Hi there,
I realized in any decent layout you have places where the polarity gets reversed. Rather than use a bunch of electrical switches I was thinking about making a minature “full wave rectifier” and insert this between the pickup wheels and the DCC board.
When the wheels roll onto a piece of track with reversed polarity, the rectifier will instantly ‘un-reverse’ the polarity. However, if diodes ‘kill’ DCC signals then no point even thinking it any further.
Thank you in advance! [:)]
The problem is not reversed polarity to the DCC board, it is the short circuit in the track, back to the DCC booster. So your solution, even if it would work (and I haven’t burned enough brain cells on it to figure that out one way or the other) does not address the root of the problem. When some of the wheels are on the reversing section and some are in the non reversing section the rails are shorted. So the solution has to be to change polarity on one set of rails, not internal to the engine.
baldwinjl is absolutely correct about needing to change the rail polarity. However, you don’t need a “bunch of electrical switches” to accompli***hat. Simply use a DCC automatic reverser. The Digitrax PM42 comes to mind (because I have one), but there are others. A few wiring connections and you’re done.
Stevert
The other thing that makes the auto-reverse module a better solution is that you only need one, rather than a circuit in every loco. You can even share one auto-reverser for more than one reversing section IF you can be sure of not crossing more than one section boundary at a time. Also, if you could do it in the loco, no one else could use their engines on your layout. Even if you choose to use a switch instead of an autoreverser you only need one DPDT per reversing section. Actually you could share one amongst several sections, with the same restriction as for the auto-reverser.
Thank you Steve! [:)]
I would still like to have my question answered…does anyone know if DCC signals/codes will (or will not) pass through diodes or will it get messed up??
Thanks in advance!![:)]
OK, I burned some brain cells. The answer may not be absolutely correct, but I think I am on the right track.
I think a diode in the DCC circuit prior to decoding is not going to work at all. I expect that decoders themselves have rectifiers as part of their power supply, but they decode the signal before the recifier. The DCC signal is dependent on detecting the transitions of the pulse from one state to the other (it is a differential signal, with the pulses determined by the changing of polarity). Since it is a rectangular wave if you use a full wave recifier on it with diodes you would lose the transitions entirely, you would just have a DC signal a couple of diode drops lower than the signal level (which is probably how the decoder powers itself). If you had just a single diode, you are obviously out of luck, since you’d have no current half the time.
So I think the answer is no, the DCC signal can’t go through diodes.
I am still not certain I understand the question. But gross generalization…
Yes, DCC signals/codes will pass through diodes. Many of the train detectors that are used in series with the track have diodes in them (in DC this would be a T detector). But these always have at least two diodes so that the signal can travel both ways. The diodes have a small current drop over them that the circuit can then sense and know a train is present on that pice of track.
No, if by diodes you mean a full wave rectifier, tapped on the DC portion. The signal instead of going plus, negative would end up being two plus with no break between. Basically it would become a DC current with no signal.
As others have said I think your idea is exactly what the auto-reversing units do. They sense the short circuit and switch a section of track. The problem with this in a locomotive would be detection on each wheel separately. ie. the rear wheels would still be on a section of track with the polarity the opposite direction.
[quote]
QUOTE: Originally posted by Texas Zepher
I did a little research, I hadn’t thought about that. Although even with this type of detection you lose around .7 Volts on the signal. If you have detectors on all blocks you could compensate by upping the booster a bit. Otherwise you lose a little bit of the margin in the system, but it will probably work fine. It looks like a better detector is a little current transformer, but we weren’t trying to design that here…
They will quite happily pass through diodes, unless the diodes are in full-wave arrangement.
John
Sorry for being nitpicky but that’s a rather presumptuous statement. I don’t beleive you HAVE to have a reversing loop or reverse ini polarity in order to have a “decent layout”.
Tom
An absolutely true point.