Shop OBEX P1 Docs P2 Docs Learn Events
capacitive level — Parallax Forums

capacitive level

hutdonhutdon Posts: 32
edited 2004-08-18 13:21 in BASIC Stamp
I am in the process of attempting to design a level. What I need is to be able to discern out of level amounts in .001-.005 inch increments with a total range of about 0.5 inches over a length of approximatly 3 feet. I have done a web search and found a couple of articles on using an ordinary bubble level wrapped with conductive tape and cut to form two or three plates of a capacitor. (2 plates form one capacitor, three form 2 caps). My initial thought is to use freqout as an ac source and use a rectifier and an ADC to measure the effect of the capacitive reactance. A stamp would then take the output of the ADC and display the results. My question is - has anyone tried this? If so I would particularily be interested in the fabrication of the capacitor(s). Any prelim advice would be appreciated. - Don

Comments

  • BeanBean Posts: 8,129
    edited 2004-07-27 01:37
    You should look at the Memsic 2125 dual-axis accelerometer. I don't think you can get down to a thousandth of an inch, but I don't think you will with the method you describe either.

    Bean.
  • ChipCircuitChipCircuit Posts: 23
    edited 2004-07-27 03:02
    I'm thinking that if you used a dielectric fluic of some sort, you could make a resistive sensor that would react to fluid level changes. If you are making a carpenter style level, inside of a pipe or 2x4 or other solid part, you could use a wheatstone bridge and compare both sides and pick up some accuracy.

    Also, an angled electrode will have a greater surface area change than one that is parallel to the fluid motion.

    Dave
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2004-08-18 09:54
    The Memsic may work very well if it is attached to a long enough straight edge and one aligns the x and y correctly. Finding a good straight edge is most important.

    With a fluid you still may have to find a good straight edge and using a fluid may oscillate from one end to the other -- this means you have to wait for it to settle down [noparse][[/noparse]like a bubble level]. The more fluid, the more time.

    The Memsic uses a heater and a vapor bubble created by the heat over a central point. Seems much easier to fabricate and calibrate. You can even recalibrate at any time if you have a read out device (LCD).

    How do you calibrate the fluid? To calibrate the Memsic, you just turn the attached device 180 degress and correct by half the error. Since you have an x and a y axis - one could be your level measure, the second could have a 'tilt warning' to increase your accuracy.

    Or, even better - MAYBE - one axis could be a horizontal x sensor (level) and one axis could be vertical z sensor (plumb, but with only one useable up direction) BY ROTATING THE MEMSICS mount 90 degrees. The BasicStamp could select which reading made sense without having to provide an electric switch.

    As a carpenter, I used to layout floors in highrise office buildings with a good long level. It was checked with a builders transit.

    My foreman could not believe that I would have the whole floor laid out within tolerance of 1/8 inch over 250 feet by using such a basic tool and a streched chalk line. (The secret is to put the level at the end of the string that moves up and down - this is where the maximum incline [noparse][[/noparse]or error] is easily recognized).

    I could do it in 5 minutes while he was setting up the builders transit.

    The forman never trusted it, though we went up 25 floors of concrete pours without him finding a mistake.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    G. Herzog in Taiwan
  • hutdonhutdon Posts: 32
    edited 2004-08-18 13:21
    The memsic looks interesting. I was under the impression that an accelerometer would only measure dynamic 'level' ie a change in condition and not static 'level' but this seems different. I'll defineatly look into it. Thanks.
Sign In or Register to comment.