Mysteries of the new LokSound in the Intermountain SD40-2

[jdcolombo]

Hi everyone.  The main thread on the new Intermountain SD40-2 has uncovered some mysterious things going on with the new LokSound board in the sound-equipped units, and I thought I would start a new thread where we can post discoveries about this new board that are not documented in the Intermountain or ESU documentation.

Here is what we have discovered so far. 

1.  There are three solder pads at the rear of the underside of the board, marked Aux3, U+ and Aux4, respectively.  These pads appear to be for connecting additional lighting or other auxiliary functions (maybe someone will produce an electrically-operated N-scale coupler?), although this has not been confirmed (the U+ pad is almost certainly a common positive - the equivalent of the blue wire on a wired decoder).  If this is correct, then these pads could be used to wire individual SMD LED's to the ditch light pipes and then program the decoder to make them alternate.  But so far, there is no confirmation from ESU regarding what these pads do.

2.  There are six (yep, SIX) LEDs already on the board.  Both the front and rear of the board have visible LEDs on the top and bottom.  The top LED's are for the front headlight and rear light, respectively.  The bottom front LED lights non-alternating ditch lights on models so equipped.  Unconfirmed speculation is that the bottom LED at the rear will be for future models that may sport rear ditch lights (e.g., NS units, maybe others).  We have confirmed that this rear bottom LED lights when the ditch lights are turned on and the locomotive is going in reverse.  In addition, there are two LEDs that are hidden by the black "tape" behind the visible top LEDs.  The LED behind the visible front LED blinks when F5 is pressed.  Here is a YouTube video of someone who has used this LED circuit to wire a stratolight (?) on his unit:

So far, no one has been able to activate the rear hidden LED using a function button.  This may be because that LED is not assigned a function from the factory, or because it is operated by a high-number function (F20, for example) that hasn't been tried yet.   We don't know (yet . . .)

3.  There is no obvious way to connect external keep alive to this board.  I believe based on what Matt Hermann (ESU's US operations head) told me at the National Train Show in Indy this past July that there is some on-board keep alive, which would be a new design feature for ESU (their current offerings do not have on-board keep alive, but DO have connections for external keep alive).  There is usually a way to connect external keep alive even to decoders that don't specifically have solder pads for this purpose, but we haven't figured that out yet, either.

If anyone discovers other hidden or mysterious aspects of this new board, please post them.  I have sent some questions to Matt Hermann via the LokSound Yahoo group, and will let you all know if I get any answers.

John C.

[peteski]

I think that it is a fairly good bet that the AUX function pads (and the common positive or "blue" connection( are wired to some DCC Function output after all that's how DCC decoders work. Unless it is some sort of SUSI or other proprietary interface.

I suspect that this decoder is based on the same electronics as other sound decoders made by ESU. So perusing a manual of the electronically-equivalent ESU decoder might reveal some clues. I'm also really surprised that more technical info about this decoder is not available (yet?). ESU or IM should have made a detailed decoder manual available (like most manufacturers do).   Maybe one os in the works but hasn't been published yet?

I'll most likely see the ESU reps at the Springfield show at the end of the month - if the questions are still not answered by the end of the month, I can ask them in-person.  I don't have a horse in this race (no SD40-2s for me), but I'm still curious.

[jdcolombo]

I think that it is a fairly good bet that the AUX function pads (and the common positive or "blue" connection( are wired to some DCC Function output after all that's how DCC decoders work. Unless it is some sort of SUSI or other proprietary interface.

I suspect that this decoder is based on the same electronics as other sound decoders made by ESU. So perusing a manual of the electronically-equivalent ESU decoder might reveal some clues. I'm also really surprised that more technical info about this decoder is not available (yet?). ESU or IM should have made a detailed decoder manual available (like most manufacturers do).   Maybe one os in the works but hasn't been published yet?

I'll most likely see the ESU reps at the Springfield show at the end of the month - if the questions are still not answered by the end of the month, I can ask them in-person.  I don't have a horse in this race (no SD40-2s for me), but I'm still curious.

Hi Peteski.

ESU isn't known for updating their documentation.  The current version of their LokSound manual dates to October 2012; they did release a one-sheet update for the Full Throttle features of their new files last year, but that's the extent of their "new" documentation.  So I'm not holding my breath on seeing technical info on this new board; I think we'll have to pretty much dig it out and document it ourselves.   The LokSound Select HO version does have six functions: front and rear headlights, plus four auxiliary functions.  So it is possible that the Intermountain board is also a six-function board.  The ditch light LEDs could be sharing a function; so could the "hidden" LEDs.  That would leave Aux3 and Aux4 to round out the six.  I have difficulty believing that this board has 8 functions, so that's why I suspect the ditch lights and hidden lights share a function.  But maybe it does - with modern locos sporting alternating ditch lights front and back, headlight, backup light, class lights, beacon, etc., maybe ESU decided to go with eight functions on this new effort. 

