Kato Big Boy K Rail DCC Install Photos Request

[JeTFiN]

Would someone who has purchased an N scale Kato Big Boy with K Rail DCC+Sound (ESU LokSound Micro 5?) install be willing to pop the shell of the tender of their locomotive and share a few pictures showing the setup?  I'd like to see a good baseline to start on for my own conversion project - I had pre-ordered the DC version back when it seemed likely it would be a drop-in install but now it looks like it may be worth making custom PCBs to perform a proper install.

[peteski]

I'm still documenting my DCC install, but life gets in a way of the project.  Will post it as time allows.
My install will be covered  in https://www.therailwire.net/forum/index.php?topic=56254.0

As for "popping the shell", on this loco, it is anything but easy popping.  This is a complex beast.
Are the DCC versions out in the wild already?  I'm also curious as to way Kato handled their install.  Whatever they did, it required lots of precision manual labor.

[JeTFiN]

I'm still documenting my DCC install, but life gets in a way of the project.  Will post it as time allows.
My install will be covered  in https://www.therailwire.net/forum/index.php?topic=56254.0

As for "popping the shell", on this loco, it is anything but easy popping.  This is a complex beast.
Are the DCC versions out in the wild already?  I'm also curious as to way Kato handled their install.  Whatever they did, it required lots of precision manual labor.

Peteski,

I am aware of your install - thank you for documenting the process!  You could correctly claim that I am being impatient  .  In this particular instance I am well aware that removing the shell of the locomotive is an arduous task, but the tender is actually quite simple.  All one must do is pull the tender from the locomotive, turn it upside down, and use thumbs to separate the shell while using two forefingers to pull the frame up and out of the shell to then take photos.  With these photos I could look at the wiring layout in the tender specifically for the K Rail installation.  I noted in several videos that there are no apparent wires between the tender and locomotive with this install (see Youtube links below).  I am currently debating the feasibility of either producing a small PCB with NEXT18 connector on one end and an undetermined port on the other end to couple with a board in the locomotive, using the ESU ribbon cable to 6 pin NEM plug to plug into a different small PCB held in the notch in the locomotive cab you showed in your thread, or some alternative option.  I consider it necessary to be able to separate the tender from the locomotive as I tend to prioritize no losses in functionality for my DCC installs.

I own an electronics enclosure design company, so I am willing to design and produce a custom part that mounts everything needed in the tender.  Right now I am considering using some variation of the ESU LokSound Micro 5 decoder (not sure which electrical interface yet) with one or two SP-1511L-2 speakers (https://www.digikey.com/en/products/detail/soberton-inc/SP-1511L-2/9924403)  These speakers have a frequency range between 100 Hz and 20 kHz so there should be a nice sound quality assuming a proper acoustic chamber is designed for it/them to fit in.  Electrically, two is favorable to bring impedance to 4 ohms, but if it is later determined that two is impractical, only one speaker can be installed instead.

Despite familiarity with the DCC world (to my knowledge, I was the first to install a DCC decoder in the Chinese Kunter CRH380AL model, see write up here: https://jnsforum.com/community/topic/17705-kunter-crh380al-dcc-installation-n-scale).  After performing that install, something tells me I should install decoders with a plug somewhere to easily replace the decoder if it gets fried and avoid extra work re-soldering.  The wiring within the locomotive itself is mostly intuitive with the only exception being the headlight PCB.  I am unsure exactly how the LokSound Micro 5 will respond with two motors.  How will it handle producing the "chuffs" for both sets of drivers, I wonder?

https://youtu.be/RA9JLiPLjyM?si=fxhlStV6dtkmlwKR
https://youtube.com/shorts/s17alMb0CP0?si=i99fOP5OqXym4kSl

In my opinion, Kato made a confusing design decision on an otherwise excellent model with regards to their lack of a supplied NEXT18 port or similar in the tender with relevant wiring/electrical routing, as models in the modern day tend to come with these features standard.

[peteski]

I guess I missed the "tender" part. I just didn't consider that part of the DCC-fication of this model to be difficult.  Since you are new here (that was your 1st post) I  had no idea about your level of knowledge, or your plans.

As far as making DCC conversion easy, due to the unconventional dual motor design, I really don't see any easy solutions Kato could have used to make this a  DCC friendly DCC install.  Funny thing is that Kato designers were most likely aware that Kato would be installing decoders in those engines and as you mentioned, except for come channels in the boiler frame for the wires, and maybe a PC board at the back of the boiler, there are no allowances for easy DCC install. The headlight is also a big issue.  I wonder if they have special headlight PC board made for their DCC installs.

