Programming to Automate your layout

[PJ_model_trains]

St1ngr4y said:

It seems to me as if the way the sounds have been 'blown' onto the Loksound decoder is wrong. It might be worth sending the loco away to a firm who specialise in programming sounds. You can have a new set of sounds blown onto

the decoder for about £15. The shop I use are very good, based in Sheffield, advertise in modelling magazines, shop named after owner's granddaughter, I think.
Ray

Hi Ray

Pity I didn't try this after purchasing but, I thought people

were having problems with programming and I really was in early days of layout then. It is now 5 month old.

Not sure who you mean, Rails or Will ies or another? (R or S?) We are going near Lincoln in a weeks time, M6, M61, M62, cutting across from Manchester,

I could take it in and leave it with them if I know which?

Thanks for your help as always. 'A Ray of sunshine' ;-)

PJ

[St1ngr4y]

The name of the granddaughter is Olivia.

Ray

[St1ngr4y]

PJ,

If you take it to them, they may be able to blow the sounds onto the decoder while you wait. I'm not sure whether they could test it using RM however.

Ray

[Fishmanoz]

Can I just do some hat hypothesising here and let me know if my thinking makes any sense. For a start PJ, your loco is under warranty and I wouldn't be reblowing any sound unless I was convinced it was going to solve the problem ....... and I'm not. Anyway,

 

let me see if I have the problem correctly specified for a start.

 

Under manual throttle control with RM, the loco produces chuffs as it speeds up, then no chuffs on idle and as it slows down. It does so correctly.

 

Under the RM accelerate command,

 

it does similarly except there are a couple of stutters in the chuffs.

 

Under RM with a series of forward commands to higher speeds then cruising at the final speed, it operates correctly.

 

Under RM decelerate, or a series of forward to a lower

 

speed commands, it operates incorrectly with chuffs appearing at intervals.

 

No does that summarise the problem?

 

Assuming it does, let's apply some logic. For a start, all that RM is doing is sending commands to the decoder to advance in speed

 

steps, remain on the same speed step, or finally reduce speed steps. RM knows nothing about the sounds being produced, it doesn't have any capability to control sounds like this, the decoder does them in response to changing throttle settings. The only sounds

 

RM can control are those via the functions, when given instruction to activate or deactivate those functions.

 

So my first conclusion is that this problem is not being caused by RM, it's happening in the decoder. In fact, if you were able to manually

 

apply changes to throttle with the same precision as RM can, you'd get exactly the same problem.

 

Now let's look at what's happening in the decoder under deceleration. Each time it gets a command to reduce by one or more speed steps, the decoder reduces

 

the voltage to the motor asking it to rotate more slowly. The momentum in the loco means that it takes a finite time before it gets down to that speed. When it does, it has to take up the load to continue at that speed. In this moment, there will be a very

 

short time when the loco will want to continue to slow and back EMF will kick in the keep it going until it then continues at the speed step rate. This will happen at every change of speed.

 

Similarly on acceleration, if the loco gets to a new speed

 

step speed before the next command to go to another speed step is received, it will start to idle for an instant before receiving the next command.

 

Summarising, on acceleration, the loco experiences short periods of idle between commands to go to the

 

next speed step. Then on deceleration, the loco has brief periods when back EMF kicks in to maintain the speed step just reached. When back EMF kicks in this way, it is exactly the same as if the decoder is briefly being asked to accelerate.

 

So it

 

seems to me that the sound is doing almost exactly what it is supposed to do - produce chuffs on acceleration and no chuffs on idle. As it instantaneously finds such periods of acceleration and idle during both the acceleration sequence, and the deceleration

 

sequence, it produces exactly what PJ is hearing.

 

If that is the case, why don't all sound locos do this. The answer is that the decoder must be set up to not make instantaneous responses, rather it waits a fraction to ensure there is an extended acceleration

 

or idle established, then turns chuff on or off accordingly.

 

So my conclusion - the decoder is faulty. It is reacting to transient changes in acceleration/idle when it shouldn't. What I don't know is whether there is a Cvs adjustment for this built

 

in so it can be fixed, or if the decoder needs replacement with a new one.

