Thanks Chris,

I was able to remove the reverse convection problem.
I placed a heating pad on top of the wood box and inside the styrofoam box=
.
=20
Hi Bryan,=20

    Check on how much power it consumes? The running cost could be signifi=
cant.

The outdoor temperatures here was 76F 24 hrs ago to 25F tonight and it is=
 staying in the 60s constant in the void between the two boxes. this stopp=
ed all the drifting or convection noise.=20

    This sounds like quite a large temperature increase. I try to provide=
 the minimum heat required to keep the air column stable inside the seismo=
meter case.=20

There seem to be a increase in the 6-second microseisms as a strong cold=
 front moved through.

    This is a common observation. Microseisms can be generated by weather=
 systems and by local storms as well as by the common deep ocean currents=
 / storms.=20

Do you have pictures of your instruments? I would like to see them.

    My original seismometer used twin L booms of 1=E2=80=9D x 1=E2=80=9D=
 x 1/8=E2=80=9D mild steel angle, about 5=E2=80=9D apart and cross braced=
 to provide the horizontal and vertical frame. It seemed to work well and=
 reliably. The bearing used a SS ball and a carbide tool counterface. The=
 top suspension was a short 10 thou music wire. The base was 30=E2=80=9D=
 long and the column 18=E2=80=9D high. The effective Al boom length =E2=80=
=98k=E2=80=99 was 22=E2=80=9D ~ a 1.5 second pendulum. These intermediate=
 size pendulums are much easier to build and to operate, than the 1 metre,=
 two second ones, while still giving an excellent performance for the 20=
 second surface waves. The practical limit to the set period is determined=
 by the tilt drift experienced locally and is usually less than 30 seconds=
.


    I later cooperated with Stewart Bullen, a secondary school physics tea=
cher, to design and test a seismometer for use in UK schools. He had been=
 using a Lehman seismometer in teaching. The British Geological Survey are=
 backing the project and SEP + MUTR produced the seismometer.

    I changed the construction of the frame to 3=E2=80=9D x =C2=BE=E2=80=
=9D section Aluminum bars. One of my prototypes can be seen at www.jclahr.=
com/science/psn/chapman.html see =E2=80=98latest seismometer=E2=80=98. Ste=
wart and I had suggested satisfactory seismometer constructions to SEP, bu=
t their commercial =E2=80=98=E2=80=99experts=E2=80=99=E2=80=99 tried to re=
design it, with very limited success. See the MuVentures photos with Paul=
 demonstrating a later prototype.=20


     A prototype, which I still use, is shown at the =E2=80=98latest seism=
ometer=E2=80=99 and it has a 3=E2=80=9D x =C2=BE=E2=80=9D section Aluminum=
 bar frame. The length is 30=E2=80=9D, the vertical bar is 18=E2=80=9D and=
 the width is 12=E2=80=9D. The arm is a bit over 22=E2=80=9D long, giving=
 a 1.5 second pendulum. Shorter pendulums are easier to construct, house=
 and operate than 1 metre 2 second pendulums.=20

    A brass seismic mass of 1 kg is clamped to the Aluminum boom. The red=
 sensor magnet frame uses =C2=BC=E2=80=9D thick mild steel backing plates=
 and 4 off 1=E2=80=9D x =C2=BD=E2=80=9D x 1/8=E2=80=9D NdFeB magnets. It=
 is bolted together with 6 mm zinc plated mild steel bolts, which fit onto=
 the outside of the base bar. The coil is wound on a rectangular former wh=
ich gives good signal linearity, as opposed to the old type chunky round=
 coils. The coil is bolted to the square =C2=BE=E2=80=9D thick mounting pl=
ate.=20

    The mounting plate is fitted with two =C2=BD=E2=80=9D OD Al rod side=
 extensions with V grooves. A V shaped 7 core SS cable made from fishing=
 trace has loop ends crimped on, which fit into these grooves. This oppose=
s any tendency of the arm to oscillate around it=E2=80=99s long axis when=
 a quake occurs. The top loop is threaded around a groove on the edge of=
 a 1.5=E2=80=9D SS panel mount washer. A 10 thou OD music wire suspension=
 is used.=20

    A horizontal 1/16=E2=80=9D thick Copper damping plate is bolted to the=
 mounting plate on the side facing the bearing. The red damping magnet blo=
ck uses 4 off 1=E2=80=9D x 1=E2=80=9D x =C2=BC=E2=80=9D NdFeB magnets and=
 it is mounted on the Al bar base. The block can be slid along the base to=
 vary the length of the Cu damping plate covered by the field and so vairy=
 the damping.=20

    The bottom bearing consists of a =C2=BD=E2=80=9D OD SS ball bearing pr=
essed into a hole drilled in the edge of the vertical column. The end of=
 the Al arm is machined flat and a section of SS razor blade is glued to=
 it with two component acrylic cement to provide the hardened counterface.=
 This is a much better glue than Epoxy for mechanical constructions and it=
 is tough when it sets, not brittle.=20

