MFX decoder sound on analogue power...?
- Thread starter Zerogee
- Start date
Greg Elmassian
Guest
A decoder can have 3 different addresses:
1 short address
2 long address
3 consist address
all 3 can be present, but you need to know which is active
please read back cv 29 and cv50
Greg
1 short address
2 long address
3 consist address
all 3 can be present, but you need to know which is active
please read back cv 29 and cv50
Greg
MTheStrong
Registered
Hi Jon,
I have been chatting with a friend in Germany about your issue. It appears that Maerklin disabled some factory installed decoders from being able to write or change certain CVs. You can read the CVs but cannot change all of them. He is not aware that there is a decoder lock as per Massoth. He also went on to say that in the case of CV50, even though you may not be able to alter the value, the decoder will look for MFX then DCC etc. There is likely to be a time lag whilst the decoder searches for the correct mode.
Regards
Martin
I have been chatting with a friend in Germany about your issue. It appears that Maerklin disabled some factory installed decoders from being able to write or change certain CVs. You can read the CVs but cannot change all of them. He is not aware that there is a decoder lock as per Massoth. He also went on to say that in the case of CV50, even though you may not be able to alter the value, the decoder will look for MFX then DCC etc. There is likely to be a time lag whilst the decoder searches for the correct mode.
Regards
Martin
Zerogee
Clencher's Bogleman
Thanks chaps.
Greg:
CV50 value reads back as 15 (00001111), CV13 reads 71 (01000111).
This is interesting, as according to the loco manual the defaults should be 10 for CV50 and 64 for CV13.
Note this was a brand-new loco purchased from a German dealer, so everything should have been set as it was at the factory - it's a little odd that some of the settings don't match the defaults in the manual.
I'll check CV29 tomorrow, had to move everything off the table for the evening...!
Martin - thanks for raising the question with your friend, but the loco manual indicates that CV50 is on the list of those that can be changed - it gives alternative values for it in the CV list, anyway.....
Jon.
Greg:
CV50 value reads back as 15 (00001111), CV13 reads 71 (01000111).
This is interesting, as according to the loco manual the defaults should be 10 for CV50 and 64 for CV13.
Note this was a brand-new loco purchased from a German dealer, so everything should have been set as it was at the factory - it's a little odd that some of the settings don't match the defaults in the manual.
I'll check CV29 tomorrow, had to move everything off the table for the evening...!
Martin - thanks for raising the question with your friend, but the loco manual indicates that CV50 is on the list of those that can be changed - it gives alternative values for it in the CV list, anyway.....
Jon.
Hi Jon,
I tried changing CV 1, and CVs 13 and 14 using Massoth MST and a PC Module on one of my LGB mxf locos, a RhB Ge 2/4, No. 24450. Just like you, I was not able to change any CVs using MST either. MST comes back with CV read or write errors. The read or write operation fails completely after the third attempt. So you are not alone!
I was able to adjust the analog operating characteristics of the mfx decoder with my CS3 as one would expect. A few of things I can verify.
First, CVs 13 and 14 are bitwise switches. I enabled cab lighting (the default is off) and it functions as expected. The forward or rear cab light now comes on in analog operation depending on direction of travel.
Sound is on by default in Märklin mfx decoders, so enabling F8, Sound on/off for analog turns sound off (just like activating F8 under mfx or DCC turns sound off). You should not have to mess with this bit if you do want analog sound.
CV 176, Minimum speed in analog DC, does change the voltage at which the motor starts operating. This acts as an offset so the motor speed still starts slowly. The CV 176 default for a 24450 is 40. Under analog operation the lights and sound fire up between 4.5 and 5.25 volts. The wheels start to turn at about 8.5V with CV 176 set to 40. This CV is one you'll probably want to change on your Borsig when you enable the operating sounds. If I remember the Borsig default was 25.
CV 177, Maximum speed in analog DC, works just like CV 5, Maximum speed in mfx and DCC. It sets an upper limit to the analog speed. CV 5 is for digital only in these Märklin decoders and does not affect analog operation. I can see this CV being very useful if kids are involved. The default value of CV 177 in my 24450 is 250 and it really flies! I reset it to 150 which is much more reasonable.
By the way, I was using a LGB 50060/110 transformer and 5007 Controller combo for analog DC testing. Max. voltage out is 22V DC. Your results may vary.
The CS3 changes mfx decoder CVs indirectly through a graphical interface so it takes a little getting use to. It is quite easy though once you get the hang of it.
While I was playing I changed the volume of some of the individual sounds in my loco. All in all this was a very useful exercise in learning more about programming mfx decoders and getting more out of my CS3.