I do like the idea of a custom PC board in the tender with a Nex18 connector (for when the decoder has to be removed/replaced)

As for handling dual motors, I'm also in the dark. From what Kelly (ex. Kato USA DCC installer) mentioned when installing ESU decoders in Kato H0 P-42 diesels which also use dual coreless motors, the decoder handles it well.  For chuff timing, as I understand it, the decoder is not "smart" enough to be able to distinguish between the revolutions of each engine's motor (since they are wired in parallel to a single motor output).  BEMF based chuff timing can be fairly accurate, but to be in perfect sync, a cam would have to be used. Even with the cams, a dual cam input would be needed for true sync of the articulated engine (or using 2 decoders - one for each engine).

EDIT: There is no separate adjustment for chuffs for both engines. You will has have to adjust for 8 chuffs per driver revolution (ant that will likely not sound very realistic since they will always be evenly spaced).

[JeTFiN]

I guess I missed the "tender" part. I just didn't consider that part of the DCC-fication of this model to be difficult.  Since you are new here (that was your 1st post) I  had no idea about your level of knowledge, or your plans.

As far as making DCC conversion easy, due to the unconventional dual motor design, I really don't see any easy solutions Kato could have used to make this a  DCC friendly DCC install.  Funny thing is that Kato designers were most likely aware that Kato would be installing decoders in those engines and as you mentioned, except for come channels in the boiler frame for the wires, and maybe a PC board at the back of the boiler, there are no allowances for easy DCC install. The headlight is also a big issue.  I wonder if they have special headlight PC board made for their DCC installs.

I do like the idea of a custom PC board in the tender with a Nex18 connector (for when the decoder has to be removed/replaced)

As for handling dual motors, I'm also in the dark. From what Kelly (ex. Kato USA DCC installer) mentioned when installing ESU decoders in Kato H0 P-42 diesels which also use dual coreless motors, the decoder handles it well.  For chuff timing, as I understand it, the decoder is not "smart" enough to be able to distinguish between the revolutions of each engine's motor (since they are wired in parallel to a single motor output).  BEMF based chuff timing can be fairly accurate, but to be in perfect sync, a cam would have to be used. Even with the cams, a dual cam input would be needed for true sync of the articulated engine (or using 2 decoders - one for each engine).

EDIT: There is no separate adjustment for chuffs for both engines. You will has have to adjust for 8 chuffs per driver revolution (ant that will likely not sound very realistic since they will always be evenly spaced).

In that case, it almost looks like two decoders are in order unless what I'm about to describe occurs.  At the NMRA show in Grapevine (Dallas) last weekend, Vincent with Kato told me they are working on a proprietary decoder with Soundtraxx (correct me if I'm wrong) to use on this model.  Intuition says it is likely that the proprietary decoder will deal with the two motor outputs.  It didn't seem to him that they initially had plans to release the decoder separately but I kindly requested it since in general I believe having replacement parts available is nice and it would help those of us who purchased the DC versions with intentions to install decoders.  Kato will evidently be deliberating soon on whether or not the decoder will be sold separately - I'm hoping it does.

I'm really unfamiliar with specifically how sound decoders decide when to "chuff".  Is it possible that the presence of two motors in parallel would ruin any ability to count rotations, or is it possible that the decoder will receive signals that can (and therefore will) be interpreted to count poles of both motors, not just one (i.e. the decoder will sense a motor running twice as fast instead of two motors running at the real speed.


EDIT: Oh, I see now.  A cam is used on a driver shaft.  This gets rid of the black magic question in my head of "How are gear ratios accounted for?" but raises more questions regarding the mounting of cams and sensing of cam positions.  However, it does seem more likely that two cams can be used in parallel to adequately simulate chuffs.

FURTHER EDIT:  I am noticing that the wires on the videos shown before seem to be disguised as brake lines - a nice implementation.  This answers most of my questions.

[peteski]

The way decoders determine the chuff triggering is not very well known to me, but they certainly use a feedback from the motor.  The BEMF feedback loop is most likely candidate for that.  There are also CVs which are used to properly set the timing with any gear ratio, but a physical cam is the most accurate way to sync the chuffs with not only driver rotation but with the correct rotation angle of the wheels.  Yes, having each engine's cam in parallel connected to a single cam sensor would generate accurate chuff sync.  But this is not something that is usually done in N scale (but it is done in larger scales).

For more details on virtual cam synchronization procedure see the decoders manual.  ESU LokSound has the process described in section 13.3 of the LokSound 5 manual.  The best virtual cam I have worked with is in the old QSI decoders.

