am testing the frequency response of the
TPS334 IR sensor. My very crude tests
agree with the datasheet, which means that the image scan time will be very
long for high contrast images. Many
low-cost microbolometer-based thermal imaging cameras have 80x80 sensors (6400
pixels). At a 10Hz pixel clock (-3dB
sensitivity from datasheet),
the frame scan time would be 640 seconds, or
nearly 11 minutes. Yikes! At 100Hz, the sensor would only produce an
output level of %10 of its DC capability, and still require just over a minute
to scan the whole frame. Obviously, this will not be a "live video"
system, but might still produce some interesting still image thermographs.
I am working on other methods of sensing
long-wave IR too. More later.
hey everyone welcome
to another
installment of my video series on making
low-cost infrared imaging thermal
imaging so today I have that sensor that
I took out of the eggs Tec IR 250 and
desolder the sensor from the board and
mounted it inside this electrical box
with a one-off amp amplifier giving us
hundred hundred times gain and then I
also mounted this shutter wheel to a
motor and the shutter wheel is between
the sensor which is aiming this way out
of the box at this hot glue gun and this
glue gun has a temperature control and
it's been running for a while so it's a
pretty it's a pretty stable temperature
reference over here so with the shutter
wheel closed you can see on the
oscilloscope over here it's almost zero
volts there's actually a tiny bit of a
residual signal which is probably coming
from the temperature of the wheel itself
but when the wheel is open the signal is
about 50 millivolts and so by turning
the wheel it's at a rate we can
determine what the frequency response of
the sensor is this information is
actually on the datasheet but of course
testing it
you know ourself is better so let me
focus in on the oscilloscope screen and
then start the motor running okay so
this reference line here is 50
millivolts just so we can see what the
original sort of DC response is so I'm
going to spin the motor just about as
slowly as I can go and you can see at
this speed we're hitting pretty much a
hundred percent signal response anything
slowed it down just a little bit more
okay so if we switch the parameters here
it's running at about 2.2 Hertz and if
we switch back to the cursors we can see
that it's hitting the hundred percent of
DC response pretty almost perfectly
there it's almost a hundred percent so
let's speed up a little bit
and you can see what happens the signal
doesn't hit the hundred percent response
mark anymore we're not getting the full
amount of a voltage level even though
the temperature is the same the thermal
mass of the sensor doesn't have time to
respond quickly enough so as that
shutter opens the thermal mass of the
sensor can only change so quickly and it
doesn't have time to get all the way up
to the hundred percent response point
before the shutter closes and it starts
cooling down again so let's see how fast
we're running now this is coming in at
six point nine Hertz about and I can
give you the rms value but I think it's
almost better just to look at it and
kind of see graphically how far off we
are so it's six or seven Hertz it's not
so bad you know we're getting about
eighty or ninety percent so let's speed
up a little bit warm
so now you can see we're getting down to
about you know maybe 60 or 70 percent
response started controlling the motor
speeds quite difficult as it's it's
nothing running at six tenths of the
volts there's not a lot of adjustment
here another thing is that as we speed
up the signal won't return all the way
to zero either for the same reason the
thermal mass of that sensor doesn't have
time to cool down all the way by the
time the shutter window comes back
around across so let's see how faster
running here about 12 and a half Hertz
and we're getting maybe you know fifty
percent or something like that like you
see the datasheet has the the response
curve for this sensor but I kind of
wanted to measure it myself just to get
a graphical feel and to verify the
datasheet so let's speed up a lot and
see what happens so now the signal is
sort of almost gone I mean it's it's
with some better I also don't have any
filtering at all in the circuitry that
the sensors connected right to the
op-amp and there's no filtering
capacitors or anything just because I
wanted to get the maximum amount of
response so the noise here is you know
almost as bad as the signal see if we
can now you can kind of almost make it
out but it's really dying so let's see
how fast we're going here you know it's
not even good well I don't think it's
frequency measurements going to be all
that accurate now it seems to be
agreeing it's almost about 50 Hertz it
looks like which sounds about right slow
down just a bit so here's what that box
looks like on the inside it's just a
really simple circuit it's just one
op-amp the 80-85 35 and I I didn't even
have to order one I couldn't believe it
I opened up my parts jump in and found
one in there just like the datasheet
suggested for using with with a
pyroelectric sensor so I didn't in the
previous experiment I didn't even use my
germanium lens here just because it was
it was too tough to focus and so I just
used the raw sensor there sticking out
through the window it was a close-up of
the board itself I even had the the
op-amp already mounted on that on that
surface mount adapter so as you can see
a really simple circuit just 100x gain
and the nice thing about that op-amp is
that it's a chopper stabilized op-amp so
the offset input offset voltage is
extremely low and this cap here is just
a decoupling cap for the power supply
has nothing to do with the signal so the
next step in this project in what I'm
going to do is get two mirrors probably
one of these cheap galvo scanning
systems from ebay with mirrors in both
axes so that you can scan the scene by
moving the nerves around and then record
the signal from the sensor so the nice
thing we will be able to change the scan
speed and sort of adjust how much image
clarity we need for a given speed
scanning so we could scan really quick
and get a low quality signal without
much contrast or scan very slowly and
get a high contrast image so this will
tie in very nicely with my entry into
the FPGA world I bought an ter ASIC de 1
an Altera board and this I'm going to
use with the scanning electron
microscope project and also this
low-cost thermal imaging project you're
both kind of similar I'm have a frame
buffer on the estrella on that board and
fill up the frame buffer with data
collected for me to the microscope or
the thermal thermal sensor and then as
the they're both scanning devices even
so I'm going to use the same code for
both and then just display the output on
VGA so this will be a really good
project because it has all these kinds
of different components digital
collection and analog filtering and all
that stuff should be pretty cool even if
it doesn't work perfectly I think it'll
still be interesting and I expect to get
an image out of the thermal system it
may not be a great image or we might
have to scan really slowly but I think
it'll be pretty interesting so let me
know if you have any feedback and I will
see you next time online jobs to earn money, best online income site, top 10 online money earning sites, easy income online, easy online earning, earn money online from home, make money online legit, earn money online free fast and easy, online earning websites list, genuine online money earning sites, online work to earn money, online surveys to earn money, earn money through internet, best online income, earn money online data entry, easy ways to make money online, best online earning websites, top websites to earn money, online typing jobs for students to earn money, earn skrill money online, earn skrill money, best way to earn money from home, make instant money online absolutely free, trusted online money making sites, online income site,
Posts
No comments:
Post a Comment