47. Our RFID Beta readers are now reliable in managing our freight stock at McKinley Railway
ฝัง
- เผยแพร่เมื่อ 1 มิ.ย. 2024
- With the development of revised sensors we now have a reliable RFID system at McKinley Railway.
Here's how .............. - บันเทิง
I don't know whether this is totally brilliant or totally crazy!
Totally brilliant and a little crazy
A brilliant idea often comes from some crazy thinking.
CRAZY, now that he is blocking out the whole 20m (14MHz) band.
Unbelievable mad, but creatively and technologically brilliant. What a layout. It’s mind blowing; would love to see it operating.
So do we.. Its taking its time, but we are getting there. Thanks for your patience.
I am a ITS manager for DART rail in Dallas Texas. I've been reading and studying using RFID as initial in the rail yard. I reached an idea of accurate positions. I kind of getting little confused in following the video but you made me think deeper. I will keep following your videos to understand your idea.
Try UHF instead of HF. UHF may give you wider range of tags and better accuracy.
Hello Yaz,
Sorry for any confusion. Hopefully we can get something out that shows what we are up to. Are you thinking about full size railways or miniature ones such as ours?
@@dattouk
I am speaking about full size train and light rail.
As software developer and long time model railway enthusiast who has been working in the material handling world -- I have been waiting for something like this ... my experience with RFIDs has been at industrial scale -- so even the 5cmx5xcm readers were good news -- but the work you have done with Eccel is going to open a whole new world to model railway enthusiasts. I for one can't thank you enough for pushing the envelope. I am especially interested in the narrow between rail readers ... just brilliant! Thank you!
Hello Paul,
Glad you enjoyed our journey. Its been a bit protracted, but the perseverance has paid off.
Wow, you've already done it.
I have been imagining something very much like this for the past couple of weeks. I've been looking at the price and range and capabilities of the different RFID frequency ranges.
I should have known I wasn't the first person to think of trying this. It is great to see that it works pretty much exactly like I hoped it might. You are doing way more with it than I need in my automation dreams.
(I'm a programmer by profession)
Well Done David and his Team
What can I say I’m stunned I’m just getting my block detection working.
I have just looked at the eccel site this looks very interesting and affordable
Keep up the Great Work
Lovely Video work from Charlie Bishop
Good to see regular updates appearing again about the layout. Love your enthusiasm for all these "dark arts""...
Amazing detection and output of the rolling stock with RFIDs.
Hello John,
It was a delightful R&D surprise for us too. When the Eccel boys finally cracked it we had a big smile.
@@dattouk :-)
I’m not first this time.HO HUM! What a letdown. But not the content. I love this stuff. McKinley kicking goals as per usual. Thanks guys.
Looking great. It will be fantastic to start to see it running.
I am running out of patience too.
Enjoyed the video today .Hope there more to come
Glad you like it. There will be.
Love this! Keep up the fantastic work!
just fascinating stuff!
Hi Dave. That’s fantastic 👍🏻
Really interesting, and pleasing to hear that a company wanted to work with you to give you the solution your wanted, or one the surpasses expectation. Dave, you sounded like a kid who had just got his dream present on their birthday or for Christmas :)
Would love to see more in-depth videos like this as things get back to normal, ideally a tad longer (Charley, less editing :) ) - Keep up the good work guys
Good day David and team.
Well done to the team!
Fantastic job..... this is really cool idea, will you be making this into kit's, for others, who would like to do the same , with software? I would be keen to do this on my one day home layout......I I'm busy with the plan at the moment..
Please, keep doing these video's, it is really appreciated!
Many thanks, Dylan.
Great to see your breakthrough moments here. If your 8 channel multiplexer can handle a shunter at say 20mph then I suspect you will be able to handle 160mph on a dedicated single channel reader. RFID certainly the way to go as long as no interference from RFID stock running by on adjacent tracks. Optical on a layout like this would be prone to dirt blocking sensors (suspect you'll end up with a few hundred dotted around )- particularly as looking upwards to read QR or bar codes under rolling stock. Great stuff, relish the moment. Stephen
I would think the on board reader is a better solution. Placing tags on tracks are much cheaper and much more reliable because passive tags are very durable.
@@groovyc4344 Here its the other way round, they have the tags in the rolling stock (suspect 1000s of even allowing for one for each of their triplets) and the permanently powered static sensors located beneath the tracks. Using the multiplexer card they had Eccel develop I suspect has also saved them both installation time and costs in addition to the space savings/ real estate elaborated on here versus single channel units. Its certainly a ground breaking and technically challenging layout!!
