I’m hoping someone here can provide some insite into a problem we’re having with our Tower 55 ES44AC. We bought it used from our Local Hobby Shop (who have been great to work with). Technically, it belongs to my 5 (almost 6) year old. You’ve never seen a kid that’s more crazy about trains. Anyway the other day we were running it in a two loco consist with a bachmann Gp40. For no apparent reason, all the trains on the layout stopped and wouldn’t move. Our controller is a Digitrax Zephyr extra which at that point was switched into booster mode. We reset the power to the layout, and eventually I was able to figure out how to restore the Zephyr to command station mode.
However, after we got all of that sorted, the Tower 55 was behaving erratically. It would only creep along even at the highest throttle setting. The lights could not be properly controlled and were flickering instead of burning steadily. Given the issues, I performed a decoder reset by writing 08 to CV08. That fixed all of the issues, but now I cannot seem to assign the loco a road number. In fact, I cannot seem to program any of the CV’s at all. I have tried using the program track (my normal method) and on the mainline.
From my research, I know the decoder in the T55 models were custom built for T55 by digitrax. However, nobody at digitrax has any knowledge of their workings. I have also read that the are issues programming the T55 models with the Zephyr. At this point I’m at a loss on what to try next. My local hobby shop does have a test track set up in their shop, so my next move (If I cannot figure out any options he
“To program the locomotive electronics to use a 4-digit address using the Digitrax Zephyr model DCS50 or DCS51 throttle, the user must manually set CV17, CV18 and CV29 to enable the 4-digit address capability of the decoder.”
I’ve got three of the T55 diesels. And had lotsa trouble programming them.
Until.
I put a sort of signal booster onto the program track. It’s screwed down, so I can’t look at the underside and see if it says who made it. But I know it’s pretty common, so your local shop should be able to figure out what you need.
Anyway, I installed the critter and easily programmed all three with my NCE system.
I was able to get her reassigned from 03 to (ironically) 55 programming on the mainline. However, I wasn’t able to write to CV29, CV17 or CV18. I was successful using the Ad2 quick programming mode. I still think there may be some issue with the decoder, but for our limited application I’ll call this a win. I still might take the loco by our LHS to see if they have any better success. My boy is already giving me static that “Dad, 55 is NOT his number…” [(-D]
I’m reluctant to try programming in blast mode, simply because we have a Ds64 that has an actucal switch programmed to switch #7. I’m not sure if that makes any difference or not.
This is the reverse of the usual problem with sound decoders - the Zephyr can usually program (even without using Blast Mode) decoders that the DCS100 can’t. It may not be a power problem, but rather one of timing. Some of the supposedly “unprogrammable” decoders (without a booster) I can’t program from my Zephyr console, but using a DT40x throttle plugged in, they program just fine. I’ve always figure the issue is that when setting a long address, the Zephyr sends out the values for CV 17, 18 and 29 in too fast a sequence for some decoders to respond. There’s a built-in delay with a DT40x because before it sets CV29 you get the AD4=? prompt and have to press the Y button to confirm - only then is the CV29 value sent to the program track.
It should be fairly easy to validate this - try setting a long address witht he Zephyr, and then even if it doesn;t appear to take, check the values of CV17 and 18 - they probably are the correct values for the long address being used, but if you look at CV29 it’s probably still set for a short address (and the loco probably still runs as address 3).
Unless I’m doing something wrong (which I’m not sure that I am because I’ve done it before), CV’s 17 & 18 read back 255, which is one of the reasons I suspect deeper decoder problems…
255 just means the decoder never responded. The decoder can’t tall the command station waht the value it, when doing reads, the command station basically asks “is the value 0” and either gets an acknowledge that it is, or nothing, then it asks if the value is 1, and 2, and 3… all the way up to 255, and if there never was an acknowledge, you see the last tested value, 255.