Re: RC Snubber question
About the best source of information on DCC signals is http://www.wiringfordcc.com/dcc_waveforms.htm. First let me say without an oscilloscope its hard to understand the problem. The site listed shows what happens to high speed switching signals. There are a couple ways to help reduce the “sign spike,” one being snubbers another is twisting the signal cable (in this case the cable providing power to the track. Another is to keep the signal cable as short as possible. Again the signal cable is the power cable going to the track. If you look at the signal on Allan Gartner’s site you will see that adding a snubber reduces the peak to peak ringing; however, the leading edge of the signal is somewhat rounded off. Add too many snubbers and you loose the leading edge completely.
What to look for if you think you are having problems is loosing control of loc’s and blowing decoders (again looking at the peak to peak voltage of the site). The voltage can be higher than what the decoder is designed for. I have seen at least 4 decoders blown when we had rather large layouts and long cable runs. At the time we couldn’t figure out the problem, but did not have an o’scope available
Another thing one should understand is that a DCC signal is NOT the same as a DC signal. A DC signal is referenced to ground. The voltage on the track is positive or negative voltage in reference to ground. A DCC signal is NOT reference to ground in any way. It is a Peak to Peak signal, Not connected to ground in anyway. If you connect your DCC signal to ground YOU WILL DAMAGE something. You do want to connect all chassis together so they are at the same potential to help ensure someone doesn’t get shocked (not all ground is the same).
This signal problem is not with just DCC units, but with any high speed switching, such as computer and computer cabling.
I hope this has belped and not added to the confusion. I have been in computers all the live and still have problems. And even a harder problem describing the problem