Jump to content

DCC Chips blowing


Recommended Posts

Yesterday I went to change a CV setting on a Hornby Diesel Shunter. It had been running well after we'd fitted a new Hornby 8 pin DCC chip to replace the factory fitted one which had blown.

 

I carefully put it on to my programming track, connected to the program output on my ELink, read the CV settings and they were all coming up 255 then "cannot be read". Now it won't run either.

 

I had previously blown several chips and I had put it down to the fact I had been using my old Select controller to quickly change ID numbers, until I'd found the voltage across the track with that connected to my piece of programming track was 30volts! So I had assumed it was that which was blowing the chips.

 

Now I am at a loss to know what is happening. I don't want to risk any more locos on the programming track until I have found out. I measured across the track with my multimeter and the voltage is zero which I assume is correct. Are there any other tests I can do to narrow it down? Or can anyone suggest what else could be going on?

 

Thanks.

Link to comment
Share on other sites

R8249 Hornby decoders

I am having the same problem. stating.Decoder not in database,

This decoder originally used and programmed using RailMaster.

Have been in touch with RailMaster support they told me decoder was in database but CV7 should read 131 Tried this with no success all my locos with this decoder installed reporting same fault and unable to to used at all.Have now gone back to Hornby support with copy of Decoder Technical Specification this states CV7 should be 013 so database is incorrect So decoder unable to be recognised

Link to comment
Share on other sites

There is no output to the programming track until the controller is actually sending data.

 

Also programming output is current limited (200mA) specifically to protect a decoder against damage in the event of duff wiring, so why yours is burning out chips is a mystery.

Link to comment
Share on other sites

Hi Chris. 

It's D4093 which came with Mixed Goods - Digital Train Set, R1075.

.

In that case, as Rob says it is a mystery. The only thing I could suggest is a DC current test to see if the motor is trying to draw excessive current through the decoder, perhaps due to a fault. The easiest way to do this would be to replace the decoder with a blanking plate (to remove the decoder from the test and make it possible to directly read the motor DC current). Using DC for the current test, rather than DCC will give a far more accurate reading. Put the loco on a small test piece of track (maybe your programming track piece for example). Then connect a 12 volt DC supply via a series connected multi-meter set to read a 1 amp DC scale to the track and measure what the loco draws. You may need to briefly let the loco wheels slip on the track just long enough to take the reading.

.

You mentioned 30 volt reading in your post. On the main track, the DCC signal if looked at with a scope will show a 'peak to peak' voltage reading typically about 28 volts. This waveform is a square wave at about 8 to 12Khz frequency. When measured with a digital multimeter set to read AC volts, this waveform should show a voltage about 15 volts give or take. You didn't say what type of instrument you used to measure this 30 volts. Rob is right, in normal use you shouldn't be able to read any voltage on the 'programming' track. The programming data signal is heavily controlled and only present when data bits are being sent to the decoder. The pulses of programming data are so short, that the voltage won't show on a normal digital voltmeter due to the measuring latency of the meter. One would realistically need a scope (Oscilloscope) to see and measure the waveform.

.

Another possibility (if the current test described above does not indicate any problems i.e current is as expected). Given that the 08 shunter has no lights ?. It may be that there is a wiring fault such that there is a partial short across one of the decoder function outputs, causing excessive current through the decoder. See if you can isolate these unused function wires. Or at least measure the resistance between the blue wire pin and function pins on the loco socket with the decoder removed. I would expect these to be 'open circuit' and not in contact with other wires or the metal parts of the loco chassis.

.

Another possibility, is there sufficient insulation underneath the DCC socket. It has been known for decoder plug pins to protrude too far through the socket and touch metal parts underneath. This could potentially provide a current path on unused function wires (see paragraph section above).

.

/media/tinymce_upload/458a3c13d69e9bb534dd87bf211eeefa.jpg

.

Drawing courtesy of Brian Lambert's web site.

 

Link to comment
Share on other sites

Chris, mixed freight locos use the 4-pin decoder and I don't believe there is a blanking plate for them.  So unfortunately BEJ, the 8-pin advice isn't relevant. 