 

Finally, if the problem is as I think, reblowing with new sound will have no effect whatsoever. All that reblowing does is put new sound files into the decoder's memory, it

 

will do nothing about how they are turned on or off.

 

Now am I talking through my hat, or are we getting somewhere?

 

If I'm right, is there any point talking to a shop specialising in sound? Most definitely, they will know if there is an adjustment

 

to fix it, if they are worth their salt. Or they will be able to say it is a fault that requires replacement, in which case you will have to go back through your warranty chain to retailer and/or manufacturer.

[Fishmanoz]

Last time I spend so much time posting this sort of stuff for you PJ! My morning coffee went cold in the meantime - yuck!

[PJ_model_trains]

St1ngr4y said:
The name of the granddaughter is Olivia.
Ray

Many thanks Ray

[PJ_model_trains]

Fishmanoz said:

Can I just do some hat hypothesising here and let me know if my thinking makes any sense. For a start PJ, your loco is under warranty and I wouldn't be reblowing any sound unless I was convinced it was going to solve the problem

....... and I'm not. Anyway, let me see if I have the problem correctly specified for a start.

Between 4 and 5am this morning that is exactly what I thought and was going to add this morning. Will re-blowing fix a fault in the chip

and what about the guarantee.

Under manual throttle control with RM, the loco produces chuffs as it speeds up, then no chuffs on idle and as it slows down. It does so correctly.

Yes with shunt and cruise buttons

and manual throttle control

Under the RM accelerate command, it does similarly except there are a couple of stutters in the chuffs.

Accelerate forward command, yes a couple of stutters to sound between 0- 50mph

Under

RM with a series of forward commands to higher speeds then cruising at the final speed, it operates correctly.

Yes it did work correctly accelerating

Under RM decelerate, or a series of forward to a lower speed

commands, it operates incorrectly with chuffs appearing at intervals.

Yes using forward to a lower speed it has intermittent chuffs or part chuffs

No does that summarise the problem?

The above

confirms we are on the same track

Assuming it does, let's apply some logic. For a start, all that RM is doing is sending commands to the decoder to advance in speed steps, remain on the same speed step, or finally reduce speed steps.

RM knows nothing about the sounds being produced, it doesn't have any capability to control sounds like this, the decoder does them in response to changing throttle settings. The only sounds RM can control are those via the functions, when given instruction

to activate or deactivate those functions.

Agree fully

So my first conclusion is that this problem is not being caused by RM, it's happening in the decoder. In fact, if you were able to manually apply changes to

throttle with the same precision as RM can, you'd get exactly the same problem.

Not possible but I agree with you.

Now let's look at what's happening in the decoder under deceleration. Each time it gets a command

to reduce by one or more speed steps, the decoder reduces the voltage to the motor asking it to rotate more slowly. The momentum in the loco means that it takes a finite time before it gets down to that speed. When it does, it has to take up the load to continue

at that speed. In this moment, there will be a very short time when the loco will want to continue to slow and back EMF will kick in the keep it going until it then continues at the speed step rate. This will happen at every change of speed.

Similarly

on acceleration, if the loco gets to a new speed step speed before the next command to go to another speed step is received, it will start to idle for an instant before receiving the next command.

Summarising, on acceleration, the loco experiences

short periods of idle between commands to go to the next speed step. Then on deceleration, the loco has brief periods when back EMF kicks in to maintain the speed step just reached. When back EMF kicks in this way, it is exactly the same as if the decoder

is briefly being asked to accelerate.

So it seems to me that the sound is doing almost exactly what it is supposed to do - produce chuffs on acceleration and no chuffs on idle. As it instantaneously finds such periods of acceleration and idle during

both the acceleration sequence, and the deceleration sequence, it produces exactly what PJ is hearing.

If that is the case, why don't all sound locos do this. The answer is that the decoder must be set up to not make instantaneous responses, rather

it waits a fraction to ensure there is an extended acceleration or idle established, then turns chuff on or off accordingly.

So my conclusion - the decoder is faulty. It is reacting to transient changes in acceleration/idle when it shouldn't. What

I don't know is whether there is a Cvs adjustment for this built in so it can be fixed, or if the decoder needs replacement with a new one.

