Jump to content
 

Has anyone practical experience on using IRDOTS?


TEAMYAKIMA
 Share

Recommended Posts

Hello

 

I am seeking serious help on using IRDOT's. I have been trying to use them for more than 3 years and have had nothing but problems. I have spoken at great length to Clive Heathcote and he has been very helpful. I thought I had finally got them working as expected two days ago but today they are giving problems again.

 

So, I do not want to bother Clive (yet) again but would rather discuss my problems with experts in this area who have practical experience of using IRDOTS because TBH the problems I'm having seem to make no sense whatsoever.

 

I think that a PM and then telephone chat would be better than a long drawn out exchange on RMweb if that is possible please.

 

Thank you

Link to post
Share on other sites

19 minutes ago, smokebox said:

But discussing it in the open on the forum may help others who are having problems or are considering using irdots and you may get an answer to your problem from someone who has never even used them.

 

OK, fair point. 

 

Let me say at the beginning that I have nothing but respect for Clive Heathcote and do not want this thread to be seen as a criticism of him or his company.

 

I am using a series of IRDOTS to control a level crossing and there have been several issues over the years and with Clive's help I have gradually eliminated all the possible causes of the problems. This week I finally got the level crossing to work perfectly.

 

There are three IRDOTS on a single line track; one a few inches either side of the level crossing and one a further 15ft away. This set up is unique as far as I know and is definitely NOT the cause of the problem. All week the two IRDOTS either side of the level crossing worked as expected. Today I went into my shed and a problem arose.

 

The problem is that the two IRDOTS are 'ON' continuously as long as my shed LED lighting is on. If I turn the lights off the IRDOTS work perfectly.

 

I have asked Clive in the past whether exhibition hall lighting would create problems - sending down infra-red light to confuse the IRDOT.  He replied, "No, because I switch the infra-red on and off at a high frequency which allows low frequency infra red to be filtered out."

 

I use LED tubes to light my shed. There are three tubes down the length of the shed. The third IRDOT 15ft away from the other two does not behave like the other two. The two that are misbehaving were not misbehaving the last time I test them and other than slightly altering ( 1ft at most) their position under the tubes nothing has changed.

 

Q. 1    Do LED tubes emit a frequency of infra red which IRDOTS cannot cope with? If so why did these two IRDOTS not misbehave before?

 

Q. 2    Assuming that it is the LED tubes emitting infra red which is the problem could it be that the slight change in position has changed the angle of attack of the rogue infra red which explains the change?

 

Q. 3   Is all of this accedemic as exhibition halls do not use LED tubes?

 

I am totally non-technical and so I hope the above makes sense.

Link to post
Share on other sites

Q1. Depending on the "drivers" used for the LED's some LED lighting actually switched on & off at a very high frequency - far too high for the human eye to "see". However, some devices react to LED lighting - you can see the flickering effect on some dasgcams when a vehicle coming towards you had LED Daylight Driving Lamps - the image on the camera "flickers" but you eye see "steady".

Q2. I have found IRDOT's unreliable in the past due to changing lighting conditions - thje only place I found them reliable was in tunnels.

Q3. nearly everywhere is changing to LED lighting so back to Q1.

Link to post
Share on other sites

15 minutes ago, SamThomas said:

Q1. Depending on the "drivers" used for the LED's some LED lighting actually switched on & off at a very high frequency - far too high for the human eye to "see". However, some devices react to LED lighting - you can see the flickering effect on some dasgcams when a vehicle coming towards you had LED Daylight Driving Lamps - the image on the camera "flickers" but you eye see "steady".

Q2. I have found IRDOT's unreliable in the past due to changing lighting conditions - thje only place I found them reliable was in tunnels.

Q3. nearly everywhere is changing to LED lighting so back to Q1.

 

Many thanks for that.

 

I conclude from this that the angle which the infra red approaches the IRDOT is critical ie if the LED light source is directly above the IRDOT there is always a problem. If the infra red approaches at an angle then things may be OK. Therefore, every exhibition hall will be an 'unknown' until we actually set up. 

 

I cannot use reed switches and magnets as I use them for another purpose for trains in one direction only and so Q. 4 is .... Is there a better IRDOT style unit which is less prone to this problem?

 

 

Link to post
Share on other sites

Can you place the IRDOT receiver into some form of cover, with a hole on the side, such that the receiver can still view the sender light, but little of any other ambient light can interfere?  If a tunnel, with two largish entrances, will allow them to work, then it suggests that a lineside hut, signal box, vehicle, ubiquitous 20ft container, mound of rubble, coal, or similar might well be placed to shade the receiver.  It would seem worth making a small card box, with a hole might make for a cheap trial run or two.

