Amber Galileo/Radiance HS MWIR 3-5um InSb

[MrSheep]

I recently got my hands on this camera. Luckily I had my power supply from my amber radiance 1 and it powered right up. It cools down pretty quickly too (about 6min). Will post more about it as I figure out how to use it. I took a pic of the power board because I was trying to figure out the pinout to power it. Luckily the power connector that it came with is the same as the amber radiance.

Does anyone know the RS232 commands to control it?

[IR_Geek]

if memory serves it may actually be RS422.  Believe there was jumper you could set for 232/422 ;   Also it wasn't simple commands but needed a specific firmware/software combo running under DOS.   Think it was called Usermenu ?

Assume you have gotten analog video out?  If so there should be a video overlay telling you the software/firmware version.   

[MrSheep]

Ah ok that is good to know.

And yes I have the software called Usermenu. I am going to install it and try it out soon so I will let you know.
And yup composite video works via J2 using a SMA to BNC adapter.

The camera is running on V1.53

[IR_Geek]

That's good.   1.53 was the last version.    The HS was (still is?) a tank of an IR camera.   Wish I could get my hands on one as I spent many hours collecting data with them years back.

[MrSheep]

Nice! Yeah hard to come by but they do pop up once in a while.
While you were collecting data. Do you remember which ports you used? There is a special Amber HSVB as seen in the circuit pic above.
I wonder if I need a special interface for it or can it use a frame grabber. I think it's RS422 but in 4 pairs.

Also are these camera's able to be user calibrated and output radiometric images/videos?
The use is very limited when only using the composite video port compared to my FLIRSC6000.

I heard that these cameras can window down too! Just very little documentation out there given how old these cameras are

[MrSheep]

I have the radiance 1 manual so I assuming they used the same pinout.
Attached is the pinout of the HSVB

[MrSheep]

Oh wow I actually found this PDF about that port in my files. I do not have the files they talk about though.

[IR_Geek]

Yes I believe the pins were the same with 16 bit digital parallel.    There is HotLink  serial digital video on the other connector with the power/coms.   Never used that.   Used several frame grabbers over the years for the parallel data till finally getting a converter box to go to CameraLink.

yes, it can subwindow down to 128, 64, and 32 with different level of dilluting or 1:1.   Also a dither ability to output a larger format, but never utilized it as you need to use and external mirror.

[IR_Geek]

As for radiometric:   Not internal to the camera, but since you have the raw it can be done external in RTools or similar.   Word of caution is to either use camera side NUC and not double apply a NUC in software.

[MrSheep]

Wow that is certainly good to know! I did a quick search for RS-422 to camera link converter. And they did exist at one point but it seems its hard to find now a days.

I was thinking of using one of these since I have an older motherboard in storage that still has PCI
https://edt.com/product/pci-dvk/

And according to the description it should work as long as I can get the drivers from them:
"The PCI DVK’s 68-pin connector makes it the best choice for cameras with standard 68-pin SCSI II type cables/connectors, such as the Kodak / Redlake MEGAPLUS i and ES models. Cables are also available for the Texas Instruments TMC-1001PF, Raytheon Infrared Radiance 1 and HSX, Cincinnati Electronics Irris 256 ST, and most Hamamatsu* cameras. Most other AIA monochrome, Category 1, extended single and dual channel digital cameras are also supportable, but may require a custom cable. See Cables for more information."


As for getting radiometric data. Correct me if I am wrong. You are saying that if I am able to get the HSVB/RS422 interface working I am able to get the radiometric information using RTools? Do you happen to know what format the files (video or image) were? Wondering if it can be used with something like FLIR ResearchIR as I have that available and is much more modern.

[IR_Geek]

Sorry, it has no radiometric stored in the camera.   You would have to build a calibration in something like RTools RCal.    The Galileo/HS would work with RTools.   Not sure which capture cards played nice with the RTools.   Bitflow and Epix come to mind.

These cameras were never 'thermography' but tailored to scientific radiometric measurements.    The thermography cameras were too black box and the Galileo/HS line allowed full deep dives into what the FPA was capable of doing.

[MrSheep]

Ok that makes sense! Reminds me of the newer cooled cameras where you can build a user calibration in ResearchIR even though it didn't have a calibration from factory.

And ok cool. I will research those brands.

That makes sense as well. I've recently learned that cooled cameras have direct access to the FPA, which is how you are able to adjust integration times vs uncooled which you can not.

[IR_Geek]

the ability to adjust integration time makes the sensor way more useful for a variety of tasks.    very short for high temperature and very long for cold temperature scenes.   Of course that makes it way more involved on performing NUC, bad pixel replacement (BPR), and temperature calibration.

[MrSheep]

if memory serves it may actually be RS422.  Believe there was jumper you could set for 232/422 ;   Also it wasn't simple commands but needed a specific firmware/software combo running under DOS.   Think it was called Usermenu ?

Assume you have gotten analog video out?  If so there should be a video overlay telling you the software/firmware version.

Hmm Do you remember the pinout of the camera? I think you might be right about RS422. So my camera has a RS232 connector on the breakout cable. However whenever I try to connect to it I get timeouts. It only uses 3 pins.

