Mysteries of the new LokSound in the Intermountain SD40-2

[Ed Kapuscinski]

Have you tried installing caps on this yet.
Pls tell us as I am sure many of us would like to know
Thanks

Yep.

I was running some around the layout last night and the experience ended in me packing things into moving boxes... (to symbolically start the tear down, move, build something new in dead rail O process).

[mu26aeh]

I must have spectacularly clean track or I'm really lucky because mine run great with no added hardware, that is after cleaning the track.  Of course, I don't run my layout very often  so a regular track cleaning is the usual regime before running trains.

[Steveruger45]

Yep.

I was running some around the layout last night and the experience ended in me packing things into moving boxes... (to symbolically start the tear down, move, build something new in dead rail O process).

Hi Ed,
Which connection points did you use ?
 Peteski did a great detective  investigation and guidance but it left me and I guess some others wondering the ideal spots to solder the ground and cap positive too seeing as there appears to be three possibilities.
I'm thinking the green and red points or the green and purple points on his great explanation.
I have two of these sd40s and am getting pretty peed off at slow speed momentary sound drop outs on clean and good track that no other loco has a problem with.
Would appreciate a heads up on where you did it.
Much thanks
Steve

[peteski]

Hi Ed,
Which connection points did you use ?
 Peteski did a great detective  investigation and guidance but it left me and I guess some others wondering the ideal spots to solder the ground and cap positive too seeing as there appears to be three possibilities.
I'm thinking the green and red points or the green and purple points on his great explanation.
I have two of these sd40s and am getting pretty peed off at slow speed momentary sound drop outs on clean and good track that no other loco has a problem with.
Would appreciate a heads up on where you did it.
Much thanks
Steve

Electrically, all of the areas indicated by the colored markings are the same. I could have just chose to indicate single spot for positive and ground. But for sake of ease of installation I showed all the best accessible locations. You then choose which is the most convenient for your specific installation. Pick the connections which will result in the neatest install.

[Steveruger45]

Ah!
I was under the impression the stages mentioned mattered a big deal.
Thanks Pete.
I'm just waiting on a couple of OEM tips (ETU and ETS) for my Weller WES51 solder station to arrive and then give this a go.

[peteski]

Ah!
I was under the impression the stages mentioned mattered a big deal.
Thanks Pete.
I'm just waiting on a couple of OEM tips (ETU and ETS) for my Weller WES51 solder station to arrive and then give this a go.

Hold on - the stages are important!  I indicated in which stage (and which color code) the keep-alive cap should be installed.  I see that I made things too complicated by showing too many details of the circuit. 

Quote from my original post:
If adding a true SuperCap-based (hundreds of thousands  of micro Farads with its ancillary circuit to limit the charging current) keep-alive circuit it should be attached to green and red marked pads of the decoder.  Since the RED pads on the decoder are very small and close to other components, one must be super-careful not to damage any components while adding the keep-alive caps.

If the external caps are less than 1000uF total then my recommendation would be to attach them to the green and purple marked pads of the decoder.  Since the current-robbing motor is isolated from that stage, the keep-alive cap will supply power to the decoder for a longer time. Hopefully the flywheels will keep the loco coasting through the intermittent contact spot while the decoder keeps on running and producing its sounds.

To further clarify this, if you are adding a gang of tantalum caps, the total capacitance will be under 1000uF.   So, use green (negative) and purple (positive) marked pads.

[Steveruger45]

Thanks Pete, that was my original understanding from you fantastic post on this.
 Green/purple being for the usual approx 400uF.

I'm thinking maybe even 200uF might be enough as there is already 200uF on the board.
I will try that first and then add additional 200uF if needs be.

I will most likely use my poly tantalum caps as they would slide under the rear end of the board pretty easily.
I'm gonna do this soon, when I get my very fine solder tips,  so I will post the results.

Thanks again for the clarification

Steve

[Ed Kapuscinski]

Hi Ed,
Which connection points did you use ?
 Peteski did a great detective  investigation and guidance but it left me and I guess some others wondering the ideal spots to solder the ground and cap positive too seeing as there appears to be three possibilities.
I'm thinking the green and red points or the green and purple points on his great explanation.
I have two of these sd40s and am getting pretty peed off at slow speed momentary sound drop outs on clean and good track that no other loco has a problem with.
Would appreciate a heads up on where you did it.
Much thanks
Steve

Oh no, I haven't done anything. And the contact issues drove me to start planning a move to a new house.