 

 

Link to post
Share on other sites

9 minutes ago, jcredfer said:

Can you place the IRDOT receiver into some form of cover, with a hole on the side, such that the receiver can still view the sender light, but little of any other ambient light can interfere?  If a tunnel, with two largish entrances, will allow them to work, then it suggests that a lineside hut, signal box, vehicle, ubiquitous 20ft container, mound of rubble, coal, or similar might well be placed to shade the receiver.  It would seem worth making a small card box, with a hole might make for a cheap trial run or two.

 

 

 

I think that you have misunderstood the set up. The IRDOTS are set in the track and face upwards and bounce off the underside to the vehicles. I know that some people mount them on their side and then the beam across the track, but this is not possible on my layout.

  • Informative/Useful 1
Link to post
Share on other sites

2 minutes ago, Ray H said:

Have you tried lowering/raising them slightly? Are both the sender and the receiver encased in heat shrink or similar?

 

Yes, both encased in the original long black tubing.

 

As far as I can judge the only problem is if the LED source is directly above the IRDOT and so that may not be the case 95% of the time in exhibition halls. However I am currently making a shield to fit to the pelmet if the need arises - photo in a moment.

Link to post
Share on other sites

1 hour ago, TEAMYAKIMA said:

 

I cannot use reed switches and magnets as I use them for another purpose for trains in one direction only and so Q. 4 is .... Is there a better IRDOT style unit which is less prone to this problem?

 

 

Could you use reeds/magnets mounted different to the ones you have in use now ?

 

e.g. if your existing magnets are centrally mounted on the locomotives/rolling stock could you not place the extra magnet to one side ?

 

There are a number of current sensing units on the market but these all seem to require an isolated section of track for analogue or digital.

 

Or, the really "old school" way - a very short (10mm) isolated section of rail & a relay.

Link to post
Share on other sites

  • RMweb Premium
1 hour ago, TEAMYAKIMA said:

 

Yes, both encased in the original long black tubing.

 

As far as I can judge the only problem is if the LED source is directly above the IRDOT and so that may not be the case 95% of the time in exhibition halls. However I am currently making a shield to fit to the pelmet if the need arises - photo in a moment.

 

I seem to recall our club layout had problems from day light. Our IRDOTs were in the "four foot" almost opposite a window that was shuttered on all but sunny afternoons and that's when we had the problems (which we never resolved).

 

Have you looked at using Light Dependant Resistors instead?

Link to post
Share on other sites

  • RMweb Gold

I used 4 of Heathcotes IRDOTs on hidden (ie in a tunnel) loops linked to LEDs, so that I could ‘see’ train occupancy. They worked perfectly, but of course there was little ambient light, and no direct light from above. In any case, the room was lit by a three conventional light bulbs, so no fluorescent or LED interference. These IRDOTs were mounted below the track between sleepers.

 

This set up was on an earlier layout no longer in existence, but I’m now using them for a similar purpose on a different layout, but (because I initially forgot about them) this time they are installed above the board, mounted sideways at the side of the track, with a screening board sited opposite them, so each is not affected by train movement on the parallel track. I have found them to be similarly effective; incidentally there is LED strip lighting to illuminate this area which is often left on whilst operating, and again, I haven’t noticed any failure due to this.

 

Edit - I also have a couple of the DCC Concepts legacy sensors for a similar purpose. Have you looked at those? They do need an isolated section, which many not be possible if your track is already in situ, and can be used to drive various things (mine simply light an LED).

Edited by ITG
Added comment
Link to post
Share on other sites

OK, I have decided that I cannot take the (very small) risk of this issue causing big problems at an exhibition. So, have taken the decision to shield the IRDOT's from any stray infra-red.  The shields will be made from foamboard and attached by velcro. Here is a test using mdf. 

 

IMG_20210822_182815.jpg.03f374344ffb54347489b6b8e37ccc1b.jpg

  • Like 1
Link to post
Share on other sites

Hi

 

The technique that I have used is to both bury the IRDOT LEDs deep, and position them apart at an angle so that they collect light reflected off the bottom of stock better.

 

As I see it there are two conflicting problems to deal with on the standard set up of the diodes together in a single hole:-

  1. Trying to get light reflected at 90 degrees with the diodes parallel means that you don't get much reflected IR so you can't limit the sensitivity of the receiver with a long tube - you need all the reflected light you can get to make it work.
  2. The receiver is pointing directly at interfering light sources on the ceiling etc.

Positioning the diodes in two angled holes means that you get a lot more reflected light, and you can therefore bury the receiver down a long tube to minimise the stray light pickup.

 

Last time I played with IRDOTs and had a problem it was with fluorescent tubes which have all sorts of flickering modes that can be too subtle to notice. Regardless of what Heathcote says pointing the receiver at a very bright light source will cause problems with false triggering. The IRDOT is able to filter out the steady light and responds well to very low levels of flickering at the right frequency!

Link to post
Share on other sites

You will need to shield the entire roof I think, then the model will be too dark, therefore you will need to put in lighting for people to be able to see it properly - and AFAIK new lighting these days is only LED which means you end up with the same problem?

