R8249 Decoder : Class 50 forward lights always on, even if reversing.

[james_harper]

I have a very nice R2428 BR Co-Co Diesel Electric Class 50 Locomotive "Illustrious".  Prior to fitting the R8249, when the loco was on track and asked to move via any controller, two white headlights and two red marker lights were always lit, at both the front and rear ends at the same time.  So, a total of eights lights were lit.  I thought, well that's just how it is, maybe going digital will sort that out.  As with other locos, going digital seems to make the motor run just that wee bit smoother, this Class 50 is now running very nicely, maybe fooling myself.  

I fitted a R8249 Decoder.  I'm using a Select controller, the two forward white headlights and two red rear lights are always on and that's fine.  However; even if I'm reversing the loco, the same lights are illuminated, ignoring the fact that the loco is in reverse.  As all of the eight lights did work, shouldn't they now operate in accordance with the direction of travel when reversing?

I've fitted a few R8249 decoders before but they were to model steam trains with no lights, so this was never an issue.  I was hoping to add more R8249 decoders to several Hornby, Bachmann and Dapol DCC Ready diesal locos but this has me a bit flummoxed.

Would somebody point me in the right direction here please, maybe I'm simplifying matters or not understanding something.  Either of these and/or other factors may very well be the case.  Thanks.

 

 

[LMSFan72]

I can confirm that that is not how it should be. the lights should be directional and only operate when you have F0 on. Directional would mean whites on at the leading end and reds at the other based on the direction the loco is moving.

 

Since the problem appears to have been there on DC I would look to the board or wiring in the loco.

 

To work out the issue will mean taking the body off and tracing the connections. There are other threads on here regarding lighting for the class 50 which may help - I have posted a traced PCB which shows what the connections should do In the post below but there other posts too:

https://www.hornby.com/uk-en/forum/class-50-dcc-connector/?p=1/#post-350759

 

But, there are a number of different boards - worth posting a picture and the board revision markings...

[james_harper]

Thanks LMSFan72.

Apologies for my delayed response, due to a family Sunday.  I never thought to set f0, always been fitting to steam locos without lights, never been an issue, so only now relevant for me.  For the record, the board number is 1621-X009R1.  I checked for connections, all seem OK except the forward moving red tail lights flickered if the cables are touched even gently, so I've re-taped things in position. I think it's a dodgy cable connection on the R8249.

Setting f0 On - Forward, 2 white headlights & 2 red tail lights; Reverse, 4 white headlights & 4 red tail lights.  Setting f0 Off - Forward, 2 white headlights & 2 red tail lights; Reverse, same lights as Forward but loco going in wrong direction to lights. I think this is where we started from.

Unless there's a miraculous quick fix, which I doubt, I'm considering swapping the R8249 for a, Christmas present to me, R8123 TTS and use the R8249 in another steam loco with no lights.  Presumably there might still be an issue with the Class 50's lighting if the problem is on the existing PCB but we're now at the limit of my skillset as it is.  I also need to be very careful with the R8249 if I put it in another loco and see how that goes.

Any further comments, observations, guidance and directions are very welcome.  Thanks again.

 

[Howbi]

JH.......just a thought - are you sure you have plugged the R8249 decoder in the right way round, it would do no harm but will upset the lighting..........HB

[96RAF]

There is a history of dodgy Class 50 lights on the forum.

I would think the R8249 decoder is fine but you can reset it just in case. Fitting a TTS decoder to this loco before sorting the lights out properly is likely to have you back here saying the TTS decoder has blown because the odd loco lighting has overloaded it.

I wonder if the loco is Second Hand and a previous owner has modified the lighting in some way in attempt to male it more prototypical.

[LMSFan72]

I can't say I'm familiar with the functionality of that decoder, but, I would try a reset - it may be that the functions have been remapped so they are not operating as directional set of lights on the white and yellow leads.

Failing that, a picture of the board connections would be helpful to check that they are as designed? RAF is correct, there is a history of dodgy class 50 lights - some faults were on the PCB, although this version PCB seems fine in mine. Someone may have put the lighting leads to the wrong connectors on the board. The lighting works by the positives and the negatives both being switched - in very over simplified explanation the light positives are switched by the direction when the function is on and the negatives are switched just by the function being on... so, if the positives are connected wrongly you may well see a problem similar to what you have....

