Hello Everyone As the title asks is anyone using the dave Bodnar design for a CT current sensor? I have built one of his circuits and it works but the locos I have need the lights on for it to signal an occupancy. I have tried testing it by laying resistors across the rails all the way to 1M ohm and cannot get it to sense an occupancy. His website said that it sensed a 2k ohm resistor but mine is no were near that sensitive. I have the arduino using the digital input_pullup sketch just to test it. Is there some math or something that needs to be used to increase sensitivity?
I'm not personally familiar with that sensor. However, current from rail to rail goes UP as resistance across the rails goes DOWN. So, if it is not working with 2k, try reducing your resistance, cuz it sure as heck won't work at 1M. Also, just laying a resistor across the rails may not make good electrical contact. Use alligator clips or similar (even a paper clip) to clamp the resistor leads to the rails.
One way to make it more sensitive is to increase the loops thought the transformer and the loops also has to be reasonable tight around the transformer. Could also use toroid in the place of the transformer.
For block detection, I found a post with a circuit that could be modified to provide digital output. https://model-railroad-hobbyist.com/node/31270#comment-302930 Good sensitivity with SFH628A-4 Optoisolator (use the -4 version). Note that the bridge rectifier results in a voltage drop to the rails; for consistent voltage, you can include just the bridge rectifier on blocks where occupancy detection is not required. For further detail, see pages 4 and 5 of the pdf file attached to prior post: https://www.trainboard.com/highball...-out-card-for-jmri.116454/page-2#post-1141719 This board has components for two blocks.
The voltage drop with a full rectifier is 1.4 v and half with a half rectifier circuit. Using a reed switch, current transformer, infra-red or hall sensor there is no voltage drop. Depending on its application it is also not necessary to cut tracks when using reeds, infra-red or hall sensors. Reed and Halls can also be used for train direction, length of train and even counting cars – yeah lots of magnets. The rectifier circuit of Ash is very similar to the detector side of S88 block detectors except they use the 1 amp round rectifier -save space - and a few resistors.
I held the resistor down with my finger tips plus rubbed it up and down the peice of track i was testing on just to make sure there was no oxidation on the rail preventing contact. I was hoping to use DIY resistor wheel sets on cars, commercial ones I have heard use 10k resistors thats why I asked about putting the various value resistors that I tried across the rails without getting any kind of occupancy signal. Unless I can go to a smaller guage wire im using 20 ga stranded I have the max wire wraps thru the coil and pulled tight then heatshrinked.
I must have had something wrong with the first bodnar sensor I built. I have built a couple more and all have worked great as soon as a loco is put on the track it senses it they dont have to have the lights on at all so I have 15 built more and all tested as they should .
David Bodnar's design is ok but not very sensitive. Have a look at this (JLC Enterprises -B. Chubb-). From that I made a prototype, and then a pcb. Sensitivity: 1mA.
