programming help

[walkingthedog]

Spoilsport.  

[St1ngr4y]

Hi Ron,

How are you controlling the speeds of your trains - do you use a series of Forward to [nn] commands then a Stop command, and therefore rely on the acceleration curves of the loco decoder? Have you tried the accelerate and decelerate commands. I find these much more reliable in speeding up and slowing down trains consistently.

You are starting the two trains off at exactly the same places each time, aren't you?

Ray

[Jailor]

Thanks for that, yes I am starting from same position each time. I am using a combination of start and stop modes so will try your suggestions and maybe stick to just one type of command. The only problem I have using accelarate and decelerate commands is that one steam loco,  (the Peppercorn from Majestic set) stalls on points under 22 mph so I have to start this from that speed which is annoying. Hornby have had it back twice with no improvement so I have given up on any.

[Jailor]

Come on guys I have only had one reply from Ray. 

I imagine using RM to write programmes is a massive part of our train running, I would think that you have all come across the same issues I am having and have a vast experience to share with myself and any other newbies reading this post. Any tips from all of you that have been used to make my program run smoothly would be really appreciated.

Thank you Ron.

[96RAF]

@Jailor

RM suggest as an aid to consistent placement that you slow a loco to a crawl just before a stop and run it into the buffers, so as to set a firm start point for the new event - crude but essential to avoid 'creep' until loco detection comes along. This is obviously of no use at all if you are attempting to stop exactly at a through station time after time.

Re speed <22 problem - I have a Class 56 that sometimes falters through a double set of points when switching from outer to inner loop (but its always OK the other way) - this then has the effect of stopping the train short which obviously causes the next event end point to be out of kilter.

As you are just getting into programming a quick way to 'write' the bulk of a program is to use the record feature then amend it to refine the events and/or timing.

[Jailor]

Thanks RAF96 I stupidly had not thought about my Peppercorn stalling making a difference to where it stops, flipping obvious now you have pointed it out --- one of my old codger moments. BTW my nephew is currently serving with the RAF in your neck of the woods.  

Thanks again. Ron.

[96RAF]

Ron

Unfortunately RM is all about time x speed = distance. Even the Sapphire decoder automatic operations are timed events.

Some Lenz decoders with Sapphire equivalent capability can be set to distance which is exactly what we are looking for.

Re your nephew - I spent several years at Akrotiri in the late '60's and 2 of my kids were born in the RAF Hospital on base. Over the next 20 years I returned on detached duty many, many times and got to quite like the place.

[Jailor]

Sounds like the distance decoders are the way forward as cash allows, I learn something every day from this forum. 

Off topic and very unlikely but did you come across Noel Green in the 60's he was based in Cyprus working in intelligence with the Navy and is my mothers brother? (appoligise using the Navy word to an RAF man)

Keep safe and enjoy your retirement. Ron.

[gilbo2]

I always find buffer stops are my best friend for programs until loco detection arrives. ie drive SLOWLY into them to reset the loco position during your program.

Crude but does get really good results and no one ever notices ;-)

[Jailor]

Understand thanks for suggestion, I do like buffet stops myself, love a bacon bun.

[Teedoubleudee]

Staying OFF topic (which I have a habit in doing) and reading your post above and your signature, I'm an ex Cosford Brat 1960-62 and was at Akrotiri Jan 68 till Jan 71, spending most of my time in the electronics bay (radar). Small world, perhaps we know each other?

Now, better add something to the thread before I get banned!

I have only today just started playing with the RM programming and so far results ref stopping position have been fairly good. However, it's important that I get it perfect so I have devised a simple circuit to ensure correct start and stop positions. My initial track plan consists of an oval with a passing loop (PL). The PL and most of the rest of the oval will actually be out of view covered by a higher level baseboard, yet to be built. The plan for the lower level activity is simple - a train (full) will travel in one direction, say left to right. This will park itself on the passing loop and the points will switch via RM allowing a different coal train (empty) to pass by in the other direction after a suitable delay. this will go on infinitum and could be adapted to more passing loops if required. The circuit mentioned uses 2 reed switches and a relay to control dead sections of track on the PL. All I have to do is ensure I time the loco to reach the dead section and stop before the prog sequence times out. The dead section will be switched back on by the other loco but won't move till RM tells it to. I've also programmed in the trains to whistle on entering and leaving the tunnels though left and right hidden speakers - good eh! ;-)

[96RAF]

Off Topic:

In those days AKR was a busy place with lots of troops.

@Ron

Noel Green - can't say I met him sorry.

@TWU

I was at there Sept 65 till Aug 68. Us grubby riggers from ASF weren't ever allowed in the Gin Palace.

On Topic:

Reed switches are a good idea to isolate a section.

Lenz decoders mentioned are their Silver and Gold series - read on their web site about them and assymetric DCC capability using their BM1,2 &/or 3 modules, which the Sapphire also supports but as I said by time not distance. You can slow or stop a train in one direction yet allow it to pass through in the other direction - fun to play with before RM came along. 