[ColinB]

I had issues with the lights on my class 50, but they were the top lights.I would suspect the decoder, if possible check it in another loco. I have had two decoders one TTS and one ordinary where the reverse light channel has been permanently on or failed. The top lights fail as the PCB plugs into the main board when the body is screwed together, on mine they weren't in line, but from what I saw it is very difficult for any of the other lights to go wrong, other than not light at all. The other thing to check is that you haven't an issue with a "short" or the rear lights shorted to supply which could be blowing up that function output (I suspect that is what damaged mine)> 

[Fishmanoz]

LMSFan, just to confirm for you the 8249 is a very basic 4-function decoder that certainly doesn't support function mapping, so that is not the issue.

 

From what I can see, particularly that the problem existed on DC before decoder installation, this appears to be a wiring or component problem in the loco.  On DC, it is simply a matter of opposing pairs of red/white lights being powered via diodes that allow current flow to one pair in one direction of travel and the other pair in the other direction.  When the blanking plug is removed and a decoder installed, the wiring instead goes via the decoder white and yellow function wires which perform the switching with the following provisos, most of which are mentioned above:

 

-  F0 must be turned on (lights are off with F0 off)

-  decoder must be mounted the correct way in the socket (if it's the wrong way around, motor functions still work but not lights)

-  decoder is set to 128 speed steps. This is the default so will be the case after a decoder reset, after which CV29 will have a value of 2 (128 speeds steps) or 6 (128 speed steps plus DC Running) depending on the decoder (it's 6 for an 8249 but 2 with TTS decoders) and

-  the wiring to the lights is correct.  Tracing that wiring from socket through the loco PCB (if there is one) should find the the blue decoder wire going to the same side of all lights (LEDs) with the white wire going to the white light at the front and red light at the rear while the yellow goes to the red light at the front and white light at the rear.  There will be resistors in series in the wiring somewhere too to reduce the current to the LEDs.

 

That last may sound complicated but it's likely Chris will be along soon to add one of his highly illuminating (pun intended) stock wiring diagrams giving a visual representation.

 

All bets are off of course if the wiring is incorrect or the decoder faulty, including if something has been blown by incorrect wiring (in that last instance, you may well blow any decoder put in the socket, as Colin suggested).

[LMSFan72]

@Fishmanoz

 Tracing that wiring from socket through the loco PCB (if there is one) should find the the blue decoder wire going to the same side of all lights (LEDs) with the white wire going to the white light at the front and red light at the rear while the yellow goes to the red light at the front and white light at the rear.  There will be resistors in series in the wiring somewhere too to reduce the current to the LEDs.

The trace of the circuit of the PCB in question is already provided by me in the link to an earlier post. You cannot easily trace the blue wire in situ because of the complexity of the circuit, but,the diagram I refer to is a full and accurate trace with the compoent values: it shows how the circuit works. That diagram is what I keep trying to refer back to and the connections to the PCB are marked to help diagnosis. It's here again in case it gets lost in amongst the posts:

https://www.hornby.com/uk-en/forum/class-50-dcc-connector/?p=1/#post-350759

 

@james_harper

As I posted earlier it  is most likely, from looking at the circuit and the symptons that the positive light connections are the issue. Without photos of the actual installation it's really tough to be any more help. 

[96RAF]

According to Bromsgrove Models conversion guide this model has common Negative to the lights, hence likely why we are seeing some of the problems. The main PCB has to do the reconfiguration of the 8-pin plug to provide switching on the positive blue wire when F0 is selected for directional lighting.

 

See here for more info

Class 50 DCC conversion and Lighting update (websitehome.co.uk)

[LMSFan72]

That's right RAF, it is common -ve. There are 3 positive and 1 negative light connectors at each end of the board. My suspicion is that there is a short between some of the positive terminals - loose wire perhaps - or as I think you mentioned earlier in the thread a possible modification...

