Sudden changes of direction

[Andmat]

Just been running my new class 37 loco R3583TTS and a strange event started occurring, while running this loco round my track, it suddenly went into reverse for a round a metre then just as suddenly returned to normal running, I can't determine whether it's the loco or railmaster causing it, so far none of the other locos of various classes have done the same, however this has only happened recently and the loco ran fine on the first couple evenings.

I know I may need to experiment with it to get more data, but if anyone has had experience of this type of fault it could prove helpful in identifying what is going on.

Equipment being used is a laptop running railmaster pro 1.65.2 connected to Elink and Ipad ( IPad was not in use on all occasions of this event happening).

[Augustus Caesar]

Welcome to the forum Andmat...

A friend of mine had a short on his system which he didn't think too much about as it was quickly fixed.

He also noted one of his locos later (the only one apparently) doing what yours is and changed CV 29 from 49 to 50. All worked again.

Now, he reckoned this happened because of that short. While I cannot 'vouch' for his 'solution' (he reads this forum but is not a member) he usually gets stuff right.

Worth checking in the meantime until another member can give a clearer solution if this doesn't work.

 

I doubt you would have more than one loco with the same ID but even that may be worth a spot check.

[Chrissaf]

Transient short circuits can cause very strange things to happen with DCC !! Usually resulting with the need to factory reset the affected decoder (write 8 to CV8), then reinstate your original decoder settings in it.

[96RAF]

Very unusual. I wondered if you had been playing with programs and the 37 was just following instructions. Dirty wheels and/or track can fool a decoder as well.

 

Turning off DC running in CV29 is always a good thing. Put the loco on a separate programming track hooked up to the eLink PROG terminals and read its CVs, then double click (or is it right click) on CV29 value and a pop up tick list will appear. Amend as desired and then write these new CV values to the decoder with the pencil icon. See if that makes a difference.

Rob

[St1ngr4y]

This morning I took delivery of a R3592TTS Class 31. I went into the Locomotives window in RM to add the new loco, but it failed to locate R3592TTS in the database so I had to program all of the functions myself. I saved the details and the loco worked fine. But this isn't the reason for this post in this particular thread. Having run the new loco for an hour or so, I decided to run a schedule of programs, and all was fine until it came to run my Bachmann A1 "Sea Eagle", which is fitted with  a Loksound 4 decoder. The program which runs this loco starts the train with a Forward to [70] command. Today, when this was executed, it shot off at 70 in reverse. I managed to grab it before a crash...!! Then I remembered that there is something odd about this particular loco. Whenever I use the Locomotives window in RM, then for the rest of that RM session, the directions of this one loco are switched around, both when controlled by a program or by the RM throttles. The value of CV29 is 2. Its address is 0014 and it is in position 4 in the list of locomotives. Closing and restarting RM restored the loco to normal running. It must be something in RM which is causing this. Has anyone had any similar experience?

 

Apologies to Andmat for hi-jacking his thread. If the moderators think I should have started a new thread, then please feel free to move this post.

 

Ray

[96RAF]

@Ray

I think it is pertinent to the OP and could well be the answer.

 

Something has been fiddled with in RM and it plays up with various other things, if not first shut down and restarted to get its ducks back in a row.

 

If Andmat shuts down and restarts RM  -  is the problem still there with his 37.

 

Sounds like it could be an RM bug worth reporting in.

Rob

[St1ngr4y]

Hi Rob,

This afternoon, I changed the loco's address from 14 to 13, and moved it up the list from 4 to 3, but it still behaves in the way described. I returned it to its address 14. Then I restarted RM. I set the loco off going at shunt speed in reverse. While it was still moving, I opened up the locomotives window, did nothing in there, not even select a loco, then pressed the X button to exit. The loco throttles list refreshed itself on the screen. Then I clicked the Forward button for loco 14. It continued to move in reverse. Then I clicked the Reverse button, and sure enough, it changed direction and started moving forwards. So simply entering the locomotive configuration window and exiting without doing anything flips the direction controls for this, and only this, loco.

Weird, eh?

 

I'll wait to see if Andmat replies, then send a help request through the system.

 

Ray

[96RAF]

Sounds like a plan Ray.

[Britannia Builder]

Hi Ray

 

The very last column in Resource.mdb is named 'Direction' and in my case all the locos have this set to F. I see that there's a feature in ProPack (which I don't have) to reverse the direction controls for a particular loco - something to do with double heading when one of the locos is back to front, according to the manual. It would be interesting to see if your 'Sea Eagle' has this field set to something other than F (you can use Excel to display the contents of Resource.mdb). It wouldn't justify the strange behaviour, but it might be at the root of the problem.

 

Regards, John

[Andmat]

Hi all, Thankyou for your suggestions, I've rechecked all track wiring, checked point motor switches (found a few things of concern but nothing relating to the problem).

I changed CV29 whiCharger was on 38 , the suggestion was to 49 or 50 however the limit was 48, so set it to 48, its now been running round the track so far with no reoccurring event, although it sometimes ran previously for 10- 20 minutes with out fault, time will tell.

This does open up another question though, what exactly does CV29 Do, I've read the hornby leaflet, sorry I'm no wiser for it, would be happy if some one could put into English.

Im new to DCC and have found the forum very useful, Thanks.

[Chrissaf]

CV29 is all explained in great detail here:

http://www.2mm.org.uk/articles/cv29%20calculator.htm

.