You easily make a 5 diode BM1 equivalent module on veroboard - see here http://www.rmweb.co.uk/community/index.php?/gallery/image/24940-assymetric-dcc-generator/

Rob

[Graskie]

I remember seeing a few UK forces housing estates in Cyprus. They all looked like bog standard council houses here. Were there any posh houses? Totally off-topic, but I just thought I would have my four pennorth.

[St1ngr4y]

I use that technique at my turntable. All but one of the pairs of opposite roads have a buffer stop at one end with an entrance/exit track at the other. So coming onto the turntable from an entrance, a loco will cross slowly till it hits the buffer at the other side, then it only needs to reverse a couple of inches very slowly to be perfectly positioned on the TT.

Ray

[Teedoubleudee]

Most of us marrieds lived in rental with the locals in Limassol. It was a peacefull period during my time there, thankfully.

[GarH]

I was fortunately too young (only just I might add) to get a call up before transcription ended so never had to do my national service, although I would have been prepared to do it had it not been stopped, but nice to hear some oldies reminiscing.

I've taken the plunge on setting up a programme but have taken full advantage of the record button on RM.  I took the time to draft something out in word to remind me what I wanted and just set the locos and points going as I used the system and hey presto RM does it for you.  I now have a five minute programme with a steam sound loco setting off from the mainline station around the outer circuit doing its thing with sound and actions, a 2-BIL EMU  reversing out of the way of a Class 08 sound diesel coming from the central loco shed and yards around the inner track and back home and the 2 BIL following round and then doing its own thing until all three locos are stopped almost precisely where I want them each time.  So my tip is use the in built recording facility.  It's not 100% accurate placing all locos back to start but close enough for me to be very impressed and you can always go in and revise the programme should you want.  Well done Hornby, excellent facility.

[96RAF]

GarH

I also use the record facility to get RM to do the basic script writing based on a hand written rough time vs event table that I prepare up front to work to.

Then once saved you can easily go back in and tweak timings and/ or add/remove events as necessary to better suit the end game.

As you say an excellent built in facility.

@ TWD - Limassol - I was in rented downtown also. My old house is still there, except it now has an upper storey.

[Teedoubleudee]

The upper story is a dowry thing - you will notice (at least in those days) that they always left the concreate rods sticking out the top for such a purpose. I've been back to Cyprus many times since but only to the Turkish held North (Kyrenia) as it remains unspoilt by high rise hotels.

Back on topic - what was the topic? Ahh yes, timing. I must admit I have been playing about with this. I have an oval with hidden loop. I want to play a whistle and train movement sound as the train passes through the "visible" section. Mine always starts in same place as I use a system of relay controlled dead track to stop them. But I've had great "fun"  trying to get the sound start and stop at the right times as the train travel time seems to vary. I've tried different ways to start it eg "accelerate to (xx)" etc but can't quite get it perfect every time.

[idlemarvel]

(on topic!) Yes I find it difficult to get sound fitted locos (with sound turned on) to respond to programs consistently in terms of stopping in the same place compared to when sound is the turned off.  This is witk LOkSound decoders.  Anyone got experience with other sound decoders like Zimo?

[Fishmanoz]

Is that because they wait for sounds to finish before responding to program commands on changing speed?

[Teedoubleudee]

Sorry I should have explained myself better (still dreaming of sunny Cyprus maybe?) . I'm not using sound decoders in the locos but rather a wave file from my PC via RM. I bought a cheap stereo amp from Maplin and connected it the headphone socket with one speaker on the left and the other on the right where the trains appear/dissapear. I then played with some downloaded train sounds and produced a couple of tracks that make the sound follow the train in whichever direction I'm driving. If I do say so myself the effect is rather good, at least it is when the train turns up on time!

[idlemarvel]

I think so fishy, yes.  The time between issuing the command and the loco starting to move seems to be variable, depending on how long the loco has been idling.  From startup it is probably consistent (haven't tested this to confirm but I expect so) but from a point after startup it is variable.  With sound turned off I can get locos to stop within a few inches consistently, with sound the variation can be a foot, which componds as you have multiple programmed operations.  I will do some more scientific testing but I would be pleased to hear from anyone else who has tried programming with sound locos.

[idlemarvel]

There is a known issue (or feature?) of using sounds from WAV files on your computer in a program.  Basically the program has to wait until the sound file finishes playing.  Or at least this is what I remember reading somewhere, but I can't find the reference.

[96RAF]

Strange that although the program has to wait for the sound to finish (the manual says allow enough time for a sound event - some sounds are longer than others i.e. >2secs -  or points action to complete before setting the next event) if I set the clock to play striking noon, I can have a loco start after that event and arrive before it has finished so there must be some overlap capability in there once an event has been commanded.

I've tried starting two locos at the same time and that fails as expected, as well as starting a train as the clock strikes, also a failure.

[Teedoubleudee]

Yes, the sound wav file will play till it's normal conclusion, in my case during my early experimentation, long after the loco has stopped!  As I mentioned in another thread on here, I use a free program called audacity to edit the wav file's length to overcome this.