On the light boards - at each end - there are 4 wires. These are in JST style plug connectors; 3 are positives - of which one feeds the sprung copper contact for the headcode lights - and 1 is the negative. There could also be a short between some of the positives at that end.

 

Now, thinking as I write. It should be possible to disconnect the jst plugs to see what happens. From recollection the right hand plug facing the direction of travel is the positives for the white lights with the left hand plug the red positive and the common negative. Unplugging the RH one should mean the white lights cannot be lit....... That should rule out a bridge at the light boards between the white and red lamp positives...

 

The other thing I would do is disconnect the board end of the positives in turn and see if all the lights stay on or off with a change to just one wire in that scenario.....

[james_harper]

I have to thank all of those who have so extensively and, to me, quite overwhelmingly responded to my quandry.  I was ready to remove a R8249 from another loco and try fitting in into the Class 50. First, I again checked that all was fitted securely, it was, but noted again that, when meddling, there was a bit of a flicker on the red tail lights.  Carrying on regardless, I removed the plug, rotated it 180 and found that the problem remained, no change.

When re-inserting the plug, I wondered if even though it was secure, I forced it more than I was happy with (What's to lose?) and more than I have done on other occassions, there remained something amiss.  For whatever reason, my amateur take on it was that inserting as normal and/or also forcing it down too much was causing the plug to be twisted/unstable within the socket.  So, I thought, very carefully, remove the plug from the socket and re-insert the pins but only by about two-thirds of the way down.  I wasn't happy, didn't think it was the thing to do, but it has worked.  The lights are finally working as they should but for how long?

I'd still like to go TTS with this model but I suspect that caution should trump valour and I'll move on and see what other diesels I can adapt.  I'm lucky to have quite a few more DCC Ready models to adapt and enjoy but I'll try and keep it simple.

I hope that I haven't taken too much of other peoples' time on this and that its not being entirely seen as a spurious waste of time.  I'm delighted that the Class 50 is running as it should and I've certainly learned a few things from this.  Thanks again everyone.

[Chrissaf]

Many Hornby locos suffer with what is either a design or quality control issue where the pins on the decoder plug go too far through the socket and short out on any metal or bare wires or something they shouldn't touch under the socket. The usual fix for this, is to see how the socket is fitted and if it is screwed down, remove it to add some insulating tape under the socket before fixing the socket back. In the case of the Class 50, this might involve lifting up the whole board.

[96RAF]

Usually the board mounted sockets have blind holes but certainly worth checking.

Also as Colin has pointed out before the diameter of the plug pins can be thinner than the bore of the socket creating poor connection, but usually things just don’t then work,  as opposed to working when not supposed to be.

[Sultan]

I recall many years ago when I bought myfleet of clas 50s there was a manufacturing fault with the pcbs.  I contacted Hornby Customer Service and they sent me new pcbs which I fitted myself.  I am not sure if it was to cure an issue with the lights or not though.  Suffice to say all of them now run perfectly with the correct lighting displays.  Most are on the basic Hornby 8 pin decoder and one is on an ESU Loksound V5 with Howes sound.  I have also disconnectd the fan drive which has really helped with smooth running atlow speeds.

[dugdug03]

I had an issue where all the wiring was correct, I used videos on line and other examples to check all wiring was correct and still the issue persisted.

After a few hours of fiddling I noticed that at one end the two little brass tab which activates the head code box were touching the metal lighting board plate causing the red and white marker lights to always stay on regardless of direction.

To resolve this i removed the lighting boards from the metal plates then I put some insulation tape on the plates and reattached the light boards.

This fixed the issue.

Be VERY careful with those pesky plates as they like to break and some were affected by metal rot. Glue dots or double sided sticky tape can be used to keep in place if the tab break away.

[Chrissaf]

I recall many years ago when I bought myfleet of clas 50s there was a manufacturing fault with the pcbs........

 

 

That manufacturing fault has been previously documented on the forum and turned out to be IIRC the tracks on the PCB associated with the blue and green socket pins being reversed. The loco functioned fine as a DCC Ready loco with a 'blanking plate fitted' but as soon as you fitted a decoder and powered up the loco, the decoder function controller chip blew.