This will be my one and only deviation from my normal train related posts. Stay safe everyone and do some modelling.
This will be my one and only deviation from my normal train related posts. Stay safe everyone and do some modelling.
I’ve had little to post about recently so have refrained from taking up precious bandwidth blathering about not much at all. However this I morning applied the final touches to a couple of Kerian Ryan Models semaphore signals which means I do finally have something worth writing about.
These brass and nickle silver kits come as a set of etches with a few small detail parts that appear to be cast in white metal. I was handed a couple by a friend quite a while ago who asked me to complete them for him. He’d already made a start on them but it took me quite a while to get around to working on them. The reason for this delay is that I tend to work on the models I need for the layout and signals tend to be well down the list of “needed” models. However since operations started on Morpeth I’ve decided to signal Queens Wharf as a test to see if I can institute a way of controlling train movements on the rest of the layout. I’m not sure I’ll be able to signal the entire layout in a prototypical fashion as the cost and time this would take would be prohibitive.
I sat with Ray Pilgrim who writes the Bylong blog and has a company called Signals Branch at a Forum not too long back and he counted the number of signals I’d need. He reached 17 signals and that didn’t include everything such as ground signals and signalling the entry to storage. I staggered away clutching my chest 🙂 I purchased one of his bracket signals online and this came in at well over $180 delivered and that was well before the Aussie dollar crashed through the floor this past fortnight. Signalling the whole layout might have to wait a little while but I’m going to run a few tests with the signals I have in hand, including the one I’ve been building along side the ones handed to me by my friend. More on that later but be warned: modellers who enjoy and know a great deal about signals and signalling will not like what I have planned. My suggestion would be turn away if you fall into this category and see any upcoming posts that have the word “signal” in the title. I will not care that what I’m doing isn’t exactly to prototype so don’t bother telling me so. You have been warned 🙂
The construction of the signals was relatively straightforward but I would not recommend this kit to anyone who isn’t experienced with a soldering iron. The kit is highly detailed, in fact I feel it’s over detailed. Working with such small detail parts might make you eel good but it can also be a one way trip to insanity and sore knees as you crawl around under the workbench trying to find something you’ve dropped. And we all drop things, don’t pretend otherwise! 🙂
I was making two of the signals operational and one that would be restricted to a static display model and I’ll be honest and say I departed quite significantly from the instructions. I cannot see the point of applying detail that no one will see, which isn’t robust (in my opinion) enough to stand up to the rigors of surviving on an operational layout or will be lost under a layer of solder, glue or paint or a combination of all three. I’m an experienced modeller with a lot of soldering, gluing and painting experience under my belt and I consider myself quite neat. In spite of this some of the detail was still lost from view. Don’t tell Keiran but I didn’t apply most of the really fiddly bits 🙂
Having said that the kits make up into an accurate and very pleasing NSWGR home signal and two of the ones I made operate and have LED lighting. I applied the coloured lenses this morning and while I’m quite happy with the red I feel the blue plastic is a bit too dark, significantly masking the LED. Also the colour of the lens appears to be shaded to provide a green light when the LED has a yellowish tinge, however the LEDs supplied with the kit are very white so the dim light showing through the lens remains very blue to my eye. This isn’t a huge issue as the semaphore arm will be the primary means of communication to drivers however if I build any more of these kits I’ll probably swap out the supplied blue lens with a small piece of green cellophane. The detail parts supplied with the kit are really quite beautiful and had very little flash and really look the part. Making the signal operate was only a small amount of extra work over building the model static and I really can’t see why you’d bother building such a kit unless you were going to place it in an operational form on a layout.
I’ve spent a couple of days working on building some turnouts and after complaining on the phone to Glenn Scott at ModelOKits that things would be so much simpler and quicker with a cast code 125 #6 frog I found myself volunteering to make a pattern so a cast version can be made available. After-all, I was making a code 125 frog for one of the turnouts I was building anyway so making two and turning one into a pattern wasn’t too much extra work. It seems to take me between 2 and 3 hours to make a single frog from scratch from NS rail. Using a casting costs you the price of the casting but you save the time it would have taken to make it from scratch and you know the flange-ways and other dimensions are consistent.
