My 1st LokSound Select Micro installations/Kato AC4400CW/ES44AC

[wbgx]

Hello,

I recently installed 2 LokSound Select Micro decoders into a couple of my Kato locomotives.  Because of the similarities between the locomotive frames and shells(Kato AC4400AC & ES44AC) I was able to create a solution that would work for both with one idea I had to improve the conversion, or at least in my mind.

In addition to the photo I've attached, I'll posted a link to the Youtube video I made of the process.  I will admit, my narration is a little dry at points in the video.

After finding the forum a couple of days ago, I've discovered what a wealth of knowledge is here and if I had some of the knowledge shared in the forum it would have lowered the slope of my learning curve dramatically.  I need to dig into 'keep alive' add-ons and try to enhance my speaker cabinets on future builds.   

Thank you to everyone that has made contributions to this forum. 

[reinhardtjh]

Looks like a good start.  Did you make your own PC board for the rear connections?  I don't have any of those model Kato locos so I'm not sure what kind of light board they have inside.

[wbgx]

Yes both installations included a crude but workable board for mounting the resistors for the 402 LED's the 16x9 speaker/cabinet and the motor clips I salvaged from the OEM board.

(edit: I added a picture of the homemade board loaded with the components mentioned)

[wbgx]

Ok, I'm ready to try my ham-fisted attempt at adding a 'keep-alive' circuit to my installation(s).  I've read a lot of prior posts on the subject but unless I misunderstood, the latest trend is to use Polymer Tantalum capacitors, please correct me if needed.  I've got a basket at Mouser loaded with a few AVX Poly Tantalum 220uf 20v and 220uf 25v capacitors.  Oh, do the capacitors need to be baked if not used right away to drive off moisture?

I recognize that I need to get the capacitors in parallel and I've seen some very nice methods of mounting the capacitors in that configuration in members posts.

Next, and where it gets foggy for me is the zener diode and resistor(s) required. 

It appears to me that in some cases the zener/resistor are used with the capacitors and sometimes they are not.  I could be overlooking some details.

My track voltage is around 12v on a NCE system as measured by a RRampMeter if that helps define my power source.

I know I've asked a lot of questions and value your input.

[peteski]

Ok, I'm ready to try my ham-fisted attempt at adding a 'keep-alive' circuit to my installation(s).  I've read a lot of prior posts on the subject but unless I misunderstood, the latest trend is to use Polymer Tantalum capacitors, please correct me if needed.  I've got a basket at Mouser loaded with a few AVX Poly Tantalum 220uf 20v and 220uf 25v capacitors.  Oh, do the capacitors need to be baked if not used right away to drive off moisture?

I recognize that I need to get the capacitors in parallel and I've seen some very nice methods of mounting the capacitors in that configuration in members posts.

Next, and where it gets foggy for me is the zener diode and resistor(s) required. 

It appears to me that in some cases the zener/resistor are used with the capacitors and sometimes they are not.  I could be overlooking some details.

My track voltage is around 12v on a NCE system as measured by a RRampMeter if that helps define my power source.

I know I've asked a lot of questions and value your input.

Yes, those appear to be the correct capacitors. I would go for the ones rated 25V (to be on the safe side), but most people will use the 20V caps.

The baking procedure is for when these are placed on circuit boards and ran through a soldering machine which heats the entire board and all the components to the solder melting temperature.  In that environment the moisture absorbed by the epoxy resin encapsulating material can quickly turn into steam and crack the resin. You don't need to bake them when hand-soldering (where you only heat up the metal terminals for a relatively short period of time).

For capacitance less than few thousand micro-Farads you should not need any extra components (diodes, zener diodes, or resistors).  Just hook the capacitor bank up to the decoder's ground and the positive common voltage.

Those extra components are needed with very large capacitance circuits to limit the charging current while still allowing enough of discharge current to supply power to the motor and other decoder functions.

Nice to see a fellow modeler who etched his own PC boards. That usually results in much cleaner and neater installs.

[wbgx]

Thank you for the reply and explanation. 

I enjoyed the process of making the etched PC board.  It allowed me a way to make some of the work modular and learn how to do something new.  I've seen a few posts where the guys have made some very nice etched boards.

This is a fantastic forum.  Thank you!

[Steveruger45]

Nicely done.  Welcome to this forum too.  As far as keep alive caps I use either 100uF x 4 at 20v or 220uF x 2 at 20v or a combination of these to get around 400uF.  I use the 2mm thick poly tantalum caps exclusively now and find their size gives added flexibility to  make a cap block fit the space. 
If your going shopping for caps and you use a lokprogrammer You might also consider getting a few small 100uH chokes too and put one in series with the caps positive. They are about 30 cents each and similar in size to one capacitor.  It is believed a choke in this circuit may solve the read and write issue the lokprogrammer has on a decoder with caps installed.   

