Gordon H

You clearly never saw 'High Gill' at any of the many exhibitions it attended during in its heyday in the 80's. We regularly ran our trains at well over a scale 100mph - with working pantographs on a tensioned 0.25mm contact wire. All fully portable too. The stock also included a full length Hornby APT, fitted with a modified Sommerfeldt 968 to simulate the version the real thing had before it was changed to Brecknell-Willis. The really unfortunate thing about all this is that it was before any of us had a means of video recording it in operation.

Depends what you mean by 'many more' and what form they might take. It is interesting that there seems to be no problem in the world of DCC thinking, whereby it is accepted that special extra units (i.e. Accessory decoders) need to be fitted out on the layout and controlled via the Power/Control bus. The only alternative that most people seem to consider is the traditional 'wire, and lots of it, back to a panel' approach. A compromise that I adopted many years ago was the notion of similar 'extra units' which do the hard work of driving solenoids locally with the high currents that entails, but with each controlled by a single low current wire - so all you need is a power bus and a 'thin' wire for each point back to the panel instead of lots of high current wiring for everything (and connectors to match). Less connections to go wrong to start with, and consequently easier fault tracing thereafter.

Originally to prove to myself that it could be done, just as an engineering challenge. Thereafter, knowing this to be the case, my aim has been to get the message across to others that a lot of these things that seem too difficult to contemplate are actually do-able with a bit of effort. In many cases, the difficulties could be eased by manufacturers helping out a bit with some of the features they might include. It all depends what aspects of the hobby you prefer - running trains round and round a bit of track just for the sake of it is not for me. I prefer to try things out that are a bit (or a lot) out of the ordinary - e.g. the Class 508 sliding door unit mentioned elsewhere.

Quite right. That can also be achieved if you put your mind to it.

If you are a committed OHLE modeller then you are more likely to be prepared to give it the time and attention required. Taking the opposing view, it doesn't make sense to me not to do it.

It certainly helps if electronics is the day job too. As you might imagine, I needed to know what facilities were going to be included, and how they were all going to work before I could embark on construction, as well as having the facilities to do it. Every coach has a unique printed circuit board for its chassis, which not only acts as the floor, but includes the circuitry required for each aspect of the control. The cab end coach circuitry is mostly concerned with the head/tail lamp lighting, the coach whose doors do not open just houses the Athearn flywheel drive mechanism (see picture below), and the fourth coach contains the microcontroller and constant current drive circuits for the memory wire operated doors. All coaches are connected using a 7 way electrical bus, again using PCBs which plug together to form solid 'bar' couplings. The door opening remote control detector is fitted as part of the underfloor equipment on the fourth coach.

Yes, it did appear for several years on the MERG stand, but not recently as we often struggle for space with the various displays we have. Construction was detailed in BRM during 1998. It was built for running on P4 track on Effingham South, which is currently stored in my garage. AFAIK it remains the only unit around having all the features already mentioned. The only change of note made since it was originally constructed has been to make the traction DCC controlled. The door power supply and operation remains as it was, using third rail pick-up and IR Remote control.

Yes, it permitted that method of working with two independent controllers, but it was by no means compulsory to use it that way. You could just use the wire supply and rail return if you wanted to. It isn't just a matter of wanting to allow more trains to run simultaneously.

Given that the manufacturer will be obliged to fit something to the roof, it can hardly be considered an additional feature. The only cost consideration should only be the incremental difference between fitting one that can never work and one that could. For their 85, Bachmann clearly went to some effort getting the dimensions of the Faiveley pantograph right, but then spoilt what could have become a useful item, even as spare parts for other applications, by moulding most of it in plastic.

If it is a cost issue, why does it not have the same effect on equivalent European and Japanese outline models? Are the expectations of modellers in those fields different? Sommerfeldt and Veissmann, to name but two European manufacturers, have complete ranges of catenary and pantographs which can be used conductively. Ironically, whilst looking around for other manufacturers, I now find that Bachmann themselves have produced DCC fitted models with working pans - and a switch to select the option: http://www.spookshow.net/loco/bachmanhhp8.html The case rests...

Difficult to know where to start with this lot. I did 'learn something from it' as you put it when our group decided to give it a go back in the early 80's. It can be done, and made to work, and shown extensively at exhibitions - and has even appeared on TV. A bit much going to the lengths of describing what electrical engineers need to consider in such a design, then following it with a reference going to 'protecting resistors on the main circuit board'. If that is what you think they are, we need go no further. If you had any practical experience of using model pantographs for real, you would know that a full metal design works best. My very first scratchbuilt cross-arm pantograph head was made using a carbon strip from a propelling pencil. Although it worked, it clearly suffered from the effects of self heating, so that experiment was abandoned. Thereafter all metal designs have been used. Thus, the impedance of that particular component has never been in question, and with the low forces involved, wear has never been an issue either. Rolling wheels, on the other hand, to this day cause all sorts of problems with intermittent and variable contact. That said, I do wonder how much effort manufacturers put into the testing of their pick-up systems generally, because it still seems to be a major issue on many loco designs. Whether the signal is DCC or not makes no difference either, other than the issues already identified. The intention is not elitism - but trying to get across to people the kind of things that can be done if sufficient effort is made to include certain design features in the first place. Over 50 years ago the Triang 81 and EM2 had pantographs that could (just about) be used, so if anything, things have gone backwards since then. As mentioned, European and Japanese manufacturers have been producing locos with (potentially) working metal pantographs for years, it just seems a shame that it doesn't happen here too. If you can buy a working cross-arm pantograph for an N Gauge Japanese loco (even as a spare part!) surely the manufacturing tolerances can't be the limiting factor. <controversial mode> All the more reason not to try... </controversial mode> :-)

Note that within MERG, the acronym 'RPI' was in use long before (over 20 years before) these recent developments which sometimes use the same shortform. To me, 'RPI' will always represent my 8031 based 'Remote Panel Interface' module. Hence, beware that any references you might come across whilst searching through MERG documentation using just the acronym may not give the results you expect !