The most common setting for CV29 is either 2 or 6 (if using short DCC addresses) OR 34 or 38 if using long DCC addresses.

.

Short DCC addresses are addresses in the range 001 to 127 and long addresses are addresses in the range 128 to 9999

.

By setting your CV29 value to 48, you have disabled 128 speed steps. This is not a good idea. The fix for "sudden changes of direction" is to disable 'DC Operation'. This is done by changing a loco with the current CV29 value of 6 to 2, and a loco with the current CV29 value of 38 to 34. I suggest you change your new 48 value to 34 instead, but only if you need the DCC address to be greater than 127, else change it to 2.

.

All these numbers will make sense, once you look at the information contained on the page I have given a link for.

.

EDIT: Since your loco is a Class 37. As a diesel it may have directional lights (or fitted later as an after market upgrade). Note that in many cases for directional lighting to work, the CV29 128 speed steps, more often than not, needs to be enabled. Hence the 'not a good idea' statement above. Another reason for changing your CV29 48 value to either 2 or 34.

[St1ngr4y]

Hi Ray

 

The very last column in Resource.mdb is named 'Direction' and in my case all the locos have this set to F. I see that there's a feature in ProPack (which I don't have) to reverse the direction controls for a particular loco - something to do with double heading when one of the locos is back to front, according to the manual. It would be interesting to see if your 'Sea Eagle' has this field set to something other than F (you can use Excel to display the contents of Resource.mdb). It wouldn't justify the strange behaviour, but it might be at the root of the problem.

 

Regards, John

Hi John,

 

Bullseye !! Well done. I looked at the Resource database using Excel and, sure enough, this loco was the only one with R in the Direction column instead of F. So then I searched the RM Guide and found this:-

 

"If you have the optional Pro  Pack installed you can set the loco to run in reverse just by pressing the direction indicator  to the left of the loco DCC ID field.  The arrow will change from right (forward) to left (reverse).  You can also program a loco to run in reverse by setting the reverse bit in CV 29 (see later in this guide). "

So here we are - this is what I found for the loco:-

/media/tinymce_upload/7a8bf7dc8c1ef48434c8d75a89c692c9.PNG

I had never noticed this small arrow on any of my numerous uses of this window. Maybe a descriptive label next to it might help 😉 Anyway, I switched this arrow to point to the right, saved the changes, closed RM and reloaded it. I tried out the loco using the throttle, half expecting that the directions would now be reversed permanently. But no, the loco behaved normally. So then I went into the locomotives window, and out again. The loco still behaved correctly. So my problem is now cleared. However, there must be a fault there somewhere, because for months, this loco has had its direction reversed by this arrow setting. Yet the only time this becomes apparent is after the locomotives window has been opened and closed. From that point, the reversed directions are actioned until the end of the RM session. But on reloading RM, this reverse directions marker for this loco is ignored.

 

Ray

[Britannia Builder]

 

Hi Ray

 

That's interesting. It's a rather odd bug, since the locos must be initialised from the data in Resource.mdb when RM is started up as well as when the loco setup window is opened. It doesn't seem to me to be a particularly useful feature, so perhaps nobody has noticed the bug before. Will you report it to HRMS?

 

I guess that this problem is not the same as the issue that Andmat reported at the start of this thread.

 

Regards, John

[St1ngr4y]

Hi John,

I agree - you would think it would be the same subroutine called at startup to that called after exiting from the locomotives window.  Anyway, I'll send a request to HRMS through the system tomorrow.

 

I also concur that this is different to problem which Andmat originally reported.

 

Ray

[96RAF]

Ray page 36 of the current RM manual talks to that blue arrow in Pro pack only.

Page 100 interestingly talks to a similar arrow in the consist set up table to reverse the direction of a loco within a double header, again Pro pack only but I just read on another forum that you can carry out this reversing action by adding 128 to the value seen in CV19, so Am pack people may be able to 'adopt' this Pro pack feature by stealth.

Rob

[Fishmanoz]

CV19 is of course consist address so how you can affect direction by adding 128 to it would seem to be a mystery.  Then again, maybe it realigns those electrical-thingumajigs and can be used instead of the freezer?

[St1ngr4y]

Ray page 36 of the current RM manual talks to that blue arrow in Pro pack only.

Page 100 interestingly talks to a similar arrow in the consist set up table to reverse the direction of a loco within a double header, again Pro pack only but I just read on another forum that you can carry out this reversing action by adding 128 to the value seen in CV19, so Am pack people may be able to 'adopt' this Pro pack feature by stealth.

Rob

Hi Rob,

If that worked, however, would it not mean that whenever the loco is used in a consist, it would always be running with direction controls switched?

Ray

[96RAF]

@ Ray

I honestly don't know without trying it and all my toys are 2000 miles away.

 

I think the CV19 setting only applies within consists and once a loco is back on solo settings the reverse arrow doesn't count.

 

I don't even know if the +128 thing even works as it was hearsay from another forum and Fishy has his doubts about it, so it needs proving on track.

 

Rob

[St1ngr4y]

 

Hi Ray

 

That's interesting. It's a rather odd bug, since the locos must be initialised from the data in Resource.mdb when RM is started up as well as when the loco setup window is opened. It doesn't seem to me to be a particularly useful feature, so perhaps nobody has noticed the bug before. Will you report it to HRMS?

 

I guess that this problem is not the same as the issue that Andmat reported at the start of this thread.

 

Regards, John

 

 

I am pleased to report that this minor bug appears to have been fixed in v1.66.

Ray