Could a camera flash interfere with the DCS signal?

Alex,

 

The flash and the crossing gates is easy to explain.  The Z-Stuff gates use optical sensors.

 

The flash and the DCS craziness is almost certainly simply a coincidence.  Nothing in the DCS control is light sensitive.

 

 

 

 

It shouldn't really, but could on occasion.

Light wouldn't be the only "wave" possibly emitted by a flash circuit.

Old single use "confetti element" flash bulbs, interference was worse.

Shielding at the board might help, as long as the signal interrupted was wire(track) transferred.

Related- Incandescent light bulbs like to throw radio waves, and many light bands just as they are burning out. Florescent bulbs too. 

Alex, try to make it happen again.  This ought to be in DCS forum.

Never had DCS issue using a flash. All that is happening is discharging high voltage to the bulb.

I would assume a guest with a camera wouldn't have a lot of somewhat specialized gear on them, and this was an odd coincidence.  

 

That said, what were they using?  There are various wireless flash set-ups out there, and I have no idea what frequency they use.  It is feasible that the camera has a built in wireless flash setting that operates at a frequency that interfered, but I tend to doubt it as that signal would be so short as not to cause problems.  

 

As for the gates and such, there is of course the light from the flash it's self that can trip optical sensors, but many cameras also use very "bright" infrared lights for the range finder for auto-focus, and that may cause problems with infrared sensors.  

 The exact situation is not here except for a mention of sparks, which this almost is, but gives an idea how easy radio interference is to produce.  

 

http://www.haystack.mit.edu/ed...king_radio_waves.pdf

I am thinking that a cell phone may have created the radio wave. I can't recall the frequency that grj told me that dcs uses from base to track, but it's a higher frequency than TMCC.

 

Infrared frequencies from a focusing device on a camera would emit a very high frequency electro-magnetic wave at very, very low power.

 

cameras with bluetooth or wi-fi radios are at a higher frequency.

 

I would lean more towards an anomaly in the recording playback and a freak coincidence.

The DCS carrier is 3.27mhz, and cell phone operates on 800mhz and higher, depending on which band it's connecting on.  I don't see cell phones having any effect on the DCS signal.

 

Cell phone operating frequency bands: 800 MHz, 850 MHz, 1700 MHz, 1900 MHz, 2100 MHz, and 2500 MHz

 

GRJ, I wonder if that particular person's flash sent out spurious emissions on a frequency that could interfere with DCS.  Curiosity would compel me to invite the photographer for a return visit, with his/her camera.

Anything is possible I guess.

 

I know that DCS is not light sensitive, and many people have taken flash pictures of the trains before without incidence. But wondered if there could be RF signals generated as a result of a flash firing, such as when the capacitor discharges, and if it happens to be at just the right (or wrong) instant, interference could possibly occur.

 

I don’t know what type of camera he was using. Actually, I believe it was a smart phone, and he also took some videos of the trains.

 

I just had a minute to read your replies, and will run the Record/Playback session again later and see what happens, and will report back. I must say, though, that it has run perfectly countless times.

 

Thanks!

 

PS. I meant to post this thread where it is now, and not where I originally posted it, under “Hi Rail, 027 . . .” Thanks, Rich.

 

 

Hi Alex

The only flash system that I use that gives off a radio transmitted signal is a Pocket Wizard Plus X. It transmits ten channels at 344.0 MHz.

     This system is mounted to the top of my Canon SLR and sends a radio signal to a receiver on a remote flash,

The Sync speed is up to 1/250 second.

I have used this system for Model Railroad Photography and no interference, with MTH / DCS Remote.

The problem may never happen again.

No way to prove or disprove any theory if the interference is random.

 

My district manager used to like to call it electrical magic.

Any "chip", from a transistor to IC can also "forget" what to do, or do something unusual that defies all expectations, then performs fine the next cycle.

These quirks get overlooked, or obsessed over. Till you begin dealing with similar circuit behavior by the hundreds and then realize sometimes "if its not broken again, don't fix it". 

 

I think the likelihood of em interference taking on the exact waveform of the  dcs remotes all engines max speed command is pretty much zero.  The commands are complex and have CRCs.

 

Enough em interference to go through TIUs metal shielding and affect the electronics in such a way that they are not so scrambled that they still respond to emergency stop?  Also unlikely.

