Peter recently assisted me in correctly setting up my L88 and S88. Last item I needed to correct was to connect my 2299 Circuit track to the L88 (via the k track rail) T connection. Once done, the shuttle works.

One problem that remains is that I am unable to have the whistle sound . I have attached screen shots programmed instructions for one of the contact points. Appreciate suggestions.

Instructions (my intent) are to stop the locomotive when it enters the station, change direction, sound the whistle and then run. Except for the whistle sound, all this happens.





Thanks
Harvey

Elsleth1

Thanks, it worked.

Harvey

I fully concur with the observation.
While 3 seconds appears a bit conservative, I think 1 second is a min and when explaining memory programming I advise 2 for emphasis.

My thinking is that the backend engine works on a poll/timer basis to check (among all its other activity), every now and then, to see if there is work to do.
If , between theses polls a function has been set on and then off - the backend discerns that there is no nett change and therefore no need to do any work.

FWIW,
for a number of current decoder devices there is an inherent extra command needed under the covers.

Imagine that you have a non digital layout with a control box with 2 buttons (red and green) for switching a turnout (curve and straight)
What actually happens is
- when you press (and hold) the red button you "energise" the red side of the turnout motor
- when you release the red button you "deenergise" the red side of the turnout motor

When using digital a separate command is set for each - i.e. a command to the decoder to energise , followed by a command to de-energise
The time delay is "usually" controlled by the controller/keyboard so that, no mater how quickly you tap the button the master looks after you and the device and ensures there is a suitable pause between commands.

However , when using a PC interface "all bets are off" , and one reason for turnouts (using a k83) overheating was that a programmer either forgot to send a de-energise command or they "sent it too quickly" and it was missed.
(The observation in the CS2 Memory function correlates with this programming without a master looking after you)

The current m83 has, designed into it, a safeguard such that it will automatically de-energise a port if the controller does not send one.

The ramifications ...
one example is the use of an uncoupler, where it used to be common practice to hold the button down while driving a train across it.
You could do that with a k83, but not with a m83 in its default mode (i.e. 4 pair mode rather than 8 port mode)