Need help fixing thermal imager.

[mike]

Need help fixing thermal imager.

I have an Argus EEV P4438 Thermal Imager.

Here's the link to the pdf.

http://www.google.com/url?sa=t&rct=...mqSQBg&usg=AFQjCNF40_N9ZFJJTSQZIqiL2uYrPy4HAA

It's designed for firefighters to see bodies in dense smoke. Built in 1998
It has no absolute temperature readout. The auto-iris adjusts the gain
and I get to see relative
temperatures. That's fine.

When I turn it on, I get an image, but the contrast is low and the image
is negative.
Hot spots are supposed to be bright. On mine, hot spots are dark.

If I stand barefoot on the carpet and step off, I can barely see the
image of my footprints.
A wall-wart plugged into the wall shows up prominently, but it's black,
not bright.
So, the unit is not dead, it's just not right.

After a minute or so, a white area appears on the edge of the picture
and slowly moves
toward the center. It's like someone poured cream on the edge and it
flows toward the center.
Then the whole screen goes white. Cycling power brings it back.

There are also some artifacts in the center of the screen. Look kinda
like black lightning
flashes. It's position dependent. Point the unit down and they all but
disappear.
Point it up and they are the most prominent. They are less prominent as
it warms up.

I've done a lot of googling and reading of patents.

The thing claims to have a TGS sensor. The sensor looks like a vidicon
with a wire coming out the front
to a shielded box that I assume is the high sensitivity input. Has WAY
more circuitry than you'd expect
and dozens of variable resistors.


There's a 6V NiCd battery that measures 3V and won't accept any current
at 6V.
I assume this stores some calibration stuff. I'm afraid to replace the
battery for fear of losing
info I can't get back. Or maybe it doesn't work because the info is
already lost???

I've found mention of a "white out" phenomena that causes the screen to
go white when you look at very hot objects.
I'm looking at room temperature stuff only.

If I can fix the negative image issue, and the white out issue,
I might be well on the way to making it useful.


It's not worth spending a lot of money, but I'd like to make it work
properly if I can.

Anybody got any ideas on how to proceed?
Any chance of finding a service manual or calibration procedure?


Thanks, mike

 

[Don Y]

Hi Mike,

Need help fixing thermal imager.

I have an Argus EEV P4438 Thermal Imager.

Here's the link to the pdf.


It's designed for firefighters to see bodies in dense smoke. Built in 1998

Have you tried talking to your local fire department?

First thought: are you operating it so far out of its normal
operating range that the rules don't apply? I.e., does the
manual claim that using it the way you are *will* produce useful
results?

It has no absolute temperature readout. The auto-iris adjusts the gain
and I get to see relative
temperatures. That's fine.

When I turn it on, I get an image, but the contrast is low and the image
is negative.
Hot spots are supposed to be bright. On mine, hot spots are dark.

I the image presented on the video out identical?

If I stand barefoot on the carpet and step off, I can barely see the
image of my footprints.
A wall-wart plugged into the wall shows up prominently, but it's black,
not bright.
So, the unit is not dead, it's just not right.

After a minute or so, a white area appears on the edge of the picture
and slowly moves
toward the center. It's like someone poured cream on the edge and it
flows toward the center.
Then the whole screen goes white. Cycling power brings it back.

There are also some artifacts in the center of the screen. Look kinda
like black lightning
flashes. It's position dependent. Point the unit down and they all but
disappear.
Point it up and they are the most prominent. They are less prominent as
it warms up.

I've done a lot of googling and reading of patents.

The thing claims to have a TGS sensor. The sensor looks like a vidicon
with a wire coming out the front
to a shielded box that I assume is the high sensitivity input. Has WAY
more circuitry than you'd expect
and dozens of variable resistors.


There's a 6V NiCd battery that measures 3V

So, you know for a "fact" that this isn't as it should be
(you know it to be a 6V battery by counting the cells?
Can you measure individual cells to verify one or
more are toast?)

and won't accept any current at 6V.

So, you can safely ^^^^^^^^^^^^ shunt it with a *good* 6V
battery?

I assume this stores some calibration stuff. I'm afraid to replace the
battery for fear of losing
info I can't get back. Or maybe it doesn't work because the info is
already lost???

Lots of pots *and* calibration constants?

 

[mike]

Hi Mike,



Have you tried talking to your local fire department?

No, but I'm not finding anything else useful, so should give it a try.

First thought: are you operating it so far out of its normal
operating range that the rules don't apply? I.e., does the
manual claim that using it the way you are *will* produce useful
results?

Yes, there are youtube links showing it work at room temperatures.
Somebody snipped the link to the info.

http://www.google.com/url?sa=t&rct=...mqSQBg&usg=AFQjCNF40_N9ZFJJTSQZIqiL2uYrPy4HAA

I the image presented on the video out identical? yes

So, you know for a "fact" that this isn't as it should be
(you know it to be a 6V battery by counting the cells?
Can you measure individual cells to verify one or
more are toast?)

I know it's six volts because it says 6 volts on the battery
and has the proper number of cells.

So, you can safely ^^^^^^^^^^^^ shunt it with a *good* 6V
battery?

Yes, I can, but it's awkward to get at and what do I learn?
It's unlikely that the battery does anything but backup some data.
If it's still backed up, shunting won't help. If it's already lost,
shunting won't help.
Second battery won't fit in the box.
Yes, it's on the list of things to try.

Thought I'd see if anybody had any input before I started messing with
it. "Measure twice, cut once."
I really need a service manual.

Lots of pots *and* calibration constants?

Ok, what's your guess at the function of a 6V 15maH NiCd battery?
Patents suggest that there might be a calibration constant for each and
every pixel. There's a jumper next to the battery, but I'm afraid removing
it will make a bad situation worse.

If it were me, I'd have a jumper that you removed while pointing it at
a known temperature surface. You'd let it do its calibration thing then
put back the
jumper...but that's just me.

This is one of those situations where it might be trivial to fix if you
knew what you're doing, but impossible if you don't.
I'm hoping to find someone who knows about this unit or at least something
about how the DGS tubes work.

Here are some images of the guts:

http://img534.imageshack.us/img534/3553/analog.jpg

http://img17.imageshack.us/img17/3662/digitalds.jpg

I'm not familiar with abbreviations in this field, so not sure what
to make of the test points and cal adjustments.

I'm guessing that the problem lies somewhere in the bias setup for the
imager tube. But I really have no clue how the tube works.

 

[mike]

Hi Mike,



Have you tried talking to your local fire department?

I went to the local fire department. The imager is older than
the oldest guy there. Nobody had a clue. They send newer units back
to the factory for repair, so nobody knows what's inside.

Need to stumble upon a service manual.