- Mon Jul 23, 2012 7:30 pm
#147725
Hi all,
I was wondering if somebody could be so kind as to do a sanity check on this circuit since I'm a complete idiot when it comes to anlog. My intent is to use the Minisense 100 as an impact sensor. I don't need to analyze the signal from the Minisense in any detail -- I'm only trying to detect vibration greater than some threshold. I'll be interfacing to a 3.3V digital input on an Amani GTX CPLD shield. I read up on op amps, fumbled around with LT Spice, and came up with the following:
"V1" on the left represents the Minisense 100. The op amp in the design is LT1169 but it seems to work the same with LM741. I suppose I'll use LM358 in the end, since that's what Sparkfun sells. I believe that I'm using the op amp as a differential amp, feeding into a full wave rectifier (since I don't care if displacement is up or down). Use of 1N914 in the full wave rectifier was random. I made a signal file for "V1" that I think is roughly similar to what I think the Minisense will produce for the sort of impact I'm interested in. This is what LT Spice shows for the "V1" signal, the op amp output, and the adapter output (across R7):
The Amani GTX would be watching the blue signal and would start a timer when it first goes "true". Timing would stop when a second, third, and fourth Minisense detect the same impact by the same means.
A few questions:
1) The Minisense can produce +/- 90V which is probably outside the input range of the op amp, but I am not yet sure how to protect against that. Or is the power associated with the theoretical +/- 90V extremes so tiny that it doesn't matter?
2) Any other glaring errors in this design? This is my first analog design, so I'm sure there are all sorts of stupid mistakes.
3) Is there any reason for me to consider the Piezo Element (http://www.sparkfun.com/products/10293) instead of the Minisense 100 (http://www.sparkfun.com/products/9198)? I'm guessing that the Minisense is more sensitive.
Thanks, in advance, for any suggestions!
I was wondering if somebody could be so kind as to do a sanity check on this circuit since I'm a complete idiot when it comes to anlog. My intent is to use the Minisense 100 as an impact sensor. I don't need to analyze the signal from the Minisense in any detail -- I'm only trying to detect vibration greater than some threshold. I'll be interfacing to a 3.3V digital input on an Amani GTX CPLD shield. I read up on op amps, fumbled around with LT Spice, and came up with the following:
"V1" on the left represents the Minisense 100. The op amp in the design is LT1169 but it seems to work the same with LM741. I suppose I'll use LM358 in the end, since that's what Sparkfun sells. I believe that I'm using the op amp as a differential amp, feeding into a full wave rectifier (since I don't care if displacement is up or down). Use of 1N914 in the full wave rectifier was random. I made a signal file for "V1" that I think is roughly similar to what I think the Minisense will produce for the sort of impact I'm interested in. This is what LT Spice shows for the "V1" signal, the op amp output, and the adapter output (across R7):
The Amani GTX would be watching the blue signal and would start a timer when it first goes "true". Timing would stop when a second, third, and fourth Minisense detect the same impact by the same means.
A few questions:
1) The Minisense can produce +/- 90V which is probably outside the input range of the op amp, but I am not yet sure how to protect against that. Or is the power associated with the theoretical +/- 90V extremes so tiny that it doesn't matter?
2) Any other glaring errors in this design? This is my first analog design, so I'm sure there are all sorts of stupid mistakes.
3) Is there any reason for me to consider the Piezo Element (http://www.sparkfun.com/products/10293) instead of the Minisense 100 (http://www.sparkfun.com/products/9198)? I'm guessing that the Minisense is more sensitive.
Thanks, in advance, for any suggestions!