When I get the SD40-2 that I just ordered from MB Klein, I'll hook it up to my LokProgrammer and see what may be revealed there - the LokProgrammer will list all the function assignments as well as unassigned ones, so there may be things we can learn just from seeing the options available on the Lok Programmer software.

I don't really have a horse in the race either - not going to be running any Nickel Plate SD40's!  But I'm curious too - especially because this board is supposed to be the basis for the new "generic" drop-in board that is coming from LokSound (maybe that's when we'll get some technical info . . .)  If it really is an 8-function board, that would be VERY interesting.

John C.

[tehachapifan]

Is there any (solid) updates on the thinner "generic" board? The last thing I saw that I believe was posted here was that it was not to be, but I'm hoping that this was posted in error.

[jdcolombo]

Is there any (solid) updates on the thinner "generic" board? The last thing I saw that I believe was posted here was that it was not to be, but I'm hoping that this was posted in error.

All I know is that when I asked Matt at the Indy show in July, he said the generic version would be out in November.  Obviously, that hasn't happened.  I've heard as recently as a few weeks ago from someone else who claimed to have talked with Matt at that time that it's still coming.  But not a word on the web site or the Facebook page, or the Yahoo group.  BTW, I don't think this was supposed to be thinner than the Intermountain board - my understanding was that it was going to be essentially the same as the Intermountain board, without the front and rear LED's (which now seem to have proliferated like rabbits).

John C.

[sp org div]

I havnt removed the shell on my Kato SDP40F to inspect, but would guess its loksound select decoder is similar to what IM is using?
As per their pdf on katousa website, it is stated to have 8 functions. There they only give some basics, and send you to the esu website for further....
I would post a link but iphone doesnt want to let me readily attach that pdf/link....
Jeff
Edit-maybe that works...
https://www.katousa.com/cgi-bin/dl/dl.pl?dl.mif,SDP40F-LokSound-Guide.pdf

[rrjim1]

I did find a complete function mapping table for the new Intermountain SD40-2, helped me map the RUN8 and Coast functions, in case anyone is interested.

http://intermountain-railway.com/customerservice/dccwebpage/ESU-Diesel-Function-Mapping-93820-IRC-N-SD40-2.pdf

[jdcolombo]

I havnt removed the shell on my Kato SDP40F to inspect, but would guess its loksound select decoder is similar to what IM is using?
As per their pdf on katousa website, it is stated to have 8 functions. There they only give some basics, and send you to the esu website for further....
I would post a link but iphone doesnt want to let me readily attach that pdf/link....
Jeff
Edit-maybe that works...
https://www.katousa.com/cgi-bin/dl/dl.pl?dl.mif,SDP40F-LokSound-Guide.pdf

Yep - that sheet does say 8 outputs at 250ma per output (jeez, that's 2 amps total just for the function outputs).  And the function mapping table linked by rrjim1 shows the headlight, rear light, and six physical function outputs (though it shows the rotary beacon mapped to Aux3 - that doesn't seem to be correct based on the experiences of the guy who wired his stratolight, but maybe I'm not reading this table correctly).  So I guess this Intermountain board probably does have 8 outputs.  Wow.  Didn't see that coming.

John C.

[tehachapifan]

All I know is that when I asked Matt at the Indy show in July, he said the generic version would be out in November.  Obviously, that hasn't happened.  I've heard as recently as a few weeks ago from someone else who claimed to have talked with Matt at that time that it's still coming.  But not a word on the web site or the Facebook page, or the Yahoo group.  BTW, I don't think this was supposed to be thinner than the Intermountain board - my understanding was that it was going to be essentially the same as the Intermountain board, without the front and rear LED's (which now seem to have proliferated like rabbits).

John C.

Thanks, John. Yeah, by thinner I meant thinner than their current Select Micro. If this indeed comes to fruition, my wallet is going to up and leave me.

[mu26aeh]

I did find a complete function mapping table for the new Intermountain SD40-2, helped me map the RUN8 and Coast functions, in case anyone is interested.

http://intermountain-railway.com/customerservice/dccwebpage/ESU-Diesel-Function-Mapping-93820-IRC-N-SD40-2.pdf

 

I might as well as be reading French....

[reinhardtjh]

I might as well as be reading French....

German, actually.   

    The ESU LokSound function mapping is a pile to learn, but it's very flexible.  Much more so than just mapping a function to a button. You can map to buttons, with qualifiers if sound or even a certain sound is playing, direction, if other functions are on or off and more.  The most information is in the V4.0 decoder manual http://www.esu.eu/en/downloads/instruction-manuals/digital-decoders/  (Check out the Full Throttle info under the Select Decoder also).  There also is some additional information in the LokProgrammer manual http://www.esu.eu/en/downloads/instruction-manuals/lokprogrammer/ and there are some hints and tips available in the files section of the ESU LokSound Yahoo group.