I am also aware that Kato is going to use Soundtraxx decoders, but I did not know that it would be custom made for this model.  Makes sense.

FURTHER EDIT:  I am noticing that the wires on the videos shown before seem to be disguised as brake lines - a nice implementation.  This answers most of my questions.

Brake lines? Which video is that?

[JeTFiN]

The way decoders determine the chuff triggering is not very well known to me, but they certainly use a feedback from the motor.  The BEMF feedback loop is most likely candidate for that.  There are also CVs which are used to properly set the timing with any gear ratio, but a physical cam is the most accurate way to sync the chuffs with not only driver rotation but with the correct rotation angle of the wheels.  Yes, having each engine's cam in parallel connected to a single cam sensor would generate accurate chuff sync.  But this is not something that is usually done in N scale (but it is done in larger scales).

For more details on virtual cam synchronization procedure see the decoders manual.  ESU LokSound has the process described in section 13.3 of the LokSound 5 manual.  The best virtual cam I have worked with is in the old QSI decoders.

I am also aware that Kato is going to use Soundtraxx decoders, but I did not know that it would be custom made for this model.  Makes sense.

Brake lines? Which video is that?

In the first video linked above watch the locomotive roll by - the wire setup is really clean.  I'll look at virtual cam synchronization soon.

[peteski]

In the first video linked above watch the locomotive roll by - the wire setup is really clean.  I'll look at virtual cam synchronization soon.

I watched the video multiple times and I didn't see any wires which would act as a cam.  Maybe you could take a screen shot of the video and point out where you see those wires?  Mechanical cams are usually hidden inside the frame.

Also, the chuffs are not set up properly.  you can clearly hear and see only 4 chuffs per driver resolution, which is correct for a 2-cylinder non-articulated loco.  Simple articulate locos will produce 8 chuffs per driver revolution because each engine with its separate set of cylinders produces 4 chuffs per its driver revolution.  But depending on the synchronization of the engines, sometimes the chuffs are distinctly audible, while other times they will blend together.  That is the effect chat could only be achieved using a cam in each engine.

However in N scale steam loco sound the chuffs are often set to less than prototypical because models are usually operated using out-of-scale (too fast) speeds and the extremely fast chuffs would not sound good.

[JeTFiN]

I watched the video multiple times and I didn't see any wires which would act as a cam.  Maybe you could take a screen shot of the video and point out where you see those wires?  Mechanical cams are usually hidden inside the frame.

Also, the chuffs are not set up properly.  you can clearly hear and see only 4 chuffs per driver resolution, which is correct for a 2-cylinder non-articulated loco.  Simple articulate locos will produce 8 chuffs per driver revolution because each engine with its separate set of cylinders produces 4 chuffs per its driver revolution.  But depending on the synchronization of the engines, sometimes the chuffs are distinctly audible, while other times they will blend together.  That is the effect chat could only be achieved using a cam in each engine.

However in N scale steam loco sound the chuffs are often set to less than prototypical because models are usually operated using out-of-scale (too fast) speeds and the extremely fast chuffs would not sound good.

I wan't referring to wires used for cams, only the wires between locomotive and tender to provide electrical connection between the decoder and lights.

[peteski]

I wan't referring to wires used for cams, only the wires between locomotive and tender to provide electrical connection between the decoder and lights.

Seems that I keep misunderstanding you. Sorry!

That is the method uses on almost ever DCC equipped steam model engine.   When the wires are not inside the drawbar then they are hanging loose under the drawbar. That nicely simulates all the hoses that are preset in the 1:1 locos.
I'm planning  on hiding the 3 wires needed in the top of the drawbar.

[dangerboy81]

There is a fella on Facebook, his page is called N scale Shenanigans. He just put 2 decoders in a Big Boy. Not sure why, I didn't really follow along, but I think it was so 1 set of drivers can slip when giving too much throttle. Here is a link to his page, or group, what ever it is. You may have to scroll down a few posts.
https://www.facebook.com/search/top?q=n%20scale%20shenanigans

[peteski]

Well, with current sound decoder technology using 2 sound decoders is the only viable way to get the chuffs getting out of sync when the drivers slip.

[JeTFiN]

There is a fella on Facebook, his page is called N scale Shenanigans. He just put 2 decoders in a Big Boy. Not sure why, I didn't really follow along, but I think it was so 1 set of drivers can slip when giving too much throttle. Here is a link to his page, or group, what ever it is. You may have to scroll down a few posts.
https://www.facebook.com/search/top?q=n%20scale%20shenanigans

Thanks for the link.  This is better than what I was looking for!