[Steveruger45]

Hi Ed,
Thanks for reply.
BTW even with the thin 2mm thick poly tantalum caps the frame is going to have to have a bit filed off to get the caps to clear the decoder. Cra@. Oh well, que sera sera.  At least I now have my fine soldering bits. Need to finish my sd45 first and then I will have a go to cure the drop outs on the IMRC sd40-2's.

[Ed Kapuscinski]

The one thing I don't understand is why on earth sound decoder manufacturers don't just include these things from the factory. It's not like they're that expensive, and they make such a world of difference.

[peteski]

The one thing I don't understand is why on earth sound decoder manufacturers don't just include these things from the factory. It's not like they're that expensive, and they make such a world of difference.

While I suspect that the increased cost is probably part of the issue, the elephant in the room is increased size (bulk) of the decoder.  Manufacturers are striving to make those decoders as small as possible (so they can be squeezed into small scale models). Having a gang of large capacitors added to the board would not be optimal. Some manufacturers (like Digitrax) do include a capacitor tethered on wires off the circuit board, but those are also awkward to deal with.  If you follow the keep-live installs here you see that there are many different sizes of the gang of caps is installed in many different locations.  But if you had those large on-board capacitors you would really be limited on your installation options.

Then there is also a difference in short-power-interruption behavior of decoders from various manufacturers.   Some brands do a full and slow "reboot" on a power dropout, while other brands recover much faster.  That is related to the hardware and firmware design of the decoder.

If you look at the sound decoders for larger scales (like 0 scale), you will often see a gangs of large electrolytic capacitors on-board.  That is because space is not an issue there.

[jdcolombo]

I would rather manufacturers NOT include on-board keep alive for precisely the reason Peteski cites: keep the decoder itself as small as possible, so that I have many mounting options.  And then I have many other options for keep-alive caps; I've seen folks stash them in the fuel tank (after milling out a space to put them), in the cab, on the front chassis shelf, on the rear chassis shelf, etc.  It's not hard to roll your own keep alive; two or more caps in parallel wherever they fit and you're home free. 

But what I DO want is for manufacturers to have an easy connection (solder pad) for whatever keep alive I use.  ESU's standard LokSound Micro has such a connection; the new boards do not, which I think is a terrible decision on their part.  C'mon Man!  It's just a pad; 1mm wide by 2mm long.  Find a spot for it. 

John C.

[Steveruger45]

I would rather manufacturers NOT include on-board keep alive for precisely the reason Peteski cites: keep the decoder itself as small as possible, so that I have many mounting options.  And then I have many other options for keep-alive caps; I've seen folks stash them in the fuel tank (after milling out a space to put them), in the cab, on the front chassis shelf, on the rear chassis shelf, etc.  It's not hard to roll your own keep alive; two or more caps in parallel wherever they fit and you're home free. 

But what I DO want is for manufacturers to have an easy connection (solder pad) for whatever keep alive I use.  ESU's standard LokSound Micro has such a connection; the new boards do not, which I think is a terrible decision on their

John C.

Well said.   My thoughts exactly

[AKNscale]

Agreed

[Ed Kapuscinski]

I would rather manufacturers NOT include on-board keep alive for precisely the reason Peteski cites: keep the decoder itself as small as possible, so that I have many mounting options.  And then I have many other options for keep-alive caps; I've seen folks stash them in the fuel tank (after milling out a space to put them), in the cab, on the front chassis shelf, on the rear chassis shelf, etc.  It's not hard to roll your own keep alive; two or more caps in parallel wherever they fit and you're home free. 

But what I DO want is for manufacturers to have an easy connection (solder pad) for whatever keep alive I use.  ESU's standard LokSound Micro has such a connection; the new boards do not, which I think is a terrible decision on their part.  C'mon Man!  It's just a pad; 1mm wide by 2mm long.  Find a spot for it. 

John C.

Ok, ok. Let me refine my point.

I agree completely. They should make it as easy as possible to add them and decoders pre-installed in locomotives should come with them.