This is the second frog pattern I’ve made: I made the code 100 version that is sold by ModelOKits something like 8 years ago. I made that one for Dave Morris at Waratah Model Railway Co and they passed over to ModelOKits when they purchased Waratah a few years ago.
Glenn told me he was probably going to retail the resulting casting so if you want some get onto him and let him know. He might have the ready for sale at the April Aus7 Forum.
My dentist occasionally gives me a lecture about practicing sound dental hygiene: by which he means brushing “properly” (which he demonstrates), flossing (I ain’t goin’ there) and the use of teeny weeny little brushes that look like tiny pipe cleaners of the type my mother used to clean her cigarette holders with when I was a sprog and she wanted to look like Rita Hayworth.
Of course, as he’s giving me the run down on “sound dental hygiene”, I diligently nod my head and as soon as I’m out of the place diligently ignore him till the day before my next visit when I use the tiny brushes to clean the crud from between my teeth.
An exchange of emails between myself and Marcus Ammann about 6 months ago put me in mind of my dentist because in them Marcus was warning me about the dire consequences of not using a common ground between the boosters on my DCC system. A common ground is really just a white connected to the negative side of the bridge rectifier in your DCC boosters that allows them to work together by eliminating electrical incompatibilities. If that last part sounds a bit vague then I’m willing to fess up and admit that this is deliberate because I’m buggered if I really understand what these “electrical incompatibilities” are. However when someone like Marcus Ammann or Mark Gurries tells you you need to install a common ground when you have a layout large enough to need electrical districts on your layout then you ignore them at your peril.
After Marcus gave me this warning and provided me with a link which led me to a bloody curdling story of a modeller who ignored this advice, he informed me that if I wanted to install this common ground wire I couldn’t use the Wangrow/System 1 booster I’d been using up till then. I diligently nodded my head (this was virtual nodding because Marcus couldn’t see me, this exchange being via email) and then I diligently ignored him. My thinking was that if I was going to implement sound DCC hygiene on my layout I would have to discard my perfectly good SYS1 booster and fork out approx $250 for a new NCE booster (a PB5 to be precise) to allow this to happen. PB5s are set up in such a way as to allow them to be connected in the required way. So nothing on the Morpeth line changed until…
During the most recent gathering of the Borderline Operators I got a recurrence of a problem with my control panels that I decided it was time to address. The problem was that about 50% of the time on starting up my DCC system my control panels simply refused to operate. I’d power the system down and start it up again and sometimes this would fix the problem but other times it didn’t. So last week I posted a description of the problem on the NCE-DCC Groups IO group and waited for a flood of responses. I got one response from Dennis DeGroff in the US who said he was having the same problem and that he’d contacted NCE about this and was waiting for their solution. Another modeller has since made contact with us and he has the same issue. So at least I could take heart from the fact that this wasn’t just my dodgy control panel building practices.
However quite a number of other responses ran along the lines of “have you got a common ground installed between the boosters?” Eventually I admitted that the answer to this was a no but could someone point in a direction of how to do this??? I was a little surprised when no one replied to this as I thought installing a common ground was common practice and that knowing how to do this would be common knowledge but evidently no it isn’t. Then Mark Gurries’ responded and pointed me to one of his excellent DCC web articles https://sites.google.com/site/markgurries/home/nce-info/nce-boosters/pb110-booster-grounding-issue which contained a description of how to modify my SYS1 booster so I could install the common ground and not have to fork out $250! Hoo-bloody-ray!
After I made this modification to my SYS1 booster I took a trip to the hardware store and purchased 12 meters (about 39′) of solid core wire to install as the common ground. When I originally installed my boosters under the layout it seemed logical to me that they needed to be as close as possible to the section of the layout they were powering, rather than all being gathered in one place. Having all your components in one spot would certainly have made running a common ground between boosters easy, as they would have been only a few inches apart, but mine were either side of an aisle with a wiring run needing about 12 meters of cable.
