Atlas MP15 HELP!!

[Barobutt]

Almost a year ago today I decided to buy my self a NICE engine. Previously all I had was an undecorated lifelike engine with rapido couplers. I was thinking of getting into DCC so when I saw this wonderful engine at the 2nd hand shop I fell in love.

 

 

I was told it was a really nice atlas mp15 that the previous owner had put about $300 of custom work into. Custom paint, custom lighting (notice the strobe on the roof) and a DCC system. I was assured it would run smooth and fine on my DC track.

 

When I got the thing home it wouldn't run. When it detected power, the roof strobe would flash once but it wouldn't move. I took it back and asked what the heck was wrong with it. We determined that perhaps the DCC chip was set to not work on DC track, we'd have to find someone with a dcc controller and change the settings. So I left the train with him to fix.

 

A few months later I finally get the train back. He reports that the problem wasn't so easy to fix. Apparently the previous owner had stripped the firmware out of the DCC chip and put some crazy home-brew firmware on, then had to spend hours pooling all their friend's dcc know-how to get the thing back to default, but now it runs!

 

Horray!!! All I ever wanted was a smooth high quality switcher!

 

I get home and put it on the track, apply a tiny tiny bit of power expecting the thing to start at a crawl...

 

the damn thing rockets off at nearly full speed!!

 

This is not good. I set my power-back from full to pulse... no change. The engine's minimum speed is extremely high, not even the worse old bachman engine performed so poorly. I'm happy to finally see the damn thing move though. It also starts and stops on a dime. Goes from full speed to a stop instantly and doesn't handle the slightly bit of dirty track well. My cheap life-like engine coasts a little bit, think it's called "flywheel" ? This engine functions worse than my bachman....

 

I go back to the store I bought it from and report. He says he'll "look into it" and ask some friends but I have no idea what he could do. I doubt the guy purposefully sold me a lemon though, he's a nice guy and just acting as a middle-man.

 

I opened the thing up to take some pics in case you guys can tell me anything! I'm really desperate here. I either want to fix the thing, OR sell it on ebay and recoup what I can (I paid $70) but I don't want to sell it until I know what I'm dealing with for purposes of full disclosure.

 

PS this is the first time I've ever opened up a train and I don't know anything about anything, but I'm not seeing a DCC chip.. or is it buried inside the metal case?

[CaptOblivious]

Looks like the decoder is embedded in the cab. I don't know of any commercially available decoder (with a few notable exceptions coming from Europe) that can be loaded with custom firmware. I will assume that your dealer simply mis-spoke, and meant that it had been set up with some crazy combination of CV settings.

 

Do you have access to a DCC system? That's the only way we can make progress with this beast. Can you shine a light into the cab and try to get a good picture of the decoder, or read any markings on it?

 

Would be good to get a readout of all the CVs, if possible; if not, CVs 2--8 would suffice as a starting point. The value of CV 8 contains the manufacturer ID, and will thus tell us how to go about reseting it to factory settings.

[Barobutt]

Ok, I was just told that perhaps my problem is in the DCC's speed table. http://www.dccwiki.com/Speed_Table

But would that effect it's performance in DC mode?  

 

And as hard as I looked I couldn't find anything that looked like a chip or anything electronic looking inside the case.  Maybe I don't know what I'm looking for, but does anyone have a DCC mp15 and does it look like mine?  I know the wires going to the cab roof are custom, but apparently this engine came stock DCC.

 

And no, I don't have DCC.  That's why I had to wait months for the shop owner's friend to "fix" my engine so it would move at all...

[CaptOblivious]

Ok, I was just told that perhaps my problem is in the DCC's speed table. http://www.dccwiki.com/Speed_Table

But would that effect it's performance in DC mode? 

 

Yes, very much so!

[CaptOblivious]

Oh, and don't used pulse-power mode on your throttle with DCC decoders :D Might break them.