    The MUTR production seismometer together with application programs and=
 technical notes are listed at www.bgs.ac.uk/schoolSeismology/seismometer.=
html
     There is also a 30 page technical manual for it, which is well worth=
 reading for potential constructors, at www.mutr.co.uk/images/seismometer.=
pdf The sensitivity range is from 5Hz to over 20 seconds period. It uses=
 crossed rod tungsten carbide bearings for both the top and the bottom sus=
pensions.
www.jclahr.com/science/psn/chapman.html see =E2=80=98latest seismometer=E2=
=80=98. Stewart and I had suggested satisfactory seismometer constructions=
 to SEP, but their commercial =E2=80=98=E2=80=99experts=E2=80=99=E2=80=99=
 tried to redesign it, with very limited success. See the MuVentures photo=
s with Paul demonstrating a later prototype.=20


     A prototype, which I still use, is shown at the =E2=80=98latest seism=
ometer=E2=80=99 and it has a 3=E2=80=9D x =C2=BE=E2=80=9D section Aluminum=
 bar frame. The length is 30=E2=80=9D, the vertical bar is 18=E2=80=9D and=
 the width is 12=E2=80=9D. The arm is a bit over 22=E2=80=9D long, giving=
 a 1.5 second pendulum. Shorter pendulums are easier to construct, house=
 and operate than 1 metre 2 second pendulums.=20

    A brass seismic mass of 1 kg is clamped to the Aluminum boom. The red=
 sensor magnet frame uses =C2=BC=E2=80=9D thick mild steel backing plates=
 and 4 off 1=E2=80=9D x =C2=BD=E2=80=9D x 1/8=E2=80=9D NdFeB magnets. It=
 is bolted together with 6 mm zinc plated mild steel bolts, which fit onto=
 the outside of the base bar. The coil is wound on a rectangular former wh=
ich gives good signal linearity, as opposed to the old type chunky round=
 coils. The coil is bolted to the square =C2=BE=E2=80=9D thick mounting pl=
ate.=20

    The mounting plate is fitted with two =C2=BD=E2=80=9D OD Al rod side=
 extensions with V grooves. A V shaped 7 core SS cable made from fishing=
 trace has loop ends crimped on, which fit into these grooves. This oppose=
s any tendency of the arm to oscillate around it=E2=80=99s long axis when=
 a quake occurs. The top loop is threaded around a groove on the edge of=
 a 1.5=E2=80=9D SS panel mount washer. A 10 thou OD music wire suspension=
 is used.=20

    A horizontal 1/16=E2=80=9D thick Copper damping plate is bolted to the=
 mounting plate on the side facing the bearing. The red damping magnet blo=
ck uses 4 off 1=E2=80=9D x 1=E2=80=9D x =C2=BC=E2=80=9D NdFeB magnets and=
 it is mounted on the Al bar base. The block can be slid along the base to=
 vary the length of the Cu damping plate covered by the field and so vairy=
 the damping.=20

    The bottom bearing consists of a =C2=BD=E2=80=9D OD SS ball bearing pr=
essed into a hole drilled in the edge of the vertical column. The end of=
 the Al arm is machined flat and a section of SS razor blade is glued to=
 it with two component acrylic cement to provide the hardened counterface.=
 This is a much better glue than Epoxy for mechanical constructions and it=
 is tough when it sets, not brittle.=20

    The MUTR production seismometer together with application programs and=
 technical notes are listed at www.bgs.ac.uk/schoolSeismology/seismometer.=
html
     There is also a 30 page technical manual for it, which is well worth=
 reading for potential constructors, at www.mutr.co.uk/images/seismometer.=
pdf The sensitivity range is from 5Hz to over 20 seconds period. It uses=
 crossed rod tungsten carbide bearings for both the top and the bottom sus=
pensions.
www.mutr.co.uk/images/seismometer.pdf The sensitivity range is from 5Hz to=
 over 20 seconds period. It uses crossed rod tungsten carbide bearings for=
 both the top and the bottom suspensions.

    The UK schools=E2=80=99 seismic initiative has proved to be highly suc=
cessful project with over 475 instruments having been sold since Easter 20=
07!=20

I hope that this description will be of help to constructors.

Regards,

Chris Chapman


    The UK schools=E2=80=99 seismic initiative has proved to be highly suc=
cessful project with over 475 instruments having been sold since Easter 20=
07!=20

I hope that this description will be of help to constructors.

Regards,

Chris Chapman
=20















Thanks Chris,




I was able to remove the reverse convection problem.

I placed a heating pad on top of the wood box and inside the styrofoam box=
..


 


Hi Bryan, 


 


    Check on how much power it consumes? The running=
 cost could be significant.


 


The outdoor temperatures here was 76F 24 hrs ago to 25F tonight and=
 it is staying in the 60s constant in the void between the two boxes. this=
 stopped all the drifting or convection noise. 


 


    This sounds like quite a large temperature increas=
e. I try to provide the minimum heat required to keep the air column stabl=
e inside the seismometer case. 


 


There seem to be a increase in the 6-second microseisms as a=
 strong cold front moved through.


