New Atlas GP7/9 (40 005 3xx series) /ESU 59925 Decoder Issues

[peteski]

Rick,
I still don't quite understand this hate and frustration. The E24 connector is just that - bunch of electrical contacts in a plastic housing.  For whatever reasons ESU had (I don't believe they ever officially came out and said it, but one can make a good guess), they started using it on their own decoders. 

They produced the "universal" LokSound5 nano decoder, and multiple adapter boards (for N and H0 scale) which I suppose could be called "motherboards", and the nano decoder itself could be called a "CPU chip".  That way they could produce a just a singe decoder which could easily be adapted for multiple applications. Sounds like pretty clever plan to me.

I again have no knowledge as to whether ESU approached other model manufacturers to see if they adapt the E24 connector to accept the nano decoder for easy installation of DCC in them.  I also read that they approached the DCC officials to see if the connector will be accepted as a standard.

Seems that some manufacturers (Atlas) decoded to go with the E24 connector, and at this point, other manufacturers are also starting to utilize the E24 connector (as defined by ESU).  You need to keep in mind that this is the latest interface connector introduced into the DCC world, so some missteps and delays are not out of the question. 

The LokSound 5 nano E24 decoder has been available from vendors for quite some time. Your gripe seems to be with the fact that the non-sound decoders weren't available til recently.  But they are now. And from manufacturers other than ESU.

As far as the problems you are having, that has nothing to do with the E24 interface.  That is just a bunch of contacts. As you mentioned, if others had problems, we would be hearing about it.  I have no problems with my E24 decoders. 

Have you tried the reset procedure I described earlier?  If that did not help, I would be happy to check out one of the non-working models to determine what is wrong.  I have a LokProgrammer.  Or maybe someone else you know has a LokProgrammer and is willing to do some troubleshooting for you.  What puzzles me is that all 3 decoders seem to be misbehaving.  This should really be as simple as plug-n-play install.

[rickb773]

Also, what brand of DCC system are you using?  Since this is a non-sound decoder, you should have no problems reading and writing any of the standard CVs in the decoder (CV 1-6 and CV29 for example).  No need for LokProgrammer for any of this.

Digitrax EVOXD Evolution Duplex Starter; DCS 210+ command station
JMRI ver 5.8

Whether you have tested them on DC or not, first troubleshooting step I recommend is to reinstall the DC "dummy" plug and test them on DC.  If the motor and lights work correctly that verifies that the light board inside the model (including the E24 connector) is working correctly.

Report your findings after the above procedure.

I am trying to follow your directions.


Shells removed. DCC plug in lower right. All are the new (Jan '24) Atlas GP7/9 engines; 2 shells (WP,B&O) are older Atlas shells to be used on the new power units; the NYC is new with the engine. All are from the Atlas Silver Series 40 005 3xx.


The underside of the board; one with original plug in, one with plug off.





The elephant moves showing the lights performing properly in each direction under DC.

As you mentioned, if others had problems, we would be hearing about it.  I have no problems with my E24 decoders.

Can anyone verify a successful installation with the new Atlas GP7/9 and 59925?

I still don't quite understand this hate and frustration

Do NOT add me to the hate category, just extremely frustrated over the 7 months of effort.

I'll add the 59925s back in and try the rest. More photos to follow.
My sincere thanks for all your help.

[peteski]

Ok, the DC test verifies the proper functionality of the Atlas light board (and E24 connector).  The same E24 contacts will be used when the decoder is installed.

[mkearns]

I successfully put an E24 ESU 59925 lokpilot in my new Atlas Silver series GP7. Two screws, plug and play, no issues. It runs great on my friend’s layout. Only does like 8-10 cars though.

I did have a second silver series GP7 from the recent batch that, prior to a decoder install, died after traveling about 24 inches on DC. It just stopped moving and nothing I tried on the layout or the bench could get it to move again. Lights still worked but it wouldn’t budge. Motor and gears could rotate by hand, so it wasn’t locked up. I exhausted my skills and it’s back with Atlas now.

[rickb773]

I successfully put an E24 ESU 59925 lokpilot in my new Atlas Silver series GP7. Two screws, plug and play, no issues. It runs great on my friend’s layout. Only does like 8-10 cars though.

I did have a second silver series GP7 from the recent batch that, prior to a decoder install, died after traveling about 24 inches on DC. It just stopped moving and nothing I tried on the layout or the bench could get it to move again. Lights still worked but it wouldn’t budge. Motor and gears could rotate by hand, so it wasn’t locked up. I exhausted my skills and it’s back with Atlas now.

