PSN-L Email List Message

Subject: Re: Morrisey's seismometer
From: Brett Nordgren Brett3kg@.............
Date: Tue, 06 Sep 2005 13:02:48 -0400


Alexander,

At 04:54 AM 9/6/2005 +0200, you wrote:
>Dear Brett:
>I sincerely thank you for your detailed technical comments, the files you 
>suggest me and specially your interest in making my first steps easier.

You are very welcome.  Always happy to help a fellow experimenter.
>You got the right sense of what I want. I don't want to merely reproduce a 
>design ignoring how it works. I absolutely need to know it before go into 
>the construction..! I already downloaded the files from your web site. 
>Considering the titles, I think they effectively cover important topics 
>for my learning purpose.

I, like you, prefer to understand how things work before I jump in and 
start building.  Unfortunately, usually that means that I only get as far 
as the analysis and planning, and rarely get around to actually building 
things.  At least I can help pass along what I have discovered so far.

I hope you understand that I am expecting people to ask questions about the 
information I have put on the Web site.  I deliberately left out some 
things that I thought might be confusing to beginners.  Other things I have 
undoubtedly left out by accident, and there are certainly some 
errors.  Please do not hesitate to ask about things that are wrong, or that 
I had omitted or didn't make clear.

In that regard, I think particularly about the STS-1 Excel spread 
sheet.  It uses a number of Excel features (macros, etc.) that need to be 
set up properly.  When I was first developing the system, it took me 
several weeks, making about every mistake possible, to finally get it to 
work.  At least now I know many things which can go wrong.  I have assumed 
that anyone who is trying to make that one work, will be asking me for 
help.  And also it is not very user-friendly.  But it does solve an 
interesing problem.

>My original interest in STM-8 was basically as a model to understand the 
>theory behind the mechanism, especially because of the use of 
>electro-magnetic resources to stabilize and define the period of the 
>mechanical system, and provide damping. Just because of my background, I 
>thought it would be not only familiar for me, but easier to understand the 
>global performance and its implementation.

A couple of comments on the STM-8 design.  One critical factor, that was 
not directly addressed much is that of mechanical resonances.  With the 
feedback design, they ultimately will be a big limiting factor on how well 
you can make the instrument work, particularly at the higher 
frequencies.  It would be nice if there could be no big resonances below 
1000 Hz, though that's going to be hard to achieve with the long, thin boom 
of the STM-8.

Also, temperature effects are going to limit performance.  Designing an 
enclosure which can be regulated to a few thousandths of a degree, with 
minimum internal convection, will help a lot.  Also keeping all heat 
sources, except for the temperature-sensitive parts, outside the enclosure 
will be a good idea.

Finally, I think that the circuit needs more gain in the region below 1Hz, 
and more current drive capability.  There was some debate about whether 
doing that would introduce additional noise, but I believe that it should 
be possible without too much difficulty.

I'm sure that you can create a workable instrument without much concern for 
the above issues, but for top performance they probably would want to be 
considered.

Regards,
Brett

-----

If my e-mail address above is not working you can always reach my mail form
at: http://bnordgren.org/contactB.html
using your Web browser. 

Alexander,

At 04:54 AM 9/6/2005 +0200, you wrote:
Dear Brett:
I sincerely thank you for your detailed technical comments, the files you suggest me and specially your interest in making my first steps easier.

You are very welcome.  Always happy to help a fellow experimenter.
You got the right sense of what I want. I donít want to merely reproduce a design ignoring how it works. I absolutely need to know it before go into the construction..! I already downloaded the files from your web site. Considering the titles, I think they effectively cover important topics for my learning purpose.

I, like you, prefer to understand how things work before I jump in and start building.  Unfortunately, usually that means that I only get as far as the analysis and planning, and rarely get around to actually building things.  At least I can help pass along what I have discovered so far.

I hope you understand that I am expecting people to ask questions about the information I have put on the Web site.  I deliberately left out some things that I thought might be confusing to beginners.  Other things I have undoubtedly left out by accident, and there are certainly some errors.  Please do not hesitate to ask about things that are wrong, or that I had omitted or didn't make clear. 

In that regard, I think particularly about the STS-1 Excel spread sheet.  It uses a number of Excel features (macros, etc.) that need to be set up properly.  When I was first developing the system, it took me several weeks, making about every mistake possible, to finally get it to work.  At least now I know many things which can go wrong.  I have assumed that anyone who is trying to make that one work, will be asking me for help.  And also it is not very user-friendly.  But it does solve an interesing problem.
 
My original interest in STM-8 was basically as a model to understand the theory behind the mechanism, especially because of the use of electro-magnetic resources to stabilize and define the period of the mechanical system, and provide damping. Just because of my background, I thought it would be not only familiar for me, but easier to understand the global performance and its implementation. 

A couple of comments on the STM-8 design.  One critical factor, that was not directly addressed much is that of mechanical resonances.  With the feedback design, they ultimately will be a big limiting factor on how well you can make the instrument work, particularly at the higher frequencies.  It would be nice if there could be no big resonances below 1000 Hz, though that's going to be hard to achieve with the long, thin boom of the STM-8.

Also, temperature effects are going to limit performance.  Designing an enclosure which can be regulated to a few thousandths of a degree, with minimum internal convection, will help a lot.  Also keeping all heat sources, except for the temperature-sensitive parts, outside the enclosure will be a good idea.

Finally, I think that the circuit needs more gain in the region below 1Hz, and more current drive capability.  There was some debate about whether doing that would introduce additional noise, but I believe that it should be possible without too much difficulty.

I'm sure that you can create a workable instrument without much concern for the above issues, but for top performance they probably would want to be considered.

Regards,
Brett

-----

If my e-mail address above is not working you can always reach my mail form
at: http://bnordgren.org/contactB.html
using your Web browser.

[ Top ] [ Back ] [ Home Page ]