You see a booster is a sort of receiving station that communicates with your command station and which has its own dedicated power supply. The power supply for the Queens Wharf section of the layout on the right in the above photo is located under that section so that power for the trains only has to travel a couple of meters, not the 12 meters that would be required if I’d installed all my DCC components together in the cupboard labelled A. At heart you have separate boosters and their associated power supplies so as to reduce voltage drop from one section of a larger layout to another. It seemed to me that for this to make any sense the booster needed to be where the power was required. Anyway what this led to is that I needed a 12 meter long common ground to connect my boosters. I did enjoy reading in one of Mark Gurries’ articles that he can’t work out why people place all their DCC components in one spot unless it’s to impress visitors with their large rack of DCC components. I don’t want to impress anyone with my large rack so I spent this morning crawling around under my layout running my common ground wire.
After I got the common ground installed I turned the whole system on and got a short on the Queens Wharf side of the layout but this was cured by simply swapping the leads from the power pack over. This brought the two DC power supplies into phase on both sides of the layout. One of the quirks of a layout with separate boosters and districts but with no common ground is problems at the track boundary between the districts. In my case I occasionally got locos running over the boundary after which they’d slow and hesitate for a moment and then continue. A common ground cures this. After I got the system running I tested a loco over the boundary and there was a very small buzz and hesitation which I put down to a difference in the voltage between the QW and Morpeth districts. I was getting a reading of just over 17 volts at QW and 12.3 volts in Morpeth. The SYS1 booster has a switch on it that allows you to select a suitable voltage for large scale, HO or N so I switched this down to HO and this dropped the voltage at QW to 12.5 volts. This isn’t exactly the same voltage as Morpeth but it’s a lot closer and it cured the buzz and the very slight hesitation of the loco at the boundary.
Did installing the common ground cure the problem with the control panels? No but at least I can sleep at night knowing I’ve followed Marcus’ advice faithfully. Following my Dentist’s advice is a work in progress.
I’d like to thank Marcus Ammann, Mark Gurries, John Parker and Dennis DeGroff for their help and advice in getting this job done and hopefully eventually coming up with a solution to the issue around my control panels. All mistakes, misunderstandings and departures from sound DCC hygiene are totally mine and should not be sheeted home to the various DCC gurus I’ve consulted in the preparation of this blog post.
This morning I was free to do a little modelling so the plan was to make some minor modifications to a recalcitrant wagon which refused to go through a couple of my Peco curved turnouts like the rest of my wagons, head up to the layout room to do a quick check of a turnout template I’d printed out from Templot and test the wagon through the curved turnouts on a train. After these small jobs I’d be free to get back to building some turnouts in the work room inside the house. It was at this point that things went awry.
So let’s back up a little and go back about 12 months to a post I made about this same wagon and another Peco turnout that it wouldn’t run through. At that stage I determined that the gauge of the track at the heel of these turnouts widens out to something like 33.3 mm. As the wagons run through the turnout some have a tendency to drop into this gap and derail. At the time I wasn’t having any problems with the other curved turnouts on the layout and they were all pretty much on unsenciked sections so leaving them in place wasn’t an issue. However this one was on the scenicked portion of the layout between Morpeth and QW and as such the simplest solution was to build a hand made equivalent and replace it. Problem solved, or so I thought.
Now jump forward to two or three weeks ago to my 3rd operating session with the Boderline Operators. As soon as the operators started running trains out of the yard they identified wagons that kept derailing at the throat of the yard where there are three of the same type of Peco curved turnout and of course one of these wagons was the little 4 wheel cattle wagon that had given trouble on the turnout that had been replaced. Other wagons occasionally had trouble on these curved turnouts but this one derailed every time. All the wagons giving trouble were removed from the track as the problems occurred and I’ve spent some time since working on them, checking wheels, couplers and back to backs. All the wagons I’ve been working on were put into a train this morning and they all ran through the curved turnouts….except the cattle wagon.