I'm going to try using my ESU programmer next to see if I can access the CVs directly.
But, getting back to your specific issue, I suggest you take your loco back to the friend with the CS3 to make the changes you want once you tire of
.
I tried changing CV 1, and CVs 13 and 14 using Massoth MST and a PC Module on one of my LGB mxf locos, a RhB Ge 2/4, No. 24450. Just like you, I was not able to change any CVs using MST either. MST comes back with CV read or write errors. The read or write operation fails completely after the third attempt. So you are not alone!
I was able to adjust the analog operating characteristics of the mfx decoder with my CS3 as one would expect. A few of things I can verify.
First, CVs 13 and 14 are bitwise switches. I enabled cab lighting (the default is off) and it functions as expected. The forward or rear cab light now comes on in analog operation depending on direction of travel.
Sound is on by default in Märklin mfx decoders, so enabling F8, Sound on/off for analog turns sound off (just like activating F8 under mfx or DCC turns sound off). You should not have to mess with this bit if you do want analog sound.
CV 176, Minimum speed in analog DC, does change the voltage at which the motor starts operating. This acts as an offset so the motor speed still starts slowly. The CV 176 default for a 24450 is 40. Under analog operation the lights and sound fire up between 4.5 and 5.25 volts. The wheels start to turn at about 8.5V with CV 176 set to 40. This CV is one you'll probably want to change on your Borsig when you enable the operating sounds. If I remember the Borsig default was 25.
CV 177, Maximum speed in analog DC, works just like CV 5, Maximum speed in mfx and DCC. It sets an upper limit to the analog speed. CV 5 is for digital only in these Märklin decoders and does not affect analog operation. I can see this CV being very useful if kids are involved. The default value of CV 177 in my 24450 is 250 and it really flies! I reset it to 150 which is much more reasonable.
By the way, I was using a LGB 50060/110 transformer and 5007 Controller combo for analog DC testing. Max. voltage out is 22V DC. Your results may vary.
The CS3 changes mfx decoder CVs indirectly through a graphical interface so it takes a little getting use to. It is quite easy though once you get the hang of it.
While I was playing I changed the volume of some of the individual sounds in my loco. All in all this was a very useful exercise in learning more about programming mfx decoders and getting more out of my CS3.
I'm going to try using my ESU programmer next to see if I can access the CVs directly.
But, getting back to your specific issue, I suggest you take your loco back to the friend with the CS3 to make the changes you want once you tire of
.
Last edited:
Zerogee
Clencher's Bogleman
Phil, thanks, that's very helpful.... and also quite frustrating, that it's not possible to adjust CVs on an MFX decoder via a DCC system....
As you suggest, I'll have to talk to someone with a CS3 - Chris (who isn't a member of GSC) is away at the moment but once he's back I'll see if he is able to come over and see me and bring his CS3 so that we can have a play around with it, with the ability to test stuff between the CS3, my Massoth and the analogue DC (which is on 22v like yours, using one of the little 50080 1-amp units).
Thanks to all for the help and suggestions.
Jon.
As you suggest, I'll have to talk to someone with a CS3 - Chris (who isn't a member of GSC) is away at the moment but once he's back I'll see if he is able to come over and see me and bring his CS3 so that we can have a play around with it, with the ability to test stuff between the CS3, my Massoth and the analogue DC (which is on 22v like yours, using one of the little 50080 1-amp units).
Thanks to all for the help and suggestions.
Jon.
The Shed
Citizen of the Republic of the North East.
Jon,
Back to basics, reading your earlier post's you were tinkering around with it on DC.
Did you put the loco back onto a DCC track and rerun the loco on it's default setting of address 3?
The 55028 (& 55029) decoder has a habit of remembering it's last protocol, and often needs a kick up the electrons to re activate itself to take account of the newest and latest protocol to use.
This oddity has been highlighted previously when from New out of the box the decoders in LGB models are by default in mfx mode, a short whirl up the track in DCC mode, decoder now functional and programmable in DCC.
Several other oddities manifest themselves when programming the 55028 decoder, one is the Navigator combination to change address,
Back to basics, reading your earlier post's you were tinkering around with it on DC.
Did you put the loco back onto a DCC track and rerun the loco on it's default setting of address 3?
The 55028 (& 55029) decoder has a habit of remembering it's last protocol, and often needs a kick up the electrons to re activate itself to take account of the newest and latest protocol to use.
This oddity has been highlighted previously when from New out of the box the decoders in LGB models are by default in mfx mode, a short whirl up the track in DCC mode, decoder now functional and programmable in DCC.