[jdcolombo]

As John points out, the reason the LokSound function mapping table is so complex is because you can assign actions to a function button that are conditional logic based, and the buttons can control sounds, stuff wired to function outputs, motor behavior, etc.  Once you understand this, it's actually pretty logical, but the learning curve is . . . steep.  To really understand it, you need to print out all 4 pages in John's link, and then tape them into one long horizontal table.  Then you will see that for each Function button (1-24), there is an assigned set of Conditions, Physical Output, Logic Function and Sound Function.  So, for example, take a look at Row 1 in the table, which gives you the mapping for the headlight.  The default function key is F0.  The Conditions are that the engine is running forward and F0 is "on" (these are highlighted by the red boxes).  Then notice there are no more red boxes until you get to Physical Output, which is the headlight (white wire).  The table also tells you that the way you get to all this is to set CV257 to 20 (add the red boxes); it also tells you that in order to set CV257, you must first set CV32 = 2.  It also tells you that CV266 is set to "1".  All the other control CV's for this particular function map (F0, engine running forward) are zero (no red boxes).

OK.  Now suppose that you wanted the front headlight to stay on whether the engine is running forward or reverse.  Look at the "Conditions" column, and you will see that "reverse" requires a value of 8 in CV257 (if you understand binary programming, then what this is really telling you is that "reverse" as a condition is assigned to bit 3 of the 8 bits - bit 0 to bit 7 - and turning on this condition requires setting bit 3 to "on" which in turn equals a decimal value of 8 - but you don't really have to know this to use the table).  To get what you want, add up 4 (for "forward"), 8 (for "reverse") and 16 (F0 "on") = 28.  Program CV 257 to 28, and now your headlight will stay on whether the engine is running forward or backward.

Or, suppose you want the front ditch lights to come on automatically when you turned on the headlight and you want them directional, just like the headlight.  To do this, you have to assign the physical output for the ditch lights to F0; so if you look at the table under the "physical outputs" columns, you'll see that the front ditch lights are assigned to the Aux1 physical output.  To have it come on with the headlight, you add 1 (the "on" setting for bit zero in CV266) and 4 ("on" for bit 2) = 5, and set CV266 to 5.  Now when you turn on the headlights by pressing F0 and the engine is running forward, the ditch lights will also be on automatically. 

And so it goes - if you wanted for fun to have the bell come on with the headlight in order to drive some operator crazy, then you'd set CV270 to 8 (take a look at the columns marked "Sound Function #1" in the first row).  Now every time you turn the headlight on and the engine is running forward, the bell will also come on.

The Lok Programmer makes this easier, though - there are drop-down boxes that assign actions to function buttons, and most of the time it is quite straightforward, and you don't have to figure out which individual CV's to program and which values are needed for each CV. 

What I plan to do when I get my SD40-2 is hook it to my Lok Programmer, see what functions are assigned to what, and then start playing around to see how you can control the different lights (all six of 'em)  as well as test out the Aux3 and Aux4 connections. With the discovery of the hidden LED's, it seems to me that doing alternating front ditch lights should be pretty easy - cut off the existing ditch light LED (bottom front) and beacon LED (hidden behind the headlight LED), and wire up two 603-size SMD LED's glued to the light pipes for the ditch lights (the pipes will have to be cut).  Then you should be able to program a single function button to turn on both LEDs and have them alternate at a set timing.  But we will need to know what physical output is assigned to each of these LED's (I suspect that the front ditch light is Aux1, and the beacon is Aux5, with the rear ditch light being Aux2 and the rear beacon being Aux6, leaving Aux3 and Aux4 unused for the pads; but I'll be able to confirm that for sure when I get my hands on a decoder-equipped loco).   I'm not a modern-era guy, so I've never actually done this programming, but it has been described in detail on the LokSound Yahoo list and is one of the most common questions asked there.

John C.

[milw156]

John, Thanx for the translation! Think I learned more from you than reading the manual multiple times, how much do I owe you?!!!!
Rick

[AKNscale]

I'll check out my 40F and see what I can find. I tried to take it apart to no avail...

[jdcolombo]

John, Thanx for the translation! Think I learned more from you than reading the manual multiple times, how much do I owe you?!!!!
Rick

Well, in real life I'm a professor at the University of Illinois, so in theory I'm supposed to be able to explain stuff - although I'm sure some of my students would dispute that 

And since I work at a public university, funded in large part by grants and contracts from the federal government (the State of Illinois appears singularly uninterested in providing state funding for its universities these days), we'll just say "Your Tax Dollars At Work!" and call it even.   

John C.