So to be more accurate from my statement earlier in this post that the other curved turnouts on the layout weren’t giving me problems I’d have to say that the other curved turnouts weren’t being used much and as such I was able to ignore the fact that some wagons invariably derail at a particular spot. Because I hardly ever run trains on my own these problems can be shunted to the side. So after having done the required checks on this particular cattle wagon and having determined that the back to back was fine on both wheels sets, the axles were reasonably parallel, that the couplers were at the correct height and had more than enough swing it came back to the turnout. As the wagon was derailing on the heel of the turnout where the switch rails come into contact with the stock rails (the same spot it had derailed on the turnout I’d replaced) it was clear that the real culprit was the 33+mm gauge at this spot and not the wagon at all.
Generally speaking I’m very happy with the Peco turnouts and track on my layout: most of it works flawlessly and does a great job in the areas where I’m not going to apply scenery and don’t have to try to pretend it looks anything even remotely similar to NSWR track. However I wasn’t at all looking forward to having to rip up the four Peco curved turnouts that lead into the storage sidings and replace them with hand built replacements that would each take me a week to construct. In working on the same problem with the turnout I’d replaced I packed out the notch and soldered some brass into the gap, filed the notch and even hammered in a few extra track pins in an attempt to overcome a small difference in elevation all to no avail. That fussy little cattle wagon would not run through that turnout without derailing.
This morning as I stood next to the layout where the offending turnout is located I reached down in frustration and pushed the rail with my thumbs and it moved. Most Peco track of this type has rather large rail chairs moulded into the plastic base and nothing short of a Dremel and a good bastard file with move it. But because the rail adjacent to the switch rails of a point can’t have rail chairs on the inside of the rail it’s not held quite so securely at this location. I suddenly realized that the offending rail gauge at this very spot might be altered enough by the use of a simple packer.
Packing the rail in a little with some folded paper pushed it over just enough to solve the problem. It turns out it wasn’t so much the wheel dropping into the gap at the heel of the turnout that was the problem but that the wheel (especially on my cattle wagon) would round the fairly gentle curve of the turnout with the outside wheel flange hard up against the outside rail and when it hit the notch the amount of slop in the gauge at this point would allow the flange to catch and let it ride up onto the top of the rail. I tested the train back and forth through the turnout 4 or 5 times with no derailments. I’ll determine the best amount of packing to insert and replace the paper with some styrene and glue this in place once I check that the problem is cured. My guess would be that something between .020 should be more than sufficient.
And just to prove that it really did work here’s a little video of the train working it’s way through the turnout. The cattle wagon mentioned in the text is the last wagon in the train.
This is a tale of how I managed to destroy two ESU L series V4 Loksound DCC decoders with 24 hours of each other. It’s not a rant against ESU because up to this point I’ve never had a problem with the decoders from this company that I’ve installed and I find them by far the best decoders in terms of sound output and quality as long as they’re paired with good hi bass speakers of the correct ohm-age. However at $AU260 a pop they’re far too expensive to stand by and watch smoke come out of them and not pass on my experience to others. So this is a post for my newest BF Rob Anderone on Facebook who asked me to share what happened.
I’d been working to install two L series deciders in the body shells of two 1:43.5 scale Auscision 45s which are large Goodwin/Alco hood diesels built for the NSWR. One of these locomotives was mine and the other belonged to a friend. I built a vero board circuit based on the 7th Heaven article written by John Parker and everything was made and ready to install when I made the decision to test all the light/sound/decoder connections on the boards prior to installing them. I had a second circuit board supplied by John that he’d built for testing purposes to try out a different variety of sockets and plugs. These circuits get a little tricky to get at once they’re installed and the wires from about 16 LEDs are running to them.
You have to picture in your mind my rather Heath Robinson “testing” station with loco chassis sitting on about 1.5 meters of test track: connections for an NCE Power cab at one end of the track, Lokprogrammer at the other end and a loco chassis sitting in between on the track. I have two LEDs wired to the correct plugs to mate with the sockets you can see in the photo and I’m working my way down the board (which is plugged into the chassis motor/track connections) testing each set of connections so I know that when I install the circuits into the loco they will actually work. All the LEDs in the loco shells have been pre tested and I know all the wires are going where they’re supposed to. In fact I’d replaced a few of the factory installed LEDs because they weren’t working when I tested them.
