faulty acc decoders??

[jbcontrol]

since my last problem with accessors decoder things were going ok, i had two sets 0f points wired up to one output port (r8247),both sets worked good along with all others from both decoders. both decoders are r8247's after a time the points operating from my second decoder stopped working. i checked that the outputs numbers were ok,and that the pulse durations were ok, which they were. the lights were working on the track plan when operated.since then i have tried several points motors but none of them operate makeing me think that the decoder has gone faulty.

  I also have a spare unused  r8247 decoder which i tried to set up, but the output port numbers were 6629 to 6632, and pulse durations were all 25500. i changed the port numbers to  5-8, and pulse durations each to 100. Then wrote to decoder. Just to check this, i read the decoder and it had reverted to the previous same very high figures.I have repeated this again, and even gone out of railmaster to try saving the inputs  but it still stays the same. could it be that i have got hold of a faulty decoder.  Any ideas,thoughts or remarks will be much appreciated. thanks.   john 

using railmaster with elink,running through windows 10 via laptop. using all hornby r8247 decoder,but a mixture of points, piont motors, even track.

 

   

[Chrissaf]

I use R8247s and find that they are rather prone to being corrupted by transient short circuits. Reading strange CV values in the 6,000 range has been reported before and seems quite a common fault. Also, re-configurations not sticking is also common.

.

A temporary fix is to perform a factory reset of the R8247 by writing value 8 to CV8, but you might find that the next time you have a 'short' on the track, the configurations of the R8247 are lost again. In the end, I replaced my errant R8247 that would not keep its configuration with a new one and added a snubber filter to protect the R8247 from further 'short circuit' induced damage. I found replacing the R8247 with a new one in addition to the snubber protection was my ONLY long term permanent solution to the problem.

.

It is not obvious how to factory reset a R8247 using RailMaster if you don't have RM ProPack. My post timed at 17:27 on the page this link goes to provides 'step by step' instructions. Of course, after performing the reset, you will have to reinstate the working configuration you want to use.

.

See my post timed at 20:32 on the page this link goes to for information on 'Snubber Filters'.

[jbcontrol]

thanks very much Chris for that reply.  i will try and do what you say. will reply again when i have a chance. the one that was playing up still says correct information upon reading the decoder. yes the fault was caused by a permanent short, which was only stopped by disconnecting the decoder. i have now refitted the same decoder, without having the short, but not checked the point motors operation.   The unused decoder i have followed the link to reset to factory settings, but it has not changed.(putting value 8 in cv8), twice tried.  thanks      John

[jbcontrol]

hi,i have tested the first mentioned points decoder which is refitted, but it is only trying to operate points, none of them actually move or change.  it's as hogh it has lost all power.  the first decoder is working ok. i.e.  output ports 1-4..  the second decoder  still has very high readings i.e 6629-6632 with pulse readings of 25500ms,although at  one stage port 4 (6632) was ticked for permanently on with a pulse reading of 1500..

 

   Chris,please what is a snubber filter, how or where can i get one, and how would i wire it in please.

thanks again  john

[Chrissaf]

Chris,please what is a snubber filter, how or where can I get one, and how would I wire it in please.

.

The answer to your question has already been posted above. See below for details of what and where:

.

Look to my post above yours timed at 11.33 - see the clickable link in it (as reproduced below).

.

/media/tinymce_upload/3f5a8c6cf7eff0f46d188168c8688ac5.jpg

.

If you follow that link, it will take you to this post timed at 20.32. The text on that page (reproduced highlighted in yellow below) provides the information you want. The clickable link (highlighted) takes you to web resource that discusses 'snubber theory' in great detail.

.

/media/tinymce_upload/d12443ba3ed6c3640665fee0e1f6f11d.jpg

.

[jbcontrol]

thank you very much for that information. i am in the process of collecting the item for the snubbers.   John

[96RAF]

I have seen odd port addresses when reading off in RM and also noted that when readdressing a decoder any previous port dwell time remained as was after the readdress.

You can try using the group address rather than the port address. Group 1  = ports 1-4; group 2 = ports 5-8; etc. This works at any level now that NMRA has deemed CV1 and CV513 are essentially the same. This CV513 thing is described in the Elite manual from memory.

I hate to ask but you are connecting the R8247 to the eLInk prog terminals for this exercise. It also helps to have one point motor connected to port 1 for programming and set the solenoid pin midway, then when programming is finished the port will throw and the solenoid will click to one side or the other indicating success.

Rob

[jbcontrol]

thanks Rob for that info, yes, i was connected to elink program, whilst doing this, and i tried to program both ways, ie,saying ports 1-4 and ports 5-8, as the latter was the numbers that i had in use before the problem.

I will try it with a points motor as you say.  i am wondering which optional make of points decoder would be recommended as i would like to try another make.  also i will be fitting a snubber to prevent shorts causing problems, as chris has advised, but still awaiting delivery

thanks to all   John