From: ChrisAtUpw@.......

Date: Thu, 21 Dec 2006 05:13:35 EST

In a message dated 2006/12/21, rpratt@............. writes: > Continuing the spring discussion, what are the practical implications of a > non-zero length? Suppose one wants a 1.5 - 2.0 second vertical similar to an > AS-1. It isn't difficult to get there with a variety of springs. Hi Randy, I suggest that you look up the LaCoste papers at http://psn.quake.net/bibliography.html As you say it is not too difficult to get about a 1 sec period, maybe a bit longer. However if you want to get longer periods, you do need the right geometry / spring load vs extension relationship. You won't get much above 5 secs period with a steel spring anyway. The large temperature coefficient of the steel eventually makes the compensation point highly unstable. You are trying to balance a gravitational load with a spring force. Professional equipment used Ni-SpanC springs which have a ~zero temperature coefficient. Also go to http://quake.eas.gatech.edu/Instruments/LPVERT0.htm If the chosen spring has zero force at say mid length how will that influence the > operation over a zero length spring? Isn't it period that matters and the > zero length is important only for getting the longest possible length in > period? A zero length spring has a plot of physical length versus load which goes through zero. It is very tightly wound and it takes an appreciable force to extend it at all. You ''don't count'' any load which simply does not extend the tension spring. If your spring has a zero force at some mid length, you just won't be able to satisfy the balance conditions - even approximately - and you won't be able to use it to give long periods. The maths is not very complicated, BUT IT IS IMMUTABLE! It sets out the conditions required to get A STABLE BALANCE POINT. Regards, Chris Chapman In a me= ssage dated 2006/12/21, rpratt@............. writes:

Continuing the spring discussio= n, what are the practical implications of a non-zero length? Suppose o= ne wants a 1.5 - 2.0 second vertical similar to an AS-1. It isn't difficult=20= to get there with a variety of springs.

Hi Randy,

I suggest that you look up the LaCoste=20= papers at http://psn.quake.net/bibliography.html

As you say it is not too difficult to g= et about a 1 sec period, maybe a bit longer. However if you want to get long= er periods, you do need the right geometry / spring load vs extension relati= onship. You won't get much above 5 secs period with a steel spring anyway. T= he large temperature coefficient of the steel eventually makes the compensat= ion point highly unstable. You are trying to balance a gravitational load wi= th a spring force. Professional equipment used Ni-SpanC springs which have a= ~zero temperature coefficient.

Also go to http://quake.eas.gatech.edu/= Instruments/LPVERT0.htm

If the chosen spring has zero force at say mid length how will that influenc= e the

operation over a zero length s= pring? Isn't it period that matters and the zero length is important only fo= r getting the longest possible length in period?

A zero length spring has a plot of phys= ical length versus load which goes through zero. It is very tightly wound an= d it takes an appreciable force to extend it at all. You ''don't count'' any= load which simply does not extend the tension spring.

If your spring has a zero force at some= mid length, you just won't be able to satisfy the balance conditions - even= approximately - and you won't be able to use it to give long periods. The m= aths is not very complicated, BUT IT IS IMMUTABLE! It sets out the condition= s required to get A STABLE BALANCE POINT.

Regards,

Chris Chapman