After many requests, I am proud to introduce a tumble down signal controller with reflective infrared detection. The infrared detection is stable for day / night operation and emits no visible light. The detector fits under the track for a truely discrete detection point. It operates with all SMD Products signals as well as SMBK signals. The operation is stand alone and one module is able to control the aspect of two signals. Multiple modules can be added for tumble down operation All the connections are RJ-11 for a truely plug and play installation for the beginner in mind. Product Page
Question I am interested in your new detectors and signalling system. I do NOT know a lot about electronics, but I want to signal my new layout when I build it this winter. Do you think I would be able to use your system without understanding the electrical background of it? Answer Absolutely. It has been designed for the beginner. Non electrical / electronic person who would like signalling. Let me give you an overview on how I tested for the beginner: I found a person who knows nothing about the system... Not even about trains. I gave him the material for 3 blocks and asked him to hook it up. He didn't even know what it was he was suppose to hook up. He followed the color coding (not depicted on the web site) and successfully wired 3 blocks.... without further instructions. I am preparing pictures and installation instructions for the system on a large layout: Serge from SMD products - he builds the signals.
Shameless self promotion, especially the Q&A post. That being said, I will be adding signals to my layout this winter and your product seems to fit the bill quite nicely.
Actually, the question came from an email I sent Allain, and he posted it for all to see the answer to.
I don't want the etiquette of being over ambitious about the promotion of a product. I have developed the board in May and June. I have waited until this time for posting. at TX pointed out, it was a real question and I thought that the answer would benefit other members reading this thread. I am excited about the product and will have many other N-scale products released in automn. The forums, mailing lists, Newsgroups are the perfect venue for a beginning manufacturer of new products to let people know there is life outside the big boxes. It is impossible for me to take out an add in model railroader. I did request the opinion on the board a few days ago for the release of new products. No one had objections. In fact I was encouraged to do so. Having gone through due process to not offend anyone, I hope that I will be able to answer all your questions. So far, the response has been great. I believe that most contemplated the idea of signals but thought that they were too complicated to operate. I believe that this is the solution to that problem.
It will be some time before I have a layout up and running. Having said that, I've been planning to incorporate signalling in the layout, and have been checking out different systems. I'll be seriously considering this system!
What about automatic operation of a double track level crossing? I need a IR detector to switch on the crossing for both directions then two more to swtich the crossing off?
The system works with Searchlights from SMD Products .... I am a supplier for the lights Serge uses. Those bicolor LEDs are common anode. It will not work with reversed bias LEDs where one polarity is red and the other is green. The obvious reason is that too many options make a product difficult to setup and operate (in general) The goal of this product is to keep things simple while being able to combine detection and signal control. One package with no need for an external computer.
Obviously the operation for crossings is not the same as for tumble down. A variation of this system will be released with crossings in mind. Detection on both sides for single track should turn on the crossing lights and an adequate delay should turn off the crossing lights. To add another track to the crossing, it should be a simple matter of adding two more detectors. UP to 5 IR detectors can be added per board. ALSO, Serge at SMD products has plans to release a crossing light this fall / winter.
Detection on both sides for single track should turn on the crossing lights and an adequate delay should turn off the crossing lights. To add another track to the crossing, it should be a simple matter of adding two more detectors. UP to 5 IR detectors can be added per board. </font>[/QUOTE]Prototype street crossing protection is designed to work with trains moving at different speeds. Using only two sensors simply doesn't look right. Minimally, for each direction there are two sensor sections (4 sensors total). These all operate on what is basically the same principal as the current sensing method of train detection many of us use for our model railroads. For each direction, there is an initial sensor section starting some distance away from the crossing (how far depends on what the speed limit on the track section is). This sensor turns power to the protection circuit on, and triggers the signalling devices. At some crossings, stopping in this section will turn the crossing lights off, and they will automatically turn back on when the train starts rolling again. There is a second sensor section for shutting the signal off. This one detects when the last car exits the roadway. This sensor ends a fixed distance from the road surface, on the opposite side of the roadway was the start sensor. This one is far enough away from the road that the last car in the train will be clear of the crossing when it is triggered, but close enough to the crossing that the signal turns off almost as soon as the roadway is clear. If you only use two sensors to do the job, you loose either advanced warning to motorists before the train gets to the road surface, or, there is an added delay to motorists after the road surface is clear. Paul
ah, thank you for the details. Obviously a more challenging task. Nevertheless, I think it is possible to make without much difficulty. The problem becomes why are off the shelf components not doing this for crossings already? I was under the impression that an inexpensive sollution existed.