The procedure I was using to test the LED connections to the decoder was to start by having both the Lokprogrammer and the Powercab unplugged from the test track. I plug the LEDs into two of the sockets (let’s say they happen to be for the short hood’s headlights). Once these are plugged in I power up the Powercab and press the horn to ensure I have a connection and then test the appropriate function button. In this case it would be the headlight and I’d need to make sure the direction is set for forward and if everything is working the LEDs should light up.
Now we get to the problem. At some point I got sloppy and forgot to unplug the Powercab before moving the test LEDs to the next set of sockets. I attempted to plug the LEDs into the next available sockets on the board with the loco powered up and watched as a component on the decoder glow red after which a tiny whisp of smoke emerged from this same component. I’m told this was the capacitor. After this I could get no sound to come out of the loco. I had a spare L series decoder in a drawer so I got this out and determined to soldier on after wiping away the tears over my dead decoder!
Testing continued on both boards and I found both were working perfectly and so installed one of them in my friend’s loco and got it running with sound and lights. Next day I placed this loco on the test track with the body shell sitting on the chassis and while I had lights and movement I couldn’t get it to make any sound. It had been working perfectly the day before and I hadn’t been testing LEDs by that stage. I hooked up the Lokprogrammer and uploaded the sound files and the CV files and while the loco moved and lit up it would not produce a sound. I swapped in my third and final decoder (which was still plugged into the circuit board of the 2nd locomotive locomotive) and it worked perfectly: light, sound and movement.
I sent both dud decoders to my friend and DCC guru John Parker to look at and he tells me both are suffering from a fried/blown capacitor. The first one I know what happened to because I witnessed it happen in graphic detail. Why the second decoder should have suffered the same fate I have no idea, however unlike the first it shows no outward signs of burn marks where the component glowed and smoked so whatever happened to it was a bit less spectacular.
So my friend has a working locomotive and I have a pile of parts and no decoders.
This morning I finally managed to get the first of the two new control panels installed and operating. The first was for the yard at Queens Wharf and I did this in conjunction with installing the mini control panel at the super phosphate siding which is outside the yard at QW but still within the control of the QW box.
In spite of putting in a lot of work on the layout and it’s new panels and electrics over the past couple of months I failed to make the deadline for the Borderline Operators gathering in early December. No trains were running and I was the subject of ridicule and finger pointing although that soon stopped when they started stuffing food in their mouths at lunch 🙂 However my friend Phil managed the same feat when the our group met at his place in November so I’m in good company. He was so embarrassed he took off for NZ for weeks in order to avoid small children pointing at him in the street 🙂
There were actually two reasons no trains were running on Morpeth and the fact that I’d essentially pulled the wiring apart on half the layout only partly explained this. For some reason I’m yet to explain one of my power packs decided to turn up its toes and as such when I flicked the switch to start things up Thunderbirds were definitely not GO! I managed to track down this problem and have replaced the offending power pack with a stand in but I’m actually going to see if I can get the dead one repaired as it cost a fair bit and has hardly done a lot of work over the years, spending 90% of its life in storage.
The other main reason for my failure to have trains running in time for the Borderline Operators arrival was that I turned a quick install into a major renovation of the layout by deciding to install new fascia right along the front of QW and down to the doorway at the top of the stairs. About 10 meters of fascia in all. I did this blithely and in the knowledge that things always take far longer than you anticipate. As I began work on the job I realized that I really didn’t want to have to come back and re do this work again later and that meant that, as everything I was going to be doing was to be attached to the outside of the fascia, I had to install some fascia first. It was this which took all the time.
I worked on both installing the fascia and making and preparing the control panels in tandem and I’m generally happy with the way the panels have come up. Actually wiring the QW panel into the layout was quite a quick job really as I’d done so much wiring of the panel and under the layout prior to actually hooking it up.
The electrical upgrade of this part of the layout is now complete. All I have left to do before Jan 8 is hook up the storage road panel. This one includes an NCE Mini Panel and this needs to be programmed so I have my fingers crossed that this isn’t beyond me.