At 11:42 AM 2/10/2004, you wrote:

Hi Chris,

Thanks for the feedback.  The pivots are made from safety pin point 
embedded in the
Plexiglas sheeting.  They rest in dents in the aluminum and in dents in the 
plastic.
All of the pivots are in compression.  I think I'll try the "ball on plane" 
joint next.

I think I need the weight to offset the weight of the sheets of 
plastic.  Maybe lighter
material would help to reduce their effect.

The coil is mounted as you suggest.  The axis of symmetry is horizontal and
although the axis of symmetry moves in a only a slight arc, the coil does 
rotate
during this movement, as it remains perpendicular to the threaded rod.

Cheers,
John



>Hi John,
>
>        I suspect that the problem lies a) in the suspension of the 
> plates, b) in the 'springyness' or hysteresis of this suspension and c) 
> in the flexure of the pendulum threaded rod. I can't quite see how you 
> have made your suspension joints. Three types of suspension give few 
> problems: 1 flexing foils or wires, preferably crossed for rigidity, 2 
> ball on plane, 3 circular cylinder crossed on circular cylinder. Point on 
> plane types are just asking for trouble.
>
>
> >> Romberg-Style Horizontal Pendulum
>
> >> February 7, 2004.  Mw 7.1This sensor is my first attempt to make the 
> compact pendulum described 
> here.  I have 
> not been able to obtain a period longer than 2 seconds, possibly due to 
> slight flexure of central threaded rod that carries the lead 
> weight.  When adjusted for a  longer period, the pendulum will drift to 
> one side or the other and is stable at either extreme.
>
>        You should be able to get a wide range of periods. I suspect that 
> the suspension is being stabilised by the spring flexing and that this 
> has a slightly different force law to the geometrical path relationship.
>        This is a 'general case' of the folded pendulum type suspension. 
> It is quite easy to get periods of over 45 sec with folded pendulum 
> suspensions.
>
> >> The magnet/coil pickup is from a PC hard drive.  It seems to have 
> sufficient gain.  One problem is that in this design the coil rotates as 
> it moves back and forth so that the amount of free travel is quite limited.
>
>        You would probably find it much more satisfactory if you suspended 
> the coil with it's axis of symmetry horizontal, perpendicular to the 
> plane of motion of the pendulum and in between horizontal magnet pairs. I 
> doubt if you really need all that lead. Plan view:
>
>                                      NNNNSSSS
>                                         C--O--I--L    <--- movement --->
>                            SSSSNNNN
>
>        Regards,
>
>        Chris Chapman

__________________________________________________________

Public Seismic Network Mailing List (PSN-L)