[CaptOblivious]

I guarantee you that your model does not have a factory-fitted decoder. If it did, it would look more like this on the inside:

 

[Barobutt]

Well then where the hell is my decoder?  When I sent this engine to be fixed after I bought it, they wouldn't have just taken the decoder out and said they fixed the problem knowing I'm running it on DC track?  If they had taken it out, would that explain the horrible performance I'm getting as the motor was designed to work with a DCC chip between it and the power source?

 

I didn't look too hard in the cab, but I think the only thing up there is the strobe, I don't think there would be enough room for a chip from the size of the one posted above, or do they make tiny ones?  I'll have to examine the cab more closely when I get home and maybe post some pics.

 

No one told me pulse ruined DCC chips... is that a sure thing or just a "may" damage it? I hope I haven't ruined my potentially non-existing DCC chip....

 

Looks like they DO make ridiculously tiny DCC decoders... maybe there is one stuff in the cab.

[CaptOblivious]

Pulse power "may" damage decoders. Depends on what the output looks like, and that can cary quite a lot; it also depends on the decoder's tolerance for non-DCC AC waveforms.

 

And yes, a Z2 would probably fit up in the cab :D There are bigger ones that would fit too (TCS M1 or Digitrax DZ125 might well fit up there too!)

[Barobutt]

Guys, thanks for all the help.  I've got 2 forums helping me now, both offering different pieces of the puzzle.  I was ready to just ebay this thing as a "potentially broken, as-is" loco at a huge loss but now I think it might be fixable.  I've always been into model railroading but I've never had much in the way of a network of people giving advise, I'd just figure things out my self as I went along.  This forum has been awesome in all things n scale, specially my new found interest in japanese trams of course.

 

I'm going to call the shop owner who sold this to me and get some more info and I'll report back.  The speed-table suggestion alone has really blown open the case.

[David]

Only 2 wires go into the cab, though there is some kind of extra wire up there (probably 2 LEDs and maybe a strobe circuit, but I can't see).

 

It I think I can be fairly confident in saying it's been converted back to DC, just without the benefit of the original lightboard, unless there is some really magic wiring involves (2 wires in isolated from each other but in the same jacket??).

 

What I see:

Faded Yellow and Purple (Dark Blue?) wire are soldered to the frame halves, providing power from the rails. Both of those wires have been split into 2 wires, with the heatshrink tubing around the split (purple becomes grey and yellow, faded yellow becomes light blue and orange). From there it gets a bit weird - it looks like the motor and front LED are sharing a wire (it's like they're run in series?), but it's hard to see from the picture. The yellow and blue split wires go into the cab.

 

Either way I do not see a decoder, and I do not see how any of this wiring could be from a decoder (track power directly to LED for example).

 

I have this particular model. It uses one of Atlas's fancy motors. I also have the original DC lightboard, I will check it to see if perhaps there is supposed to be some kind of resistor or other doohicky between the power and the motor. Though that weird LED+Motor in series wiring that the picture seems to suggest is equally bizare.

[CaptOblivious]

The wiring is rather odd. I wondered if they were using common-return on the LED, in which you wire the anode to

the blue wire on decoder, and the cathode to one of the rails (black or red wire).

 

There's a green wire in the cab that goes nowhere?

 

At any rate, the whole thing does look like quite a mess!

 

Also, presuming there is a decoder, the speed table may not be the only cause of your problems. I agree that it might simply be a messed-up speed table, but I've seen damaged decoders behave in the same way.

 

I am at a loss to explain the running behavior if there is no decoder, however. I can't imagine a way of wiring up the various bits in a decoderless loco so as to ensure top speed at small DC voltages. (Indeed, now that I think about it, I can't imagine a way of doing that with a decoder, either). Could you post a video of the suspect behavior? (And more, better-lit photos of the interior?)

