What gives with this Tsumani on the Prog. Track ? Happy Ending!!!

[Gerome]

I thought I had a dead decoder.  It didn't respond after a while on my Zephyr controller....just the dnd decoder not working error message.  I had already sent it in once for repair.

So when I set it on the mainline, it started up.  It responds to OPs mode programming but with a Zephyr, I can't programme the address on the main.......(or can I, if I calculate CV 17 and 18 I should be able to do it right?)

But back to the decoder.  Is it the Tsunami or the Zephyr that is causing the problem.  I know Zephyrs are not the top of the line in performance, but I don't know what to think.

[selector]

I can't answer the Zephyr question with respect to why it won't seem to deal with your Tsunami, but I would try setting CV 29 to 34 in Ops Mode, and then dial in the new address to see if you can blow the whistle.  If you get the toot, you are in business.

My Digitrax manual for the SEB says to use Paged Mode preferentially, but they don't tell us to never use Ops Mode to assign addresses.  In fact, in the DT400 manual section dealing with addressing decoders, it says, "If you are using Ops Mode, you can program anywhere on the main."  Mabye the Tsunami will behave for you if your CV 29 setting doesn't kick it into life if you then try it all over again in Ops. (?)

[CSX Robert]

...
So when I set it on the mainline, it started up. It responds to OPs mode programming but with a Zephyr, I can't programme the address on the main.......(or can I, if I calculate CV 17 and 18 I should be able to do it right?)
...

With the Zephyr, you can program the address on the main, but if you are using the built in throttle, you do have to program each CV(17, 18, and 29) individually, the automatic programming of 4 digit addresses in OPS mode does not work from the Zephyr throttle. If you have a DT300 or DT400 throttle, then you can program the 4 digit address in OPS mode automatically, even if your command station is a Zephyr.

I'll point this out too. If you get into a situation where you don't know what address the decoder is set to respond to, you can do OPS mode programming to address 0(you'll have to remove all other engines from the layout to keep from reprogramming every one of them).

[Gerome]

Thanks for the tip CSX Robert but this one is a bust.  I have seveal other Tsunamis that work just fine on my setup. This one is going back.

It will not respond on the programming track [just the dnA error on the Zephyr] and setting the Zephyr to blast mode did not do the trick.

Moreover, after trying to set CV 29, 17 and 18 on the main to the correct numbers for the loco address IT LOST ITS ADDRESS COMPLETELY, and only using the trick of programming to address 0 on the main to reset the default worked.  So it goes back to address 03 but will not take any CV address changes after that. It just will not respond to CV 29 extended or CV 17 and 18.  It is like it is stuck on address 03.

I can, however change other CVs...like motor control and sound.

The diagnostic lights are good on the decoder itself.

Back to the maker.

[betamax]

The solution is probably a programming track booster. I found them to be erratic, and even downright stubborn, without a booster.

After getting a booster, no problems programming the Tsunami.

[CSX Robert]

...after trying to set CV 29, 17 and 18...

Did you try to change them in this order? If so, if you are programming to address 03, after setting CV29, the decoder would stop responding to the short address so that CV17 and 18 would not take. If you are programming to address 0, the Tsunamis do not let you change the extended address with OPS mode programming if extended addressing is enabled, so if you change CV29 first, it still will not take the changes to CV17 and 18. In other words, always change CV 17 and 18 before enabling extended addressing.

[wjstix]

It sounds like you're putting in a three or four digit address, but the decoder isn't accepting it. Rather than accessing CV 17 and 18 directly, try reading the four-digit address thru the Zephyr in regular programming. If 0003 pops up, put in the number you want and hit the button on lower right to save the change.

[trainnut1250]

Tsunamis (and other sound decoders) draw lots of current due to the sound amplifier in the decoder.  This draw often overwhelms the weak current supplied to the programming track by the command station and messes with the programming commands.  It also makes it difficult to read back CVs using the program track. 

I have worked around this problem with my CVP system by programing CVs 17, 18 and 29 in OPs mode using hexidecimal values.  This method also solved some headlight issues with some QSI decoders.  This suggestion came from a soundtraxx tech.  

