I was in the bicycle business in my past life and there was a brand call Wald, a real cheap brand. We were all Campagnolo snobs at the time. At one international bike show the entire Wald group showed up with t-shirts that read “when the best just won’t do”. Actually the Wald company was bigger and more profitable than the snob brand, and for the most part they got the job done.
Well, more on picking an instrument for your seismic network. Once you know what frequency range you want then you have to pick a digitizer. Yes, yes 24 bits. Everyone wants 24 bits. Even if you decide you need 24 bits, not all 24 bit digitizers are the same (see Advertised and Effective Bits: Don’t Be Fooled).
Sometimes 12 bits is enough. Sometimes it is more than enough. A while back someone was trying to make a case for using a Q330 24 bit digitizer to replace and National Instruments 16 bit digitizer to digitize a signal that came in over VHF telemetry. The dynamic range of the signal is limited by the dynamic range the analog radio signal to about 11 bits and using a 24 bit digitizer simply makes no sense even if you had a bunch of Q330s just kicking around.
Here we had a hardware developer that wanted to digitize a fluid level sensor that had a resolution of 1 cm over a 10 meter range. He was arguing for high-bit digitizer. I had to explain that the limit of the sensor was one part in a thousand and that maybe a ten bit digitizer was OK. If all you are going to do is pick P, S and coda you don’t need much bandwidth. If you are going to measure temperature and pressure you also don’t need much bandwidth. See the chart below:
One of our favorite digitizers is the Earthdata 209. It just a digitzer, true true 24 bits. It digitzes 6 channels and has a native Seedlink server. Not much else. Some digitizers do all sorts of things. Some digitizers have little TV screens to show wave forms. Then you put them in a vault and hopefully never go back.
But here in the lab we use a true 16 bit 8 channel digitizer for most of our work. It does wonders. And most of what we do in the lab has a limited dynamic range.
When the best just won’t do, look around for what will get the job done.