Link to post
Share on other sites

2 hours ago, WIMorrison said:

You will need to shield the entire roof I think, then the model will be too dark, therefore you will need to put in lighting for people to be able to see it properly - and AFAIK new lighting these days is only LED which means you end up with the same problem?

 

That's an interesting point which needed checking as I do use LED lighting strips behind the pelmet. However, it appears to me that it is only light coming directly downwards which is a problem and so according to my theory it only needs screening directly above the IRDOT.

 

So, I put to the test and I can confirm that my LED lighting does not trigger the LED's. All of this will be put to the test at our first exhibition at Tolworth in November.

 

Thanks to you all for your input.

 

Paul

Link to post
Share on other sites

Having just seen this thread, I'd agree with the comments on background lighting. I used a Logic Rail US grade crossing module with two pairs of IR detectors on a Freemo module and was forever adjusting the sensitivity to compensate for ambient lighting. So much so that I ended up building a nighttime layout so it would only work in the dark. 

 

I'm currently planning fiddle yard detection to automate a future project, and would either have to shield the detectors more closely or choose a different system.

Link to post
Share on other sites

On 23/08/2021 at 14:15, TEAMYAKIMA said:

 

That's an interesting point which needed checking as I do use LED lighting strips behind the pelmet. However, it appears to me that it is only light coming directly downwards which is a problem and so according to my theory it only needs screening directly above the IRDOT.

 

So, I put to the test and I can confirm that my LED lighting does not trigger the LED's. All of this will be put to the test at our first exhibition at Tolworth in November.

 

Thanks to you all for your input.

 

Paul

 

I am not sure how many sensors you have to protect, but could the overhead screens might be reduced in size and placed on the layout directly over each receiver, footbridge, gantry {with horizontal floor}, over-track Signal Box, part made Network Rail {other era} temporary works with scaffolding over the tracks, Crane Jib, model plane, hot air balloon, hung from the Pelmet, err- clouds{?}, or perhaps a transparent diffusing, cloth / film across the layout at Pelmet height?

 

  • Like 1
Link to post
Share on other sites

9 hours ago, jcredfer said:

 

I am not sure how many sensors you have to protect, but could the overhead screens might be reduced in size and placed on the layout directly over each receiver, footbridge, gantry {with horizontal floor}, over-track Signal Box, part made Network Rail {other era} temporary works with scaffolding over the tracks, Crane Jib, model plane, hot air balloon, hung from the Pelmet, err- clouds{?}, or perhaps a transparent diffusing, cloth / film across the layout at Pelmet height?

 

 

There are three sensors on the scenic section and I whilst I accept that those are good suggestions I regret that they cannot be used on my layout - the scenic section is finished. The sensors are up to 12 inches from the line of the pelmet and so I will use foam board attached to the pelmet to protect them. It already takes four people 3/4 hours to set the layout up at shows and I don't want to add to that time if I can avoid it.

Link to post
Share on other sites

I have given up on IR sensors because of their un-reliability and unpredictability and have gone to hall sensors.  However, since you dont want to use magnets so they are not appropriate although it may be possible to do something since Hall sensors are pole sensitive..   I also identified a problem with overhead lighting and the problem did improve when I went to a monocromatic LED.  Did not completely go away though.  Fluorescent lights can be a problem in that unless they are really good quality the temperature can change over relatively short time periods.    It might be worth purchasing some good quality LED monochromatic 'fluorescent' strips.   A friend is using IR-Dots with some success, but  he put a 'tunnel i.e. a black straw' over each LED and adjusted the height so that external light was minimized but it was a pain.   If you are prepared to introduce an Arduino it might be possible to use a radar proximity sensor. 

Link to post
Share on other sites

  • RMweb Gold

Just out of curiosity, are the infra red emitters in IRDOTs driven from plain DC - or are they pulsed?

 

This might sound like a crazy question - but, about 20 years back, one issue of Elektor included a design for a R/C car speedo.

 

I'm not sure if I've still got the circuit - but it was based on bouncing an IR beam off a reflective pulse painted onto a tyre. Its designer made a point of running the IR emitter using a pulsed signal - then putting the detector output through a homebrew band filter, before smoothing out the pulses - all to "knock out" ambient light.

 

Most of the circuit used was based around a counter using CMOS logic - but the output was passed through a suitably biased transistor, to allow it to drive an electronic bicycle speedo.

 

I'm just wondering if a pulsed supply / filter combo might work here for taking ambient light out of the equation. Otherwise, I don't know what to suggest.

Link to post
Share on other sites

5 hours ago, Theakerr said:

I have given up on IR sensors because of their un-reliability and unpredictability and have gone to hall sensors.  However, since you don't want to use magnets so they are not appropriate although it may be possible to do something since Hall sensors are pole sensitive..   

 

That's very interesting. I've never heard of hall sensors.

 

You say that they are pole sensitive - can you get two types, + sensitive and - sensitive? That might do the trick for me.

 

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...