Summary from Equipment Meeting on September 18th, 1996

Attendees: Paul Russell, Matt Daum, Kevin Howard, Kevin Main, Greg 
Hoshal, Jack Dority, Yongquan Zhou, Mark Doutt, Bill (?) from Lansmont 
(Bill, what is your last name? sorry I didn't catch it)

Most of the time was spent on defining the differences between 
Piezoelectric and Piezoresistive accelerometers.  IST uses primarily 
piezoresistive accelerometers, with a correction algorithm for offset 
problems.  Voltage mode piezoelectric accelerometers are also used in 
some products.  Lansmont uses piezoelectric accelerometers in Saver.  
The two types of accelerometers capture data differently, so the same 
event may look different from each recorder.  Both resistive and 
electric have application strengths and weaknesses.  Resistive 
accelerometers appear to be the choice for longer events (drops), since 
it uses a "DC" response.  This also allows the accelerometer to detect 
tosses since it can detect vertical and horizontal velocity changes.  
Lansmont has apparently found ways to obtain the same results with 
certain algorithms.  This difference is a concern for me, since we want 
as much consistency and accuracy in the data we give to the Analysis 
group.  We don't want to "pass judgment" on particular products, but at 
the same time we need to have the most accurate and reliable way of 
collecting data.  The group discussed some ways to overcome this issue, 
including lab testing to characterize how each recorder captures 
certain events.  No decisions were reached.  This is a topic we need to 
brainstorm more on.

Jack agreed to send out surveys to find out exactly what recorders are 
available to use within the MADE group.  Our group will be responsible 
for communicating what is available, and which recorders will be used.  
I suggested we limit the recorders to the Saver and EDR-3 (to ensure 
the latest technology is used).  We may need to refine this based on 
the survey or other factors.

Please spend some time considering a plan of action for resolving the 
accelerometer difference between Saver and EDR3.  The goal is to be 
100% confident the data captured is what exactly happened; otherwise 
the analysis is unreliable.  Let me know if you have any ideas before 
the next meeting.