Ask The Search Engineer 12560

Dear Search Engineer: We're interested in developing a computer test system for various automotive electromechanical relays. The parameters include nominal voltage, coil resistance, pick-up voltage, drop-out voltage, insulation resistance, dielectric strength, contact resistance, operate time, release time. Any other suggestions or advice? -A.K., India

Dear A.K.: It won't be easy, but it can be done. Try a PIC with at least 5 A/Ds and lots of control channels (D/O) with an RS232 port for communication to a PC. For switching and release time, select a line on the PIC to switch on the relay, then start and timer, and use one of the interrupt ports to see when the contact switched, stop the timer, and use that value to determine operate time. For coil resistance, use a shunt resistor and the A/D of the PIC. For drop-out and pick-up voltage, try to use a DAC with a buffer to supply the coil and monitor the contact to be switched. You can store the results in a buffer to send to the PC via RS232. Delphi and USART can control the tests being done.

Search Engineer: When using our FEA program for cursory evaluations, we are typically exploring the results on metals within their elastic region. Occasionally, we need to explore stresses that put the materials into their plastic region where there will be permanent deformation. Most textbooks on materials give general curve shapes for various materials, but I am looking for a reference that has real stress vs. strain values for common metals (and other materials if possible). -M.K. in MI

Hey there M.K.: Try looking in ASM Handbooks or search for stress/strain curves at www.asm-intl.org or www.tms.org. For common metals, you may be successful in finding information in graduate textbooks, such as Plastic Deformation of Metals by Honeycombe, Dieter's Mechancial Metallurgy, or any theory of plasticity books. Good luck!

Dear Search Engineer: Is there an industry standard for the torque values of screws (#2,#4,#6,#8,#10). The intent is to decrease the number of torque tools on the production floor. I have come across many engineers, design books, and articles that do not agree with each other. -B.Z. in VA

Dear B.Z.: "Shigley's Mechanical Engineering Design" says that for static loads, the minimum preload should be 90% of the screw's proof load. The reason there are no industry standards for torques is that they can vary widely depending on screw/bolt material, clamped member material, lubrications use of gaskets, and if there are alternating loads (which makes torque even more critical!). Try calculating the torque requirements for all sizes and applications, then reduce your tool selection based on the overlapping of torque values.

Dear Search Engineer: We're prototyping solenoid designs and would like to prototype in our shop. We're looking for a magnetic material with relatively high permeability and a very high saturation value that does not require annealing after machining. Can you recommend a material and resource to locate? -R.K. in NY

Hey R.K.: Generally, any material will have improved permeability after machining. If work is kept to a minimum and done judiciously, you may be able to avoid post-process anneal. The easiest to obtain is grain-oriented electrical sheet steel, which is sold in a fully developed state and used in transformer laminations. Depending on the amount of machining, this may be the answer. Try Tempel Steel at (www.tempel.com) and materials in anything other than sheet form can be found at Carpenter Technology (www.cartech.com).

Search Engineer: What is the difference in winding types for step motors (i.e., bipolar winding and unipolar winding)? -K.Y. from Hong Kong

Dear K.Y.: In a unipolar motor, current always flows in each winding in the same sense. In a bipolar motor, current circulates in each winding in both senses. Most common step motors can be driven in either unipolar or bipolar modes. This can be done because the two windings (two phase motor) are each divided in two halves making their center terminals accessible, so you actually have a total of six terminals instead of four. If you drive the motor in a bipolar mode, each winding is powered with a H bridge driver capable of making the current circulate in both senses. In this case, the winding's center terminals are not used. If you drive the motor in a unipolar mode, the winding's center terminals are tied to the power supply and the two half windings are powered one at a time. In both bipolar and unipolar modes a proper sequence of current switching in the windings has to be generated in order to obtain a circulating magnetic field.

Dear Search Engineer: Do you know if there is a level sensor on the market that will sense through thin starch water? We want to be able to see the level of diced potatoes in a tank. We would like to have something with 4-20 mA output. -R.C. in Montreal

Dear R.C.: Consider using a capacitative level sensor, such as the Model 9080 manufactured by Arjay Engineering. It has a 4-20 mA output and works with a wide range of media. A limitation is that the sensor must have a ground plane to work. This is not a problem if the vessel is metallic; otherwise a sheathed sensor configuration can be provided where a metal tube is installed coaxially with the sensor (www.arjayeng.com/arjayeng_level_controls.htm).

Dear Search Engineer: Where can I find an extensive table of coefficient of friction between all kinds or most materials?-J.L. in IA

Dear J.L.: When I need a rough idea of a friction coefficient, I usually consult the "CRC Handbook of Tables for Applied Engineering Science" (CRC Press). It has tables of friction coefficients. "Machinery's Handbook" (Industrial Press, Inc.) also has a section on friction coefficients. Coefficients of friction can vary widely depending on the materials in contact, load, lubrication, and surface finish. There is also a difference between the static and dynamic coefficients of friction. It's best to test for critical applications. Check out this website for an ASTM test spec applicable to your conditions: www.astm.org/cgi-bin/SoftCart.exe/DATABASE.CART/REDLINE_PAGES/G115.htm?E+mystore.

Dear Search Engineer: I'd like to know what type of sensor I could use to measure the reaction force on a timing belt tensioner, taking into account conditions (oil, heat, water etc). -K.G. in CA.

Dear K.G.: An easy way to measure the reaction force is to use a strain gage built into your belt tensioner. I assume your belt tensioner is mechanically set to give a certain tension. Even if it is spring loaded, you can still build in a strain gage. The strain gage, if properly installed, would not be affected by any liquid. You need to do some work in correlating the output signal to the actual tension unless you are just looking for relative change.