Locked in forward

One other idea, Gregg, is to check the continuity of the wires between the battery and the board.  Without the battery in the circuit the engine will not retain the setting change on the direction lock.

FYI, setting or resetting the direction lock in DCS has no effect on the conventional direction lock.  The conventional mode direction lock is stored in the locomotive.  The command mode direction lock is stored in the DCS remote.  With the direction lock set to forward in DCS the remote simply stops issuing reverse direction commands and only sends forwards.

If the battery was disconnected, the sounds should immediately cease if you drop power to the track, that would be an easy test.

So far I've done all that,  Sounds continue on for at 6 or 7 seconds after track power is removed. I can change the engine's ID and have it stick after a layout power down. After  command factory  reset  the engine adds  at the lowest available ID (slot) i don't think it a battery issue.

 Dave are you sure a command factory reset doesn't over ride the conventional lock. I'm pretty sure it does, well  with at least 3 volt proto-2s.  Can you try it  again.

Which board controls the motor Driver?? I'm not sure driver is the proper term 

Since the engine goes in reverse in command mode, I don't think it's the motor driver electronics (which is on the bigger board).

After you performed the fast whistle-bell conventional reset, you presumably got the double-toot acknowledge(?).  Did you then try conventional control reversal BEFORE you shut down power?  That is, deactivating direction-lock if successful should take effect immediately.  If you instead let the engine shut down by removing power for more than 6-7 sec or whatever, you then depend on the engine's non-volatile memory logic to properly save/recall the direction-lock setting.  There is a slim chance the problem is in this memory function though the solution is not something you'll like...

Gregg,  That is an interesting issue.  I have not come across it before.  Sounds like Stan has some inside information so maybe he wants to expand on his post.

If it was my board here is what I would try.  I would swap the top power supply board with a known good one.  It has one logic chip that does send information to the bottom board.  It may register a power change and send that to the processor.  I am guessing at this, but it is a simple test.

If that doesn't fix it, then I would be at a loss.  I know where the FET that controls the relay coil, I know which chip controls that fet, but I don't know where the input comes from.  So it very well could have 2 inputs.  One for DCS and one for conventional.  Now why reloading the program doesn't fix it means there must be a hardware fault of some sort.

I could take the board only and test and repair, but in this case I would not want to replace that chip since it runs fine in command.  So unless Stan provides more details this may be something you need to live with.   G

Here's a little more info.... I did swap one of my good boards for a friend that only runs conventional, Steamer) . It fixed his engine except for a rear headlight issue??  although the rear  light works in command.  Now I want to fix my engine just because slipping into conventional results into a run a way I did this a couple of years ago so I'm going from memory,(not great). I don't mind purchasing a new bottom board if possible or maybe even  the top board. I have no way of testing a board and shipping the circuit board is pretty easy compared to the whole engine, Interested G?.

I tried everything with the conventional resets, immediately before  and after power down. I can't remember  for sure but think I got the 2 whistle blasts.   Yes I can buy a new upgrade kit but we have some talented repair guys on here that don't mind helping which is great.

If by "repair" you mean sending the board set to G and he's willing to do so, then that's what I'd do next.  That is, there are tests that might narrow the problem to specific circuits but to what end since you'd be sending the boards to someone who has a board tester which should more quickly isolate the problem.

I don't think this is something you can live with since, as you point out, if the engine slips into conventional with 18V command voltage on the track it will take off like a rocket if stuck in forward-lock conventional mode.

You may want to re-confirm the 2 whistle blast confirmation which should occur even if you only do a direction-lock toggle (w-b-b-b) rather than a reset (w-b-b-b-b-b).  If we're talking an F-3, and since it appears you have the shell off, note that you can pull the motor connector (yellow/white 5-pin cable) from the board.  This can make it easier to see/hear what is going on without the engine moving.  That is, if you successfully get the engine out of lock, you should be able to hear the faint relay click and the front/rear headlights toggle as you press the DIRECTION button

Gregg,  E-mail is in my profile and you can contact me.  I can test and evaluate.  So not only locked in forward but run away too, but only in conventional?  Interesting.  G