Changing CV's Sound and Speed

[PJ_model_trains]

I am currently reading and writing CVs

 

CV63 Sound - I am reducing the sound level on the Class 37 as it over powers the Steam loco with sound.

 

CV** Acceleration

CV** De-acceleration

 

There are a number of CVs for acceleration and

 

de-acceleration does anyone know which ones I should alter?

 

I have a Class 37 with sound which is very slow and a Class 37 which is to fast.

 

I may want to double head them at some stage so it's better if I level them up now.

 

Once I know

 

which CV to alter I will be looking at the other locos also, the DMU is far to fast and some of the other locos need tweaking a little.

 

Many thanks for your help as always

 

PJ

[Fishmanoz]

A range of questions here PJ, some pretty basic, some more advanced and some specific to your sound decoder.

 

You find the basics lic accel and decel are covered adequately in the RM manual. Then go to the old trusty source at http://www.brian-lambert.co.uk/DCC.html#Decoder_1

 

to see all CVs defined (that looks like it is the lighting link, CVs are immediately above that).

 

Note particularly the groups 33-46 and 47-64 which are used for sounds in your sound chip. You'll need to look at the chip spec to see what these are

 

used for so you can understand and adjust them. Clearly CV63 is volume as per previous thread discussion.

 

Now there is a lot of Cvs relating to speed - CV3,4,5,6,10,23-27,66-95. Again, some of these are simple and covered in the manual, some more complex.

 

For the more complex, DCCWiki is a good resource and you might start with speed tables at http://www.dccwiki.com/Speed_Table

 

Just on your Class 37s running at different speeds - this may be due to mechanical or electrical maintenance issues, but then

 

again it may be that you haven't set one or both up correctly in RM. Unless you have set them up as the correct model by costs loge number, RM will not be setting the correct shunt and cruise speeds for them. So double check them.

 

That lot should give

 

you hours of fun to read about and play with. Then come back with any specifics as there are some here with detailed knowledge on these matters.

[Fishmanoz]

Just correcting, lic should be like, and costs loge should be catalogue.

[Fishmanoz]

And remember, with the potential to make a large mess playing around with so many CVs, writing 8 to CV8 will get you back to scratch.

 

And if it's only the 8249 you are playing with, not much chance of getting in a mess as it doesn't support adjustment

 

of most of those advanced CVs. Don't know what your sound chip supports, again look at the spec, but you can have lots of fun with the 8245.

[PJ_model_trains]

Thanks Fishy

 

I saw there were large groups of speed Cvs in various areas, hence raising the question.

 

I compared some and the differences they were miles apart.

 

I will check RM tomorrow to make sure the problem doesn't lie there also.

 

It

 

would be so much better if we could print out the CVs or export then as csv files and compare one with another. It has been asked for previously, we wait patiently.

 

Mean while bye for now, looks like I'e been given some home work.

 

PJ

[Raven_Electric]

May I offer a suggestion PJ_model_trains. Before you alter anything read all the CVs into a spreadsheet. Perhaps use the left column for the CV numbers. Then a column for each decodors spec and then one for each loco pre-tweaking setting and then as many

 

columns as you want for sets of changes you make. I use column 2 for the NMRA spec then blocks of columns for each manufacturer with one column for each chip spec followed by a description column. I then just copy/paste columns for any loco/decoder setup so

 

it is next to the manufacturers spec to make changes easy. It sounds like a lot work and it is a first but the payback is invaluable once you are using it.

Between the CV rows you will need "sub" rows for bitmapped CVs eg:-

Hornby R8249 R8245

CV29

 

Decoder config

bit 0 Bit 0 set = reverse YES YES

bit 1

bit 2

bit 3 Bit 3 set = RailCom NO YES

etc...

bit 7

CV30 Fault indication NO YES

 

Another advantage of this is you can add cell comments for complex setups (which you need now)

 

for such things as CV33/36 the Function matrix (NO on R8249 but YES on R8245)

 

Another advantage for things like speed curves is you can use Excel functions to draw them with a simple copy/paste eg =(B2*1.07)/$B$1 and just paste the formula down the

 

range. Anf of course you can use your own or Excel's built in formulae for logarithmic changes etc.

 

I am sure Hornby will get there with automating flowing all this in Railmaster, there are a lot of demands on them for peoples' preferences as they

 

develop it. Meanwhile you can flow out to a spreadsheet and into DecoderPro from a spreadsheet either directly or via the JAVA data files. This makes reading/writing very big changes more or less instant.