 


    This is a common observation. Microseisms can be=
 generated by weather systems and by local storms as well as by the common=
 deep ocean currents / storms. 



Do you have pictures of your instruments? I would like to see them.



    My original seismometer used twin L booms of 1=E2=80=9D=
 x 1=E2=80=9D x 1/8=E2=80=9D mild steel angle, about 5=E2=80=9D apart and=
 cross braced to provide the horizontal and vertical frame. It seemed to=
 work well and reliably. The bearing used a SS ball and a carbide tool cou=
nterface. The top suspension was a short 10 thou music wire. The base was=
 30=E2=80=9D long and the column 18=E2=80=9D high. The effective Al boom=
 length =E2=80=98k=E2=80=99 was 22=E2=80=9D ~ a 1.5 second pendulum. These=
 intermediate size pendulums are much easier to build and to operate, than=
 the 1 metre, two second ones, while still giving an excellent performance=
 for the 20 second surface waves. The practical limit to the set period is=
 determined by the tilt drift experienced locally and is usually less than=
 30 seconds.





 


    I later cooperated with Stewart Bullen, a secondar=
y school physics teacher, to design and test a seismometer for use in UK=
 schools. He had been using a Lehman seismometer in teaching. The British=
 Geological Survey are backing the project and SEP + MUTR produced the sei=
smometer.


 


    I changed the construction of the frame to 3=E2=80=
=9D x =C2=BE=E2=80=9D section Aluminum bars. One of my prototypes can be=
 seen at www.jclahr.com/science/psn/chapman.html see =E2=80=98latest=
 seismometer=E2=80=98. Stewart and I had suggested satisfactory seismomete=
r constructions to SEP, but their commercial =E2=80=98=E2=80=99experts=E2=
=80=99=E2=80=99 tried to redesign it, with very limited success. See the=
 MuVentures photos with Paul demonstrating a later prototype. 






 


     A prototype, which I still use, is shown at=
 the =E2=80=98latest seismometer=E2=80=99 and it has a 3=E2=80=9D x =C2=BE=
=E2=80=9D section Aluminum bar frame. The length is 30=E2=80=9D, the verti=
cal bar is 18=E2=80=9D and the width is 12=E2=80=9D. The arm is a bit over=
 22=E2=80=9D long, giving a 1.5 second pendulum. Shorter pendulums are eas=
ier to construct, house and operate than 1 metre 2 second pendulums. 


 


    A brass seismic mass of 1 kg is clamped to the Alu=
minum boom. The red sensor magnet frame uses =C2=BC=E2=80=9D thick mild st=
eel backing plates and 4 off 1=E2=80=9D x =C2=BD=E2=80=9D x 1/8=E2=80=9D=
 NdFeB magnets. It is bolted together with 6 mm zinc plated mild steel bol=
ts, which fit onto the outside of the base bar. The coil is wound on a rec=
tangular former which gives good signal linearity, as opposed to the old=
 type chunky round coils. The coil is bolted to the square =C2=BE=E2=80=9D=
 thick mounting plate. 


 


    The mounting plate is fitted with two =C2=BD=E2=80=
=9D OD Al rod side extensions with V grooves. A V shaped 7 core SS cable=
 made from fishing trace has loop ends crimped on, which fit into these gr=
ooves. This opposes any tendency of the arm to oscillate around it=E2=80=
=99s long axis when a quake occurs. The top loop is threaded around a groo=
ve on the edge of a 1.5=E2=80=9D SS panel mount washer. A 10 thou OD music=
 wire suspension is used. 


 


    A horizontal 1/16=E2=80=9D thick Copper damping pl=
ate is bolted to the mounting plate on the side facing the bearing. The re=
d damping magnet block uses 4 off 1=E2=80=9D x 1=E2=80=9D x =C2=BC=E2=80=
=9D NdFeB magnets and it is mounted on the Al bar base. The block can be=
 slid along the base to vary the length of the Cu damping plate covered by=
 the field and so vairy the damping. 


 


    The bottom bearing consists of a =C2=BD=E2=80=9D=
 OD SS ball bearing pressed into a hole drilled in the edge of the vertica=
l column. The end of the Al arm is machined flat and a section of SS razor=
 blade is glued to it with two component acrylic cement to provide the har=
dened counterface. This is a much better glue than Epoxy for mechanical co=
nstructions and it is tough when it sets, not brittle. 


 


    The MUTR production seismometer together with appl=
ication programs and technical notes are listed at www.bgs.ac.uk/schoolSei=
smology/seismometer.html


     There is also a 30 page technical manual for=
 it, which is well worth reading for potential constructors, at www.mutr.co.uk/images/seismometer.pdf The sensitivity range is from 5Hz to over=
 20 seconds period. It uses crossed rod tungsten carbide bearings for both=
 the top and the bottom suspensions.


 






    The UK schools=E2=80=99 seismic initiative has pro=
ved to be highly successful project with over 475 instruments having been=
 sold since Easter 2007! 


 


I hope that this description will be of help to constructors.


 


Regards,


 


Chris Chapman