PSN-L Email List Message

Subject: Re: How much mass ?
From: ChrisAtUpw@.......
Date: Thu, 14 Feb 2008 09:49:04 EST


In a message dated 2008/02/14, shammon1@............. writes:

> Hi, Jon, I like to use the old style Lehman design. The device I built with 
> the best results was the PSN San Jose Lehmans which used a 1-in square 
> aluminum rod L= 80cm and 80-ounce lead mass. They had a natural period of 
> 20-seconds with minimal setup effort. The total boom length was 100cm and they used a 
> brass plate mounted at the end of the boom for the damping. I was very happy 
> with the performance. One of the issues you will face is the selection of 
> the gauge of the upper support wire. I tried to use #8 machine (piano) wire but 
> it kept breaking during use. I increased the wire gauge to #10 machine wire 
> and the wire life was about 1-year before rust causes the wire to snap under 
> tension. 

Hi Steve,

       If you use D'Addario piano wire, it comes protected with Nickel 
plating.
       Use either Aluminum or Copper plate for the damper. Al works, but 
1/16" Cu is better.
       I use 1" square x 1/8" thick NdFeB magnets for the sensor and 1" x 
1/2" x 1/4" thick NdFeB magnets for the damper, four off in both cases. A N/S 
pair on one 1/4" backing plate faces a S/N pair on the other backing plate. The 
backing plates are held in position by 1/4" zinc plated mild steel bolts. This 
greatly reduces any stray field. 
       Put the coil on the arm and the magnet on the base. If you put a 
magnet on the arm, you will pick up a lot of magnetic noise from the house wiring, 
passing trains and lorries etc.

> When I moved here to Aptos, California I was forced to reduce the length of 
> the boom to L=60cm because of limited space as seen in the photos in the 
> link below and the resulting natural period is now typically 10-12 seconds. 
> Needless to say, the performance these Lehmans are marginal and I live near the 
> ocean and the wave action causes them to osculate during storms. 

       You should be able to get 30 seconds out of a 60 cm arm quite OK. Most 
period limitations are caused by a poor lower suspension, such as a point in 
a cup or knife edge. It improves the period if you put the mass at the end of 
the arm. 
       Use 3/4" square or round Al tube for the arm - definitely not solid. 
You want to keep the arm light but rigid compared to the mass. I use 15 mm SS 
water pipe + plumbing fittings
       The bottom hinge should be ball on a plane or crossed cylinder. You 
put the SS ball on the upright and the SS / WC plane / SS blade the end of the 
arm. For WC rod, you can use 1/8" Tungsten Carbide drill shanks, with the 
vertical on the upright. See www.smallparts.com. You can also buy type 416 SS 1/4" 
shoulder bolts from McMaster Carr. www.mcmaster.com
       If you construct a double T frame similar to 
http://www.bgs.ac.uk/education/school_seismology/seismometer.html you can set up the system and then 
trim the response - all adjustments are sequential, not inter related like on 
the psn designs. This makes setup and adjustment far easier. The red block is a 
support, with a horizontal damping blade and the sliding damping magnet is 
placed on the bottom frame. The coil is fixed to the outside end and swings 
between the other magnet block. The diagonal Al support tube prevents rotation. 
       I used 3" x 1" Al U Channel with triangular 6" x 6" x 1/8" Al plates 
to support the joint in the prototype. This is easy to make and to set up. I 
use 1/4" SS bolts.

>  My suggestion is to use the calculation seen on this list to determine the 
> boom length you desire. Then plan the design based on that calculation. You 
> said you wanted a device with a period of 20-second and as pointed out below 
> by Chris, L is based on the length of the boom measured from the pivot point 
> to the center of the mass weight (Chris, do you have this calculation?). I 
> have found that setting the device up this way also reduces the impact of 
> ground deformation (boom does not remains level and centered) over time which 
> improves the overall operation of the device keeping the boom off the stops. 
> http://pw2.netcom.com/~shammon1/AptosStn.htm

       If the distance between the hinge and the centre of mass is L metres 
and the suspension angle is A, the period T = 2xPixSqrt(L / (9.81 x sin(A))) I 
use 0.56 m which would give a 1.5 sec pendulum if held vertically. A is ~1/3 
degree.
       Alnico magnets are now quite expensive. NdFeB magnets are much cheaper 
and give a much higher output. For damper and coil designs - click on Lehman 
at http://jclahr.com/science/psn/chapman/ Make the damping blade wide so that 
it covers both magnets at all times as shown.
       There is also an alternative design under muventures.
       The boom does not have to be level, but it should be adjusted parallel 
with the lower frame. Then you can simply slide the damping magnet further 
over the horizontal damping blade to adjust the damping.
       You adjust the whole rigid frame to centre the arm and to set the 
period. I use three SS plates glued to the concrete floor for the screw adjusters. 
I glued SS bearings into the end of the adjust bolts. Then they don't move 
about / wander when you adjust them. The SS mounting nuts are glued to the 
underside of the frame.

