Brett
I'm an engineer but more familiar with small deflection theory. I'll go bac=
k to my college notes :) Thanks for the info on the stainless.
Barry



--- On Sun, 8/16/09, Brett Nordgren  wrote:

From: Brett Nordgren 
Subject: Re: Real time traces
To: psn-l@..............
Date: Sunday, August 16, 2009, 6:32 AM

Barry,

Thanks for the kind words.=A0 I'll try to keep more coming.

Regarding spring thickness:=A0 There is a direct relationship between the s=
tress level in the spring, its thickness and how sharply it is bent as defi=
ned by its radius of bending.

Approximately:=A0 max stress =3D Elastic modulus * thickness / (2 * min ben=
d radius)

So for a given max stress we get=A0=A0=A0thickness =3D 2 * max stress * min=
 bend radius / Elastic modulus

This obviously works for any units, so long the units of thickness and radi=
us are the same, and stress and Elastic modulus are also.

Putting some numbers to that:
desired max stress 72,000 psi
min bend radius =3D 2.5"
Elastic Modulus from a table of metal properties =3D 3E7 psi (or a little l=
ess)

So=A0=A0=A0thickness =3D 2 * 72,000 * 2.5 / 3E7 =3D 0.012"

Bending the spring more sharply requires thinner material to stay below the=
 same max stress.

For a spring bent to this radius, hacksaw blades and taping blades are quit=
e a bit too thick.

One issue is that you want to keep the stress level much lower than the yie=
ld stress of the spring material to stay far away from any issues like cree=
p or hysteresis losses.=A0 Clock makers seem to be designing their springs =
to stay under 40% of yield.

When hardened , the 17-7 stainless appears to achieve a yield strength abov=
e 280,000 psi, so we're in great shape, only 25% of yield.=A0 Even without =
hardening, the 17-7 is not so bad (198,000 psi), but from Dave's reports th=
e hardening process for this material is not all that difficult.....bake it=
 at 900 +/- 10 deg F for 1 hour, then let it cool.

If you want all the gory details about the material, and more, see.=A0 =A0 =
http://bnordgren.org/seismo/17-7_Stainless.zip

You can watch the seismo make wiggles on Larry's page=A0 http://psn.quake.n=
et/currentseismicity.html=A0 =A0 Thanks, Larry
Big quake this morning.

Regards,
Brett

At 04:18 AM 8/16/2009 -0700, you wrote:
> Dave
> I'm interested in finding out more and impressed also. I reduced the size=
 of the STM style vertical 12" but like your design. Didn't you try a hacks=
aw blade(s)=A0 before? I don't remember the thickness but would the taping =
blade material STM mentioned work? I'm not sure about my heat treating abil=
ities (proper annealing etc). I have read Bretts excellent documentation in=
 the past.
> Regards
> Barry


__________________________________________________________

Public Seismic Network Mailing List (PSN-L)

Brett
I'm an engineer but more familiar wi=
th small deflection theory. I'll go back to my college notes :) Thanks for =
the info on the stainless.
Barry



--- On Sun, 8/16/09, =
Brett Nordgren <brett3nt@.............> wrote:

From: Brett Nordgren <brett3nt@.............>Subject: Re: Real time traces
To: psn-l@..............
Date: Sunday,=
 August 16, 2009, 6:32 AM

Barry,

Tha=
nks for the kind words.  I'll try to keep more coming.

Regardin=
g spring thickness:  There is a direct relationship between the stress=
 level in the spring, its thickness and how sharply it is bent as defined b=
y its radius of bending.

Approximately:  max stress =3D Elastic=
 modulus
 * thickness / (2 * min bend radius)

So for a given max stress we ge=
t   thickness =3D 2 * max stress * min bend radius / Elastic=
 modulus

This obviously works for any units, so long the units of th=
ickness and radius are the same, and stress and Elastic modulus are also.
Putting some numbers to that:
desired max stress 72,000 psi
min=
 bend radius =3D 2.5"
Elastic Modulus from a table of metal properties =
=3D 3E7 psi (or a little less)

So   thickness =3D 2 *=
 72,000 * 2.5 / 3E7 =3D 0.012"

Bending the spring more sharply requi=
res thinner material to stay below the same max stress.

For a spring=
 bent to this radius, hacksaw blades and taping blades are quite a bit too =
thick.

One issue is that you want to keep the stress level much lowe=
r than the yield stress of the spring material to stay far away from any is=
sues like creep or hysteresis losses.  Clock makers seem to be designi=
ng
 their springs to stay under 40% of yield.

When hardened , the 17-7 =
stainless appears to achieve a yield strength above 280,000 psi, so we're i=
n great shape, only 25% of yield.  Even without hardening, the 17-7 is=
 not so bad (198,000 psi), but from Dave's reports the hardening process fo=
r this material is not all that difficult.....bake it at 900 +/- 10 deg F f=
or 1 hour, then let it cool.

If you want all the gory details about =
the material, and more, see.    http://bnordgren.org/seismo/17-=
7_Stainless.zip

You can watch the seismo make wiggles on Larry's=
 page  http://psn.quake.net/currentseismicity.html    Th=
anks, Larry
Big quake this morning.

Regards,
Brett

At 0=
4:18 AM 8/16/2009 -0700, you wrote:
> Dave
> I'm interested in
 finding out more and impressed also. I reduced the size of the STM style v=
ertical 12" but like your design. Didn't you try a hacksaw blade(s)  b=
efore? I don't remember the thickness but would the taping blade material S=
TM mentioned work? I'm not sure about my heat treating abilities (proper an=
nealing etc). I have read Bretts excellent documentation in the past.
&g=
t; Regards
> Barry


_______________________________________=
___________________

Public Seismic Network Mailing List (PSN-L)
<=
br>To leave this list email PSN-L-REQUEST@SE=
ISMICNET.COM with the body of the message (first line only): unsubscrib=
e
See http://www.seismicnet.com/maillist.html for more information.

[ Top ] 
[ Back ] 
[ Home Page ]