 

I setup a script file (Excel macro actually)

 

to read everything into a spreadsheet from any new decoder I buy and then AND every bit by way of a test. Then the script file just loads what I want if things are OK. For such things as loading different train load profiles such as 400 ton 500 ton etc into

 

a few Gresley A4's this works like a dream because you only have a few tweaks to make which you can calculate and run out in cell functions.

 

Finally take care if you have a range of decoders there are huge differences between say a Lenz Gold and an

 

R8245 which you can see easily looking across the rows of a spreadsheet (and include pointers to the other chips equivalent if it has one) but manually - you'd go crazy.

 

 

 

 

 

 

 

 

eg:-

 

[Raven_Electric]

Sorry my post hasn't parsed properly. the column titles Hornby, R8249 and R8245 should be above the CV description column and then the YES, NO columns for each decoder showing if it is implemented.

[PJ_model_trains]

Raven_Electric said:

May I offer a suggestion PJ_model_trains. Before you alter anything read all the CVs into a spreadsheet. Perhaps use the left column for the CV numbers. Then a column for each decodors spec and then one for each loco

pre-tweaking setting and then as many columns as you want for sets of changes you make. I use column 2 for the NMRA spec then blocks of columns for each manufacturer with one column for each chip spec followed by a description column. I then just copy/paste

columns for any loco/decoder setup so it is next to the manufacturers spec to make changes easy. It sounds like a lot work and it is a first but the payback is invaluable once you are using it.
Between the CV rows you will need "sub" rows for bitmapped

CVs eg:-
Hornby R8249 R8245
CV29 Decoder config
bit 0 Bit 0 set = reverse YES YES
bit 1
bit 2
bit 3 Bit 3 set = RailCom NO YES
etc...
bit 7
CV30 Fault indication NO YES

Another advantage of this is you can add cell comments

for complex setups (which you need now) for such things as CV33/36 the Function matrix (NO on R8249 but YES on R8245)

Another advantage for things like speed curves is you can use Excel functions to draw them with a simple copy/paste eg =(B2*1.07)/$B$1

and just paste the formula down the range. Anf of course you can use your own or Excel's built in formulae for logarithmic changes etc.

I am sure Hornby will get there with automating flowing all this in Railmaster, there are a lot of demands on them

for peoples' preferences as they develop it. Meanwhile you can flow out to a spreadsheet and into DecoderPro from a spreadsheet either directly or via the JAVA data files. This makes reading/writing very big changes more or less instant.

I setup a script

file (Excel macro actually) to read everything into a spreadsheet from any new decoder I buy and then AND every bit by way of a test. Then the script file just loads what I want if things are OK. For such things as loading different train load profiles such

as 400 ton 500 ton etc into a few Gresley A4's this works like a dream because you only have a few tweaks to make which you can calculate and run out in cell functions.

Finally take care if you have a range of decoders there are huge differences between

say a Lenz Gold and an R8245 which you can see easily looking across the rows of a spreadsheet (and include pointers to the other chips equivalent if it has one) but manually - you'd go crazy.
eg:-

At last someone who thinks as I do,

and someone who is miles ahead of me!

Firstly spread sheets, I use Excel for everything, well almost. I learnt with the old SuperCalc, every cell was hand coded none of the easy stuff we take for granted in Excel. (Although I never got in to Makros)

Then

the miles ahead bit, I have been asking for exportable csv for CV's for months, I am sure Hornby will do this but we just have to wait. I would lay out the worksheet similar to what you say, I used to do similar when I was checking search engine optimisation

for our website. It is a great way to store and compare data.

Is there another way to export them?

[Fishmanoz]

Unfortunately I think it is called data entry at the moment.

 

I remember an Office style program that I used in the 80s called Open Access. The spreadsheet part was far more capable than anything else going at the time, but you had to set all the

 

cell parameters like you say PJ. Part of it is still around today because MS bought the database part and called it .......

[Raven_Electric]

Hi PJ_model_trains

 

Yes, spreadsheets are the "kryptonite" of the modern world! Absolutely everything in my life is done in spreadsheets, hyperlink driven correspondence databases, my diary, reminder pop-ups, stock control of everything and on....