@@stephenpike3147 You have the logic correct Stephen The readers are expensive and need quite a lot of power and space which isn't available in the wagons themselves. Its been an interesting journey just to get reliable reads. The big challenge is metalwork. It interferes with the signal more than track power does. I have no idea if it will work at a scale 160 MPH. I have nothing that will go that fast. We have found that certain MUX readers work better that the dedicated ones. Go figure! Anyway, we now have to test the hardware working with the software. I envisage a typical operations session of 3 hours will have about 900 reads in one station alone. If we can get the system to behave with say 3 or 4 hiccups for the operators to resolve I will be super happy.
regards
David
Interesting g to hear about your problems. If the rails are the problem interfering with the magnetic field have you tried using rails of a different composition. E,.g. relative permeability of iron circa 1000 versus brass circa 10, copper 1 at 50Hz (brass depends on composition rest I believe varies on frequency), a simple trial set up might help evaluate. Problem is track manufacturers do not declare composition of alloy used. Just wondering if this might be root of problem (I am not an RF or antenna man!). Stephen
I always enjoy your very informative videos but I didn't know that the 'Flying Banana' was around in McKinley's time period.
Well spotted. We felt it was important to have the real High Speed Test train as well as my Mk3 Lima coach set. It will be making guest appearances when nobody is looking. Just to check the cant on the rails you know.
Cool. Looking forward to see it in action once all the nodes (how many?) are up and running.
I think we will be using between 20 and 30 antennas per location depending on the yard size and complexity. When we get to showcase each one I will share the RFID location logic with everyone.
Looks really great, ultimately though the point selection of the sidings could determine whether a reader needs to be powered to read wagons as they pass over.
Hello Andrew,
You raise a valid idea that we did consider. We decided to put the sensors on each track rather than rely on the point logic to make the determination about the siding just for simplicity at this stage. It could be done and in time we might get there, but its a bit of a crawl-walk-run process at the moment.
David
Interesting!
thank you for share great detailed video on your RFID system
hope further give us the detail of what use there on the layout because maybe looking into this my operation in Australia
You may consider using UHF technology. It is offering a wider range of RFID tags and better accuracy in term of reading. Also it allows future technology expansion.
Will do
Hi David.
This just takes "playing trains" to a totally new level. It is surely what all those of us with large layouts have always wanted. For those of us not wanting to go to your levels of operation, but who would like to use this purely as a stock detection system in say a fiddle yard, is there any future commercial development to turn this into a stand alone system?
Great video as always (and I followed this one!!) for which thanks David,.
Best regards
Bob
Hello Bob,
we have been playing with ideas as to how RFID could be easily used for modellers without needing all the kit we have installed. We are working on a demo layout for Southampton 2022 show, but I am not sure if that is going ahead yet? More to come in time.
David
Fantastic stuff! I'm particularly a fan of the cargo registry for automatic unloading of appropriate goods, it's amazing what new ideas a "small" change can produce!
I'm interested though how the initial setup works for each wagon/loco - do you have to tell it where it started initially, or will it work it out once it's gone through a couple gates?
The system will know where stock is when it goes over a reader. However, its not quite that simple. We have to get the engine right in terms of pre-locating things as the software wouldn't know which locomotive to start from track 10 so to speak. Once the layout locations have been recorded by the RFID system, the two worlds (database and the layout) should stay in sync. The assignment of waybills is entirely automatic. At the moment, they are being generated on a random A to F destination basis to test the software.
Hi Dave.
What with Charlie's new electronic signalling system, and now your new rolling stock location system, I've started developing seizures and find myself staring into space for no apparent reason. I told my doctor I only wanted to play trains, then he insisted I put on this white jacket with sleeves that are tied around the back of me. What's going on?
Dear Paul,
Your comment had me in stitches. I promise the end result will be rewarding. A core mission for McKinley is that visiting railway clubs should be comfortable operating a part of the layout within 30 minutes of arriving for a day's operating session.
Most of the followers I have met blame me for their wallets looking really empty, but they have a huge smile on their faces. Take comfort that if they haven't committed me yet, they will leave you alone for another year at least.
David
Thanks Dave.
Judging by the odds, I'd say we are both safe from being committed. As my ol' Mum used to say: "There are more out than in!".
Could you provide the references for the various boards and antennas?
You mentioned it was tuned...
So is this special or off-the-shelf readers / tags? I'm curious, because I've been trying to use RFID tags myself, but I have been unable to read the tag fast enough to detect a running train.
They are off the shelf. Our results show they are excellent for HO/OO speeds of 65 MPH/105 KPH. Maybe faster, but that is the max speed we will operate our trains at.
Is there a published list of what you used and where you got it? Could you put parts and contact in the description?
Not as such. We will start to use the discussion tab soon to document that.
I hope that the freight software will be made publicly available or its all for naught! Bloody brilliant though.