[Barobutt]

UPDATE

So I called the owner and I told him what you guys told me.  First of all apparently the chip is "under the hood" in the shell and yes it still exists.  He didn't seem to think the speed table would be the issue because he said "we set it to dc mode" but when I told him people on !THE INTERNET! told me the speed table still does effect how it performs in DC mode he said for me to bring it down again and he'll get a friend to take it home and look into it and hopefully adjust the speed table.  He also said perhaps there's a faulty component and didn't outright say he'd replace it, but mentioned "we'll see about taking care of it".  

[CaptOblivious]

DO NOT SEND IT BACK TO THIS SHOP! I am quite sure they are screwing you.

 

I am now inclined to agree with David's analysis that there is no decoder in there. Photo to follow.

[CaptOblivious]

Here is a photo in which I tried to trace the wires to show common connections. The original author of this rat's nest didn't use wire colors consistently at all. I caution that, outside of concrete continuity-check evidence that this is just a hypothesis. But it looks like the motor, headlight, and strobe are wired in parallel, and in parallel with the track—exactly how a DC loco is wired up.

 

HOWEVER, there might be a decoder hidden inside the frame just above the motor. The wires all get jumbled together at that point, and what I'm identifying as a junction at the motor tabs might actually be seperate wires from a decoder hiding in there.

 

Grain of salt and all that.

 

But I'm telling you, this store you're dealing with? They're not being straight with you.

[Barobutt]

Really doesn't seem in this guy's nature to do something like that.  He's bent over backwards for me in the past fixing other people's stuff, given me free replacement parts and replaced a bunch of couplers for me that he even gave to me for free.  HATES N scale but he sat there for a good 30 min chatting with me while occasionally swearing in frustration at how tiny N scale is as he worked on my rolling stock.  He's an old friendly hippie with a big gray braided beard who won't carry anything military related in his store as it's against his personal morals.  I'd believe a mix up and bad photography before I thought the guy was trying to scam me.  And even if he was, if he finally fixes the damn thing I'm happy.

[David]

This is what I came up with.

 

Like Don the only possiblity I have for a decoder is that it just below the orange shrink wrap - is that a circuit board there, or a flat black surface? Even if it is a decoder the LEDs don't seem to be connected to, unless there is a really weird job under that heatshrink.

[Barobutt]

Haha this is such a crazy mystery, I'm almost more interested in this as a "who done it" than actually fixing my train.  Very curious!  You guys are good detectives!  I wish I had even one of you guys here to actually examine my train.  I'd picture you all with Sherlock Holmes caps and magnifying glasses getting to the bottom of this.

[David]

Really doesn't seem in this guy's nature to do something like that.  He's bent over backwards for me in the past fixing other people's stuff, given me free replacement parts and replaced a bunch of couplers for me that he even gave to me for free.  HATES N scale but he sat there for a good 30 min chatting with me while occasionally swearing in frustration at how tiny N scale is as he worked on my rolling stock.  He's an old friendly hippie with a big gray braided beard who won't carry anything military related in his store as it's against his personal morals.  I'd believe a mix up and bad photography before I thought the guy was trying to scam me.  And even if he was, if he finally fixes the damn thing I'm happy.

 

I wouldn't accuse him of 'scamming' you. He may just not know much about what he talking about in this case (homemade decoder firmware?). There are plenty of people of guys in model railroading for whom DCC just means 'something complex under the hood that I don't want to know about'.

[David]

Also, based on my diagram/guess - what happens when you try to run the locomotive in reverse?. Does it move, do any of the lights come on?

[CaptOblivious]

This is what I came up with.

 

Like Don the only possiblity I have for a decoder is that it just below the orange shrink wrap - is that a circuit board there, or a flat black surface? Even if it is a decoder the LEDs don't seem to be connected to, unless there is a really weird job under that heatshrink.

