Page 2 of 2
Re: FLIR baseline on Development Board
Posted: Tue Jul 25, 2023 8:26 pm
by GTR3QQ
Wavicle wrote: ↑Sun Jul 23, 2023 10:30 pm
The "orange thing" is the audio mute relay. It produces the "click" sound when you reset the computer. It actually looks like maybe the "hot spot" is the standoff for VERA connected to the PCB, not the relay. Maybe it is reflecting something.
U10 is the 3.3V linear regulator. 1/3 of all power drawn from the 5V rail is dumped as heat within that part. This is pretty much expected.
U11 is the video amplifier for composite and s-video. That one surprises me a little bit. Not sure what's going on there.
Relays… why? why not use SCR? or cut power to the amp? Some amp comes with a Mute pin.
LDO, I think I saw a switching PSU circuit, I don't quite understand why not using muti voltage switching chip for 3V3 as well.
Sure…LDO is cheap, some how I think there is more reason to using LDO for IO bank that I missed.
Ripple?
well, he is using composite so there will be current on the output driver of the amp which I think is the reason for heat gen.
Do note, tho looks scary, it's only 30ish degree.
Re: FLIR baseline on Development Board
Posted: Thu Jul 27, 2023 1:53 pm
by spargue
voidstar wrote: ↑Fri Jul 21, 2023 3:15 pm
There may be "false-positives" due to reflections - for instance the metal casing around the keyboard/mouse are also a large reflective surface. At certain angles, I noticed the FLIR was "seeing itself" (or some other object in the reflection - such as my own body, which ends up appearing "hot").
The SD showed "hot" only in one image there - I think because it caught a reflection at that angle.
I tried closer shots - and I ended up just covering the SD metal housing with an index card for now.
I would expect some heat thru the PS/2 connectors - power flowing for those LED on those accessories. But notice in the last image 199, it ends up showing as "dark" near the PS/2 ports - so not sure what happened there.
This FLIR superimposes two images together - at closer ranges, I'm thinking the alignment of the two isn't as accurate. So the static images are one datapoint, but it makes a little more sense to see the results as the camera is being moved/shifted slightly (in terms of seeing which bright spots stay, and which ones are "iffy").
It doesn't record video, but maybe I can somehow attach an action-cam to it, hmmm.
Anyhow, around the VERA, there is definitely "warmth" near that "orange component" (at least it's orange on mine), then two smaller components at the top.
EDIT: And haven't gotten a look at whatever is below the VERA itself.
There's a lot of parallax error on the image, more obvious when it's closer to the aperture.
If you look at the Relay you can see it's outline and thermal shadow don't line up. makes it difficult to align the 2 layers. Might be best to position the camera such that the PCB is flat to the camera.