Simond Posted January 26, 2015 Share Posted January 26, 2015 I built a 7mm Acorn kit of a 57xx pannier some years ago. It's a lovely model, and well weighted, runs very smoothly and pulls as it should. I need to take it to bits, but I believe it has a Mashima motor, and a fold-up gearbox. About a year after I built it, I was running it at the late Ken Stansfield's Swanage layout, and, for reasons I cannot explain, it stalled, and the decoder (an old Lenz one) let all its smoke out. This was put down to "one of those things", and the decoder was replaced - in those days, Lenz gave an unconditional guarantee (may still do?) and that was that. I haven't used the loco for ages, but when I got my Lionheart one at Christmas, I had to put the two together, so it came out of the glass cabinet, and onto the Greater Windowledge Railway for a run - all good. Then, horrors, it did it again! Another dead decoder. I've never had a problem like this on any of my other locos, or the locos I have built/DCC fitted for others. I am 99% sure that it is not jamming, but I'll check. I'm fairly sure that there is no obvious way to create a short between decoder output and track supply, which is pretty much guaranteed way of killing them, but I'll check that too. In th end, I suspect the motor - there was a topic posted on RMWeb a month or so ago which I now can't find, referring to possibly over-long screws attaching the gearbox to the motor, but also interfering with the armature as it rotated - I will look for this, but I'm not sure how to, without taking the motor apart which I don't want to do. I can pull the screws out, one at a time, and look for wear, and I can also meter between the motor connections and the chassis. Does anyone have any other ideas? I've posted here, because I don't think it's a DCC issue, though I'll put a link in the DCC forum too - I think its the motor or the loco. Any suggestions of things to look for or things to test/try out before I fit another decoder, would be more than welcome! thanks Simon Link to post Share on other sites More sharing options...
Adrian Wintle Posted January 26, 2015 Share Posted January 26, 2015 It is possible that, if the motor/loco actually stalled, the stall current exceeded the limits of the decoder. I'd look to see if there is something that can jam the motor/drivetrain for a moment - it wouldn't take long for the stall current to rise to a level where the decoder could be damaged. Adrian Link to post Share on other sites More sharing options...
Simond Posted January 26, 2015 Author Share Posted January 26, 2015 Thanks Adrian - you are right that a stall would cause the highest current, and it was an old Lenz chip, so possibly not as well self protected as the more modern ones. Not sure which chip, so not sure what current limit it could handle. I stripped the remnants of the harness (it had a multi-pin socket) out of the loco this evening, and connected it up as plain old-fashioned DC - it ran fine, as expected. I can't see any obvious signs of anything catching - it doesn't even have dummy inside motion, and whilst equalised, there really doesn't seem to be anything for the rods to catch on. Your question/suggestion reminds me of a mishap some years back - I tap my wheels and screw in a 10BA screw from the back, secured with a drop of araldite or superglue. One came undone on another loco, caught on the chassis, and caused a very abrupt halt. I can't think of any reason why this should have undone itself, but worth a check, I guess. I did find one other potential area - the loco has Slaters plunger pickups, and it is just possible that one of these could touch the chassis - but there is no obvious way that one of the motor wires could do so as well, so this seems an improbable cause too. I'll have another dig when I come to put a new chip in - don't want a hat trick! Thanks again Simon Link to post Share on other sites More sharing options...
34theletterbetweenB&D Posted January 27, 2015 Share Posted January 27, 2015 ... it stalled, and the decoder (an old Lenz one) let all its smoke out. Then, horrors, it did it again! Another dead decoder... Does this mean it was exactly the same both times, running smoothly, stall, smoke and dead decoder? The obvious questions about the DC current draw, decoder rating, any mechanical problem or the possibility of track power reaching a motor terminal need to be looked at, but there is something simpler. Is the decoder position such that it is vulnerable to contacting any metal? Doesn't need to be live. That'll kill a Lenz decoder in flash. Link to post Share on other sites More sharing options...
chaz Posted January 27, 2015 Share Posted January 27, 2015 There's a coincidence. On Sunday morning I was operating Dock Green at the Eastleigh show and loco 69727 (my N7) was running nicely. It stopped abruptly and within seconds a lot of blue smoke and an acrid smell told me that I would have to stump up for a new decoder (ESU LokSound - expensive). I had run the loco into a point that I had forgotten to throw and the crossing had stopped the loco with a full short. Normally this would cause the DCC to shut down but for some reason on this occasion it didn't. The point is one of those which has its blades on one board and the crossing on another so there is more wiring length than usual between the switch and the crossing - a slightly higher resistance may have prevented the DCC detecting the short circuit. Chaz Link to post Share on other sites More sharing options...
