> On Tue, 9 Jun 2009 22:25:02 +0100, Andy Pugh wrote: >> 2009/6/9 Steve Blackmore <[email protected]> wrote: >> >>> Andy - timing belts are not bungees ;)
>> Yes they are, everything is. >> I have a PhD in materials science. Your move :-) The PhD is right - everything stretches. Every material has a Young's Modulus. It doesn't matter whether the material is rubber (a few thousand PSI), HDPE (200,000psi), MDF (500,000psi), oak (1,600,000psi), aluminum (10,000,000psi), steel (30,000,000psi), tungsten carbide (80,000,000psi), or diamond (150,000,000psi). Every single one of those materials stretches under load. The diamond obviously stretches a lot less than the rubber, but it still stretches, and the Young's Modulus can be used to calculate how much. Steve Blackmore wrote: > I am a qualified engineer and machinist of many years > In practical terms there is no stretch, provided the design criteria is > not exceeded. The engineer is also right, in that many times you can assume that the stretch is negligible without calculating it. That is partly why experience and engineering intuition are valuable - if you stopped to calculate every single thing, you'd never get anywhere. But experience and intuition will get you in trouble when you attempt to stretch them too far, and that is exactly what is happening here. This discussion is NOT about timing-belts transmitting rotary motion from one pulley to another. Pulley-to-pulley drive is what timing belts are designed to do, and in such normal applications, you can follow the "design criteria" and conveniently ignore stretch. But we are talking about linear motion. This started when the PhD mentioned the ServoBelt (http://www.bell-everman.com/servobelt.html) linear motion actuator, and its approach to reducing the effect of belt stretch. The "engineer" immediately dismissed all concern about stretch with > What belt stretch? Once they are adjusted up, that's it. Such a blanket dismissal, unsupported by any hard data, is not what I'd expect from a "qualified engineer". When I did the math for a hypothetical case, I got a stretch of 0.020". Whether that is "negligible" or not depends on the application. For a plasma cutter, it is probably fine. For a wood router, maybe. For a metal cutting machine, certainly not. I'm an engineer too, but I'm sticking with the PhD on this one. Regards, John Kasunich ------------------------------------------------------------------------------ Crystal Reports - New Free Runtime and 30 Day Trial Check out the new simplified licensing option that enables unlimited royalty-free distribution of the report engine for externally facing server and web deployment. http://p.sf.net/sfu/businessobjects _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