[MrSheep]

I found this in the case the camera came with. The Connector is a DB25 Connector which I assume plugged into a host machine. But to where the wires leads is still a mystery.

[IR_Geek]

Interesting.   Assume the photo with the red, white, and black exposed wire on the DB9 connector is on the breakout cable coming from the camera?   If so then it may already be setup as RS232      If that is the case then be sure the firmware and software version are the same.   Not sure the GUI with RTools worked on every firmware version.  Try the usermenu DOS software.   It may require a null modem to swap the Tx/Rx.

Are there any other connectors besides power and the DB9?

[MrSheep]

Hmm interesting. I will have to do more experimenting.

For the Breakout it goes:

-DB9 (Labeled RS232)
-SYNC- (BNC)
-SYNC+ (BNC)
-Power (AMP MIL style Connector)

[MrSheep]

Wait a second. Is this the NULL modem you were talking about? It is DB9 Male on both Ends. This came in the case too!

[IR_Geek]

That's just a gender changer, but that could mean they used a null modem cable.   All a null modem does is swap the Tx and Rx.   try swapping those to signal lines.

[MrSheep]

Yeah I will try swapping the lines in a bit. Fingers crossed it is just as easy as that haha.

[IR_Geek]

Found this on the WayBackMachine from the original website.  The Galileo/HS was on another level from the Radiance 1

"The Galileo camera is a fully self-contained infrared camera. Features such as compact size, single DC input voltage operation, and remote control enable Galileo to be compatible with ta variety of applications, notably remote sensor functions.

This camera incorporates Amber's 256x256 InSb Snapshot focal plane array into a high reliability, cryo-cooler/dewar assembly with at-dewar electronics and dedicated signal processing electronics. The AE173 FPA furnishes a detector pitch of 30 µm, and implements a centroid windowing feature, allowing full 256x256 operation at 120 Hz, or 128x128 at 480 Hz, or 64x64 (1,400 Hz) subarrays.

Galileo's signal processing electronics are based on a three-board set that includes uniformity correction, power, and video. The camera is designed to accommodate separation of the sensor engine from the signal processing electronics via an interconnect cable up to ten meters in length with the appropriate shielding.

The camera furnishes composite analog video in RS-170 or PAL formats, and 12-bit digital data in parallel differential or serialized ("hotlink") high-speed formats. Additional features such as FPA control via external synchronization signals promote camera use in a variety of special applications. "

[MrSheep]

Yeah wow! Another reason why I want to get this fully operational again. The ability to sub window down! Those frame rates are impressive given this was from 1992/1993.

Also i've never heard of HotLink before, I wonder what the data rates are like. I believe those pins are also on the 15 pin D-Sub Connector.

[MrSheep]

ok no dice. Switching TX/RX didn't work. The RX led didn't light up on my tester so it is prob not the right way. Switching it back made it light up again.

I'm wondering if the RS232 was there originally and they switched to RS-422? There were two sets of these connectors in the case

One labeled Galileo one Radiance HS

[IR_Geek]

Only other thing is to check the baud rate and other settings on the COM.   Not all USB to COM play nice.    They originally came with 25 pin COM cables and it appears it was 'cut' so they could replace with a newer standard 9pin.   I'd assume 232 is correct since there are only 3 wires on that 9pin cable.   The analog video should say the baud rate.

[MrSheep]

Yeah I believe the baud rate is correct 9600, and it says 9600 on the analog video. I will post full settings in a bit.

Side Note: I contacted the frame grabber company https://edt.com/product/pci-dvk/
Fingers cross they still have the firmware files for the Amber Cameras (Galileo/HS/Radiance 1). They said they discontinued this card in 2014. Which isn't that far back.

[MrSheep]

Some interesting shots of the boards.

[MrSheep]

Some interesting shots of the boards. Pt 2

Very cool to see the Linear Stirling Microcooler located towards the top of the camera.

Looks like it was manufactured in 1997 by Hughes

[MrSheep]

Still a long shot but FLIR ResearchIR 4 has a selectable pixel type for the Amber Radiance 1!! (It even has selection for Amber 4128 which was their LN2 Dewar unit, as well as Phoenix cameras)

In Theory that means If I convert the 12-bit RS-422 Amber HVSB port to CameraLink, Then Camera Link to GigE it may be possible to connect to ResearchIR 4?
Would be a huge breakthrough if that is remotely possible haha.

[MrSheep]

Still no dice on communicating with the Galileo. There is no response when sending this command. It could be a software thing where other programs used a different hex command.
On the radiance 1 byte 3 is 01 instead of 0a. So that might play a role.

[IR_Geek]

Had any more luck with the Galileo?   

[MrSheep]

No unfortunately. Been pretty busy with other work/projects haha. I will prob get back to it after the holidays

[IR_Geek]

Roger that.  I've got some time off and playing around the internet archives.  I'll see what more can be dug up.    Maybe there are some cards that handle the HotLink.   Never used it but sure seemed a lot easier on the cabling.  Of course it probably wouldn't play nice with any of the FLIR capture software.

[MrSheep]

Yeah I was looking into hotlink because it only required 2 wires rather than 20 or so pairs of RS422. But yeah it prob wouldn't play nice since FLIR usually looks for Pleora or cameralink