When I started work on what eventually turned out to be my scratchbuilt Class 314 back in the mid 80's, the cab was the first item produced, basically to prove the principle of the resin casting method used. One of the first cab castings was then used to produce the inner coach end master. It was some years later that my Class 508 was produced from the same moulds. The cab on the left is the master, fabricated from styrene.

Strictly speaking, the decoder doesn't 'isolate' the motor as such in the electrical sense, it just creates an internal DC supply from the track feed, which is used then to control the motor using PWM. The two processes are entirely separate functionally, but tied together in voltage terms by use of a simple bridge rectifier to generate the DC supply from the DCC waveform, i.e. the voltage available for the motor is directly related to the voltage on the track. The 'nett' voltage value on the track could indeed be said to average to zero as the track data waveform is notionally square using two different full cycle periods for 1's and 0's. The bias applied for the DC 'Loco 0' option is not additional voltage to the track as such, but a distortion of the equality of the square wave periods - specifically during '0' bits. One half of the ''0' period is allowed to be stretched up to about 10ms, from a nominal 100us or so, thus the waveform starts spending more time at one polarity than the other, changing the average value in the process. Apart from the average loco motor not really liking this method of control very much, it also drastically slows down the transmission of DCC commands to other items if done to excess. A typical approach to minimise this delaying effect is to only stretch certain '0' bits in the data stream, say, the 'Start' bits in a byte (which are always '0's).

You can do that with any radio based system. Nothing to do with battery operation as such.

It shouldn't cost any more to have a working pan than non-working. If anything, the pantograph is the most obvious feature of any OH electric, so certainly will be seen, and if it's there why not use it? Not much point modelling electric outline if you don't do it properly.

I am guessing you have no practical experience of this aspect of operation, as a properly pivoted pan head has at least two points of contact. As Andi has said, if anything the wiping action of the head on the wire tends to keep the wire clean anyway - far better than a rolling wheel which simply accumulates dirt.

You might struggle to understand, and not care, but some of us do! There should be no insurmountable difficulties using DCC either. You just need to know what you are doing and design the system accordingly.

I would guess someone has at some time, but it is not something to encourage without knowing what kind of output stage the particular controller has, what frequency it is running at etc, etc. There is no single definition for PWM controller designs as such, they just employ a time based variation concept rather than direct voltage level. If it is a simple emitter follower output or similar, i.e. a single ended drive, a big capacitor on its own will likely charge to the full output voltage very quickly, assuming the driver can take the strain, so little or no benefit there. An H Bridge output (like a DCC decoder has) would likely suffer even more, and would need a bipolar capacitor to be used. If you want to try anything like this, it needs to be a properly defined filter circuit that you add, set up for the PWM rate being used, and rated for the currents to be encountered. Could be as simple as an LC, RC or LCR type.

I am a member, but still unsure what document you are referring to. What is it a 'handbook' for? Can you provide a link to where you found it? As you say, it probably needs updating if it is still available.

I am intrigued as to where you obtained this 79Hz figure. What does it relate to? While it is certainly true that DCC loco decoders use PWM as a method of driving the motor for reasons of efficiency, the PWM frequencies in use these days are typically into the tens of kilohertz range, so usually above the audible range. In the early days of DCC the processors available for use on decoders were rather limited in their capabilities and speeds, so lower PWM rates were often encountered but that time is long gone. The PWM rate the motor sees via the decoder bears no relationship to the waveform on the track, which is simply information encoded into raw power.

You would probably be better off trying to coincide with an exhibition during your visit. Unfortunately, there don't seem to be any in the London area in February. Perversely, about the best model shop I know of in London now is Broughdale - but that is London, Ontario, just up the road from Ingersoll, but I expect you know of that already.

Not that many would want to do it anyway, but if you had a DC loco with sufficient on-board storage to run a cab light in the first place it would be relatively trivial to construct an extra piece of circuitry to make it fade to off when the track voltage rises, so yes it could be done in DC. Some years ago I was involved in producing a similar effect for an HO Dutch electric 1600 class loco, whereby its second pantograph would drop once the DC track voltage was raised above a certain level to simulate the real thing doing this once the train was in motion and drawing less current.

A simple solution to this would be an RFID set-up, where each train carries a small unique tag whose information is read out whenever it passes a reader device beside the track. The information stored on the tag can then be interpreted and an appropriate sound produced by whatever means you have for this. On the MERG demonstration stand we have such an arrangement which works with an Excel spreadsheet to display an image and play sounds as each tag passes.

How are you powering the unit? There is a specific note in the Smartswitch manual that implies the requirements without being explicit enough: "The SmartSwitch can use DCC track power as its input voltage, but it will not work with direct AC current." This implies to me it will not work from the standard 'Aux' output (AC or DC) from an older style 'DC' controller as it does not include sufficient smoothing capacitance on-board. You are OK with a smoothed and/or regulated DC PSU - or DCC power which doesn't need much capacitance to keep things alive due to its waveform, but anything that 'dips' slowly through zero 100 times a second probably won't work.