 

And with conditional formatting and cell functions instantly to hand with DDE, RTD and OLE comms AND the power of either the included, interpreted VisualBasic Editor (The best free gift that ever came with anything, the documentation is awesome) or straightforward

 

calls to compiled programmes there is no limit to what you can do in terms of data processing. None of this is advertising though because Microsoft has no competitors in this field, Excel truly is now an industry standard and all credit is due to them.

 

 

Microsoft

 

ran with Multiplan on the MS-DOS launch Fishmanoz and the big daddy of the 1980s that revolutionised the world's business on MS-DOS and caused PCs to fly of the shelves was Lotus 1-2-3. Lotus failed to see GUIs importance and Microsoft ported their own Excel

 

to the PC from the Mac more or less for Windows 3.1 launch with Excel 2 and history was made and Lotus lost out. Excel quite literally did change the business world and private users have got the benefits too.

 

I have always liked Basic, it's in-built

 

text handling is handy but I like it most because of the no hassle interpreted run tests you can do before compiling. Because of the way compiling C works that is not easy in C the standard OS computer language. It is horses for courses, sometimes the efficiency

 

of C code is needed but sometimes Basic with super fast assembly machine code embedded does the job best. As a certain actor would say "Not a lot people know this" but one of Microsoft's first successful products before even MS-DOS was a Basic interpreter

 

programme and I was told B Gates played a major part in writing it, perhaps he will let us know if he reads Hornby Forums!

 

I don't buy into what the promoters claim about "Device" independent higher level interpreted languages, often glorified scripting

 

in my view. They are totally dependent on the speed and efficiency of the compiled operating system and applications are often limited, almost toy like because they often cannot use operating system resources properly!

 

The interpreted Basic in Excel

 

is no CLI simple language that schools got their heads round, it is Object Based and gives instant access to the entire resources of the Operating System AND, with care, it can be written to port to a Mac as fast as any "higher level" language with any meaningful

 

capability - I've done that with biggish progs more than once. I've gone on about this because it is the way to get the power of spreadsheets at your fingertips - have a good play around with it as a route to directly reading and programming your decoders

 

- fast and automatically. For the computer interfacings NETSTAT and TCP/IP utilities and their documentation will point you in the right direction,

 

You can quickly manually copy/paste the data in decoder manufacturers Pdfs on their decoders into a text

 

file of course then sort it into character separated fields to get it into a spreadsheet or just paste it into a spreadsheet cell (Into the formula bar box so it is read as text) in lumps and sort it from there. Some offer the information in xls/csv files

 

anyway. That way you can build up super accurate data (more than some manufacturers pdf files, but not Hornby I hasten to add, I have found their PDFs data sheets are pretty accurate) on decoders and what they can and can't do and where they do it.

 

Using

 

Import External Data in the Excel Data Menu (If you have it loaded for Tools|Customise) will guide you though mastering this using a text file. VITAL: ensure you add a comment to the decoder column's title cell with the VERSION number of the decoder's data

 

sheet you entered in your spreadsheet - if you don't you will cry later!!

 

Many decoders have (publicly) undocumented spare and private memory because obviously the things are build out of standard components. Once you automate read/writes of chips

 

you can see if that memory is used for anything - sometimes it is.

 

Building up the instructions/data from the read of the square wave timing requires memory of some kind and more complex control chips handle this in sophisticated routines using what

 

amounts to system memory. There is a lot of (publicly) undocumented stuff to enjoy playing with in this area! If you write to this memory, run the decoder and look for overwrites as there is sometimes interesting stuff and if you know the processor instruction

 

set you can easily figure out what is spare and what is happening - needless to say, this is at your own risk! By way of example I store loco names as ASCII on some decoders and a reference code for the profile in my spreadsheet that the decoder currently

 

has loaded. And on one make of decoder I can switch a profile matrix as the train is running by changing a pointer, useful if you decouple a carriage rake!

 

What you want of course is to be able to stream ALL CV settings into and from a decoder to

 

a spreadsheet or text file ASAP. Well, in time I am sure Railmaster will offer that facility, a key basic facility probably of more use to keen modellers than overriding decoder settings to control how fast a J39 goes in a hump yard!