What did you use to program it?

[mkearns]

All I did was plug the decoder in, screw the board back on, and then change the DCC address to the road number. Everything else is stock to the ESU default settings and it runs great.

 I don’t know exactly what DCC system my friend uses. He has a programming track on his workbench with a laptop and JMRI decoder pro. That’s all I know and all I did. I’m not too DCC savvy. I run DC only at home. But like taking my C&O stuff to show off on his B&O railroad.

[peteski]

All I did was plug the decoder in, screw the board back on, and then change the DCC address to the road number. Everything else is stock to the ESU default settings and it runs great.

That's exactly what I would expect from a factory-fresh ESU LokPilot decoder (or from any other brand of factory-fresh decoder).  From the basic operations perspective, ESU is not different than any other factory-fresh decoder.  Sure, it has some advanced lighting capabilities if one chooses to configure them, but a factory-fresh decoder only needs the address changed from the default "3".  If it doesn't behave that way (and the "light board" is in working order) then decoder is either defective, or someone messed around with it's CVs.

[jagged ben]

FWIW 'dcc ready' used to just mean that the motor was isolated from the frame and connected to the light board.   Then manufacturers copied the light board dimensions to make their decoders.  None of this having to match tiny proprietary connectors. Personally I'm not at all enthused about the E24.    I can't imagine needing 24 pins for any DCC install, and in my opinion they are too tiny to be robust enough for N scale motor currents and hobby handling.  6-pin and Next18 are better.   

[peteski]

Personally I'm not at all enthused about the E24.    I can't imagine needing 24 pins for any DCC install, and in my opinion they are too tiny to be robust enough for N scale motor currents and hobby handling.  6-pin and Next18 are better.   

True, in N scale we usually don't need all those different outputs, but in larger scales they come in handy.
Here is a link to Digikey for that connector: https://www.digikey.com/en/products/detail/molex/5050702422/9693946

Looking through the tech. specs, the pins are rated for 0.3A each, and the end connections (called "nails") are rated for 3A.  Motor output pins are doubled. While that does seems a bit weak, contemporary motors don't need a lot of current, and ESU even piggybacks the E24 nano onto their H0 motherboard: https://www.esu.eu/en/products/loksound/loksound-5-dcc-direct/  and https://www.esu.eu/en/products/loksound/loksound-5-dcc-direct-with-integrated-powerpack/

Next18 connector capability would be quite sufficient for most uses, but it has quite a bit of higher profile, which is not a good thing for tight N scale installations.  But I agree that if Next18 connector had a lower profile, I would prefer it to the E24 connector.

The 6-pin connector is very simple and robust, but it doesn't even provide the common positive function output, and only 2 functions.  How abotu if one wants extra lights like ditch lights and also sound (speaker). Those all have to be hardwired, making the connector sort of useless as an easy way to install the decoder.

[rickb773]

If that's the case, reinstall the LokSound decoder, then let's go back to basics.  Instead of using DecoderPro, just put it on your  command station's programming track and reset the decoder by writing 8 to CV8.   Then tilt the loco for couple of seconds to disconnect one set of wheels from the track (not needed, but won't hurt to do).  After this you should be able to address and run the loco using short address 3. It should also have the standard directional headlights turned  on by F0 button on the throttle. Just like any other brand of DCC decoder.

Also, what brand of DCC system are you using?  Since this is a non-sound decoder, you should have no problems reading and writing any of the standard CVs in the decoder (CV 1-6 and CV29 for example).  No need for LokProgrammer for any of this.

Report your findings after the above procedure.

When I set CV8 to 8 and run on DCC address 3:
Lights work when running in each direction
EXCEPT in one of the 3 locos the forward light won't go out.

And I have very, very little speed control range.

My goals are:
1) Set locomotive number
2) Lights in proper direction
3) Log Curve speed table for mainline engines with adequate high end speed.

I lose everything (including lights) when I attempt to program with JMRI (the only way I've been taught to program a DCC engine)
It just tells me the locomotive will not acknowledge the commands (error 308)

If that's the case, reinstall the LokSound decoder, then let's go back to basics.  Instead of using DecoderPro, just put it on your  command station's programming track and reset the decoder by writing 8 to CV8.

Does that mean using JMRI Tools>Programmer.Single CV Programmer?
That is the only other thing I know how to do

[peteski]

Rick,  the one decoder where the forward light does not turn off likely is a defective decoder.  I have heard of several ESU decoders where some AUX function output (like the headlight) is either always stuck on or off.  Well, it might be corrupted firmware on the decoder, but since you have no ability to upload firmware to the decoder we have to just call it defective.

