Atlas SD45's shipping

[spookshow]

BTW, didn't one of the companies you mention burn down or went out of business several years ago, or was there a 3rd Chinese contractor making model trains, which is no more?

IIRC, Sanda Kan was purchased by Kader.

-Mark

[jagged ben]

...  That also sort of tells us that the mechanisms are basically designed in China, with likely very little input from the American companies.

I don't think it's that simple.  In the 90s, Atlas had clearly taken the Kato design principles and asked a Chinese factory (or more than one) to copy it essentially verbatim.  And it was like that with Atlas for a couple decades.  It's only in recent years that we've apparently seen the Chinese factories push back and modify certain design elements, or seen Atlas apparently commission a new model from a factory that does things very differently.   So that suggests to me that something has been shifting in the factories' relative power to negotiate.  Or Atlas feels the pressure to produce more models and diversify its manufacturing sources.  Or both.  We've seen Scale Trains be transparent about some of the process, and I'm sure that Atlas can and does do the same, to whatever extent they feel makes business sense. 

[jagged ben]

...

And of course, if the sub-par model is the only one of that locomotive available, will you really not buy it, or will you hold your nose and buy the inferior model anyway?

Depends what it is.  Sometimes we have choices.

[bbunge]

.... Interestingly, two SD40-2s walked up the grade pulling 45 cars without wheel slip, so let's call me mystified for the time being.

My kid got into Bachmann 44s.  We ended up with four or five of them.   He learned quickly the math is something like this:

1 loco - 4 cars
2 loco - 10-14 cars
3 loco - 20-25 cars

The pulling curve isn't linear.

Bob

[bbunge]

At Altoona I asked the guys at the Atlas table if they were looking or considering looking at moving their production out of China.  A couple of things I picked up on from the conservation:

1) At least one of the companies they deal with is starting to set up a factory in another unnamed country, or at least _trying_ to set up a factory;
2) Sort of sounded like they were dealing with a middle man and not the factory directly. 
3) One of the gents said he had been asking for management to move production for sometime
4) COVID has had a major impact on labor.  Laborers would come from the countryside to work for a couple of years, save money, and then go home.  After the COVID lock downs, these people don't want to move for fear of not being able to go home.

To get a sense of how quickly things could go down the tube, watch last week's 60 minutes segment on the South China Sea.

Every time, BLI asks for questions, I always ask the "china" question.  I have yet to see them answer it in their newsletter

Assume any divorce from China is messy and results in the loss of molds, etc.  Made more difficult since it appears much of the mechanical design is done overseas. 

Bob
 

[peteski]

Interesting observations.

German Fleischmann/Roco company has their locomotive models made in Romania, so yes, there is a movement to get out of China.

[basementcalling]

Interesting observations.

German Fleischmann/Roco company has their locomotive models made in Romania, so yes, there is a movement to get out of China.

Not surprising. Wages in China have been rising for some time as the pool of working age factory employees actually shrinks. China has birthrate issues, if you believe Peter Zehan.  Even if it's Covid related, wages going up there makes our trains more expensive even more so than inflation here.

As for the truck assembly issues, I blame sound equipped engines. I agree with Peteski, the the Kato mechanical design with needle point axles and cupped pickups was idea from a rolling and pulling power perspective, but sound decoders are more sensitive to micro interruptions in electrical contact, and decoders don't have an equivalent to flywheels to smooth performance. Thus the need for hardwired electrical pick ups, keep alives non withstanding in N scale. All this over the least prototypical feature of the model itself, as N scale sound just doesn't sound like a real engine to me. YMMV.

[Ed Kapuscinski]

decoders don't have an equivalent to flywheels to smooth performance.

Except they DO, and I don't understand why decoder manufacturers continue to to release products without them.

[dem34]

Maybe perception. Since a lot probably want the HO equivelants that will let the engine run with sound for 5-6 seconds after leaving the track. While I personally am fine with something like 1-2 seconds like an ISE Run-N-Smooth.

[peteski]

Except they DO, and I don't understand why decoder manufacturers continue to to release products without them.

As I see it, because for N scale and smaller the added circuitry on the decoder's PC board would make it too large to easily fit in models.  Modelers add those external keep-alives by shoe-horning them in various places on the model. Not a plug-n-play install.

Decoders all have some capacitors on-board, but they only supply power for milliseconds.

I'm also puzzled by those who want the loco running without track power for several  seconds.  Is their track that unreliable?!  Yes, a second or two should be more than adequate for smooth operation.

[jagged ben]

As I see it, because for N scale and smaller the added circuitry on the decoder's PC board would make it too large to easily fit in models.  ...

Nah, I don't really buy that. We don't need a lot to make a big difference.   And somehow Digitrax (non-sound) and ESU decoders already do a much better job of this than, say, NCE and Tsunami.  I think there are numeruous examples where various manufacturers could surely add some more small capcitors to PnP decoder boards and the results would be noticeable.  Or at least just provide documented, easy to solder pads for adding your own.

[peteski]

Nah, I don't really buy that. We don't need a lot to make a big difference.   And somehow Digitrax (non-sound) and ESU decoders already do a much better job of this than, say, NCE and Tsunami.  I think there are numeruous examples where various manufacturers could surely add some more small capcitors to PnP decoder boards and the results would be noticeable.  Or at least just provide documented, easy to solder pads for adding your own.

Yes, including solder pads for adding keep-alive circuitry would be very helpful.

What I found out by examining the ESU LokSound decoders, there are multiple stages of on-decoder keep-alive capacitors.  Their location (on 12V, 5V and 3V internal voltages can vary from decoder to decoder).  The caps on the 3V power are for making sure the microcontroller stays running (doesn't reboot during brief outages).  In my experience Digitrax small non-sound decoders have a capacitor only in that circuit and it is enough to make you consider them as acceptable.  Microcontrollers use relatively little power compared to other parts of the decoder, so even small value capacitors can make a difference.  Of course we are only talking about very brief power "blips", nothing like what a "real" keep-alive circuit can handle, but often that's all that is needed for improving the running characteristics.

I have not done any analysis of NCE or Soundtraxx decoders, so I can't comment on those.

[Scott Nixon]

A question for anyone who has opened up their SD45(s): How is the lubrication?  Overdone, or OK?

[wazzou]

A question for anyone who has opened up their SD45(s): How is the lubrication?  Overdone, or OK?

Sounds like a good question for Diddy. 

[jagged ben]

A question for anyone who has opened up their SD45(s): How is the lubrication?  Overdone, or OK?

I don't recall noticing any.  Which I consider a good thing.