 

But, THIS IS Hornby's

 

website and I did mention another shareware open source programme in my first post on this topic, DCC discussions will range wide but I think this needs to be said, Railmaster would be a massive project for even a specialist software company. People reading

 

this are enthusiasts but for us all to enjoy Hornby products the company has got to be able to ensure Railmaster and eLink work for everybody straight out of the box and that the product is appealing to all users or we will all loose out. Those of us who have

 

needed to contact them know Hornby back up Railmaster with quick, concise one to one support too. So I say we should hold back on criticism if they don't supply what we want immediately. Give the company headroom to develop Railmaster in response to what they

 

see as market demand, keep buying Hornby products and supporting the company, I am sure they will deliver most requests as development continues. Never forget UK outline OO is a tiny market in this global world and Hornby have kept faith with us even in the

 

face of very large international competitors. Hornby is a listed company so there is another way enthusiasts can offer support that appears on the bottom left of website menu if you get my drift!

 

So, for now yes there is a way of directly streaming

 

for enthusiasts with some computer skills who can automate linking it with spreadsheets and all credit to those enthusiasts behind it. It isn't a commercial product, it doesn't compete with Hornby in any way, nobody will hold your hand but for those that want

 

to try it it will get the job done. For myself, and only with the help of those enthusiasts' work I have written and compiled my own bit of software to automate the process in both directions. I have tried the most expensive controllers on the market and I

 

have ended preferring the Hornby Elite which is a solid product gives me tactile throttles at the rail side and despite comments on one would be "up market" retailer's website, works in a perfectly logical, practical way and is a sound investment with a degree

 

of future proofing and expandability. I am sure the eLink is likewise but for me trackside knobs to twirl are part of the fun. But because of the above there is a further comment I should make. The NMRA spec is licenced open source but the software link between

 

a computer and a manufacturer's controller is part of their proprietary product and belongs to them. Hornby and other manufacturers have every right to keep it that way and change the interface commands anytime and anyway they want to protect their product

 

so they can prosper and provide us with what we want. After all, Bill Gates has.

 

That is why my little programme indulges my interest in industrial control software and is password (and system) protected and goes no further than me. I am sure manufacturers

 

will not object to enthusiasts interest in these things provided we respect their rights. And all credit to Hornby for Railmaster, the more they show us the future the more we demand it now.

 

And a final cautionary point when using spreadsheets. Be

 

aware of the distinction between a cell's display format and what is actually in the cell. A Number may be 1.9 but the cell can be set to display rounded down to 1. If you don't handle this in code your data could round up to 2! Also, the cell could actually

 

contain an address pointing to a cell containing an address pointing to a cell containing the number 1. Cell A1 contains 1, Cell B3 contains =A1, cell D12 contains = B3 and displays 1. If your code picks up the cell content you get "=B3". Believe me this is

 

nothing to what you may have if you start using implied addressing, names and lookup functions,

 

Hope you find a few pointers here to help you to construct your decoder spreadsheet PJ_model_trains.

 

 

 

[PJ_model_trains]

@Raven_Electric:

 

Many thanks for your input, go have a brew now you deserve it.

 

A lot has been absorbed, some has gone over my head for now.

 

I remember when MS-Dos and PC-Dos were fighting for the market we all know the outcome. I

 

started with a Tandy green screen computer, then the BBC micro and added a second processor running PC-Dos. I then moved to Windows 3.1 and a newer machine and have had every version of windows from then to Windows 7. The only one I haven't had is Windows

 

8. I have also used Microsoft Office from the beginning with Excel being the most important piece of the package.

 

I am aware of the open source software but have not used it yet, I guessed that would be probably what you may be referring to, I will

 

play with it one day but have a lot going on right now.

 

A lot of praise for Hornby products, that is good and not just because it is their forum. I am sure they will be more than pleased with your comments about RailMaster and eLink, it has to be good

 

for future readers of the thread also. It is a good piece of software which can only get better but, as you say, the more they give us, the more we want but, it all takes time, we live in a 'want it now' age.

[Graskie]

Crikey, LC. I had to speed read that one, knowing that, if I read it at leisure, I still wouldn't understand it. I programmed in Atari Basic and then machine code but.....

 

Incidentally, because she mainly used it in her employer's finance department,

 

my wife persisted in writing letters in Excel rather than Word. I've cured her of that now, though.

[PJ_model_trains]

Fishmanoz said:

Unfortunately I think it is called data entry at the moment.

I remember an Office style program that I used in the 80s called Open Access. The spreadsheet part was far more capable than anything else going at the time,

but you had to set all the cell parameters like you say PJ. Part of it is still around today because MS bought the database part and called it .......