Several other oddities manifest themselves when programming the 55028 decoder, one is the Navigator combination to change address,
Last edited:
Thank you Shed - you have provided the key! Put the loco on a DCC track and give it a run before you try to make any CV changes. You'll need to do this even if you run it on a analog system. Otherwise, you'll get nothing or garbage back when trying to read or write CVs using a DCC system.
I turned off the mfx protocol on my Märklin CS3 central station and ran my mfx chipped LGB 24450 using DCC. I then tried reading and changing some CVs using MST (Massoth Service Tool) and my Massoth PC Module. It worked both reading and writing individual CVs! I then successfully tested the loco using an analog controller to verify the changes.
However, when I went to restore the CVs to what I originally wanted, again using MST, it had difficulty and was returning garbage on CV reads. Running the loco using DCC again reset whatever was going on in the decoder. I was then able to read and write CVs using MST again, putting them back to what I wanted.
I again verified the things were back to what I wanted using an analog controller.
So, run that Märklin mxf chipped loco on a DCC system before trying to make any CV changes using DCC!
I turned off the mfx protocol on my Märklin CS3 central station and ran my mfx chipped LGB 24450 using DCC. I then tried reading and changing some CVs using MST (Massoth Service Tool) and my Massoth PC Module. It worked both reading and writing individual CVs! I then successfully tested the loco using an analog controller to verify the changes.
However, when I went to restore the CVs to what I originally wanted, again using MST, it had difficulty and was returning garbage on CV reads. Running the loco using DCC again reset whatever was going on in the decoder. I was then able to read and write CVs using MST again, putting them back to what I wanted.
I again verified the things were back to what I wanted using an analog controller.
So, run that Märklin mxf chipped loco on a DCC system before trying to make any CV changes using DCC!
Last edited:
PhilP
G Scale, 7/8th's, Electronics
As Phil say's.. Run it on DCC first.. What address will it respond to? - You said it was set to '7' but displayed '3'??
Do NOT use the Massoth 'Change address' function to change the address of MFX decoders.. - The 'Change address' function change a number of CV settings, as well as CV1. Fine for a Massoth decoder, but not a MFX.
This means you have to change CV1, and if you want a long address, CV's 17, 18, and 29. - All having done the calculations 'long-hand'.
PhilP.
Do NOT use the Massoth 'Change address' function to change the address of MFX decoders.. - The 'Change address' function change a number of CV settings, as well as CV1. Fine for a Massoth decoder, but not a MFX.
This means you have to change CV1, and if you want a long address, CV's 17, 18, and 29. - All having done the calculations 'long-hand'.
PhilP.
Zerogee
Clencher's Bogleman
Thanks Shed and both Phils, this is all very interesting.... I am familiar with the need to run a new MFX loco on DCC in order to set it onto that protocol before trying to reprogram anything including the loco address, but in this case I HAVE been running it on DCC (alternating with testing on DC) in my attempts to get the sound to work - or to keep working - on DC....
This is the sequence of events, as best I can recall them:
1) Brand-new, box-fresh loco from German dealer; on arrival, I'm pretty sure that I briefly tested it on Massoth DCC and found it to be running and all functions working fine, didn't at that point try to change loco address or do anything else with it; back in the box, until....
2) recently, took loco to friend's line and ran it on his new CS3 MFX system; all working fine, sound ON all the time we were running it.
3) some days later, took it to another friend's line with DC only, loco ran perfectly but NO SOUND.
4) brought it home, tried it on my switchable DC/DCC test track - first on DC, still no sound; switched to DCC (Massoth) and was able to turn sound ON with F6. Switched back to DC, sound CONTINUED to operate (note that the loco was in all cases STOPPED before switching over power supplies, not just toggled over in mid-run!), but after loco was then stopped again and re-started the sound defaulted to OFF. Was able to repeast this several times - switch to DCC and turn sound back on via F6, stop loco and switch to DC, sound works when first re-started but then ceases again after loco is stopped and left for a few seconds.
5) moved loco to Prog track (my test rig has one track dedicated to programming and connected only to the prog outputs of the 1200Z, and a separate "running" track with rollers that is switchable between the DCC track output of the 1200Z and the LGB 50080 DC power pack) and attempted to read some CVs on DCC - read attempts successful, only oddity being that CV1 (loco address) reads back as "7" while loco responds to address "3"! All attempts to WRITE to CVs result in no change and an "X" instead of the expected tick agains the CV number on the Navvy screen.
Later today I will try everything again from first principles, running the loco on DCC on the test rollers for several minutes to see if that re-sets anything.... if still no luck, Martin has very kindly offered to pop round later in the week and bring his CS2 so that we can have a go at it with that....
Jon.
