Post by thread:Sensors

Hello, I am presently designing a magnetic sensor device using the PIC16C71 as the controller for the project. What I am looking for is a linear output Hall effect sensor. In the past I've used Allegro Inc. for these parts but presently they are sold out! Any sources that you are aware of will be considered. Other magnetic sensor types will also be considered. As is typical, cost is an issue. To save bandwidth please drop replies to my Email address. If anyone has run across a list or net site that focuses on sensors I would also like to know about them. Thanks in advance. Mark Herold Email: spam_OUTheroldTakeThisOuTresearch.bfg.com

---------- >Hello, >I am presently designing a magnetic sensor device using the PIC16C71 as the >controller for the project. What I am looking for is a linear output >Hall effect sensor. In the past I've used Allegro Inc. for these parts but >presently they are sold out! Any sources that you are aware of will be >considered. Other magnetic sensor types will also be considered. As is typical, >cost is an issue. To save bandwidth please drop replies to my Email address. > >If anyone has run across a list or net site that focuses on sensors I would >also like to know about them. > >Thanks in advance. > >Mark Herold Email: .....heroldKILLspam@spam@research.bfg.com > Whereas the world is Analog, it is also true that it is non Linear. So a linear device may be a proublem. A way around this is to plot the output of the device vs input and derive a formula that reflects this. Then use the formula in a pc program to generate 'data' statements to put in the pic program. I think the 'load w reg and return from subroutine' instruction could be used as a table lookup. As this would take 256 instructions, perhaps it would be better to have a sixteen byte lookup table for the high nibble of the value and then a table with the 'interperalation' for the low nibble of each value of high nibble. It may be necessary to have the low nibble table contain a mixed value. that is a value in which the bits do not represent one, but represent 1/2 or 1/4 or 1/8 of the value. With this it would be possible to have the interpralation be 3/8 at points where the change per unit input does is less than one. A word of advice, if you do this, don't loose the formula. You may need it latter for something. All that said, how expensive is a magnetic pickup for a guitar? --------- Bill Cornutt billcornKILLspaminfoserv.com Located in Ione California USA. A small town in Northern California. Sitting against the foothills of the Mother Lode. ----------------------------------------------------

Temperature: For such a high temperature a thermocouple is (probably) the most sensible choice. Depending on requirements a 'Pyrometer" may be justified - measures wavelength of radiated infra red. Thermocouples are MUCH cheaper. Even the common "Type K" thermocouples sold in most electronics stores will work to over 1300 Celsius. The "sensor" is Nickel-Chromium/Nickel-Aluminium wire and is potentially very cheap. Designs for thermocouple amplifiers are available from various semiconductor makers. Pressure /Direction / Altimeter: Look at the archives of recent PICLIST discussions. There have been various mentions of all these in the past few months. Someone here will be able to tell you how to access the archives. . . -----Original Message----- From: Neil Calitz <EraseMEncalitzspam_OUTTakeThisOuTOLDMUTUAL.COM> >I am looking for sensors for a project that I am working on. > >1. What I need is Temprature sensors (up to 1000 deg Celcius) >2. Presure sensors. >3. Directional (like a compas) >4. Someting to act as a Altimeter.

For the temp sensor, you probably want a nickel-chrome to nickel aluminum thermo couple. The cost of these arises not so much from the wire junction, but the housing that is used to protect the lead wires and the junction itself from physical damage. I use one of these thermocouples to measure the temp of molten aluminum, and the sensor I use is in a 3 foot long protective tube to keep the aluminum off the wires - cost around $90. If you have a more controlled environment (maybe not immersing in a molten metal or other corrosive material) so that you do not need such a large protective housing, the thermocouple will be cheaper. To use a thermocouple you need a rating curve in mV/degree. But note that 1) at some point the thermocouple wires will transistion to copper wire, and 2) your reading of mV from the copper leads, when translated to degrees C, is actually the difference in junction temperatures. Unless the nickel alloy to copper junction temp will be constant, you will want to monitor that secondary junction with something like a DS1620 located at that junction. You can then compare that secondary junction temp to the temp at which the thermocouple rating curve was developed. If the thermocouple was calibrated with the secondary junction at 0 degrees (in an icewater bath), and your secondary junction temp is measured as 25 degrees C, just add the 25 degrees C to the temp reading you get from the thermocouple's calibration curve. You may need an amplifier. These junctions can be read easily with a DVM, but I havent tried reading with a PIC A/D, which wants a relatively low source impedance. Don't have any info for the other sensors. What are you building anyway, a Venus probe or some other sort of re-entry vehicle? -Nicholas -----Original Message----- From: Neil Calitz <ncalitzspam_OUTOLDMUTUAL.COM> To: @spam@PICLISTKILLspamMITVMA.MIT.EDU <KILLspamPICLISTKILLspamMITVMA.MIT.EDU> Date: Wednesday, September 30, 1998 5:21 AM Subject: Sensors {Quote hidden}>Hi all > >I am looking for sensors for a project that I am working on. > >1. What I need is Temprature sensors (up to 1000 deg Celcius) >2. Presure sensors. >3. Directional (like a compas) >4. Someting to act as a Altimeter. > >Please help > >Niel Calitz >