ON 20050620@1:10:37 PM at page: http://www.sxlist.com/io/sensor/temp.htm#38523.0406597222 James Newton[JMN-EFP-786] removed post 38523.0406597222 |Delete 'labview1958@yahoo.com asks: " I intend to make my own home-made circuit to measure the temperature of liquid nitrogen using a silicon diode? What shall I do? Anyone can help?"

' ON 20050622@6:00:45 PM at page: http://www.sxlist.com/io/sensor/pos.htm#38525.6096064815 James Newton[JMN-EFP-786] published post 38525.6096064815 with comment: 'Very happy to hear from you again!' A 'spiffy' position sensing system is air. It's sometimes used for precision gages, usually in the form of a gage to measure the difference between desired size and actual size.
A simple application would be measuring the diameter of a hole, like a gas engine cylinder. A test plug is inserted in the hole. Around the periphery of the plug are several small holes that lead to a central air supply. A T from the supply also leads to a fine needle valve. A sensitive pressure sensor between the arms of the T is nulled out by adjusting the needle valve, with the plug inserted in a standard cylinder. When the plug is inserted in a larger cylinder, the air pressure in this leg will be reduced. The pressure sensor can be calibrated for size.

This technique has been superseded for many applications by the linear differential variable transformer. It might make more sense to move this comment to 'proximity'.

Anne Ogborn |Delete 'P-' before: '' but after: ' Current shunts are a great application for a 4 terminal resistor.
4 terminal resistors are a bar of some slightly resistive stuff,
with a connector at each end for the high current. They have a small tap contact at a point near each end. It's the resistance between these tap contacts that's well known.
Since the tap contacts are in a high Z circuit, the contact resistance is negligable. Since the high current contacts are in series with the measurement shunt, they don't affect the reading.

Anne Ogborn |Delete 'P-' before: '' but after: ' Direct collimated light into a photocell. Spin the direction the light comes from. (a robot might be able to spin itself occasionally for this, or you could spin a mirror.
Run the output voltage through an ADC and into processor.
If you're doing the spinning mirror, you'll also need a hall effect or slotted disk or something so you know when the mirror is 'dead ahead'.
Have the processor note the orientation of the brightest light heading.
In many environments this is 'good enough' for direction.
For example, outside the brightest light would normally be the sun. The sun is a fixed target during a short activity.
A room with a single, naked light bulb might work as well, if the lamp is in one wall. An overhead bulb would have 'cone of silence' problems.

Anne Ogborn |Delete 'P-' before: '' but after: '