[markw]

A question for some of you techies

today i was setting up a new loco with my ecos. on the programming track i set the address to 3110 then moved the loco to the main and tried to run it but had no response, lights not working, basically nothing

i checked out the values of cv1, cv17 and cv18 which read back as 38, 192 and 100 respectively so selecting the address 38 i could then control the loco. using the calculator on this forum it shows 17 and 18 should be 204 and 38.

questions

what should the value of cv1 be.
why are 17 and 18 values different to the calculator
how do you calculate what the values of all 3 cv's should be for any 4 digit address.

i know about setting cv29 bit 5 so i don't think there was an issue with that. and i have succesfully setup most of my other locos with 4 digit adressing with no problem but most of those have either loksound or lenz decoders. This loco and another i tried have bachmann ez 2 function decoders (36-552 i think) cos thats all i had to hand.

Anybody out there care to help me try and understand this issue ?

Cheers

Mark

 

[Makemineadouble]

long address pdf

This should help explain

 

[markw]

QUOTE (Makemineadouble @ 20 Mar 2008, 23:53) <{POST_SNAPBACK}>long address pdf

This should help explain

Thanks very much, great document it explains quite a lot.

Now just need to find out about waht ahppens with CV1 and what the hecks going on with my ecos and these bachmann decoders.

 

[neil_s_wood]

If it's any consolation Mark, I have had the same issue on occaission. If I redo the address it will usually take after a couple of go's. Most times it works fine straight away but every now and again it seems to have an issue with cv17/18 long address.

 

[Richard Johnson]

QUOTE (markw @ 21 Mar 2008, 08:38) <{POST_SNAPBACK}>A question for some of you techies

today i was setting up a new loco with my ecos. on the programming track i set the address to 3110 then moved the loco to the main and tried to run it but had no response, lights not working, basically nothing

i checked out the values of cv1, cv17 and cv18 which read back as 38, 192 and 100 respectively so selecting the address 38 i could then control the loco. using the calculator on this forum it shows 17 and 18 should be 204 and 38.

questions

what should the value of cv1 be.
why are 17 and 18 values different to the calculator
how do you calculate what the values of all 3 cv's should be for any 4 digit address.

i know about setting cv29 bit 5 so i don't think there was an issue with that. and i have succesfully setup most of my other locos with 4 digit adressing with no problem but most of those have either loksound or lenz decoders. This loco and another i tried have bachmann ez 2 function decoders (36-552 i think) cos thats all i had to hand.

Anybody out there care to help me try and understand this issue ?

Cheers

Mark

***Hello Mark

Its confusing isn't it - thankfully all modern controllers take away the need to play with CV's 17 and 18 other than for curiosities sake. I have to confes I haven't done it manually for many years.

Basically you shouldn't need to even have to worry about this with ECOS or any current controller. however reading the above it looks to me like you have ended up with wrong settings in CV29, 17 and 18 (if you had truly programmed it to respond to long address, it shouldn't have responded to short address 38!)

I do sense some confusion as to where the "38" should have been - did you perhaps put it in CV1 instead of CV29 or CV18?

For the loco to respond to #3110, the settings should be as follows. (ONLY CV29, 17, 18 are relevant to long address...The setting in CV1 has no bearing on the long address setting - CV1 is simply the short address)

CV29 = 38 (making bit 5 high by adding decimal 32 to setting when short address is used tells decoder to respond to the long address)
CV17 = 204
CV18 = 38

The process for calculation is a pain - the following calculator will be useful - found at the top of the Moidel Rail Forum DCC section:

http://www.modelrailforum.com/forums/index...ress-calculator

Regards

Richard
DCCconcepts

PS: Neil - I am very surprised you have programming issues with ECOS - My experience with ECOS owners is that they often forget to carry out one step in the programming sequence, but if all boxes checked/actions carried out, it never fails as long as the wheels and programming track are clean.

REJ

 

[Doug]

When a controller sets the long address, CV1 is left alone, so you can set it yourself - in case you use the loco on a system that only handles short addresses or leave it at the default address of #3.

