Fried Decoder?

[CDRC]

Some help please. I hardwired a decoder into a Hornby 0-6-0 Pannier (metal chassis) with a Type 7 motor following the instructions at this site

https://www.modelrailforum.com/threads/decoder-hardwire-install-on-j83-tank-loco.19339/

At the motor insulation stage using a continuity tester I confirmed the motor was indeed isolated from the chassis at all points. To test further I re-connected the relevant wires on each side of the loco and ran it on my DC track. The loco ran perfectly. I connected a Laisdcc decoder following the accepted guidelines and placed the now DCC wired loco on my programming track (Hornby Elite). I programmed the loco to 2783, the locos running number and the Elite confirmed the changed in the usual way. I then placed the loco on the main track. The Elite ran through its usual self test and then went straight into the ‘ERROR’ configuration. I removed the loco rebooted the Elite and the Elite again went through the self test and powered up ok. I placed a different loco I had hardwired previously on the track and it functioned as expected, running well. I replaced the 2783 back on the track and on this occasion the Elite didn’t finish the self test before the ERROR notification showed.

I can only assume I have not insulated the motor from the chassis sufficiently well, unless someone knows differently and can advise or made and error in the wiring of the decoder. If so does the ‘ERROR’ message on the Elite indicate the decoder fried?

[Howbi]

Just to be sure, did you wire Orange and Gray to the motor then Red and Black to the pick-ups? Also did you fully isolate the remaining wires.?

[96RAF]

From the term ‘self test’ it seems that you are placing the loco on track then powering up the Elite. Is there a reason you do this rather than simply place your locos onto a pre-powered track.

Usually if the controller can read-back an address the decoder wiring is OK.

Unfortunately by hard wiring in the decoder rather than fitting a socket you lose the chance to easily swap out a decoder or indeed test the loco installation without the decoder installed or even test the decoder in a different loco.

It does however seem likely that the decoder is toast.

[ColinB]

Yes it sounds like it has died. Generally with LaisDCC decoders they generally smoke so have a look at it to see if the sleeve around it has melted. As it programmed perfectly I suspect a motor current issue. The 0-6-0 Hornby motor takes a reasonable amount of current so it might be that the decoder has died through "overcurrent". I have had many go that way. Generally when I convert my locos I put a socket in the wiring, it makes replacement easier. In the case of an 0-6-0 I use a 6 pin socket, the Bachmann or DC Concepts ones seem to be the most reliable, I found the LaisDCC ones tend to pull out the wires too easily. I use a Zimo 8 pin decoder, which has a much higher current limit.

[CDRC]

Just to be sure, did you wire Orange and Gray to the motor then Red and Black to the pick-ups? Also did you fully isolate the remaining wires.?

Hi and thanks. The unused wire were isolated by cutting each back at different lengths and then applying shrink tube. I checked the wiring before programming the chip and the chip has been wired correctly

[CDRC]

From the term ‘self test’ it seems that you are placing the loco on track then powering up the Elite. Is there a reason you do this rather than simply place your locos onto a pre-powered track.

Usually if the controller can read-back an address the decoder wiring is OK.

Unfortunately by hard wiring in the decoder rather than fitting a socket you lose the chance to easily swap out a decoder or indeed test the loco installation without the decoder installed or even test the decoder in a different loco.

It does however seem likely that the decoder is toast.

 

 

I’m working in my kitchen at the moment as the layout is in the garage and it’s to cold out there at the moment. I have a small piece of track on the kitchen table on which I test the loco on DC to ensure good running before conversion. On each occasion prior to this I install a socket as you have suggested. In hindsight’s I’m rather happy I did as had I not I might have replaced the chip thinking it was a dud and fried a second and fried that one. Re the suggestion of testing the installation prior to putting in the decoder in. I’m assuming this is by putting in a blanking plate before the decoder and running it on DC. As I said in my original question I tested the wiring by connecting the wires appropriately and running on DC. Thanks for your thoughts. I believe the chip is fried also……so back to the drawing board.

[96RAF]

By test the installation I meant using a multimeter pin to pin on the socket, as a blanking plate normally only bridges pins1-8 and 4-5 and masks any cross connection twixt the inner four pins and/or with the outer four.

This is where a decoder tester is handy as it can prove the fault lies with the decoder or if not it has to be with the loco.

[ColinB]

As I said in my original post LaisDCC decoders are easy to fry, I have even had them die when previously they have run the loco perfectly. Do a stall test on the loco and check what the maximum current is. Then add about 10%. Now check what the maximum current limit on your LaisDCC decoder is. Generally with mine, works perfectly on a straight piece of track, put it on the layout with curves and especially points, they die. I figured that to go round bends and cross points requires more effort by the loco, hence an increase in motor current. Surprisingly, I got the same effect with Hattons and Hornby ones when fitting them to old ringfield motor driven locos, although in both cases the current was supposedly within specification.