 

 

 

 

3.27 MHz would have that transmitting on 80 meters of the ham band. I thought from what I have read on here that DCS is around 900 mhz. The antenna to transmit at 3.27 MHz would be extremely long.

I recall 900 also.

Just some questions.  When is guest took his picture, was he close to the lead "A"?  Does this ABA set have screens along the sides?  Could it be possible the flash penetrated the shell through the screens and momentarily overwhelmed or blinded the tach reader.  I would think if the tach reader cannot read the stripes, it would accelerate the engine.  Just some thoughts.

 Just as possible I suppose, but reflection is more likely than penetration.

It should resume right away though, cutting power to adjust to the new load. Unless momentum, or one of the other features prevented this from happening by the averaging of opto input/power numbers they do.

  If an opto receiver suddenly fails on that loco, can that happen?

   Or will it "shut down" after a predetermined time, if it doesn't get a pulse signal once it tries to move with "full power"?

 

 

That should be reproducible with another camera & a good angle 

 

If that cameras flash dies soon, that would point an extra big finger at RW interference.

 

I don't think a cell would be as likely a candidate. That design would be far more thoroughly tested for radio wave interference than a camera. 

 

 

On DCS, if an engine is powered up with the DCS signal, then loses the signal, does it stop or does it run conventional, ex. full track power?  I don't know how the engines act in this circumstance, but if they do revert to a conventional operation when no signal is present, then whatever interference does not have to issue a command, but simply block the signal.  

 

(Edit:  upon thinking about it, I seem to recall that DCS engines keep doing whatever they were last doing if they lose signal, so scrap the above if that is the case.)

 

Also to note, if it was a smart phone, it would almost certainly only use 1.9GHz or above for cellular, and 2.4GHz for wireless internet... Neither of which should have any effect on the DCS signal.  I can't be certain that the flash isn't using some sort of pulse-width modulation to control brightness, but in the end it is just a very bright LED on a phone, and should not produce any tangible interference. (I'm sure it's there if you have sensitive enough equipment to measure, but it isn't going to effect a track signal.)

 

Honestly, I'd love for there to be something to the theory that the camera did it, but I just don't believe it to be so.  

 

 

It should continue with the last command received.

The signal I meant, was the one from the opto.

What if that stops off,or on? The "numbers" counted per second stops. That changes averages

 

This is a grasp at straws sure. It always is. But..

   At the gameroom there were approx. 200 lights of various types on every day. We had some radio(not much), and lots of optical devices for the lights to mess with. When a machine acted up that didn't usually, I would start trouble shooting by looking at the ceiling .

Radio glitches, I also looked outside for a cab, cop car, ambulance, semi-truck, or train. Most of those transmitters let out a ton of sweeping radio interference waves when they key the mic. At another job, with up to 200 cars in a lot, those also set off the better car alarms that sense sweeps, and in lower grade units might unlock, lock, or set off the panic mode in it. 

 

Actually, in a game room, cameras probably caused more "intermediate" opto issues than I had thought of before. I usually made myself scarce and put a "coin tech" on the floor before "party time" crowds stopped my work in its tracks.....Mister, mister, um, um, um, um, um, um, um............um I lost my coin.... no I lost three coins, and my brother put all his in till they wouldn't fit anymore, then we tried pounding them in.... then that kid poured his soda in it. Can you fix it before you leave its my favorite?

(Saying no to that is harder than it sounds. Lots of weekend OT if your not careful )     

I believe that 3.27MHz the track signal frequency. So, you do have a long antenna.

Mrmarcher,  that sounds viable.

Light will affect the tach reader if it is exposed.  It will also affect Lionel's IR drawbar.

 

Neither will run correctly in direct sunlight.

 

Dave

 

Remember that the 3.27 is not being broadcast in the traditional sense.  The wiring and track is serving as an RF transmission line, hence the talk about filters and magic light bulbs for DCS signal issues.  There should be very little radiated 3.27mhz signal, and any signal that would be able to affect the track signal would have to be pretty strong as the impedance of the transmission line (track) is pretty low.

On DCS, if an engine is powered up with the DCS signal, then loses the signal, does it stop or does it run conventional, ex. full track power?

Neither.  It continues to run at the last setting until the DCS signal is regained and a command is sent.

Now with that info from GRJ & RJR, it sounds like it might be more likely the light threw a feature "count" off, or it was "magic".