 

But there is still a simple way to isolate the problem. Swap your decoders between locos. If it is the decoder, the fault will move to the swapped loco. If it is the loco, the fault will stay put. I'm assuming the other loco works although you haven't said. 

Link to comment
Share on other sites

Thanks for all the advice. The friend who helped me install the new decoder on this loco (the soldering inolved was way beyond my capabilities) and who is a lot more savvy on these issues than me is coming back tomorrow morning to give me a hand so we will have a close look at it.

 

Thanks again.

Link to comment
Share on other sites

Can I ask a supplemental? The decoder we fitted to the shunter was a Hornby R8215. Can this be reset by writing to one of the CVs? The instructions don't mention this and it seems CVs, 1, 3, 4 and 29 are the only ones in the "supported" list. 

Link to comment
Share on other sites

Strange,  Hattons show the R8215 that BEJ installed as an 8 pin decoder. Not 4 pin as Fishy intimated. I assume the orignal was a 4 wire decoder and BEJ's loco has since been significantly modified to fit the R8215.

.

As BEJ infers that the decoder is hard wired in to the loco and not connected to a socket, then I think my previous conmments about function wires touching something they shouldn't is still potentially relevant.

.

Someone can correct me if I am wrong (again), but I was under the impression that the R8215 had loads of issues with it and should be avoided. If you can read CV8 on the R8215 (value 48?) then you should be able to reset it with writing value 8, but again only a "should" as I have never owned a R8215

Link to comment
Share on other sites

My mistake Chris, missed BEJ has said he replaced the factory fitted with the old type 8215 which must have been hanging around in someone's odds and ends bin.  Would have been simpler to unplug the 4-pin X9___ that is fitted in these locos, at least it was in the set I bought, and plugged in a new one.  And would not make sense to then add an 8-pin socket to which you could fit a blanking plug as you would very quickly be running out of room, and no chance of doing it with the Jinty in that set as there would be even less room (both have the same chassis).

Link to comment
Share on other sites

The R8215 can be reset by writing value 8 to CV8 but it is a poor decoder, susceptable to self resetting to address 03 at the drop of a hat and it would be better to replace it with a newer variant, either the proper four pin one Fishy speaks about or at least an R8249 decoder If you pal is hard wiring it in.

 

Rob

Link to comment
Share on other sites

Sorry to report, no progress. Basically because we got sidetracked onto another issue (non-working Windsor Castle!). This brought to light the fact that my programming track was not working properly. I think it was down to the fact I was using a powerclip and cable (as supplied by Hornby with the R8247 decoder) where the pins don't fit well into the sockets of the clip. A new cable has now been soldered to the rails and the programming track is now... back on track!

 

However, this proves that neither the diesel shunter, nor the Windsor Castle decoders are working. The Windsor Castle one was (once the program rail problem was sorted) until we put the loco back into its case. Must be a loose connection somewhere.

 

I'll report back on the diesel shunter as soon we get time to look at that. Thanks as ever for the help. 

Link to comment
Share on other sites

An example of the benefits of using directly soldered wires to track rather than clips. Particularly on 'programming' tracks where signal power levels are so low that they cannot burn through any oxidation. 'Programming' tracks need IMHO extra special care and consideration with regard to a good clean electrical signal path.

.

BEJ - keep us updated. Everybody likes a story to progress to a happy ending. 😀

Link to comment
Share on other sites

Indeed. We were both a bit crest-fallen this morning at the lack of success, though there was one positive - I can now trust the programming track at last.

 

I now have to chose whether to get on with my new layout or the two sickly locos. As I have a number ot healthy ones to keep me going, I suspect I will be getting on with my new layout, as shown here https://www.hornby.com/uk-en/forum/railmaster-layout-plans-share-yours-here/?p=9/#post-203252.

 

I'll certainly report back on any progress in the casualty ward! Thank you all for all the advice.

Link to comment
Share on other sites

That is exactly the decoder they come fitted with!  There is a socket connected to the chassis into which the plug on the decoder goes.

 

You can check it out by doing a site search, not forum search, on mixed freight which brings up the set and from there to the service sheet with the decoder and wiring arrangement shown.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
  • Create New...