 ·Consolidated Com..
·Republic Wireless
 ·Hollis Hosting
2 recommendations |
SMT process videoIt has been a long time since I've done production level design. Interesting tour of a modern SMT manufacturing facility. Board stuffing has come a long way from the days of hand taped thru-hole technology. » www.eevblog.com/2014/11/ ··· ry-tour/I found the moving probe tester very interesting, a far cry from using a bed of nails fixture. The other interesting techno tidbit came near the end. The tour was at a security alarm company and they mentioned the problem of IR motion detector generating false alarms is not primarily due to IR but rather RF ingress since the sensor output is so small. The power/signalling wires make a great antenna and we live in an increasing strong RF environment. That came as a surprise. I had assumed (know what they say about assume) false alarms were due to inability to properly filter the IR signal. /tom |
|
| |
OldCableGuy3
Anon
2014-Nov-19 3:23 pm
Oh man, flashback to the years I spent as a temp at Rockwell Collins. |
|
 SparkChaser
Premium Member
join:2000-06-06
Downingtown, PA
1 recommendation |
to tschmidt
said by tschmidt:It has been a long time since I've done production level design. Interesting tour of a modern SMT manufacturing facility. Board stuffing has come a long way from the days of hand taped thru-hole technology.
»www.eevblog.com/2014/11/ ··· ry-tour/
I found the moving probe tester very interesting, a far cry from using a bed of nails fixture.
The other interesting techno tidbit came near the end. The tour was at a security alarm company and they mentioned the problem of IR motion detector generating false alarms is not primarily due to IR but rather RF ingress since the sensor output is so small. The power/signalling wires make a great antenna and we live in an increasing strong RF environment. That came as a surprise. I had assumed (know what they say about assume) false alarms were due to inability to properly filter the IR signal.
/tom Thanks for that. That's the 3rd eevblog I've link to this week. Fortunately, by the time I was doing SMD, I was working for a large enough company that took the schematic and had different departments that did all the hard work of layout, producing and testing. Only got involved when there was a particular circuit that needed TLC. As for the RF, that was interesting. With sensing you never know what the source of the signal is. As you know Tom, I worked on sensing paper documents both optical and magnetic. What no one tells you is that paper going through a plastic path can develop a few thousand volts on it.  |
|
lutful... of ideas
Premium Member
join:2005-06-16
Ottawa, ON
1 recommendation |
to tschmidt
My first proper job was as a "test engineer" at Northern Telecom (later Nortel) in 1992 when they were starting the transition from plated-through-hole to surface-mount technology. |
|
| |
to tschmidt
said by tschmidt:I had assumed (know what they say about assume) false alarms were due to inability to properly filter the IR signal. said by SparkChaser: As for the RF, that was interesting. With sensing you never know what the source of the signal is. Lots of sensors now come with an integrated amplifier, AD, whatnot, and a digital output, and more resistance to interference is one of the reason. When everything is inside an IC, the connections can be much shorter than otherwise. And if extra protection is needed, a metal can can provide additional shielding, and should be able to prevent most of the stray RF reaching the circuit. |
|
SparkChaser
Premium Member
join:2000-06-06
Downingtown, PA |
Even back it the day they came that way  Real world problem: I have a product that has a build cost of $120. Do I use a fancy sensor that cost $10 or a photo diode that I can get for $0.40. If I can make it work with the 40 cent sensor, we have a product. It not, ummm nothing good. |
|
1 recommendation |
I don't doubt that, but I think there are much more integrated sensors nowadays.
I know what you mean by cost, and I'm sure that's applicable in many cases. However, the sensors I'm very familiar with are usually cheaper if you buy them integrated in comparison to rolling your own from the components. And that's even before design, real estate, and insertion cost come into play. |
|
lutful... of ideas
Premium Member
join:2005-06-16
Ottawa, ON
1 recommendation |
to tschmidt
said by tschmidt:IR motion detector generating false alarms is not primarily due to IR but rather RF ingress since the sensor output is so small. Tom, I have been unable to duplicate that on several PIR modules at home with as much as 1W RF emission from 900Mhz/2.4Ghz/5.8Ghz at very close range. But it is plausible from higher power AM/FM/VHF transmitters. |
|
 alphapointeDon't Touch Me
MVM
join:2002-02-10
Columbia, MO |
to tschmidt
I just finished watching that video. You're right, that flying probe tester was sweet!
(now, if our dispatch wifi would come back up soon (firmware updates...), I'd be able to watch the latest one with the Rigol scope with an attitude (trigger jitter) |
|
·Consolidated Com..
