I have a friend who is a huge advocate of current detection and has resistors on one of the two axles of a truck for all cars except the cabin (caboose) car. For the cabin car, all 4 axles get a resistor. He has used 15k ohm for the resistors on all resistor axles. Locomotives have their own current draw as they have decoders. His system is the NCE on his layout. We are now converting the area's club layout to resistor axles with the same structure - 2 of the 4 axles get the resistor and the cabin cars gets all 4 axles. Why that? well it shows the end of the train as a slightly larger current draw so the end of a train is always marked and identifiable. Most of the trains are in the 20-40 car length. The club's system is Digitrax. really like your discussion on using IOTT for trains. Keep up the good work!
Thanks for the comment. This sounds very interesting. What I wonder though is how your friend is detecting the slightly increased current of the cars with 4 resistors. To my knowledge the NCE BD20 does not output an analog signal, but only occupied vs. not occupied. Or am I not up to date with that?
@@IoTT The reason for all 4 axles is to ensure that the train's end is detected. Loco's will always be detected, but the end also needs to be detected. Dirt, poor contacts may not get detected, but if the end is detected, this will ensure that a block is cleared. Or that is what he is trying to do. BTW: here is his layout: pamodelrailroads.com/smallery/default.htm
You covered block detection very well. I have experimented with the 5 Amp version ACS712, and was unable to reliably detect under about 100mA track current. They are noisy below this figure. The data sheets cover this and their sensitivity. However, they are great for detecting higher currents and very usable for over current protection such as booster sections and reversing loops.
Thanks for the information. That's exactly what I thought might be the case when I saw the low current to voltage ratio in the datasheet of the ACS712.
The Okkie8 is an eight channel current sensor module from arcomora.com in the Netherlands. From a functional point of view it is identical with the Mollehem Occupancy 8 Diode board I reviewed in Video#68. However, the Okkie8 current sensor circuit does not use comparators for detection, but only the opto coupler LED's, which means less predictable sensitivity as it depends on the current transfer ratio of the opto.
@@IoTT Thank you for the quick response. I like doing a lot of diy stuff with microcontrollers so I have looked around a few projects. Sometimes it can be hard to find information. I did settle on DCC++EX for my command centre as my layout is only small at the moment. Any suggestions on a good way to add loconet to a DCC++ based controller?
I am not really focussed on 3-rail track, but as far as I remember the standard way of doing block detection is simply isolating one side of the track and use the axles to bridge them to the other rail. A decoder then simply detects whether the isolated rail is pulled to ground or not. Very prototype like, really. Common decoders to process the input signal is the s88, which has been around for decades, or similar products from other manufacturers.
Is it possible to add the ABC braking function found in some locomotive decoders like ESU, that use a dccwiki.com/Automatic_Brake_Control with current sensing block detectors ? This way the loco will stop automatically if the signal is red and then go when the signal is green.
Yes, that should be no problem. The asymmetry for the ABC system is a voltage asymmetry and - if at all - will only cause a very small difference in current. ABC actually is a very smart way to stop locomotives in front of a signal, it eliminates the need for knowing the address of the locomotive that is approaching the signal. Too bad it never made it into the main stream.
![[LIVE] : ONE ลุมพินี 84 วันนี้!! คู่เอก "ก้องศึก vs เมืองไทย"](http://i.ytimg.com/vi/3kuzsAIDf1c/mqdefault.jpg)
I have a friend who is a huge advocate of current detection and has resistors on one of the two axles of a truck for all cars except the cabin (caboose) car. For the cabin car, all 4 axles get a resistor. He has used 15k ohm for the resistors on all resistor axles. Locomotives have their own current draw as they have decoders. His system is the NCE on his layout. We are now converting the area's club layout to resistor axles with the same structure - 2 of the 4 axles get the resistor and the cabin cars gets all 4 axles. Why that? well it shows the end of the train as a slightly larger current draw so the end of a train is always marked and identifiable. Most of the trains are in the 20-40 car length. The club's system is Digitrax. really like your discussion on using IOTT for trains. Keep up the good work!
Thanks for the comment. This sounds very interesting. What I wonder though is how your friend is detecting the slightly increased current of the cars with 4 resistors. To my knowledge the NCE BD20 does not output an analog signal, but only occupied vs. not occupied. Or am I not up to date with that?
@@IoTT The reason for all 4 axles is to ensure that the train's end is detected. Loco's will always be detected, but the end also needs to be detected. Dirt, poor contacts may not get detected, but if the end is detected, this will ensure that a block is cleared. Or that is what he is trying to do. BTW: here is his layout: pamodelrailroads.com/smallery/default.htm
You covered block detection very well.
I have experimented with the 5 Amp version ACS712, and was unable to reliably detect under about 100mA track current. They are noisy below this figure. The data sheets cover this and their sensitivity.
However, they are great for detecting higher currents and very usable for over current protection such as booster sections and reversing loops.
Thanks for the information. That's exactly what I thought might be the case when I saw the low current to voltage ratio in the datasheet of the ACS712.
Thanks for sharing such an Awesome video.
Glad you enjoyed it
Have you heard of a chip / circuit called an Okkie8? How does this work compared to the other devices you highlighted?
The Okkie8 is an eight channel current sensor module from arcomora.com in the Netherlands. From a functional point of view it is identical with the Mollehem Occupancy 8 Diode board I reviewed in Video#68. However, the Okkie8 current sensor circuit does not use comparators for detection, but only the opto coupler LED's, which means less predictable sensitivity as it depends on the current transfer ratio of the opto.
@@IoTT Thank you for the quick response.
I like doing a lot of diy stuff with microcontrollers so I have looked around a few projects. Sometimes it can be hard to find information.
I did settle on DCC++EX for my command centre as my layout is only small at the moment.
Any suggestions on a good way to add loconet to a DCC++ based controller?
@@EsotericArctos Have a look at Video#79 th-cam.com/video/BG7DtPcBpxU/w-d-xo.html
Could you make a video about Ground blocs detectors often used in 3-rail (marklin) setups?
I am not really focussed on 3-rail track, but as far as I remember the standard way of doing block detection is simply isolating one side of the track and use the axles to bridge them to the other rail. A decoder then simply detects whether the isolated rail is pulled to ground or not. Very prototype like, really. Common decoders to process the input signal is the s88, which has been around for decades, or similar products from other manufacturers.
Is it possible to add the ABC braking function found in some locomotive decoders like ESU, that use a dccwiki.com/Automatic_Brake_Control with current sensing block detectors ?
This way the loco will stop automatically if the signal is red and then go when the signal is green.
Yes, that should be no problem. The asymmetry for the ABC system is a voltage asymmetry and - if at all - will only cause a very small difference in current. ABC actually is a very smart way to stop locomotives in front of a signal, it eliminates the need for knowing the address of the locomotive that is approaching the signal. Too bad it never made it into the main stream.