EEVblog #668 - Mailbag + FLIR TG165

[EEVblog]

Help beat Congenital Muscular Dystrophy:
http://www.beat4life.com/
(Pozible campaign live tomorrow)

A look at the FLIR TG-165 visual spot thermometer and (unfair) comparison with the FLIR E8 thermal camera.

High density Mitra 125 16KB magnetic core ferrite memory from the 1970's
http://feb-patrimoine.com/english/mitra.htm

Eric LPRS low power radio modules:
http://www.lprs.co.uk/easy-radio/eric/

[resistor]

Okay, I must be stupid.  How is the rotating laser thing possible?

[EEVblog]

Okay, I must be stupid.  How is the rotating laser thing possible?

They are just slightly angled in both axis.
In this case angled to be vertically aligned at about half a meter, and then lower and that and above that you get an apparent "rotation" in either direction.

[coppice]

Those strips on the memory controller board are a mezzanine bus. They have very little capacitance, so they have very little decoupling effect. I used to use them a lot in the 70s and early 80s.

Hand assembling those fine ferrite rings through a microscope wrecked people's sight in a surprisingly short time. They were not just assembled in South East Asia for labour cost reasons. It was also done to avoid health and safety regulations.

[Ebivetar]

So basically, FLIR TG165 is equivalent of Fluke VT02 ?

[nowlan]

Here is the comparison sheet form the other thread.
http://www.eevblog.com/files/FLIR-TG165/TG165_Comparison_FINAL.pdf
Comparison to Vt102 Vt104 Flir One.

[mikeselectricstuff]

TG165 teardown?

[mikeselectricstuff]

Was I the only one shouting "Tagarno" at the screen during the core memory bit....?

[EEVblog]

So basically, FLIR TG165 is equivalent of Fluke VT02 ?

In term of target market, yes, but the VT02 is pathetic in comparison.

[EEVblog]

Was I the only one shouting "Tagarno" at the screen during the core memory bit....?

It's still packed up from the trade show.
In any case it's less magnification than the macro lens.

[bktemp]

Was I the only one shouting "Tagarno" at the screen during the core memory bit....?

No. I waited all the time until the video switches over to the tagarno shot but then suddenly the mailbag video was over.

By the way, what is the advantage of using two different sensors on the TG165? Is the spot sensor more accurate than the lepton core?

[mikeselectricstuff]

Was I the only one shouting "Tagarno" at the screen during the core memory bit....?

No. I waited all the time until the video switches over to the tagarno shot but then suddenly the mailbag video was over.

By the way, what is the advantage of using two different sensors on the TG165? Is the spot sensor more accurate than the lepton core?

Lepton needs a reference shutter to do absolute temp measurement

[nitro2k01]

How often does the TG165 recalibrate? Given that the sensor is not used for absolute measurement, they should able to get away much less frequent calibration, maybe even just once on startup.

Regarding the Fluke, are you sure it wasn't just out of focus? There was a dial that I didn't see you touch or mention in the video.

[ryanmoore]

You wondered about the size of the core memory right at the end of the video. The convention at the time was to specify memory by number of words rather than bytes or bits, so it's most likely 16K x 18-bit words and not 16kB or 16kBits.

[mikeselectricstuff]

How often does the TG165 recalibrate? Given that the sensor is not used for absolute measurement, they should able to get away much less frequent calibration, maybe even just once on startup.

Regarding the Fluke, are you sure it wasn't just out of focus? There was a dial that I didn't see you touch or mention in the video.

Lepton does not use shutter calibration, and a thermopile sensor shouldn't need it

[EEVblog]

By the way, what is the advantage of using two different sensors on the TG165? Is the spot sensor more accurate than the lepton core?

Not necessarily so, both ultimately need to be calibrated. In this case it's a pure marketing move, giving the traditional user the look'n'feel of a PIR spot based device they are familiar with, whilst not encroaching upon the market of a "real" thermal imaging camera like the E4.