My concern as of thoughts in the early hours! faulty sound chip. The fault is worse in deceleration

than acceleration, the problem seems as though a mathematical difference of a fraction of a second e.g. 50 to 45, 45 to 40, 40 to 35 it is alright, then the timing is a fraction out... a second or part second chuff or part chuff occurred at 30, 25, 20, 15

and 10mph

Finally, if the problem is as I think, reblowing with new sound will have no effect whatsoever. All that reblowing does is put new sound files into the decoder's memory, it will do nothing about how they are turned on or off.

Agreed, that was my concern and the reason I wanted the shop off Ray. I can contact them by email first. I am not concerned about £15 if it solves the problem. I am concerned if we do anything and it affects the guarantee.

Now

am I talking through my hat, or are we getting somewhere?

No you are not talking through ya hat mate! You have removed them there corks.

If I'm right, is there any point talking to a shop specialising in sound?

Most definitely, they will know if there is an adjustment to fix it, if they are worth their salt. Or they will be able to say it is a fault that requires replacement, in which case you will have to go back through your warranty chain to retailer and/or manufacturer.

Agreed,

hence the reason an email to the shop first will same all that and leave my favour steam sound loco (at the moment) untouched.

I have not contacted the supplier yet. It felt we had to run the mill so to speak. HRMS involved, Blue Box Co contacted, forum

discussions here and elsewhere and finally we wait to hear from ESU.

Many thanks for your help (everyone)

[PJ_model_trains]

St1ngr4y said:

PJ,
If you take it to them, they may be able to blow the sounds onto the decoder while you wait. I'm not sure whether they could test it using RM however.
Ray

Hi ray

See my reply to Fishy above.

I work up 4-5am and was thinking. No problem I wake up in the night every night nothing new. It is usually pain or toilet.

My concerns are included below, one the guarantee and two would they be able to fix the problem or would re-blowing do just that

leaving the problem in a chip/in the sound decoder.

To quote Hornby a weird one.

But, a problem is only a problem until fixed one way or another and through discussions someone (me included) has learnt more through it.

Thanks for your

help

PJ

[St1ngr4y]

Hi PJ,

 

Although I'm not an expert on the Loksound decoder, my impression from reading various articles and forum posts is that reblowing the sound project on the decoder is much more involved than simply replacing the sound files. Also, I may be

 

wrong but I don't think reblowing the sound would invalidate the warranty. After all, if you buy a pc, you don't invalidate the warranty by storing some files on its hard disk. Why not give the shop a ring, ask them to read this thread, and get their opinion?

 

Ray

[PJ_model_trains]

Hi Ray

 

I have kept options open, I would like to hear from ESU but will probably contact Olivia's and see what they say also.

 

Thanks for your comments

 

Just setting up new computer and have touch screen etc in place.

 

Loading

 

software and setting network ;-)

 

PJ

[PJ_model_trains]

Hi Ray

 

I have kept options open, I would like to hear from ESU but will probably contact Olivia's and see what they say also.

 

Thanks for your comments

 

Just setting up new computer and have touch screen etc in place.

 

Loading

 

software and setting network ;-)

 

PJ

[PJ_model_trains]

Hi Ray

 

I have kept options open, I would like to hear from ESU but will probably contact Olivia's and see what they say also.

 

Thanks for your comments

 

Just setting up new computer and have touch screen etc in place.

 

Loading

 

software and setting network ;-)

 

PJ

[Graskie]

You certainly kept a few options open just then, PJ. Best of luck with it all.

[PJ_model_trains]

What on earth happened there?

[Graskie]

You suddenly triplicated.

[PJ_model_trains]

Oh no, Oh no, Oh no

[St1ngr4y]

If you feel a sneeze coming, let go of the mouse!!

[Fishmanoz]

Have just spent a little time reading the Loksound manual. This decoder is highly capable, flexible and can operate in many modes with many different systems and makes of loco. Consequently, the manual is long and technically complex and takes some time

 

to determine the things that apply to DCC operation (yes it is able to operate in a number of systems including DC), and what of these may be relevant to your problem PJ.

 

Briefly, my conclusions run along these lines: the chances are the decoder isn't

 

faulty, rather needs some fine tuning; there doesn't appear to be any adjustment to delay before chuffs start or stop, as I suggested might be the problem; and the most likely area for adjustment would seem to be in load control which is set in CVs 53-56.

