Hi All,
I have had a professionally built 7' x5' L shaped layout that I bought secondhand some thirty years ago. Life intervened and it has been awaiting my attention for all those years until recent retirement gave the time required to start work on it. I got it running fine on 12volts, but on upgrading to Dcc the running of my three Arnolds was very poor over the Peco setrack points, although a new modern 0-6-0 gave no problems at all. I believe the problem is that the wheel flange is too deep, and when running through the points the flange runs over the plastic base between the rail/check rail, and lifts the wheel off of the track. Can any member confirm this?, and if so, what was done to reduce the flange size?
Whilst writing, my electrical test meter gives a fluctuating continuity continuity across the two rails in Dcc, but no continuity on DC. Is this normal?
Any help would be much appreciated.
If it was just the big flanges on these chassis then you should also have the same problem on DC as DCC - you do not mention that when running on DC the Arnold chassis don't run through the points. If you have also been getting issues with the Arnold chassis on the point when running12v DC then you have correctly identified the problem as the deep flanges and comparitively shallow flange ways on the points.
If this is only happening with DCC then it might be related to the flanges, but not at the common crossing (frog). I have a feeling that if unmodified Peco points (particularly 30 year old ones) have the same polarity on both point blades with the result that on the 'open' side the stock rail is one polarity and the blade is the opposite which means that with 'fat' flanges (or a tight back to back) it is possible for the wheels to short across between the stock rail and open blade. Whilst you would still get shorting on 12v DC the controllers tend to be much less sensitive and do not shut down (like the DCC ones do).
In theory it should be possible to reduce the amount of flange by running the loco at full power upside down on the workbench and attacking the wheels with a needle file as they rotate, but make sure that you cover the motor to stop the metal filings getting into it and the rest of the gear train - in fact it may be best to do a deep clean (i.e. strip down, clean & rebuild) on the chassis once the flanges have been cut down to be on the safe side.
The other approach would be to take a fine file to the points to deepen the flange ways, but if the problem is shorting between the point blade and stock rail then some more drastic modifications will be needed that may involve lifting the points to cut through rails (to isolate the blades) and add wiring (to electrically link the blade to the nearest stock rail). Though if going down that route then it might be best to replace the points with new ones that are DCC compatible - and you may still get problems at the frogs - designed with modern finer wheelsets in mind - with the Arnold 'pizza-cutter' wheels (probably 1960's vintage!).
I suspect that the final option would be to replace the Arnold chassis with something by Neville Kent who uses a much finer wheel profile - and this may be easier in the long run especially if one orders the DCC friendly versions of his chassis (just needing the chip to be linked to the 4 wires provided on the chassis). There might be a wait as Neville is a one man band who also holds down one (if not two!) full time jobs but they are worth the wait!
The results from your electrical test meter may be due to the fact that DCC uses an alternating current (16-18 v AC iirc). The DCC controller communicates with the chip in the loco by superimposing a signal on this AC current and the chip converts the AC to DC to feed to the motor, with the polarity and amount of power fed to the motor being controlled by the signals received by the chip. As such even when nothing is running the AC current will always be there, unlike with an analogue (DC) system when if nothing is moving then the current is off somewhere, either by using an isolating section or turning the controller to zero - in both cases on analogue / DC the result will be an 'open circuit' and therefore no current / continuity (unless all the sections are on and the controller turned up from zero).
The fluctuating reading may be due to the meter picking up any communication signals from the controller (I assume they 'scan' for chips periodically). If the meter was set to read DC current however then you will either get a wildly fluctuating reading as the meter tries to follow the AC wave form or it will take the 'average' of the wave form which will be zero.
I hope that helps...
I turn down the flanges on the Arnold 0-4-0 chassis, but I do not recall having to do so on the 0-6-0's, certainly not on the later versions with the plastic rods, but I will have to check the early 0-6-0 chassis I have to see if there is a problem with them. If not you may have back-to-back issue?? (ie the gap between the wheels.)
I remove about 0.5mm from the flanges of the 0-4-0's, it is not rocket science and can be done with the minimal of equipment.
What you need is a chuck that can hold the wheels but as most drill chucks are too small you may will have to find something larger.
I simply remove the complete axle assembly from the chassis, mount one wheel in the chuck up to the flange, spin it up and use a hand file to file down the flange on the other wheel, removing about 0.5mm from the flange but keeping the shape of the flange and making sure I do not overheat the plastic wheel centers. It takes about 10 mins. per axle.
Brian Meldon
Thank you both for your input, this was my first post, and the very prompt, helpful, and insightful suggestions show just what an excellent medium this forum is. I think it could be the shorting issue giving the problem due to the more sensitive DCC controller, however I will reduce the flange sizes on an old Arnold chassis that I have and try that as an experiment, then moving onto the points. Thanks also for the source of DCC compatible chassis, I will contact him and discuss
I have now checked the older Arnold 0-6-0's and the flanges are indeed too large for the Peco 009 points and they do indeed ride up on the plastic just as you say.
Again removing about 0.5mm off the flanges should do the job. I will do this with the chassis I have and report back.
(The flanges are 1.9mm deep.)
However I would still check the back to back measurements on the loco wheels as even with DCC they should not short out in the points.
Brian Meldon