 

Yeah, the strobe is clearly wired to the rail/motor. But it looks to me like the headlight could conceivably be wired to the function leads of a decoder inside the frame—Look at the wire coming off the resistor. It might wind under the blue wire, and into the red shrink wrap. OR, it could be that it disappears under the blue wire headed to the cab; and a different wire is going to the red shrinkwrap. Evidence: On the right of the blue wire, it looks white; on the left it looks gray. White wire might or might not be soldered to the motor tab. The blue wire from the LED does clearly seem to dissappear into the frame, where it might or might not be soldered to the motor tab. If this hypothesis is right, it has the advantage that the proper colored wires are being used: White/blue to LED, orange/gray to motor…yeah, nevermind, that can't be right, because the motor really does look clearly wired to the rails, and that isn't proper wiring for a DCC installation at all.

 

Hrm.

 

I wouldn't accuse him of 'scamming' you. He may just not know much about what he talking about in this case (homemade decoder firmware?). There are plenty of people of guys in model railroading for whom DCC just means 'something complex under the hood that I don't want to know about'.

 

Again, I'm inclined to agree. Not that it's a scam, but this guy is representing himself as knowing more than he does. Or his friend.

 

Getting to the bottom of this will require a continuity checker for sure, a willingness to crack this baby open, and if at all possible a DCC system…

[CaptOblivious]

Haha this is such a crazy mystery, I'm almost more interested in this as a "who done it" than actually fixing my train.  Very curious!  You guys are good detectives!  I wish I had even one of you guys here to actually examine my train.  I'd picture you all with Sherlock Holmes caps and magnifying glasses getting to the bottom of this.

 

This is fun!

[Barobutt]

Yeah, I trust him when he says he took it to one of his regular's with N scale DCC and the worked on it.  I can't remember the specifics but basically they had to put in a couple hours time to figure out what the heck was going on because he says it was a very unusual setup.  He didn't tell me this as a sob-story, he in fact said him and his friend had a blast solving the mystery and figuring out what the hell was going on inside.  Maybe his friend "fixed" it by taking the decoder out and then they tried to re-wire the strobe and couldn't...  he mentioned when he gave it back to me that they couldn't get the strobe to work in DC mode but it should still work when I upgraded to DCC.  But if that was the case he's the type of guy that would have said "Sorry dude, to get it to run on DC track we had to take the decoder out, we tried to re-wire the strobe but it doesn't work.  Here's your decoder, when you upgrade to dcc we can put it back in".  I even told him when I gave it to him that I'd be fine if they just took the chip out but he said it wouldn't come to that.

[CaptOblivious]

Also, based on my diagram/guess - what happens when you try to run the locomotive in reverse?. Does it move, do any of the lights come on?

 

Dammit, David, you are not allowed to suggest clever, simple tests like that. If David is right, and there is no decoder and everything is wired as our diagrams suggest, then if you ran the thing backwards, the headlight won't light. Which would pretty much either nail his hypothesis, or indicate that the decoder was programmed to turn the headlight off in reverse. Which is often the default behavior. But if it remains on, we know we've got a deocder in there somewhere… (LEDs are polarized devices; they light when voltage is applied one way, but not the other)

[Barobutt]

I'm not at home at the moment so I can't run any tests.  The moment I get home though I'm going to test as you say and take some better pics with a flashlight and identify every wire.

 

 

Here's a quick picture I made.  I'm mostly interested in that mystery wire in the cab.  The train does run forward and backwards fine but at the exact same wayyyy too fast minimum speeds.  The strobe never works, and I can't remember if the lights stay on or not.  One thing I did notice is that the lights stayed at an almost totally consistent level of brightness regardless of the power applied, contrasting to my life-like engine that's light dim or get brighter as you apply power.

[CaptOblivious]

Green wire in cab is indeed quite the mystery. As is the Kapton tape on the frame under the cab (the yellow-orangish tape); it's usually used in decoder installs to electrically isolate bare metal, but it's not at all clear what's being isolated from the frame.

 

The two "prongs" coming off the resistor just looks like a sloppy soldering job to me.

 

And if there is a decoder, it probably is hiding here you ask if something could be hiding in there.

 

Incidentally, LEDs have the interesting property of appearing to have a fairly constant brightness over a range of voltages, so your description of the headlight isn't surprising, nor is it evidence for any of our hypotheses, sadly.