In this video I configure some infrared proximity detectors as feedback in improve the accuracy of the stopping positions within blocks on my n-gauge model railway.
Nice to see an update from you😁. Great progress with the opto sensors they are certainly going to do the trick, and make the sidings more accurate. Look forward to seeing your future progress all the best and a Happy New Year Tony.
another great video phillip you got some interesting comments maybe that person should spend time and do his own video on this subject , look forward to your next video , ps i do remember you use traincontroller software $$$$$$ have you purchased itrain if not there is a software called rocrail you just make a small donation / year ...................tony
Thanks for your encouragement. In the end I bought iTrain rather than train controller. As for 'interesting' comments - they all help with the TH-cam algorithm, regardless of what they say :)
I wonder if a metallic sequin glued on to the underside of the Loco would provide a simpler and more consistant reflective surface ... I'm greatly enjoying the series as I'm considering an N guage layout ... ny very first foray into railway modelling. I have very limited space ... I may not be able to avoid 1st radius curves.
Glad to hear you have found my efforts helpful. Small layout is always the best way to start (did not do this myself!), in order to learn the stuff you can only learn on the job. Short radius is fine, if you are willing to accept the restrictions. Most rolling stock can go around on it, however: 1) the over-hang at front and back is larger which can cause problems with clearance and coupling. 2) there is greater resistance, so while a loco might make it around the curve, it can not pull a long train around it. I am reworking my layout plan to remove all 1st radius if I can. Live and learn.
If you decide you need your IR detection, rather than painting the underside white could you use a white sticky label cut to size which would be more easily removed?
Hi. Got there eventually. Does this mean the first optical sensor is now redundant in its original position? You seemed to be "fixed" on the 10cm distance from the end of the block. Is this some arbitrary distance, or does it ensure that a loco/train stopping there doesn't block another loco/train crossing through the same point? Hope you had a great Xmas and all best wishes for the New Year.
Both optical sensors are actually redundant because if iTrain was configured correctly the train would have stopped in the correct position without them :)
I have not done any ballasting yet, but I expect to have to ballast around them. So far all the places I have used them are not scenic anyway, so perhaps I will use current sensing in the scenic sections.
@@naivegauge Ah! ok, I have followed you chanel since you started, also use digikeijs and n-scale peco. Havent anything set up as you with blockdetection or sensors etc. But thinking of doing it some day. Im no way near the size of you layout, but mine is done so it will be easy to expand the day I get a room free from one of the kids that still lives at home. :D
There are unfortunately several errors in this video that contribute to the initial "inaccurate stopping" of the train, and correction of these errors would mean that the optical sensors are not required, here or anywhere else on the layout, as trains would stop very accurately. The first is that after remeasuring the block length was correct but NOT the feedback length, which shows later as the feedback is 1.3 cm longer than the block. The biggest failing is that the reaction delay of the loco has not been measured or entered and remains at the default setting of 200 ms, which is why the train is not stopping correctly. Setting this value accurately will get the train to stop with a couple of millimetres off the desired position. It is also possible that speed has not been measured correctly. As the block is a 'long' block and accurate stopping is wanted, the easiest way to achieve this (after correction of the errors mentioned) would have been to create a second occupancy feedback 30 - 40 cm long before the turnout and entering this into the block. The train would then stop 10 cm (or whatever was wanted) before the end of the block without any additional effort because the position would be reset upon entry to the feedback. I hope this helps people with getting the best performance from iTrain, and saves the video producer from unnecessary expense and complexity that isn't required :)
You are quite right Iain - a big question that needs to be asked, is whether all this is necessary / helping. Part of it is me learning by experimentation, and I make videos to document the process, either good or bad, and I am pleased to have got to the stage where I can use an IR proximity detector to add an extra feedback. But of course, it needs to be considered if that is a useful tool to have in the toolbox, and I am planning to make a video to consider just that point. However, I can imagine some situations in which I might want to add an IR detector - perhaps near to an un-coupler electromagnet for example. I also noticed the discrepancy between the block length and the feedback length while video editing, but I was not motivated to rerecord it. Full marks for observation though. Calibrating the loco will also be the subject for a future video. I have done it, however I do not yet have a suitable stretch of track that is all on the level, so I suspect there there is a slight error in my calibration. Thanks for you interest.
@@naivegauge what is vitally important is that you do not demonstrate anything that is incorrect and that provides inaccurate information to people considering to purchase or use ITrain. Unfortunately your video gives the very clear impression that ITrain cannot stop trains accurately without additional help which is completely wrong.
I understand what you say. However, making itrain look good is not vitally important to me - I am just documenting my progress. Learning how to get the most out of itrain is all part of my journey, and as I learn from my mistakes I will make videos about those mistakes and the lessons learned. I am looking forward to getting the most out of the software. However, to be fair, this video does not make itrain look bad. Indeed - it is a feature of itrain supporting different types of feedback that makes it possible for me to do what I am doing.
Great to see you making progress. Very interesting, looking forward to further updates. Happy New Year.
Nice one hope to see more in the new year. Thanks for sharing.
Nice to see an update from you😁. Great progress with the opto sensors they are certainly going to do the trick, and make the sidings more accurate. Look forward to seeing your future progress all the best and a Happy New Year Tony.
