Digitrax TL1 frustrations - solved

[SP-Wolf]

Hey Guys,

Can use some assistance. I have installed a TL1 into an N scale Kato Daylight observation (Wanting to add the red oscillating signal light).

What I have done so far:

I cannot seem to program the darn thing. First, I tried my ESU decoder tester -- Through JMRI - no good. Errors reading and writing. Doesn't see a locomotive.

So- I then tried just JMRI - same thing.

Ok, so then I soldered a  1K ohm 1/4 watt resister between the blue and white wires -- same thing.

So, then I put an analog locomotive on the track -- Frustrations going through the roof.

Up to now the decoder has yet to be installed in the car. Wanting to program it first.

OK, -- so, all that didn't work. I have since installed it in the car. With the same results -- JMRI doesn't see a locomotive--- WTH???

I then tried to program it using my DT400 --- same results -- doesn't see it.

However -- the LED flashes when I input a command -- weather to read a CV or when I try to program it. Why does the LED flash then It tells me it doesn't see a decoder - ?

I did try all of the different programming modes.

I had similar results with a TCS function only decoder -- I have terrible luck with function only!!

At wits end,
Wolf
 

[RBrodzinsky]

With it in the observation car, have you tried putting it on the programming track, and putting a resistor across the track?

[mmyers]

Have you tried it to see if any of your programming worked?
IIRC, I set one of these up with a decoder'd engine on the track. Both decoders got programmed then I reprogrammed the engine by itself.

Martin Myers

[SP-Wolf]

Have you tried it to see if any of your programming worked?
IIRC, I set one of these up with a decoder'd engine on the track. Both decoders got programmed then I reprogrammed the engine by itself.

Martin Myers

Hi Martin,

Yeah -- I re-entered the values I wanted into the decoder. I entered 038 into CV29 for 4 number address - then the 4 number address and 02 into CV49 for the mars light.  After each entry, it (the DT400) displayed "NoPg". And - the LED I installed in the car flashed after each command was entered into the DT400. I then put the car on my layout - called up the road number -- hit "0" ---- and.....nothing.

So, any other ideas - ?

Thanks,
Wolf

[peteski]

It is a function-only decoder driving low-power LEDs (no motor).  Most likely those LEDs are not enough of a load for the command station to get the write-acknowledgment from the decoder (when programming on programming track).  This is a tough one. I don't think a 1k ohm resistor will help.

One thing you might place a DCC-equipped loco (preferably which has a Digitrax non-sound decoder) on the programming tracks along with the car with TL1.  Then program both at the same time. Hopefully the TL1 will take the programming.  But this process will reprogram all the CV you change in the loco too.

[SP-Wolf]

It is a function-only decoder driving low-power LEDs (no motor).  Most likely those LEDs are not enough of a load for the command station to get the write-acknowledgment from the decoder (when programming on programming track).  This is a tough one. I don't think a 1k ohm resistor will help.

One thing you might place a DCC-equipped loco (preferably which has a Digitrax non-sound decoder) on the programming tracks along with the car with TL1.  Then program both at the same time. Hopefully the TL1 will take the programming.  But this process will reprogram all the CV you change in the loco too.

Thanks, at this point, I am willing to give anything a try.

I'll report back as to what happens.

Wolf

[SP-Wolf]

So, with Marc Fournier's (Yahoo Digitrax group) help, I was successful in programing the Tl-1. Digitrax ought to set these decoders up like their mobile decoders.

What he had me do: (Hopefully this will help someone else)

CV29 = 38   
 CV17 = 194
 CV18 = 232  (this gave me 744 as the 4 digit address).
 CV49 = 2  (Simulated Mars light)
 CV61 =  1  (This puts Function 0 in control of output.  If CV61 = 0,  you end up with function 1 in control)
 CV64 = 0  ( This defines WHO controls the single output on F0/F1, F2, F3 or F4.   CV61 differentiates between  F0  or F1.

 Be aware that F0  is NOT directional meaning the head light will not go out on a change of direction.
 Since F2 is momentary on most throttles, setting the TL1 output to F2 will be momentary.

Thanks,
Wolf

[C855B]

I take it from your CV list without method comments that you programmed it "blind", that is, without expecting feedback from your programmer.

If that's the case, I apologize for not jumping in earlier because I don't know the TL1 specifically. However, I have run across a couple of other "simple function" decoders that send no confirmation back to the programmer. One common attribute is they flash an LED on receiving a command, your only "confirmation". Once getting familiar with a specific board (Team Digital SHD2, for instance), sometimes I program before installation, on the bench. I send commands on faith, without being connected to anything other than the programming track. This sorta kinda works, but in the case of the SHD2 I eventually built a pluggable output board for testing and especially to verify that the address "took".

[peteski]

So, with Marc Fournier's (Yahoo Digitrax group) help, I was successful in programing the Tl-1. Digitrax ought to set these decoders up like their mobile decoders.

Like Mile said, it sounds like you just programmed the decoder blindly.  My suggestion was utilizing another decoder to supply that acknowledgement pulse (to give you feedback that a CV was programmed).

Another alternative would be to program the decoder on the bench (not in the model) and use a standard 12V Grain of Wheat bulb (like the ones used in older models for headlights) hooked up to the function output.  But the way this decoder is constructed (with flat contacts on both sides of the board) would be cumbersome to temporarily set this up (without the need to solder some wires to the contacts).

What manufacturers should do in function-only decoders is to add a circuit which would generate a high enough current drain to provide the acknowledgement pulse.  But that would require "wasting" a function output to only be used for programming.

I have successfully added a capacitor/resistor circuit to the  outputs of a function-only decoder which will produce a current spike large enough to acknowledge the command station's read/write operation yet not affect the regular functionality.  But that requires adding 3 components outside of the decoder.

[SP-Wolf]

Thanks, guys

Programming blind is exactly right.

I did try all of the other techniques with this decoder. I have an ESU decoder tester - no bueno. I tried the resister thing - no bueno. I tied a regular (small) light bulb - no bueno. I tried putting an analog locomotive on the programming track - no bueno. All of this was done on my JMRI programming track - and, outside of the car. The only thing I didn't try was to put a DCC equipped locomotive on the programming track with the TL-1. 

Then I tried all of these things with the TL-1 installed in the car. -- still, no bueno.

So, I next took it up to my train room and tried it with my DT400/DCS200. I also tried programming it like a mobile decoder. Using the 4 digit address aspect of the programmer. (I found out - that doesn't) - it worked using the technique Marc showed me. I don't know why Digitrax set the function only decoders up like this -- ?

OH well -- live and learn.

Thanks again!!
Wolf