       Regards,

       Chris Chapman   
In a me=
ssage dated 2008/02/14, shammon1@............. writes:

Hi, Jon, I like to use the old= style Lehman design. The device I built with the best results was the PSN S= an Jose Lehmans which used a 1-in square aluminum rod L=3D 80cm and 80-ounce= lead mass. They had a natural period of 20-seconds with minimal setup effor= t. The total boom length was 100cm and they used a brass plate mounted at th= e end of the boom for the damping. I was very happy with the performance. On= e of the issues you will face is the selection of the gauge of the upper sup= port wire. I tried to use #8 machine (piano) wire but it kept breaking durin= g use. I increased the wire gauge to #10 machine wire and the wire life was=20= about 1-year before rust causes the wire to snap under tension.

Hi Steve,

       If you use D'Addario piano wire, it com= es protected with Nickel plating.
       Use either Aluminum or Copper plate for= the damper. Al works, but 1/16" Cu is better.
       I use 1" square x 1/8" thick NdFeB magn= ets for the sensor and 1" x 1/2" x 1/4" thick NdFeB magnets for the damper,=20= four off in both cases. A N/S pair on one 1/4" backing plate faces a S/N pai= r on the other backing plate. The backing plates are held in position by 1/4= " zinc plated mild steel bolts. This greatly reduces any stray field.
       Put the coil on the arm and the magnet=20= on the base. If you put a magnet on the arm, you will pick up a lot of magne= tic noise from the house wiring, passing trains and lorries etc.


When I moved here to Aptos, Ca= lifornia I was forced to reduce the length of the boom to L=3D60cm because o= f limited space as seen in the photos in the link below and the resulting na= tural period is now typically 10-12 seconds. Needless to say, the performanc= e these Lehmans are marginal and I live near the ocean and the wave action c= auses them to osculate during storms.


       You should be able to get 30 seconds o= ut of a 60 cm arm quite OK. Most period limitations are caused by a poor low= er suspension, such as a point in a cup or knife edge. It improves the perio= d if you put the mass at the end of the arm.
       Use 3/4" square or round Al tube for th= e arm - definitely not solid. You want to keep the arm light but rigid compa= red to the mass. I use 15 mm SS water pipe + plumbing fittings
       The bottom hinge should be ball on a pl= ane or crossed cylinder. You put the SS ball on the upright and the SS / WC=20= plane / SS blade the end of the arm. For WC rod, you can use 1/8" Tungsten C= arbide drill shanks, with the vertical on the upright. See www.smallparts.co= m. You can also buy type 416 SS 1/4" shoulder bolts from McMaster Carr. www.= mcmaster.com
       If you construct a double T frame simil= ar to http://www.bgs.ac.uk/education/school_seismology/seismometer.html you=20= can set up the system and then trim the response - all adjustments are seque= ntial, not inter related like on the psn designs. This makes setup and adjus= tment far easier. The red block is a support, with a horizontal damping blad= e and the sliding damping magnet is placed on the bottom frame. The coil is=20= fixed to the outside end and swings between the other magnet block. The diag= onal Al support tube prevents rotation.
       I used 3" x 1" Al U Channel with triang= ular 6" x 6" x 1/8" Al plates to support the joint in the prototype. This is= easy to make and to set up. I use 1/4" SS bolts.


My suggestion is to use the c= alculation seen on this list to determine the boom length you desire. Then p= lan the design based on that calculation. You said you wanted a device with=20= a period of 20-second and as pointed out below by Chris, L is based on the l= ength of the boom measured from the pivot point to the center of the mass we= ight (Chris, do you have this calculation?). I have found that setting the d= evice up this way also reduces the impact of ground deformation (boom does n= ot remains level and centered) over time which improves the overall operatio= n of the device keeping the boom off the stops.
http://pw2.netcom.= com/~shammon1/AptosStn.htm


       If the distance between the hinge and=20= the centre of mass is L metres and the suspension angle is A, the period T=20= =3D 2xPixSqrt(L / (9.81 x sin(A))) I use 0.56 m which would give a 1.5 sec p= endulum if held vertically. A is ~1/3 degree.
       Alnico magnets are now quite expensive.= NdFeB magnets are much cheaper and give a much higher output. For damper an= d coil designs - click on Lehman at http://jclahr.com/science/psn/chapman/ M= ake the damping blade wide so that it covers both magnets at all times as sh= own.
       There is also an alternative design und= er muventures.
       The boom does not have to be level, but= it should be adjusted parallel with the lower frame. Then you can simply sl= ide the damping magnet further over the horizontal damping blade to adjust t= he damping.
       You adjust the whole rigid frame to cen= tre the arm and to set the period. I use three SS plates glued to the concre= te floor for the screw adjusters. I glued SS bearings into the end of the ad= just bolts. Then they don't move about / wander when you adjust them. The SS= mounting nuts are glued to the underside of the frame.

       Regards,


      
Chris Chapman

[ Top ] [ Back ] [ Home Page ]