Thanks Tony. It has been a struggle to get back to the hobby, but I am pleased to have finally pushed through and made some progress.
If ITrain had been correctly configured then the train would have stopped very accurately with the need for any additional optical sensors.
brilliant video bud. keep it up
Thanks, will do!
That will do the trick :) Job well done!
This is cool
thanks
another great video phillip you got some interesting comments maybe that person should spend time and do his own video on this subject , look forward to your next video , ps i do remember you use traincontroller software $$$$$$ have you purchased itrain if not there is a software called rocrail you just make a small donation / year ...................tony
Thanks for your encouragement. In the end I bought iTrain rather than train controller.
As for 'interesting' comments - they all help with the TH-cam algorithm, regardless of what they say :)
I wonder if a metallic sequin glued on to the underside of the Loco would provide a simpler and more consistant reflective surface ...
I'm greatly enjoying the series as I'm considering an N guage layout ... ny very first foray into railway modelling.
I have very limited space ... I may not be able to avoid 1st radius curves.
Glad to hear you have found my efforts helpful. Small layout is always the best way to start (did not do this myself!), in order to learn the stuff you can only learn on the job.
Short radius is fine, if you are willing to accept the restrictions. Most rolling stock can go around on it, however:
1) the over-hang at front and back is larger which can cause problems with clearance and coupling.
2) there is greater resistance, so while a loco might make it around the curve, it can not pull a long train around it.
I am reworking my layout plan to remove all 1st radius if I can. Live and learn.
If you decide you need your IR detection, rather than painting the underside white could you use a white sticky label cut to size which would be more easily removed?
alu tape
Hi. Got there eventually. Does this mean the first optical sensor is now redundant in its original position? You seemed to be "fixed" on the 10cm distance from the end of the block. Is this some arbitrary distance, or does it ensure that a loco/train stopping there doesn't block another loco/train crossing through the same point? Hope you had a great Xmas and all best wishes for the New Year.
Both optical sensors are actually redundant because if iTrain was configured correctly the train would have stopped in the correct position without them :)
alu tape works best , do you get troubled with sunlight ? next to it all , do you get the same behavior with a very different (make) train ?
Aluminium tape is an excellent idea.
How will these opto sensors work when ballasting the tracks?
I have not done any ballasting yet, but I expect to have to ballast around them. So far all the places I have used them are not scenic anyway, so perhaps I will use current sensing in the scenic sections.
@@naivegauge Ah! ok, I have followed you chanel since you started, also use digikeijs and n-scale peco. Havent anything set up as you with blockdetection or sensors etc. But thinking of doing it some day.
Im no way near the size of you layout, but mine is done so it will be easy to expand the day I get a room free from one of the kids that still lives at home. :D
There are unfortunately several errors in this video that contribute to the initial "inaccurate stopping" of the train, and correction of these errors would mean that the optical sensors are not required, here or anywhere else on the layout, as trains would stop very accurately.
The first is that after remeasuring the block length was correct but NOT the feedback length, which shows later as the feedback is 1.3 cm longer than the block. The biggest failing is that the reaction delay of the loco has not been measured or entered and remains at the default setting of 200 ms, which is why the train is not stopping correctly. Setting this value accurately will get the train to stop with a couple of millimetres off the desired position. It is also possible that speed has not been measured correctly. As the block is a 'long' block and accurate stopping is wanted, the easiest way to achieve this (after correction of the errors mentioned) would have been to create a second occupancy feedback 30 - 40 cm long before the turnout and entering this into the block. The train would then stop 10 cm (or whatever was wanted) before the end of the block without any additional effort because the position would be reset upon entry to the feedback.
I hope this helps people with getting the best performance from iTrain, and saves the video producer from unnecessary expense and complexity that isn't required :)
You are quite right Iain - a big question that needs to be asked, is whether all this is necessary / helping.
Part of it is me learning by experimentation, and I make videos to document the process, either good or bad, and I am pleased to have got to the stage where I can use an IR proximity detector to add an extra feedback. But of course, it needs to be considered if that is a useful tool to have in the toolbox, and I am planning to make a video to consider just that point.
However, I can imagine some situations in which I might want to add an IR detector - perhaps near to an un-coupler electromagnet for example.
I also noticed the discrepancy between the block length and the feedback length while video editing, but I was not motivated to rerecord it. Full marks for observation though.
Calibrating the loco will also be the subject for a future video. I have done it, however I do not yet have a suitable stretch of track that is all on the level, so I suspect there there is a slight error in my calibration.
Thanks for you interest.
@@naivegauge what is vitally important is that you do not demonstrate anything that is incorrect and that provides inaccurate information to people considering to purchase or use ITrain.
Unfortunately your video gives the very clear impression that ITrain cannot stop trains accurately without additional help which is completely wrong.
I understand what you say. However, making itrain look good is not vitally important to me - I am just documenting my progress.
Learning how to get the most out of itrain is all part of my journey, and as I learn from my mistakes I will make videos about those mistakes and the lessons learned. I am looking forward to getting the most out of the software.
However, to be fair, this video does not make itrain look bad. Indeed - it is a feature of itrain supporting different types of feedback that makes it possible for me to do what I am doing.