I agree with Betamax that a programing track booster should solve the problem...Some newer systems have beefed up the programming track current and don't need the booster, don't know about the Zephyr.

 

Guy

[Gerome]

The solution is probably a programming track booster. I found them to be erratic, and even downright stubborn, without a booster.

After getting a booster, no problems programming the Tsunami.

Yes, Betamax, I think you are right. This one probably needs a booster and I am ordering one now. (BTW do you know if regular DSDLCs can be as fussy without a booster? I have a couple that act odd as is).

...after trying to set CV 29, 17 and 18...

Did you try to change them in this order? If so, if you are programming to address 03, after setting CV29, the decoder would stop responding to the short address so that CV17 and 18 would not take. If you are programming to address 0, the Tsunamis do not let you change the extended address with OPS mode programming if extended addressing is enabled, so if you change CV29 first, it still will not take the changes to CV17 and 18. In other words, always change CV 17 and 18 before enabling extended addressing.

I love happy endings. Thanks a big bunch to CSX Robert. Why didn't I know that I had to set CV 17 and 18 then CV 29 on the Tsunamis? Why that order? I dunno, but it did the trick and I was able to program in the correct loco address.

As with all these puzzles, I now would like to know why I didn't know this. Is it written in the Tsunami literature somewhere? My other decoders don't work that way as far as I know. I have done a few NCEs without worrying about this order, I believe.

[CSX Robert]

...
I love happy endings. Thanks a big bunch to CSX Robert. Why didn't I know that I had to set CV 17 and 18 then CV 29 on the Tsunamis? Why that order? I dunno, but it did the trick and I was able to program in the correct loco address. 

As with all these puzzles, I now would like to know why I didn't know this. Is it written in the Tsunami literature somewhere? My other decoders don't work that way as far as I know. I have done a few NCEs without worrying about this order, I believe.

I love happy endings too, I'm glad you got it going. The order is important to all decoders, but only when doing OPS mode programming. Here is what happens: The decoder has the default primary( short or two digit) address of 3. You select address 3 and program CV29 to enable the extended(long or four digit) address. At this point, the decoder will no longer respond to it's primary address but instead be looking for commands that match the extended address, so when you try to program CV17 and 18, they do not take. This is not a problem when doing page mode programming(useing the program track outputs) because you do not send commands to a particular address. Whatever loco is on the progam track will respond to all program commands, so if you program CV29 first, the loco still responds to the commands to change CV17 and 18.

[locoworks]

   AMEN to that.   very interesting.

[Paul3]

Last night, I installed a TSU-1000 diesel Tsunami decoder for an Alco 251 in an Atlas Classic RS-11.  The unit was programmed and tested out on a DCS50 Digitrax Zephyr with a DT300 throttle (all programming was done with the DT300).  There was no programming booster.

I was able to program a 4-digit address into the RS-11 using Ops Mode.  I didn't try just using Z's own controls.  I also was able to read the decoder using the programming track without any trouble at all.

BTW, this was my second attempt at installing a TSU-1000.  The first one was DOA right out of the package.  It'll be going back tomorrow.  Fortunately, I had another decoder and was able to complete the installation.

Paul A. Cutler III
************
Weather Or No Go New Haven
************

[wjstix]

BTW with a Zephyr you can program an "extended" three or four digit ID number without going thru CV 17 and 18. Just go into Page programming mode, and call up "Loco" and then set it to  four digit ID and have it read the four digit ID number (probably be 0003). Then put in what you want and hit the program button.

[CSX Robert]

BTW with a Zephyr you can program an "extended" three or four digit ID number without going thru CV 17 and 18. Just go into Page programming mode, and call up "Loco" and then set it to four digit ID and have it read the four digit ID number (probably be 0003). Then put in what you want and hit the program button.

True, but in this case the original poster had a decoder that was not responding to page mode programming, and when programming in OPS mode, the Zephyr throttle does not automatically do the extended address(although a DT300 or DT400 throttle using the Zephyr as the command station will).