Edit: PhilP - yes, I'm familiar with the need to change CV1 rather than use the "set address" function on the Navvy - CV1 reads back as "7", but loco still responds only to address "3"....
This is the sequence of events, as best I can recall them:
1) Brand-new, box-fresh loco from German dealer; on arrival, I'm pretty sure that I briefly tested it on Massoth DCC and found it to be running and all functions working fine, didn't at that point try to change loco address or do anything else with it; back in the box, until....
2) recently, took loco to friend's line and ran it on his new CS3 MFX system; all working fine, sound ON all the time we were running it.
3) some days later, took it to another friend's line with DC only, loco ran perfectly but NO SOUND.
4) brought it home, tried it on my switchable DC/DCC test track - first on DC, still no sound; switched to DCC (Massoth) and was able to turn sound ON with F6. Switched back to DC, sound CONTINUED to operate (note that the loco was in all cases STOPPED before switching over power supplies, not just toggled over in mid-run!), but after loco was then stopped again and re-started the sound defaulted to OFF. Was able to repeast this several times - switch to DCC and turn sound back on via F6, stop loco and switch to DC, sound works when first re-started but then ceases again after loco is stopped and left for a few seconds.
5) moved loco to Prog track (my test rig has one track dedicated to programming and connected only to the prog outputs of the 1200Z, and a separate "running" track with rollers that is switchable between the DCC track output of the 1200Z and the LGB 50080 DC power pack) and attempted to read some CVs on DCC - read attempts successful, only oddity being that CV1 (loco address) reads back as "7" while loco responds to address "3"! All attempts to WRITE to CVs result in no change and an "X" instead of the expected tick agains the CV number on the Navvy screen.
Later today I will try everything again from first principles, running the loco on DCC on the test rollers for several minutes to see if that re-sets anything.... if still no luck, Martin has very kindly offered to pop round later in the week and bring his CS2 so that we can have a go at it with that....
Jon.
Edit: PhilP - yes, I'm familiar with the need to change CV1 rather than use the "set address" function on the Navvy - CV1 reads back as "7", but loco still responds only to address "3"....
Greg Elmassian
Guest
It seems that Phil can read and write CV's but you cannot. I would keep trying, hard to believe the decoder cannot be put into a state where the DCC standard of reading and writing CVs in service mode does not work.
Greg
Greg
PhilP
G Scale, 7/8th's, Electronics
Ah, but you are forgetting two points:
1. It is a 'MFX' decoder, playing at supporting DCC.
2. German logic!
Greg Elmassian
Guest
Sorry: Phil S says he can read and write CVs
Jon can't...
So just commenting that Jon should try to get where Phil S is... run on DCC, then put on DCC programming track (NMRA service mode) and see of CV's can be written and read properly.
Greg
Jon can't...
So just commenting that Jon should try to get where Phil S is... run on DCC, then put on DCC programming track (NMRA service mode) and see of CV's can be written and read properly.
Greg
Zerogee
Clencher's Bogleman
Going to try that (again) when I get the chance - kind of run out of time today with mix of work and Real Life(tm), so may have to be tomorrow evening....
Jon.
MTheStrong
Registered
I have just setup my own LGB 20752:
Here with a screen print of the data taken from CS2:
Here with a screen print of the data taken from CS2:
MTheStrong
Registered
Note in my case the address is shown as 11. However on my system I only use MFX and so the actual address, for me, is irrelevant because I select the loco from its image. Apologies for the reflection on the CS2 screen as I took the photo in the conservatory
MTheStrong
Registered
Hi Jon,
Just a thought, 995015 is my 11th MFX loco. I wonder if yours is your 7th MFX loco?
Regards
Martin
Just a thought, 995015 is my 11th MFX loco. I wonder if yours is your 7th MFX loco?
Regards
Martin
Zerogee
Clencher's Bogleman
No, only my second (though I now have a third....) - but as I don't have a CS controller, I don't think it would apply in my case anyway.
Though actually, thinking about it, we put the Borsig onto Chris' CS3 controlled line... I wonder if it might possibly be the seventh loco that he had registered on his system?
Jon.
Greg Elmassian
Guest
I have to comment that adding a "loco number" that can vary by system is nuts.... the DCC address is constant and easy, especially since 4 digits can usually map to the road number easily.
Extra unneeded complexity.
Greg
Extra unneeded complexity.
Greg
Mfx decoders retain whatever DCC address a person assigns to. It always has the same address on a DCC system. It is only when a mfx loco registers itself on a mfx system that it receives a "unique" id. The ID number is pretty much irrelevant as locos are selected by icon or name. Also this assures the loco does indeed have a unique id and does not replicate one already registered.