RMweb Premium Grovenor Posted January 27, 2015 RMweb Premium Share Posted January 27, 2015 So how did this short circuit current get to the decoder? The rails should only be connected to the decoder input connection which is designed to withstand full DCC voltage, with a short on the rails this voltage would be reduced substantially if not cut off so the decoder should not have a problem. The short circuit current on the loco would just be passing from one wheel to the other via the pick up wiring. There is something else amiss if the decoder has blown which the location of the damage on the decoder may help to determine. Regards Keith Link to post Share on other sites More sharing options...
chaz Posted January 27, 2015 Share Posted January 27, 2015 So how did this short circuit current get to the decoder? The rails should only be connected to the decoder input connection which is designed to withstand full DCC voltage, with a short on the rails this voltage would be reduced substantially if not cut off so the decoder should not have a problem. The short circuit current on the loco would just be passing from one wheel to the other via the pick up wiring. There is something else amiss if the decoder has blown which the location of the damage on the decoder may help to determine. Regards Keith It's not quite as simple as that Keith. I have heard of other decoders dying in similar circumstances. I agree that in theory "The short circuit current on the loco would just be passing from one wheel to the other via the pick up wiring". But there must be something happening with the AC nature of the DCC square wave that upsets the decoder. I seem to recall reading somewhere that it is vital that DCC shuts down promptly when short circuited to protect decoders. Hence the need for a very low resistance current path from the DCC to the rails. I suspect the point crossing in question has a bit more resistance in its wiring. The Tortoise that switches the polarity is on another baseboard to the crossing (The point was cut into two pieces) and there is therefore a jumper cable and plugs and sockets in the current path. It seems one hell of a coincidence otherwise that a loco that had worked well all the previous day failed at the moment when it stalled on the wrong-polarity crossing. The two things look certain to be cause and effect to me. I'm not sure if my knowledge of electronics will allow me to identify the cause based on the damage to the decoder - but I will gladly post a photograph of the wreckage when I have dismantled the loco - this will wait on receipt of a replacement decoder - which I am ordering this week. I will post a report later but not in this topic - which is not mine. I will put it up on my Dock Green thread. Chaz Link to post Share on other sites More sharing options...
Simond Posted January 27, 2015 Author Share Posted January 27, 2015 Chaz I don't know... From memory, in the first case, the loco was trundling a fairly short (15 wagons?) goods train around a garden layout that was otherwise operating normally. I wired the layout, using Lenz controllers and boosters, and we were all confident that there wasn't a layout problem. There wasn't a fat lot of choice about DCC in those days - it was probably '99 or 2000. The train stopped. I turned up the wick, and the loco smoked, sufficiently to be visible from the other side if the garden. Ooh, there's realistic I thought... Second time, it was on the Greater Windowledge, shunting - near, but not on, a crossover. Lack of forward motion, followed by smoke. In neither case was the loco doing anything that might indicate it was being heavily loaded, nor was it going anywhere near top speed. It's the apparently random stop that worries me - like it jammed, but I have checked very carefully, and it seems fine. I've also taken out the screws that hold the motor & gearbox together, one by one, and there is no sign of any mark on them, so I'm confident they are not interfering with the armature. I also metered between motor connections and chassis - no connection, even turning the motor. I also double checked that the motor wires cannot touch the body or chassis. I'm with Keith regarding the failure not being caused by what should be a track short, caused by running over a wrongly set trailing point - if anything, the decoder should be seeing zero volts (an effective short across its input) in this case. That should be safe. Is what happened to you merely a coincidence, or at least, not causality? I'm sure it doesn't feel like it from your position, but I have driven over incorrectly set crossings many times, the locos I have fitted with KA caps just ignore the problem and keep going. The controller shuts down, as it should (also Lenz set 100, but updated to v3.6). Locos without KA stop as expected. Regarding posing on here - please feel free - it's all related to the topic, so might help someone in a year or two. 34theletter... Maybe... The decoder was mounted on double sided sticky tape on the bottom of the ashpan - I made a flat bottom for it, which has a couple of stiff wires soldered across it, fitting into holes in the bottom edge if the firebox sides - it just clips in. When I came to take it apart, it was clear that the tape had gone a bit manky - whether through age, or being cooked by the decoder, I cannot say. Neither can I say with any confidence that the decoder cannot have touched, metal, nor can I say that it definitely did... Is the decoder-killing property of touching metal a short from point to point on the decoder, or an ESD effect? Of course, if there had been contact between a pickup and the chassis, and between the decoder and the chassis, it would be very likely to kill the decoder Keith. Thanks for the thought regarding location of fault on decoder. I'll look. Though whether I have the necessary understanding to draw a rational or useful conclusion... Perhaps I'll follow Chaz' lead & post the photo. It's my birthday today. MrsD & the kids have very kindly purchased a Zimo sound chip & speaker loaded with a 57xx sound project. I would really like to eliminate any risks before installing the new toy. Will check again for potential jamming, and ensure that the plungers are "safe". Thanks to all for your thoughts & advice - much appreciated Simon Link to post Share on other sites More sharing options...
34theletterbetweenB&D Posted January 28, 2015 Share Posted January 28, 2015 The decoder killing property is an internal short. If track power gets connected to an output stage, phut! But having the decoder neatly stowed on tape argues against that being the cause. Suggest taking the motor out, putting an ohm meter over the brush terminals in the lowest range available, and going round in very small increments to see if it has a zero impedance somewhere. Seen this once in a can motor, and that had been a 'decoder eater'. Link to post Share on other sites More sharing options...
Simond Posted January 28, 2015 Author Share Posted January 28, 2015 I did that, but not with the intent of finding a short within the motor - I'll add it to the list thx Simon Link to post Share on other sites More sharing options...
Simond Posted February 5, 2015 Author Share Posted February 5, 2015 NFF - ie no fault found. Checked motor for shorts to ground and across terminals, checked chassis for binding, locking, jamming and other irregularities. Nothing, nix, nada, niente, nought, zip, zero, zilch. But I figured that it's better to spend £20 on a new motor than £100 on replacing a new chip, so motor ordered. Thanks to all for the thoughts & suggestions Best Simon Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.