I lose everything (including lights) when I attempt to program with JMRI

That to me shows that Decoder Pro (JMRI) does in fact modify bunch of decoder's CVs, scrambling things.

Does that mean using JMRI Tools>Programmer.Single CV Programmer?
That is the only other thing I know how to do

This tells me that you have been taught to use a higher level DCC programming (via Decoder Pro) and never got familiar with basic programming by directly using your Digitrax DCC system (going into programming mode on the Digitrax throttle and programming one CV at a time on programming track).  That method would be needed to take the DecoderPro out of the picture. 

Exactly how were you able to reset the decoder if you aren't using your command to program CVs directly on the programming track?

I don't use Decoder Pro (or Digitrax DCC), so I can't really provide good troubleshooting info. That is why I suggested (after resetting the decoder) to try using the command station directly to program few CVs directly. Like the 3-point speed curve (CV2, 5, and 6).

[Dupesy]

Peteski, is it possible this is a power issue?  I don't know how he's set up his programming track, but maybe its something like a PR4 not having the juice to properly read/program an ESU decoder and needs a programming booster, similar to how a booster is needed for programming some sound decoders?

[peteski]

Peteski, is it possible this is a power issue?  I don't know how he's set up his programming track, but maybe its something like a PR4 not having the juice to properly read/program an ESU decoder and needs a programming booster, similar to how a booster is needed for programming some sound decoders?

Well, I agree that there seems to be some sort of compatibility issue, but in this instance ESU LokPilot is just a plain motor-control-only decoder, not like a LokSound decoder. It doesn't require more power than any other motor-only decoder (like Digitrax DZ126 for example). I think there might be something with the Decoder Pro going on, but I'm not really up on that facet of DCC.

[Lackawannae8]

I went through hell and back with a pair of brand new FA-1s one gold one silver which i dropped my own decoder into and had nothing but problems... they tried to tell me i was installing the board wrong...LMFAOOO I can do esu hardwire jobs in my sleep. they wouldnt admit the sisterboard had an issue which as well had i narrowed it down to being after grief with the motor tabs and speaker tabs that sideline the motor and frame i isolated all those probable issues and still blowing decos out of the box... the new stuff sucks! im scared about the next run of sd45s coming out soon

[jagged ben]

True, in N scale we usually don't need all those different outputs, but in larger scales they come in handy.
Here is a link to Digikey for that connector: https://www.digikey.com/en/products/detail/molex/5050702422/9693946

Looking through the tech. specs, the pins are rated for 0.3A each, and the end connections (called "nails") are rated for 3A.  Motor output pins are doubled. While that does seems a bit weak, contemporary motors don't need a lot of current, and ESU even piggybacks the E24 nano onto their H0 motherboard: https://www.esu.eu/en/products/loksound/loksound-5-dcc-direct/  and https://www.esu.eu/en/products/loksound/loksound-5-dcc-direct-with-integrated-powerpack/

It's not the amp rating under normal operation that worries me (although I'm still bit curious how that amp rating would hold up to the heat build-up after a couple hours continuous running in a heavy draw loco, say an Athearn FP45).  What bothers me is the fact that such tiny connector pins would be more easily damaged and impossible to repair when hobbyists maintenance their locomotives, which is something MRR manufacturers should expect.  I've already seen other complaints here about how difficult it is to line and up and install the E24 connector if you choose to buy a DC model and add a non-sound decoder.   In my opinion it's technology designed for factory assembly of non-user-repairable throw-away consumer electronics being used in a limited run collector model that must be the opposite of all those things: hand (re)assembled, user-repairable, and never thrown away.  I'm concerned for the number of shelf-queens there may be in our collective future.

The 6-pin connector is very simple and robust, but it doesn't even provide the common positive function output, and only 2 functions.  How abotu if one wants extra lights like ditch lights and also sound (speaker). Those all have to be hardwired, making the connector sort of useless as an easy way to install the decoder.

Kinda disagree with you here.  ESU makes a 6-pin connector with extra wires for extra functions, and I plan to use them in my FVM GP60s where the extra function LEDs will be attached to the shell, so the 6-pin connector should still make a nice way to disconnect the shell from the chassis and motor when maintenance is needed.  So it won't be pointless at all.  But yeah, the 6-pin isn't going to make sense for sound so I don't expect to see new models designed for it nowadays.