People now a days don't realise how long it took to code every cell in a spread sheet. All

this drag to cell *** and copy and paste makes it so easy.

[PJ_model_trains]

Fishmanoz said:

Unfortunately I think it is called data entry at the moment.

I remember an Office style program that I used in the 80s called Open Access. The spreadsheet part was far more capable than anything else going at the time,

but you had to set all the cell parameters like you say PJ. Part of it is still around today because MS bought the database part and called it .......

I always thought R2D2 and C3P0 were spread sheet codes... then I watched Star Wars!

[PJ_model_trains]

Fishmanoz said:

A range of questions here PJ, some pretty basic, some more advanced and some specific to your sound decoder.

You find the basics lic accel and decel are covered adequately in the RM manual. Then go to the old trusty

source at http://www.brian-lambert.co.uk/DCC.html#Decoder_1 to see all CVs defined (that looks like it is the lighting link, CVs are immediately above that).

Note particularly the groups 33-46 and 47-64 which are used for sounds in your sound chip.

You'll need to look at the chip spec to see what these are used for so you can understand and adjust them. Clearly CV63 is volume as per previous thread discussion.

Now there is a lot of Cvs relating to speed - CV3,4,5,6,10,23-27,66-95. Again, some

of these are simple and covered in the manual, some more complex. For the more complex, DCCWiki is a good resource and you might start with speed tables at http://www.dccwiki.com/Speed_Table

Just on your Class 37s running at different speeds - this

may be due to mechanical or electrical maintenance issues, but then again it may be that you haven't set one or both up correctly in RM. Unless you have set them up as the correct model by costs loge number, RM will not be setting the correct shunt and cruise

speeds for them. So double check them.

That lot should give you hours of fun to read about and play with. Then come back with any specifics as there are some here with detailed knowledge on these matters.

dccwiki page not available

I

am searching other items

PJ

[PJ_model_trains]

Some may find this worth reading...

 

http://www.dccconcepts.com/index_files/DCCdecoderfunctionsdesc.htm

 

WHAT IS A CV?

CV's 1 to 29

CV29 Is also the last of the “Basic set-up CV’s” you need to worry about.

 

[Fishmanoz]

PJ_model_trains said:

dccwiki page not available

I am searching other items

PJ

Just Google DCCWiki, then use their search function for speed table, which is what I did to get the link for you.

[Fishmanoz]

Both Brian Lambert and DCC Concepts have the basic info on CVs. DCCWiki has the more advanced stuff that will show you how to trim the decoder for better operation. Some will say that to get this stuff right, you need to use JMRI DecoderPro which allows

 

you more direct control and visibility of Cvs. Always worth checking out to see what you think. Using a combination isn't out of the question. Certainly Rm am is easier for the average user.

 

You might also Google CV29 calculator for a site that does

 

decimal to binary conversions of this Cvs for you, and more.

[Fishmanoz]

My most common typo on this site - Rm am instead of RM. Don't know how I do it but happens all the time.

[PJ_model_trains]

Fishmanoz said:

My most common typo on this site - Rm am instead of RM. Don't know how I do it but happens all the time.

Not sure if you are using pc, laptop or tablet. Looks like you have automatic capitalisation. So you

end sentence with .
You have capital R then like my tablet has just done it reverts to lower case. On my tablet if I want CAPITALS I have to hit shift twice and it changes colour.

[Fishmanoz]

iOS 7 autocorrect is the culprit. Yes I have to keeping touching the shift key unless I want to switch it to caps only, but why it adds the am when I fail to hold shift is what puzzles me.

[PJ_model_trains]

Fishmanoz said:

iOS 7 autocorrect is the culprit. Yes I have to keeping touching the shift key unless I want to switch it to caps only, but why it adds the am when I fail to hold shift is what puzzles me.

am
not sure either,

sorry.

[RDS]

@Raven_Electric

Fascinating. Thanks.

[PJ_model_trains]

Fishmanoz said:

PJ_model_trains said:

dccwiki page not available

I am searching other items

PJ
Just Google DCCWiki, then use their search function for speed table, which is what I did to get the link for you.

Try

it again Fishy... google search dccwiki come up no problem

Click google clink for site... no site?

[Graskie]

Using my favourites list I couldn't get that page either on http://www.dccwiki.com/CV. - " The server cannot find the requested page."

 

7am o'clock, Fishy and PJ?