Uploading a new sound file to loksound select decoders

[Steveruger45]

Hi Guys,
Well I just bit the bullet and got me a lokprogrammer and a decoder tester.  Just installed with latest firmware etc on my laptop too.
Now I want to put a different sound file on a hard wire select micro and I am not sure of the process, can anyone give me a brief procedure on how to do this without me screwing it all up?

Secondly, same query on a new hardwire and slide in variants of these decoders that are new and not installed in a loco already.  I can see I just plug the 8 pin plug in the tester for the hard wired select micro but how to connect the slide in variety?

 I am a bit paranoid about messing things up and would appreciate some guidance before I do.
I have looked on YouTube university and manuals etc but am still a little unclear.

[jpwisc]

I follwed the steps in this video for my first file swap. It worked like a charm.

[jdcolombo]

For the engine with a decoder already installed:

Connect a piece of track (we'll call it the "programming track") to the LokProgrammer.  Connect the LokProgrammer to your computer, fire up the LokProgrammer software, and open the new sound file in the software.

Now put your loco on the programming track.  If it has a keep-alive installed, disable it (e.g., unsolder one of the wires from the keep-alive), because I have found that keep-alives sometimes cause communication errors between the LokProgrammer and the decoder.

Now at the top of the LokProgrammer software, you will see icons for "read decoder data" "write decoder data" and "write sound data."  Click on the "write sound data" icon.  A dialog box will come up, asking you if you also want to write the decoder data and if you want to set the new file as the default.  To make the correct choice, you have to understand the difference between "decoder data" and "sound data."  The "sound data" is just the sound file; writing a new sound file will NOT affect any CV's that you might have already programmed, such as the engine address, speed curve, and so forth.  These programmable CV's are the "decoder data" - even the setting for the various sound levels are "decoder data" NOT "sound data".  So if you want to do so, you can replace just the sound file, and leave everything else alone.

HOWEVER . . . that's often not a good idea.  The reason is that the new sound file may have new sounds (surprise!) or have stored sounds in different sound slots from the old file.  If you just write the sound data without writing the decoder data, then strange things can happen, with oddball sounds coming on when you press different function keys (which were set up to operate with the OLD sound file).  So my recommendation would be to write BOTH the decoder data AND the sound data.  This, unfortunately, will reset any CV's you previously programmed to factory defaults - e.g., the default address will be 03, the function keys will all be set to their default actions, the decoder will have the default ESU speed curve, BEMF settings, sound volume settings and so forth.  But in my experience, this is the best way to load a new sound file.  BTW, if you have made several CV changes in your decoder, what you can do is do is open the LokProgrammer, and then BEFORE loading the new sound file, have the LokProgrammer read the decoder data from your engine, and then save that file somewhere.  Now you will have a record of all your CV values and can use this record to reprogram your engine after loading the new sound file.  Note that I would not simply wholesale write these old values to the newly-programmed decoder, because then you can end up with the same problems caused by doing a "write sound data" only.  But at least you will have a reference to go back to in setting up the reprogrammed decoder.

After you are done with all your programming, reconnect your keep alive (if you have one).

For the drop-in boards, I can think of two options.  The first is to solder some wires to two of the track power pickup pads, the motor pads, and the speaker pads and then connect these to the terminal block on the ESU decoder tester and then program.  The second option is to wait until you've done the installation in an engine, and then program it.  This latter option isn't optimal, because if the decoder is bad before install, you won't catch it this way.  But I haven't used any of the new boards, so perhaps someone else has a better idea . . .

John C.

[Maletrain]

The first is to solder some wires to two of the track power pickup pads...

I am not (yet) familiar with these decoders, so I am wondering if it is not possible to use small "alligator" clips to attach these wires to the track power pick-up pads in order to do testing.

[RBrodzinsky]

First, welcome fully to the dark side!   I can spend more time playing with LokProgrammer than I actually do running trains!

Loading files:  There are really two distinct parts to this question, depending on if you want to preserve custom function mapping, but it comes with a significant caveat (more on this following), but let's take the simple case, first -- loading a new project into a decoder

Step 1 - make sure LokProgrammer hardware is connected to PC via USB  (duh, but....)
Step 2 - make sure the output of the LP hardware is connected to either LokTester or programming track (another duh, but... )
Step 3 - load the project into LP software; make any customization changes (address, function maps, etc).   SAVE PROJECT WITH NEW NAME
Step 4 - with decoder on tester (or in loco with any non-ESU keep alive disconnected) -- update decoder firmware (Tools | Update decoder firmware)
Step 5 - press Write Sound data.  Select Write decoder data and make default, but some folks will do this in two steps.
Step 6 - find something else to do for approximately 30 minutes while the sound file writes.
Step 7 - use the Control Cab in the software to test the install.

The biggest gotcha is the parenthetical remark in step 4.  You may not get a good load if you try to write with the capacitors installed, and usually an error message.  I have found that I can write CVs (write decoder data) with <600uF of keep alive, but cannot read or write sound (ymmv).

Now - if you want to preserve all the custom matching, speed tables, BEMF calibration, etc. you have painstakingly already done with the loco, you need to either first READ DECODER DATA into the project, or copy all the data manually into the proper tabs before writing.  Normally, the read data is ok, but it is important to study the sound slots and function mappings of the two projects, first.