 

 

 

These Cvs are adjusted to ensure smooth operation of the loco without jerkiness as it runs or changes speed. You might remember there has been recent discussion in the forums on adjustment of these with blue box decoders, so taking a look at these

 

threads might be useful.

 

But if this is the problem, it will only be happening because there is some jerkiness in the operation of the loco, and the chuffs will be responding with this jerkiness. So the question PJ is whether you find the loco runs

 

as smoothly as your locos that have correctly operating sounds, or is it less smooth?

 

And in this regard, is there anything in the operating instructions about adjustments to CVs 53-56?

 

And Ray, having read the manual, I'm yet to be convinced

 

that reblowing does more than replace sound files themselves. When I first read yours saying it is more complicated, I thought this may be right as you can envisage that it might include the way sounds interact with movement. But beyond the fact that things

 

will be different for steam, diesel and electric, I think these interactions are pre-defined and so built into the decoder rather than part of what comes with particular sound files.

[Fishmanoz]

Just checked and the thread I'm referring to is https://www.hornby.com/forums/hornby-forums/hornby-digital/6615/ ignore that this thread is particularly about a diesel and concentrate on the adjustments needed.

[PJ_model_trains]

Sorry for not replying just set up new computer and new monitor ;-)

 

For more details including the benefits of the larger screen see details at

 

http://www.rmweb.co.uk/community/index.php?/topic/83539-rael-dale-valley-railway-building-modelling-signalling-and-more/

 

There

 

is a screen shot of what I had with a 50% view layout and what I have now with a 100% view layout (after adjustment)

 

Also we can see more trains ;-)

 

If Hornby do the improvements to loco boxes and icons it will be fantastic. Lots of potential.

 

I

 

have set up every loco again in RM more on this shortly.

 

Comments always welcome

 

PJ

[Ullswater]

Just read through your posts on rmweb PJ. Inspirational stuff! A very nice layout indeed and such rapid progress. Great job!

[PJ_model_trains]

@Ullswater

 

Many thanks for your kind comments. There is still lots more to do but it has been fun.

 

I am really chuffed with the higher resolution screen and to see more trains, what we need (I feel) is plenty of room for the layout (a little

 

crammed but I can see at 100% now) and more loco's so that we can sway and change with ease.

 

Oh and the new RM update, and the RM Pro and the Loco detection LOL and lots more too.

 

We are easy pleased ;o)

[Graskie]

Absolutely amazing, PJ. I just don't know how you manage to achieve so much in such a short time. Congratulations. Yes, it does look so much better. I just wonder how you're going to operate those 4.5 locos, though (lol).

[PJ_model_trains]

Hi Graskie

 

I have done it so many times before. Plan first and work through my list ;o)

 

4.5 loco's won't be hard. The Royal Signals with Sound is still not right. I will add a separate message about that shortly. (This is my half loco)

[PJ_model_trains]

@HMRC and all forum members who have helped to date

 

UPDATE RE: Royal Signals B*mann loco with Sound

 

NO REPLY from ESU!

 

I think this has been a good test but was surprised at the outcome.

 

NEW computer, NEW larger resolution touch

 

screen, NEW installation of RM

 

I have one by one re-installed all loco's back into RM

 

I did the Royal Signals steam loco first.

 

ALL loco's except the one above worked correctly.

 

Now the Royal Signals steam loco.

- when I got

 

it I had to change CV29 as forward was reverse and visa-versa.

- prior to uninstalling and reinstalling on a new computer you may remember I reset the chip to default settings and forward was forward?

- having installed it in the new computer from scratch,

 

forward is reverse again and visa-versa, Very strange?

 

As a final test I have run the accelerate and decelerate command program (basic program). Although the loco went in reverse instead of forwards, I ran it to the same program and speeds as was in

 

the program. It accelerated backwards and missed sound once or twice. It decelerated and missed sound 4 or 5 times, so the problem still exists but the changing of forward and reverse is a mystery and only leads me to feel now it is faulty.

 

I will compile

 

an email for eH and send it to them tomorrow. I do not feel not with this intermittent change it is worth considering re-blowing the sound.

 

Many thanks everyone.

 

PJ