Thread: Re: Calculating the CONSTANT to use to remove instument response

Hello,

Reftek 130 has 24 bit digitizer. and the min and max count values are +-2^(24-1) = 6291456 count for +- 20V
20/(2*6291456) = 1.5895microVolt/count.

We use ref2segy and seg2sac programs to convert the files.
segy2sac (without -g parameter) removes the Reftek gain and its output is in Volts.


best regars
Onur

Dr. Onur TAN |------------------------------------------------------------
------------- |
| TÜBİTAK Marmara Araştırma Merkezi | TUBITAK Marmara Research Center |
| Yer ve Deniz Bilimleri Enstitüsü | Earth and Marine Sciences Institute |
| Gebze - Kocaeli | Gebze - Kocaeli - TURKEY |
--------------------------------------------------------------------------

________________________________
From: sac-help-bounces<at>iris.washington.edu [sac-help-bounces<at>iris.washington.edu] On Behalf Of Chad Trabant
Sent: Friday, April 10, 2009 5:19 AM
To: sac-help<at>iris.washington.edu
Subject: Re: [SAC-HELP] Calculating the CONSTANT to use to remove instument response


Melaku,

I can't help you verify the RefTek gain value. If the equipment is from the IRIS PASSCAL Instrument Center perhaps you can ask them for more details of the digitizer. If the data was processed by PASSCAL you might also have the data in SEED format in which case you can use the rdseed program to create the SAC poles and zeroes file for you. If you have a manual or any documentation for the digitizer it might be in there as well.

Chad

On Apr 9, 2009, at 5:46 PM, Melaku Ayenew wrote:

Hi Chad,
Thank you so much! It help me a lot!
One thing I am not sure about is the digitizer gain, I got this value from IRIS, PASCAL, High Resolution Acquisition systems, under
Reftek R130 , 24 bit, 3 channel digitizer bit weight 1.589e-6 volts @ X1
gain.
Thank you once again.

Melaku

On Thu, Apr 9, 2009 at 5:28 PM, Chad <chad<at>iris.washington.edu<chad<at>iris.washington.edu>> wrote:

Hello Melaku,

I am forwarding this to the sac-help list in the hopes that it may garner the attention of more expertise.

Your calculation of CONSTANT looks generally correct with the exception of the 2*pi, it should not be included.

A standard gain STS-2 should be nominally 1500 Volts/meters/second, which needs to be scaled if you want nanometers. I have no idea what the digitizer gain for a RefTek 130 is, but lets assume your number is correct for now.

CONSTANT = A0 * SensorGain * Digitizer Gain

CONSTANT = 5.92e+07 * 1500 / 1.589e-06 = 5.588e+16 (so it looks like 2*pi is not in there after all).

That CONSTANT results in units of meters. So scale it by a factor of 1e9 for nanometers = 5.588e25

Hopefully someone more knowledgeable will correct any mistakes I've made.

On a slight related note, the poles and zeros you are using are the "truncated"/"simplified" STS-2 response and not the STS-2 nominal responses (you can access the nominal responses for each of 3 generations of STS-2 here: http://www.iris.edu/NRL/sensors/streckeisen/streckeisen_sts2_sensors.html)http://www.iris.edu/NRL/sensors/streckeisen/streckeisen_sts2_sensors.html%29. This is not a critical problem unless you are working with high frequency data, I have heard that it's not important below 35 Hz.

Chad

On Apr 9, 2009, at 3:11 PM, Melaku Ayenew wrote:

Hello Chad,
Thank you for your response
I am just starting to learn sac for my research, I was trying to remove STS-2 broadband instrument and convolute the Wood-Anderson response. This is the STS2.pz file I came up with would you check it for me please

ZEROS 2 (rad/sec)
0.000 0.000
0.000 0.000

POLES 5 (rad/sec)

-0.03701 0.03701
-0.03701 -0.03701
-251.3 0.0000
-131.0 467.30
-131.0 -467.30

CONSTANT 5.5884E+16

This is how I calculate the the Constant
CONSTANT=A0 X SensorGain X Digitizer Gain X 2*pi
Where A0 is normalization factor =5.92 E+07
The digitizer is REF TEC 130 data logger with bit weight 1.589E-06volts which I figure the digitizer gain would be 1/1.589 E-06 ( I am not sure about this step)

And from SAC>transfer from polezero subtype STS2.pz to WA

when I plot this I got a waveform with amplitude in the order of 10E-4 nm (transfer returns values in nm)

I really appreciate your help.
Thank you in advance.!!



Melaku Bogale
New Mexico State University
Department of Physics
Las Cruces NM 88003-8001




On Wed, Apr 8, 2009 at 10:13 PM, Chad Trabant <chad<at>iris.washington.edu<chad<at>iris.washington.edu>> wrote:

Hello Melaku Bogale,

You need to know the total sensitivity of the digitizer, this is the value which relates digital counts to ground units. The poles and zeros only represent the sensor. The CONSTANT in the SAC poles and zeros file should be the total sensitivity multiplied by the normalization factor for the poles and zeroes. The value for CONSTANT depends on the units desired also.

Chad
IRIS DMC


On Apr 8, 2009, at 3:50 PM, Melaku Ayenew wrote:

Hi All,
I am trying to remove instrument responses from STS_2, CMG3T. CMG_ES broadband instruments. I got the poles and zeros from IRIS Pascal Instrumentation. But I have difficulty figuring out what constant I should use in the *.pz file that has to read by the SAC TRANSFER command. For example, I was trying to synthesis a WA seismogram by removing the STS_2 response and convoulating the WA response. I used a constant of 1 in this calcultion. The synthesis seismogrm looks good but the displacemnt values don't make sense (they are several hunderd meters). I understand it is because of the CONSTANT=1 I put. My Question is How can I calculate the correct CONSTANT that I need to put with the zeros and poles in the SAC *.pz file. The IRIS PASCAL instrumentation page gives the zeros and poles and a normalization factor.
I greatly appreciate your help in this matter.

Melaku Bogale
New Mexico State University
Department of Physics
Las Cruces NM 88003-8001
_______________________________________________
sac-help mailing list
sac-help<at>iris.washington.edu<sac-help<at>iris.washington.edu>
http://www.iris.washington.edu/mailman/listinfo/sac-help