Hi Brett,

You may recall that I suggested a rolling “rocking chair” like sensor a 
while back.  Well I built one out of two ½” floor pipe flanges and a ½” 
x 3” pipe nipple.  I took care to machine the flange edges round and 
somewhat smooth with valve grinding compound.  A couple of pieces of ¼” 
threaded rod at right-angles to the axis of the nipple balanced the 
“rocking chair.”  I always had in the background the thought that this 
would then be a rolling band bearing pivot.  But to simply the first 
tests, I wrapped a single 0.001” stainless steel wire around each flange 
edge and placed the rolling affair on a piece of glass.  I adjusted to 
about 5 seconds period and could get a couple of swings.  Bah humbug. 
But thinking about it, I wondered if the “softness” of the SS wire might 
be the problem, so I switched to some 0.01” diameter music wire – much 
harder.  Things were considerably improved.  The whole thing was still 
not tremendous because now I could see actual problems with smoothness 
of the roll – probably specks of dust in the path of the bearing points.

So conclusions I took away from this test and the discussions in this 
thread.
1) Even very small amounts of friction, such as the compressions of 
non-elastic soft steel, can be detrimental.  Use only hardened wire or 
foil such as spring steel for the bands.
2) I feel wire is preferable to foil as the ability to “cut” through 
specs of dust will help in real life environments.
3) Use the smallest diameter pivot commensurate with fatigue failure 
properties of the wire vs. pivot diameter.  This in order to give 
maximum ratio of leverage of the pendulum bob to the pivot roll wire 
diameter/friction.

While on the subject of pivots, I’m going to start another thread with a 
question for you Brett.

Regards,
Charles R Patton


Brett Nordgren wrote:
> Gary,
> 
> I agree with Chris that the decay times are likely being affected by the 
> rigidity of your setup.  Precision pendulum clock makers learned long 
> ago that they needed to have an extremely rigid mounting from which to 
> hang the pendulum, or the energy loss could be considerable.  Air 
> resistance is a smaller effect at our long periods, but with a good 
> enough pivot, you might be able to see its effect.
> 
> It's my opinion that a decaying exponential is a decent approximation to 
> what you will normally observe.  There are undoubtedly second-order 
> nonlinearities due to all sorts of things, as observed by Dr. Peters.  
> Plotting the difference between what you observe and a theoretical 
> exponential curve might be interesting, but I argue that the difference 
> will not be all that large.  Possibly worth some more experiments?
> 
> It would be interesting to see how a rolling foil pivot would perform.  
> See:  http://bnordgren.org/seismo/RollingLehman.pdf  and 
> http://bnordgren.org/seismo/zerohng2.pdf    In theory that should have 
> virtually no friction.  It might not be easy to implement as the hinge 
> would need to be on the back side of the support from the boom, and you 
> would need some sort of way for the boom to wrap around the support.  
> That is shown in "RollingLehman.pdf", which is a top view of two 
> possible ways of designing the pivot.  That pdf also includes a sketch 
> of the hinge design which I got from Chris Chapman.  Also, you would 
> need to carefully adjust the support wire length so that the tension on 
> the upper and lower foils will be roughly equal.  With that arrangement, 
> most damping would likely be from motions of the top of the support 
> rod.  The foils should be as thin as you can make them.  If the mass 
> isn't too great, you might be able to use .001" foil.
> 
> Regards,
> Brett
> 
> At 04:07 PM 7/3/2008 -0700, you wrote:
> 
>> I have been taking my time to get my first Lehman up and running. 
>> First just to learn how the thing works and then to optimize the 
>> operation. In the last month we have had lots of discussion on various 
>> Tungsten rods and other materials for the 2 horizontal rods and ball 
>> bearing support system. I first tried various rod materials by trying 
>> to get as long resonant period, but came to the conclusion this was 
>> difficult to repeat. Meredith on his 6/25 message got me thinking to 
>> take a different approach. I reset the Lehman for a short period 
>> (about 10 seconds) and then connected the sensor to the amplifier and 
>> watched the display as the amplitude decayed. Now we have something 
>> that can be easily measured and repeated. I standardized on 
>> measurements for 5 minutes or 300 seconds. The decay equation is Y= A 
>> e- t/T where t is time and T is the time constant of the system. For 
>> materials I used hardened steel, stainless steel, tungsten carbide (as 
>> received), tungsten carbide (mirror polished) 1/8 diameter rods. The 
>> ball bearing is ¼ diameter silicon carbide.
>>
>> Here is what I found for the time constants:
>>
>> Hardened steel: 154 seconds
>>
>> Stainless steel: 125 seconds
>>
>> Tungsten carbide ( as received): 155 seconds
>>
>> Tungsten carbide (mirror polish): 191 seconds
>>
>>
>>
>> The goal is to have a high time constant, indicating lower friction.
>>
>>
>>
>> Polishing: The tungsten carbide as received actually had a very good 
>> polish when received, but not quite a mirror finish. I obtained 3 
>> diamond polishing grit sizes, starting out with 35u, then 15u, and 
>> finally 3u. The 35 and 15u actually made the tungsten carbide rod 
>> rougher. I followed Chris Chapman s method for polishing using a bent 
>> sheet of copper with the diamond paste.
>>
>>
>>
>> Conclusion: Polished tungsten carbide rods have the lowest friction.
>>
>>
>>
>>
>>
>> Gary
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Gary Lindgren
>>
>> 585 Lincoln Ave
>>
>> Palo Alto CA 94301
>>
>>
>>
>> 650-326-0655
>>
>>
>>
>> www.blue-eagle-technologies.com
>>
>> cymonsplace.blogspot.com
>>
>>
>>
>>
> 
> 
> 
>               My e-mail address above should be working, but if not
> you can always use my mail form at: http://bnordgren.org/contactB.html
>                            using your Web browser.
> 
> __________________________________________________________
> 
> Public Seismic Network Mailing List (PSN-L)
> 
> To leave this list email PSN-L-REQUEST@.............. with the body of 
> the message (first line only): unsubscribe
> See http://www.seismicnet.com/maillist.html for more information.
__________________________________________________________

Public Seismic Network Mailing List (PSN-L)