EMF can do strange things.  Just like a spark on a layout will start up DCS locos all over the layout, who can tell if the camera emitted some EMF, from just the right angle and distance, that just hit the DCS system in the right way to trigger some action.

 

Or there could have been something else that occurred unnoticed which triggered the result encountered.  Unless the issue can be re-created, we shan't know.

Thanks GRN

I did some searches last night and came across TMCC using 455 kHz signal. I did tune my HF radio down to 3.4 MHz and only received random clicks when operating the layout(radio is not general coverage receiver so couldn't tune to 3.27 mhz). I think I will try A dead key of 100 watts into the dummy load next to the layout and see what happens to the loco.

 

The flash going off when the engine is running I would think could blind the tach reader. It would be so bright that all it seen was white or black. Then it would think that it was not moving. The only way to see if this is the case is to remove the body from a loco and try it(more so with the lights off). Not trying on mine as the body is too much of a pain to take off.

 

I can't really imagine the flash having anything but a very transient effect on the speed control.  It's a momentary flash of light, and then normal operation resumes.  I'd be VERY surprised if anyone proves that the camera flash has any effect on the PS/2 operation, other than maybe a very brief hitch in the motor running.  I'd be pretty surprised if you can even see any effect.

 

To test that theory, I connected a PS/2 board to my test set and repeatedly fired my camera flash at close range at every angle at the tach strip and reader on the exposed motor.  There was never an observable change in the motor speed of any kind.

 

 

I had a engine sitting in TIU #4 's track  and was re-railing an engine on the turntable in TIU #1s zone. The engine in tiu 4s track took off.   I   repeated  the take off   a couple more times which leads me to  think that loose connections or shorts can cause runaways. Maybe even loose center rail rollers on passenger cars that then to wobble all over the place at times.. Yep in super mode.

I figured the unit "thought" too fast to notice a change. Does change in momentum setting have much effect on it when it is a "long count" on one or two stripes?

 If it cant be reproduced "the magic show" may be over. 

 

Shorts could be interpreted as a signal Gregg. Or modified it in route.

  They can in theory reprogram some chips too.

  Normally it only resets base parameters. Other times it will appear "dead" because it now functions differently. These types of IC's are not often used for consumer goods. But in general, that's one more reason to kill power with electronics vs the PW habit of letting the breaker trip if there are going to be lots of sparks. Just don't blame the electronics if that's "just the way it is" with you & your power. (I do it too)

Cameras and trains?  All we need is a big block Lincoln and all my interests will be involved...But, I'd have to lean toward the camera being incidental to whatever the problem was.  A standard camera flash can be measured in hundredths of a second, so is unlikely to do anything given how relatively slow the flywheel is spinning.  

 

On a phone the light may remain on for a few seconds, so it'd be worth repeating the test with both a very bright light, and an infrared light, as some phones may use infrared for the autofocus.  

The IR auto focus, that is a good one.

 

Camera Mystery Solved!!!

So this is why TCA does not allow cameras at York meets...

First of all, THANK YOU ALL for your replies! I read them all, and as we are well aware, this is further testament of how knowledgeable our friends in this forum are.

 

I ran the sequence four more times, and took a picture (once) with my Nikon camera when the ABA was in the same location as when the anomaly had occurred, and nothing happened; i.e., DCS continued to execute the remaining commands in the Playback, including the PFA, SSD, throw several turnouts back to neutral, etc.

 

As Adriatic suggested (above)

The problem may never happen again.

No way to prove or disprove any theory if the interference is random.

Hopefully, this will be the case.

 

About four years ago, when I was starting to lay the track and testing, one of the DCS engines that was 'parked' on a spur, took off on its own, full speed, and on to the partially finished main. My wife Judy and a friend were watching, and Judy exclaimed, "There goes Bush!" She was referring to the SD70ACe Union Pacific Cab 4141, which was the engine that had taken off. I shut the power off, and no harm was done.

 

This early on engine runaway engine experience, in addition to a couple of other occasions when derailments have caused some DCS locos to come alive (though not take off running) further convinced that my decision to be able to turn off siding, spurs, and even main line districts when not in use was a sound one. So I have installed toggles switches at the control panel, and relays close to the tracks, and can shut off any section I want.

 

Thanks again to all, and if by chance this happens again, I will let you know. (Hope it doesn't happen.)

 

Alex