For speakers I do try to get as much cubic capacity as I possibly can on the enclosure.  I’m just wondering if you have some more space to use for enlarging that speaker enclosure. But maybe you are considering using that space behind the speaker for the caps?.
FYI
For an 8x12mm speaker I try for a minimum of about 500mm3 and for a 9x16mm a minimum of about 800mm3.  Just about any combination of depth, width and length is ok but in n scale it is usually the width that’s limited to that of the shell.

[peteski]

If your going shopping for caps and you use a lokprogrammer You might also consider getting a few small 100uH chokes too and put one in series with the caps positive. They are about 30 cents each and similar in size to one capacitor.  It is believed a choke in this circuit may solve the read and write issue the lokprogrammer has on a decoder with caps installed.   

Yeah, I recently recommended that in https://www.therailwire.net/forum/index.php?topic=44673.msg585207#msg585207 . It worked for me with a QSI sound decoder and added caps, but I have not yet tried it on ESU decoders.  Like I mentioned that type of circuit is recommended in the ZIMO decoder manual, in the section where they discuss adding extra capacitors (but the circuit could be used for other brand decoders).

[wbgx]

Thank you Steve and Pete. 

I've looked at getting an ESU programmer, but I've been told it will not directly interface with the Mac.  However, at the National Train Show just a few weeks ago, the ESU representative told me they were working on a solution.

I will definitely look into the chokes for the circuit as I'm sure I will be looking into programming at some point.

Thank you for the additional information on the speaker box.  I made both boxes just large enough to enclose the speaker.  I will rebuild the speaker box and follow your suggested volume recommendations.  I imagine I have enough space under the board for new speaker box and capacitors.  I can now see how a double sided board would be helpful.

Thanks again.  I appreciate any suggestions.

[peteski]

The LokProgrammer device uses a standard old-fashion serial interface to the computer. The problem is with the software it needs to run - there is no Mac version of that software available - only MS Windows.  But the bottom line is that like you said, it is not compatible with a Mac.  I wonder if you could install a virtual Windows machine on your Mac, then run the LokProgrammer from it?

[reinhardtjh]

The LokProgrammer device uses a standard old-fashion serial interface to the computer. The problem is with the software it needs to run - there is no Mac version of that software available - only MS Windows.  But the bottom line is that like you said, it is not compatible with a Mac.  I wonder if you could install a virtual Windows machine on your Mac, then run the LokProgrammer from it?

You can.  Many of us do.  I use Oracle VirtualBox and Windows 7 Pro.  Others user VMWare Fusion and some use Parallels.  I tried to get WINE (or Crossover on the Mac), but, while it runs the LokProgrammer GUI fairly well, it won't talk to the USB port correctly to run the hardware side.

A Mac native LokProgrammer software would be nice if that's what the ESU rep was hinting at.  It would be a lot of work as the current software is based on MS .NET and there is no real alternative for the Mac.  There is a Linux open source project to make the equivalent, but I"m not sure how ready for prime time it is.

[railnerd]

You can.  Many of us do.  I use Oracle VirtualBox and Windows 7 Pro.  Others user VMWare Fusion and some use Parallels.  I tried to get WINE (or Crossover on the Mac), but, while it runs the LokProgrammer GUI fairly well, it won't talk to the USB port correctly to run the hardware side.

A Mac native LokProgrammer software would be nice if that's what the ESU rep was hinting at.  It would be a lot of work as the current software is based on MS .NET and there is no real alternative for the Mac.  There is a Linux open source project to make the equivalent, but I"m not sure how ready for prime time it is.

I wouldn't hold my breath.  I think a Linux version is far more likely, but that would be an interesting starting point for a macOS port.

[wbgx]

Thank you for the helpful suggestions.   I've known of ways to get the ESU programmer to work with my Mac, but I hesitate to move too far that way on this computer.  I might invest in a inexpensive student MS laptop to allow for a dedicated train hobby pc with the ESU programmer and JMRI as I find a greater need to dig into the programming.

Maybe I shouldn't speculate but, is it possible that ESU could be working on a web based programmer but require the user to have proprietary hardware to complete the package(?)

[peteski]

Maybe I shouldn't speculate but, is it possible that ESU could be working on a web based programmer but require the user to have proprietary hardware to complete the package(?)

I really dislike the today's trend to move your data and applications to the cloud (even though I fully realize that there are many benefits of this arrangement).  I like to at least feel that I "own" the software I purchase by having it completely installed on my computer.  Adobe, Microsoft, and many others are moving to the web-based computing. I know that we will eventually be assimilated, but for now I run my apps locally and like to keep it that way. 

[Mike C]

The borg welcoms you !