waiting for someone to do it as a DIY I'm too much of a noob to figure it out

[mike_biddell]

In theory, if you have three channels in a plastic tube (a centre channel sandwiched between 2 outer channels) and you put two narrow flat plates in the outer channels. Then you let water rise up the centre channel as the water in the tank rises. You have a very sensitive variable capacitor in which the capacitance varies with the fluid level. A capacitance bridge could measure the variation very accurately. This method would work for any depth of tank. The water acts as a dielectric between the plates. Oh, the outer tubes are sealed.

[jstrack2]

Aztecphoenix, what I did is a DIY way! haha If you want I can help you build it. I think that if the Peachy team makes laser cut parts then that will be best. It will cost almost nothing and work perfectly, with virtually 0 set up time. What I did was super sloppy (I used random metal, plastic and styrofoam glue gunned together and a random cup as a counterbalance), yet surprisingly still worked well!

Again though the Kalman filter will need to be implemented too. This will lead to the extreme z-axis control. Like Mike Biddell I was amazed at how well it works when I first implemented it for something else. And the beauty is that it is just software. Once written it is free!

[jstrack2]

Mike Biddell, I agree that your idea would work in theory (water has a huge dielectric constant!), but in practice I don't know if it is worth doing. Very accurate measurement methods that take very little work can be purchased for just a few dollars. This method strikes me as something that will take a lot of work to get right. Besides getting it manufactured which is non-trivial, I think that there are other things to watch out for. A temperature sensor would be needed since the the dielectric constant of water varies pretty strongly with temperature. Also it varies a little with different levels of salinity.

That said I might be wrong! If you could make it work that would be great since it is pretty hard to beat the simplicity of dropping a tube into the water!

[mike_biddell]

Jstrack..... agreed..... there could be a lot of graft involved........ the cheap calipers are the best option.

[jstrack2]

Yeah there could be a lot of money exchanging hands working on something like that! Haha jk

[rylangrayston]

Some how ive been missing all the convo in this thread until now.... Yet Ive been working on much of the same stuff.

1. jstrack2.. wow great job!! That last video really shows that the caliper method can work! amazing how smooth it is! and the read out already hacked into the arduino!

2. We got capacitive feed back working quite well here at peachy a few weeks ago, its showing real promise.
I had never heard of the Aqua Plumb before, it definitely looks like capacitive feed back too but I dont know how they did it with such small wires!
We have to use large plates to get the resolution we need. Sorry Im too pressed for time to post about it right now but the time will come.

3. Thanks for posting about the Kalman filter, ive sent an email about that to James Townly so he knows about it.

[Aztecphoenix]

Ok, for all the techies out there, to compensate for the salinity variables and having to calibrate prior to each print, what about having a sensor in the salt water tank which would automatically read the saline level to calibrate the capacitive sensor in the build tank?

[rylangrayston]

Ok, for all the techies out there, to compensate for the salinity variables and having to calibrate prior to each print, what about having a sensor in the salt water tank which would automatically read the saline level to calibrate the capacitive sensor in the build tank?

hmm .... google ing .... yep!
http://web.cecs.pdx.edu/~eas199/B/ho...or_wiring.html

[BrockMcKean]

If the saline solution is mixed well enough, yes. If it's not because you didn't physically disturb it enough or you used cooler water, the salinity in one area is going to be different than another area. You would have to assume the person followed instructions to ensure it was properly mixed to a homogeneous state. In my opinion it would be simpler just to come up with an way of using an absolute amount of water and absolute amount of salt at an absolute temperature. That is to say, ensure the error in mixing is very small due to the simplicity and rigidity of the mixing process. I'm sure there's something like this out there, but I'm a computer engineer, not a chemist.