[EEVblog]

How often does the TG165 recalibrate?

It doesn't, because it doesn't take absolute readings from it.
The Lepton sensor doesn't need to recalibrate anyway, unlike the E4 et.al

Regarding the Fluke, are you sure it wasn't just out of focus? There was a dial that I didn't see you touch or mention in the video.

That was shot a long time ago, I don't recall. In any case, the resolution is awful.

[nitro2k01]

Lepton does not use shutter calibration, and a thermopile sensor shouldn't need it

Doesn't it? As far as I remember form your videos you have to periodically close a manual shutter on the Flir One to be able to use the absolute temperature measurement function. It doesn't depend on it for general use, but it can use it, as seen in the One. At about 6:50 into Dave's video when he turns his unit on, you can see the image freeze, and Dave also mentions an auto shutter. I also think I remember seeing the image freeze later on in the video, but I'm not going to back and look for those instances. Unless that's just a laggy user interface, I would presume it does use a shutter.

[mikeselectricstuff]

Lepton does not use shutter calibration, and a thermopile sensor shouldn't need it

Doesn't it? As far as I remember form your videos you have to periodically close a manual shutter on the Flir One to be able to use the absolute temperature measurement function. It doesn't depend on it for general use, but it can use it, as seen in the One. At about 6:50 into Dave's video when he turns his unit on, you can see the image freeze, and Dave also mentions an auto shutter. I also think I remember seeing the image freeze later on in the video, but I'm not going to back and look for those instances. Unless that's just a laggy user interface, I would presume it does use a shutter.

Lepton needs a shutter to do absolute measurements, but as they are using a seperate sensor to do temp measurement, they don't need a shutter - probably cheaper to do it that way, maybe also more accurate.
The other issue is field of view
 - you want a fairly narrow FOV for something like this, and a narrow FOV on the Lepton would mean a bigger, more expensive lens, whearas to get the FOV for a single-pixel thermopile they can use a cheap polythene fresnel lens

[bktemp]

The datasheet says: Automatic Shutter (calibration)
Therefore the lepton core should be more accurate than in the FLIR one. So why an additional sensor? With the higher resolution of the lepton core it should be possible to have a much smaller measuring spot than with the PIR sensor which would be a big advantage.
Does anybody know how large the measuring spot of an E8 is? Does it use a single pixel or avaraging of a group of pixels?

[nitro2k01]

Two documents from Flir themselves claiming that it's using a shutter.

http://www.eevblog.com/files/FLIR-TG165/TG165_Comparison_FINAL.pdf
http://www.flir.se/uploadedFiles/Instruments/Products/TG165/FLIR-TG165-Brochure-EN.pdf

Edit: That is not to say that they're using the image array for temperature, only that they thought that using shutter calibration improved the image enough to make it worth it.

[dentaku]

All that old magnetic core technology is fascinating.
I've watched Jeri's video on "how to make logic gates with square hysteresis ferrite" twice already. Very educational.

[EEVblog]

Two documents from Flir themselves claiming that it's using a shutter.

The Lepton sensor datasheet says it is shutterless for non-uniformity corrections, but can use an external shutter if required.
It also has automatic internal temperature compensation.
http://www.flir.com/uploadedFiles/CVS_Americas/Cores_and_Components_NEW/Products/Uncooled_Cores/Lepton/FLIR-Lepton-DataBrief.pdf

[nitro2k01]

*sigh*
Yes. However, I'm talking specifically about the TG165, as that was what was being discussed in the video. Go back and read my original question on page 1. Everything points to it having a shutter, the image freezing at least once in the video, and the advertising material mentioning a shutter. Out of curiosity, I'm wondering how often the TG165 will recalibrate, or if it will only do it once every time the unit is turned on, or perhaps even less often than that. It seems like you're trying to prove something I already know and never asked about.

[R_Gtx]

Flir One resolution 80x60 px.
New boy on the block at half the price, resolution 206x156 px.

Looks like the market is hotting up.

Further details here.