 

[Chinahand]

QUOTE (Richard Johnson @ 21 Mar 2008, 09:41) <{POST_SNAPBACK}>***Hello Mark

Its confusing isn't it - thankfully all modern controllers take away the need to play with CV's 17 and 18 other than for curiosities sake. I have to confes I haven't done it manually for many years.

Basically you shouldn't need to even have to worry about this with ECOS or any current controller. however reading the above it looks to me like you have ended up with wrong settings in CV29, 17 and 18 (if you had truly programmed it to respond to long address, it shouldn't have responded to short address 38!)

I do sense some confusion as to where the "38" should have been - did you perhaps put it in CV1 instead of CV29 or CV18?

For the loco to respond to #3110, the settings should be as follows. (ONLY CV29, 17, 18 are relevant to long address...The setting in CV1 has no bearing on the long address setting - CV1 is simply the short address)

CV29 = 38 (making bit 5 high by adding decimal 32 to setting when short address is used tells decoder to respond to the long address)
CV17 = 204
CV18 = 38

The process for calculation is a pain - the following calculator will be useful - found at the top of the Moidel Rail Forum DCC section:

http://www.modelrailforum.com/forums/index...ress-calculator

Regards

Richard
DCCconcepts

PS: Neil - I am very surprised you have programming issues with ECOS - My experience with ECOS owners is that they often forget to carry out one step in the programming sequence, but if all boxes checked/actions carried out, it never fails as long as the wheels and programming track are clean.

REJ

Hi Richard,

Just found the CV calculator your link pointed to. What a useful litle tool that is. Is there any way of saving this calculator to my PC ???

Regards,

Expat.

 

[Doug]

QUOTE (Expat @ 21 Mar 2008, 09:18) <{POST_SNAPBACK}>...
Just found the CV calculator your link pointed to. What a useful litle tool that is. Is there any way of saving this calculator to my PC ???
...

You can bookmark the link

See the other DCC resources at the top of the page.

 

[markw]

QUOTE (Richard Johnson @ 21 Mar 2008, 05:41) <{POST_SNAPBACK}>Basically you shouldn't need to even have to worry about this with ECOS or any current controller. however reading the above it looks to me like you have ended up with wrong settings in CV29, 17 and 18 (if you had truly programmed it to respond to long address, it shouldn't have responded to short address 38!)

I do sense some confusion as to where the "38" should have been - did you perhaps put it in CV1 instead of CV29 or CV18?

Hi Richard

The initial programming was done by simply setting the address as 3110 in the add a new loco screen. It was only when it didn't respond to that address that i decided to read the cv's to try and fathom why. As i said i've done this a good few times before with no problem but last night two locos same problem. The only things different were the bachmann decoders (instead of lenz or esu) and i think it's the first time i've programmed since updating the ecos with the 1.1.1 firmware installed.

I'm going to have another play hopefully later today, i'll let you know what happens.

Thanks

Mark

 

[dwb]

QUOTE PS: Neil - I am very surprised you have programming issues with ECOS - My experience with ECOS owners is that they often forget to carry out one step in the programming sequence, but if all boxes checked/actions carried out, it never fails as long as the wheels and programming track are clean.

I have seen the same problem as Neil on the ECoS where the change from long to short or vice versa does not seem to "stick". Finger trouble? Maybe. Dirty track, more likely as I have found the ECoS programming track to be very pernickity about absolute cleaniness. I find it helps a lot to have pickup tenders attached to steamers to maximize pickup.

The ECoS appears to remember the current setting, so if it didn't work the first time, you need to change it back on the ECoS before trying again. It can be very frustrating.

David

 

[neil_s_wood]

QUOTE (Richard Johnson @ 21 Mar 2008, 16:41) <{POST_SNAPBACK}>PS: Neil - I am very surprised you have programming issues with ECOS - My experience with ECOS owners is that they often forget to carry out one step in the programming sequence, but if all boxes checked/actions carried out, it never fails as long as the wheels and programming track are clean.

REJ
Hi Richard, this issue has been around for a while. While I cannot completely rule out dirty track as I have not chemically cleaned it, it is pretty clean, and clean enough to do long addresses and all other sorts of programming 75% of the time.

My best guess was that as the bulk of my decoders were Lenz there were some issues with programming Lenz. Ultimatley I have just accepted this as a quirk of the system.