·Republic Wireless
·Hollis Hosting
|
to lutful
said by lutful: I have been unable to duplicate that on several PIR modules at home with as much as 1W RF emission. Interesting - that would be another interesting Dave talk. Post the result of your experiment in the comment section and see if he responds. I recently upgraded our outdoor lighting with a bunch of 12V LED floodlights and PIR sensors. As far as I can tell false triggers are due to IR - but then I live in a pretty rural area so RF benign. /tom |
|
lutful... of ideas
Premium Member
join:2005-06-16
Ottawa, ON
1 recommendation |
lutful
Premium Member
2014-Nov-20 9:52 am
The last diagram in this app note shows how Paradox counteracts RF noise, so it was definitely a concern in the past. » www.paradox.com/Products ··· ?CATID=8But newer design PIR modules usually have the sensor output connecting directly to a micro-controller or custom IC, in a very compact footprint, so there is probably less chance of corruption by external RF noise. |
|
| |
to tschmidt
I work for a company that makes carrier tapes and transport trays for components. A lot of specialized plastic extruder machines and thermoforming machines. |
|
TheMG
Premium Member
join:2007-09-04
Canada MikroTik RB450G
Cisco DPC3008
Cisco SPA112
1 recommendation |
to lutful
said by lutful:Tom, I have been unable to duplicate that on several PIR modules at home with as much as 1W RF emission from 900Mhz/2.4Ghz/5.8Ghz at very close range. But it is plausible from higher power AM/FM/VHF transmitters. It's not just the RF field strength but also may be frequency dependent due to the circuit design, trace and wiring lengths, etc. Sometimes you can pump 100W and it will do nothing but a few milliwatts at just the "right" resonant frequency can cause a device to go haywire. |
|
lutful... of ideas
Premium Member
join:2005-06-16
Ottawa, ON |
lutful
Premium Member
2014-Nov-24 11:10 am
said by TheMG:a few milliwatts at just the "right" resonant frequency can cause a device to go haywire. FCC/IC regulations do not actually allow "a few milliwatts" outside the 900Mhz/2.4Ghz/5.8Ghz bands I mentioned and the 460Mhz band for FRS walkie-talkies, which I also tested since that posting. The other frequencies allowed in a home, like 27Mhz or 300-500Mhz for misc wireless control, are limited to very low power, mainly to avoid interference. I mentioned it is plausible with FM/VHF/UHF/Cellular transmissions ... if the transmitters are very close. I could not see any effect from the Manotick tower which carries a dozen strong FM and TV channels. It is only a few km from my home. The airport's VHF landing system is even closer. |
|
|
public
join:2002-01-19
Santa Clara, CA |
to tschmidt
said by tschmidt:As far as I can tell false triggers are due to IR - but then I live in a pretty rural area so RF benign. Consumer grade pir detectors tend to randomly trigger after several months in outdoor installations. Perhaps the sensor output drops, and the circuit increases gain trying to compensate. |
|
·Consolidated Com..
·Republic Wireless
·Hollis Hosting
|
said by public:Consumer grade pir detectors tend to randomly trigger after several months in outdoor installations I hope that is not true. ne of my summer electronics projects this year, installed 5 12V LED flood lights and 4 12V PIR sensors. If a sensor fails I can disable it but I'd hate to have to go out in the dead of winter to replace them. So far the system has been running for several months without a hitch. They run off a DC supply. I tied the controllers (one in the house other one in the shed) to a photocell so the PIR sensors are only powered when when it is dark. If one of them locks up hopefully it will get reset by the power cycle. /tom |
|
lutful... of ideas
Premium Member
join:2005-06-16
Ottawa, ON |
to public
said by public:Consumer grade pir detectors ... There are just a few sensor ICs available ... and you will find exact same sensor inside consumer and industrial grade products of same company. They justify the price difference using bells and whistles, not sensor performance. probably the best: » ca.mouser.com/new/panaso ··· cnapion/ |
|
 SmokChsrWho let the magic smoke out?
Premium Member
join:2006-03-17
Saint Augustine, FL
1 recommendation |
to tschmidt
said by tschmidt:they mentioned the problem of IR motion detector generating false alarms is not primarily due to IR but rather RF ingress since the sensor output is so small. I've got PIR's at Several AM & FM transmitter sites, and they haven't had any problems with the RF. Now I did make the alarm techs use shielded wire when they did the installation. One night I received several false trips at a site, so I just sat down inside with the lights off waiting for the sensor to trip. After about 20 mins sure enough it lit up. When it did I could see the problem with the glow from the led inside the sensor. It was one or those pin head size spiders crawling around on the inside surface of the PIR lens. After the cover was removed and the spider dispatched, no more false alarms. |
|
·
·