[CDRC]

By test the installation I meant using a multimeter pin to pin on the socket, as a blanking plate normally only bridges pins1-8 and 4-5 and masks any cross connection twixt the inner four pins and/or with the outer four.

This is where a decoder tester is handy as it can prove the fault lies with the decoder or if not it has to be with the loco.

I see, thanks.

[CDRC]

As I said in my original post LaisDCC decoders are easy to fry, I have even had them die when previously they have run the loco perfectly. Do a stall test on the loco and check what the maximum current is. Then add about 10%. Now check what the maximum current limit on your LaisDCC decoder is. Generally with mine, works perfectly on a straight piece of track, put it on the layout with curves and especially points, they die. I figured that to go round bends and cross points requires more effort by the loco, hence an increase in motor current. Surprisingly, I got the same effect with Hattons and Hornby ones when fitting them to old ringfield motor driven locos, although in both cases the current was supposedly within specification.

 

 

[CDRC]

As I said in my original post LaisDCC decoders are easy to fry, I have even had them die when previously they have run the loco perfectly. Do a stall test on the loco and check what the maximum current is. Then add about 10%. Now check what the maximum current limit on your LaisDCC decoder is. Generally with mine, works perfectly on a straight piece of track, put it on the layout with curves and especially points, they die. I figured that to go round bends and cross points requires more effort by the loco, hence an increase in motor current. Surprisingly, I got the same effect with Hattons and Hornby ones when fitting them to old ringfield motor driven locos, although in both cases the current was supposedly within specification.

 

 

[96RAF]

LiaisDCC decoders are widely known to be reverse-engineered versions of other makes of decoder, but without any guarantees they have built in all the original features or achieved the donor spec, such as self protection against overload. From Colin’s experiences of burnout it would seem there is no such self protection.

[CDRC]

As I said in my original post LaisDCC decoders are easy to fry, I have even had them die when previously they have run the loco perfectly. Do a stall test on the loco and check what the maximum current is. Then add about 10%. Now check what the maximum current limit on your LaisDCC decoder is. Generally with mine, works perfectly on a straight piece of track, put it on the layout with curves and especially points, they die. I figured that to go round bends and cross points requires more effort by the loco, hence an increase in motor current. Surprisingly, I got the same effect with Hattons and Hornby ones when fitting them to old ringfield motor driven locos, although in both cases the current was supposedly within specification.

I have stall tested all the DC conversions prior to fitting. I chose Laisdcc as they show in their blurb a continuous of 1amp and stall of 1.5amp. The XO3 and 4’s I’ve converted to date have shown 6-8 ma although this has always been on straight track. This particular motor, a type 7, only showed 5 so I think from that perspective the draw was low enough. I think that I have not insulated the motor from the chassis sufficiently carefully so I’ll have another look.

The advice to test the pin connections is something I have not thought about and will follow that in future conversions.

Thanks again

[CDRC]

LiaisDCC decoders are widely known to be reverse-engineered versions of other makes of decoder, but without any guarantees they have built in all the original features or achieved the donor spec, such as self protection against overload. From Colin’s experiences of burnout it would seem there is no such self protection.

I suppose like most things in life, you get what you pay for. Laisdcc are cheap, I accept that and in general I only fit them in the smaller 0-6-0 or 0-4-0 as a bit of an experiment. The type 7 was just that. I saw the way the loco and motor were configured and decided to try and convert it.

[ColinB]

To be quite honest LaisDCC decoders are no worse than certain other makes. When I first ventured into DCC I tried all sorts. I eventually found the following to be the most reliable, Zimo, Hornby and LaisDCC. I don't think I found any of them had decent motor protection, in that if the motor has a short circuit or draws more that the device can source then bye bye decoder. Seeing as I have quite extensive experience in electronics I did try to figure out why this was the case seeing as everything I designed had decent overload protection. Eventually I came to the conclusion that most of the overload protection worked for a little while before it blew, measured in seconds. I did contact YouChoos about how they did it and the answer was basically there is only so much you can do. Surprisingly I have never managed to blow up a Zimo decoder even when the valve gear has jammed on a loco, so obviously they are doing it right. Seeing as their economy versions are probably one of the cheapest, I have stuck with them ever since. I will add I have nothing to do with Zimo, I just like their products.

[Jim-R855]

Does your 0-6-0 with Type 7 motor have a sprung rear axle with a lot of movement possible? If so check if the wheel rims can touch the metal chassis within the range of articulation of that axle, I had a bit of an issue with that which I solved with some kapton taope in the intermittant contact area. If the axle isn't the sprung type with long travel, this suggestion is not relevant, but I thought I might as well put it all together rather than wait for your answer...

If you suspect this could be your problem set your multimeter on resistance, loco on the bench, attach probe to opposite wheels, decoder unplugged, rock chassis and make sure it always reads open circuit.

Good luck