Wait, it gets even better. On the prototype, you want a constant timeout period from the point where the indicators start flashing and the gates drop to the point where the train gets to the crossing. A simple timer would work, IF you guaranteed that the train would only operate at one speed through that section. If the train is going faster, you want to turn on earlier, and vice versa for a slower train. I saw a circuit to do this, IIRC it was a couple of integrators. You need two detectors before the crossing to get a time/speed reading, then you set the timeout delay from that. Clear as mud? A microcontroller would make this task really simple, but the circuit I saw was all op-amps.
OK, to get the most accurate would require 2 approach sensors and one end of train sensor on either direction 6 sensors at all. still able to do that but the main question now is are there lots of people who require this level of realism. How about other scenarios?
Allan, I really like your products and would love to use them on my railroad but, honestly, they are priced well beyond my budget. Take this product for example. It is priced at $100 and controls two signals. I am thinking that I would use about 12 signals on my 16' x 7' layout. For the signal controllers, I would need 6. That's $600 and I haven't even bought the signals or power supply yet (add another $300). I know that each person has their own price level and perceived value for a product but, your products seem a little out of reach for the average modeler. I am not trying to tell you what to price your products at, but I think you are losing a few potential sales.
I had actually left out the speed dependent potion of the sensors, because you need to have those further away from the crossing than I have room for. I think there is only one thing we left out. What if the train stops between the start sensors and the crossing? Some crossings will continue to flash, until the equipment is clear of the section of track protecting the grade crossing Other crossings will return to an idle state until the train starts moving again. I beleive this is actually triggered by the track circuit, though it could be triggered by either a crew member going to the electrical cabinet, or the dispatcher remotely. Also, the grade cossing itself may need to be triggered by an arbitrary number of tracks. If any of these tracks is occupied, the crossing should be flashing. Paul
I think there is only one thing we left out. What if the train stops between the start sensors and the crossing? Some crossings will continue to flash, until the equipment is clear of the section of track protecting the grade crossing Other crossings will return to an idle state until the train starts moving again. I beleive this is actually triggered by the track circuit, though it could be triggered by either a crew member going to the electrical cabinet, or the dispatcher remotely. Also, the grade cossing itself may need to be triggered by an arbitrary number of tracks. If any of these tracks is occupied, the crossing should be flashing. Paul </font>[/QUOTE]I agree on the extra space issue. The real thing has almost infinite space to add sensing equipment, but we aren't so lucky. However, with a fast enough integrator or micro, you wouldn't need more than an inch or two of separation to get a speed reading, but you'd still have to put the sensor pair pretty far from the crossing. When you say the signals you've seen, is that prototype or model signal modules? I have to believe that the proto will adapt the signal behavior to the location. The timeout method would be fairly easy, but getting the signal going again when the train resumes adds yet another sensor, most likely. This gets complex quickly, doesn't it?
I understand that things can get quite expensive. The reality is that signals and electronics are made in small production run and made in Canada. I have made a starting run of 300 boards. So you see, quantities are well below mass production. I don't think I will be able to recover the design, programming and tooling for a while. Already in the first run, there are alterations to be made for the second run. There goes my tooling. I'm not trying to make anyone cry for me but I'm simply saying that developing a new product needs to be worth while in the $ sense in order to develop more products and be able to offer more cool stuff. The SMD products signals which I sell do not compare in price to digitrax SMBK... $5.00 each. But at least they look like signals. I guess what I'm saying is You get what you pay for. I hear similar stories from various users stating that things seem to be extremely expensive - locos - track - kits - rolling stock - hell I haven't seen inexpensive stuff in N-Scale except inexpensive crap. I know a person that paid over $5.00 for cut straws on ebay when he thought he was buying aqua pipe loads.