I doubt anyone would want to use our system. It is very bespoke.
We are checking out how the RFID system will interface with the freight manager system in JMRI through.
David
@@dattouk I would I think, I'll have that many wagons when I'm done later and like you I'd really want to know where they all are as the "off scene" section wont be easily viewable.
Interesting stuff David! Your approach, with the latest updates and reworks obviously performs well and the database integration opens up a whole load of new modelling dimensions as you touched on. However, What are your thoughts on costs? I think my main reservation would be the cumulative cost of RFID on a layout. RFID costs have been a major barrier to the adoption of RFID tagging of 100% of goods in supermarkets and its use only on comparatively high value products. Aside from that reservation this is great work!
Vehicle location schemes are of course most valuable on large layouts where there may be multiple operators or where significant portions are not easily viewable. I have been investigating using micro-QR codes (Just printed on a general purpose Brother label printer and measuring no more than 7mm square) stuck on the bottom of vehicles, and using cheapy little QR code readers with serial outputs under the track. QR codes are massively more robust than 1D barcodes and can be read very fast. So far I have 2 readers talking to a single Arduino pro (also very tiny) to receive and marshal the serial data. The Arduino send all data back to a central control point (a PC at the moment) via serial cable (eventually it will be just a CAT5 Ethernet). As yet, I lack a layout to do proper testing on though, so no idea how reliable it would be with trains passing at speed and with many in operation at once. However just pushing stuff along the rails fairly fast it all works well so far. Now you have me wondering if I should go RFID instead!
Thanks again
Alan T.
Regarding cost: RFID tags are really cheap if you get them in bulk, like a cent per tag for 1000 tags
To be honest, whilst Dave is cautious with expenditure, I don't think they worry too much about the costs. Look at the project, there's probably tens of thousands of pounds invested in track and wagons alone. A few hundred on RIFD tags isn't much in the overall scheme of the project.
plus, it looked to me one RFID per 3 cars/wagons. Perhaps more reliable at speed.
@@goawaygoawaynow Before I investigated the micro-QR route I did play around with 1017 RFID readers and NFC tags, but the range was a problem and the stick-on micro NFC tags were nothing like as cheap as you instance. You can certainly buy the Mifare medallion tags for that kind of money, but they are too large for this work and - as Henry instances below - the range is not great for this purpose anyway. I have never found stick on ones of the right size for this application anything like that cheaply - more like £1 per tag last time I looked. Maybe I should look again now. :-)
I have been trying RFID with an Arduino but the standard mifare RFID doesn't work (can't get the tags close enough to the reader and metal weights in the wagons interfere). I like your idea of QR codes but I want to know if the reader uses IR or visible light to read the code.
Why not considering UHF technology. Also using a 2g/3g/4g on board UHF unit would be a cheaper solution. Putting RFID tags on the tracks are cheaper. With an on board unit, you will be able to expand the application such as speed of train, etc.
Hello groovy,
We are not RFID experts and Its taken us a long while to get this far. We don't need to monitor train speed as that is controlled by the software and it does a good job at it. Our real challenge is to identify where the wagons are as the primary goal is to manage freight movements.
I believe putting readers onto wagons would be extremely expensive and very challenging re power and space. The tags are cheap, but the readers not. Bear in mind we have 10 times more wagons that we have locations we want to record. So our orientation of the world works for us (i.e. tags on wagons and readers under the track). We looked for practical solutions provided by manufacturers that could be considered to be Off-the-shelf. Good luck with your ideas there.
One step past rivet counting
Thats what floats our boat.
Triplets will make rather odd-looking goods trains. Mixed units of 2, 3 or 4 would look better.
There's nothing to say that the three wagons must be the same, is there?
Funny enough, our triplets are length driven. So Many of the sets a doublets and some are just singlets. Seeing it in operation will be the test. The triplet count was driven mostly by shunting times and trying to have trains of reasonable length.
David
Are you going to be selling this stuff?
Nope... Golden rule. It has to stay as our hobby. You can buy it yourself though.
Very interesting and exciting! Following your progress closely.
Do you still have the sam RFID-tags you showed two years ago or have you found any updates to the tags?
th-cam.com/video/xcWixqpF44A/w-d-xo.html
I'm looking for tags that would fit under an N-scale waggon.
Thanks for at great channel!
You may try UHF PCB metal tag that uses screw to mount.
@@groovyc4344 Thanks!
Do you have any suggestions on brand and model or where to buy them?
@@alexanderehn3729 try this
sid-global.com/rfid-card-readers/rfid-tags/metal-tags/
Model: 79x20x3
@@alexanderehn3729 they may send free samples
Thank you very much @@groovyc4344! Much appriciated. I'll check them out.