As to the new micro direct boards -- if you want to use on the tester, attach track and motor leads to pads on the board (I use an old snipped off 8 pin from one of my hard-wired installs).  And attach speaker leads, too.

[Steveruger45]

Thank you So much everyone for the great responses.  I don’t feel as stupid as I did before to have asked.
For the slide in type I am thinking of making some sort of plug in receptor that I could connect to the tester terminals to avoid soldering leads on pads on a new decoder.  But that would depend on how many I plan to do. 
Perhaps esu would have such an animal? Or if not think about making one.
Thanks again guys this will help me get my feet wet without too much angst.

Oh, and yes, me too, I seem to be playing more with decoder installations and computers now rather than trains, but I enjoy it so I guess that’s why we all do it.

[reinhardtjh]

I bought a set of these mini test clips to clip onto the 73100 and 73199 boards.  They work okay for the rail power and motor pads but getting to the sound pads or any of the AUX outputs is tough.  I've been thinking about buying some chassis parts from Atlas or maybe Kato to create a jig for holding the boards but I haven't determined which model would be the best yet.

https://www.amazon.com/gp/product/B06Y14QHCQ/ref=oh_aui_detailpage_o08_s01?ie=UTF8&psc=1

[peteski]

I am not (yet) familiar with these decoders, so I am wondering if it is not possible to use small "alligator" clips to attach these wires to the track power pick-up pads in order to do testing.

Yes, I prefer this type of a hookup rather than depending on the track/wheel/chassis contact during programming. And of course, when the decoder is not installed then this is the way to go for programming and testing it.  If the decoder is not installed it is a good idea to also use a pair of jumper wires to temporarily attach a 12V motor to the decoder (as it is used to generate a load needed for the electric load which acknowledges certain programing operations). And I  attach the jumpers directly to the LokProgrammer. It has banana-type sockets.

[jdcolombo]

I wonder if someone might do a 3D printed "jig" that would mimic a frame, with some kind of spring contact (phosphor-bronze strips) for the track power and motor pads with wires attached to the contacts.  On the 73199, you could also have these spring contacts for the speaker pads, which are on the bottom.  For the 73100, you'd still have to solder wires to the top-mounted speaker pads, but you're going to have to solder wires to those pads anyway eventually.

The "jig" could be just a quarter-inch thick and just long enough to hold the board (66mm).

@Lemosteam - I see a market opportunity!

John C.

[Steveruger45]

I bought a set of these mini test clips to clip onto the 73100 and 73199 boards.  They work okay for the rail power and motor pads but getting to the sound pads or any of the AUX outputs is tough.  I've been thinking about buying some chassis parts from Atlas or maybe Kato to create a jig for holding the boards but I haven't determined which model would be the best yet.

https://www.amazon.com/gp/product/B06Y14QHCQ/ref=oh_aui_detailpage_o08_s01?ie=UTF8&psc=1

Many thanks.  And yes, what you are planning is exactly what I had in mind too.

[Steveruger45]

I wonder if someone might do a 3D printed "jig" that would mimic a frame, with some kind of spring contact (phosphor-bronze strips) for the track power and motor pads with wires attached to the contacts.  On the 73199, you could also have these spring contacts for the speaker pads, which are on the bottom.  For the 73100, you'd still have to solder wires to the top-mounted speaker pads, but you're going to have to solder wires to those pads anyway eventually.

The "jig" could be just a quarter-inch thick and just long enough to hold the board (66mm).

@Lemosteam - I see a market opportunity!


John C.

Yep, agree..
 In fact the top of an IMRC SD40-2 has such an arrangement already and could be used as a jig in itself, if you had one of these locos lying about, just slide out the fitted decoder and slide your new one in for testing/ programming.   Wouldn’t work for the 73100 though.
If Atlas will use the 73100 in there soon to be released gp39’s or others then similar could be done there too, I guess.
Anyhow I would prefer a jig for this stuff. I don’t like disturbing locos that are set up unnecessarily.

[peteski]

I wonder if someone might do a 3D printed "jig" that would mimic a frame, with some kind of spring contact (phosphor-bronze strips) for the track power and motor pads with wires attached to the contacts.  On the 73199, you could also have these spring contacts for the speaker pads, which are on the bottom.  For the 73100, you'd still have to solder wires to the top-mounted speaker pads, but you're going to have to solder wires to those pads anyway eventually.

Speaker is not required for uploading sound files or to read CVs, but if one wants to fully test the decoder that would come in handy.

[Lemosteam]

So the point is wire the whole thing up, install sound file. test, if OK install in loco?

Who wants to mail me one of each board?

[peteski]

So the point is wire the whole thing up, install sound file. test, if OK install in loco?

Who wants to mail me one of each board?

I posted measurements of each decoder in https://www.therailwire.net/forum/index.php?topic=44324.0 . Maybe that'll give you a head start?

[Lemosteam]

@peteski , just need pad dimensions (Length and width) and thickness at pads (could you measure each pad on each board to establish a range?) added to those great views!

Don't do a lot of DCC yet so I was not aware of that nice writeup.