MTH Sound File Issue

Check the settings on the remote. (Make sure everything is turned on.)

Check the tether.

This was not caused by the sound file.

If you can grab it with the remote and run it, the board is probably OK.  It is seeing the DCS signal.  

I have to go with Dave, hard to believe this is the sound file if the load goes OK.

Everything turned on on the remote, tether plugged in fully. Unplugged it, tried moving it around while it was plugged in to check if contact was bad or tether was having issues but both check out. I can grab it with DCS and control forward/backward, couplers, but cannot do anything else. Leading me to think it is the tether but I doubt it, it was running fine up to that point 

John and I are on the same page here.  If the the file loaded OK, there is a hardware issue (tether/socket)

Does your tender have a roller on it?

I would not guarantee that.  That is a flash board for small smoke I believe.  MTH has seen issues with boards when folks use the editor to modify a sound file.  I had a repair just like that.  The edit sf (and this was on a Trolley) corrupted the Processor board.  So even with a reload  of original file it would not work.  So.....???  G

@David Minarik posted:

John and I are on the same page here.  If the the file loaded OK, there is a hardware issue (tether/socket)

Does your tender have a roller on it?

 

Dave ,  does the roller on the tender really help ? 🤔       I haven't tried to install one yet.

I've had a hard time trying to change out different size roller assemblies  on the engines to facilitate smooth operation of a couple of steamers moving through back to back turnouts in the staging yards.    Only partial success .

@GGG posted:

I would not guarantee that.  That is a flash board for small smoke I believe.  MTH has seen issues with boards when folks use the editor to modify a sound file.  I had a repair just like that.  The edit sf (and this was on a Trolley) corrupted the Processor board.  So even with a reload  of original file it would not work.  So.....???  G

It is possible to corrupt the board, even with a file right from MTH!  I just ran across that, another MTH tech came down with an issue and we were trying to load a file into a steam board.  It trashed the board firmware.  Tried another board, and it nailed that one as well.  Those had to go back to MTH for a factory FLASH procedure to revive them.

Thanks for all the insight, guys. On this one I may have to go with GGG as I just can't get this engine recovered. Just a stroke of bad luck I guess. I appreciate the help, and if I ever find a resolution (Aside from sending it to MTH) I will for sure follow up.

@GGG posted:

I would not guarantee that.  That is a flash board for small smoke I believe.  MTH has seen issues with boards when folks use the editor to modify a sound file.  I had a repair just like that.  The edit sf (and this was on a Trolley) corrupted the Processor board.  So even with a reload  of original file it would not work.  So.....???  G

GGG

What is a flash board?  Do you mean like the PS2 quillable boards?  I was unaware of anything like that for different smoke units.

I do not see how modifying the original sound file would change any of the performance features if only the addresses with sound bits are replaced.  If the function addresses remain untouched, there should be no problem.  I still think this is unrelated.

If Joe sends this off to you, please post your findings. 

Joe,

I asked about the roller because you would be able to somewhat test the board functions (sound) even if they were not attached to the locomotive.  Just trying to eliminate the possibility of something wrong in the boiler or the tether.

Good luck and please post your findings!  This is something I have not seen before.

Yes, in the later PS-2 3V they had flash boards that MTH could reprogram. Had a special chip that could be read/write vice write once.  If you had a GG-1 operating panto, you needed the GG-1 flash board.  If QW you need that, if small smoke you needed that.   In the 5V world it was special boards for GG-1, trolley, and articulated chuff.  3V had more memory so several flash codes could be combined.  PS-3 still needs those codes.  That is why I pushed MTH(Jason) to get those PS-32 flash codes out.  GG-1, small smoke, trolley, COORS etc...

I was doing so many PS-2 5V and 3V repairs, but needed those special features that MTH had to get them on the fast track.

As far as the newer boards, MTH has seen processor boards get corrupted.  If the FPGA (john correct me if I get it wrong) has to store memory items in the correct spot.  If it doesn't the board can no longer function correctly.  It does not know how to start up right.  I have had a few PS-3 do that with just a normal load early on.  Had to send board back to MTH so they could reflash the board.   Remember the document about Morse code type after a reflash?  Certain Morse code meant a corrupted memory.  G

I just lived through a PS/3 steam issue with a sound file that got corrupted with a bad sound file.  Another tech came down and we were trying to sort out the issue, and even getting the sound file again from MTH still corrupted the boards.  Finally had to send all of them back to MTH for them to beat them into submission. 

It's a bad feeling when you load a sound file and the board craps out, then you load another one and it does it again!

I attached the code audio for good and bad flash.  Luckily, I normally hear the good one.

I received a whole 6 pack of Proto3/2 boards last year - every one of them would not flash.  Had to send back to MTH to do the direct programming fix.  We need a way to do this in the field in the future, especially how hard it is to get boards currently.  It's a killer when you have a board and think you can do the repair to have this happen.

Jim

Any idea which specific files have been corrupting boards when edited? I've edited all the PS2 3V sound files in my Premier steam as well as the Sound files in my PS3 converted (PS32 board) steam with the ADPCM editor with no issue. I've just changed whistles, bells, and the announcements in the PFA sequence. They've all loaded fine. 

We had one corrupt the board right from the MTH website, no editing in sight!  I believe they were "supposed" to change it for the "good" file, but who knows.

All this information is truly quite interesting.....Who knew how in depth you could go with all this. Here I was thinking it's a tether issue but I'm glad (Or not so glad) to see others have had similar issues

@GGG posted:

Yes, in the later PS-2 3V they had flash boards that MTH could reprogram. Had a special chip that could be read/write vice write once.  If you had a GG-1 operating panto, you needed the GG-1 flash board.  If QW you need that, if small smoke you needed that.   In the 5V world it was special boards for GG-1, trolley, and articulated chuff.  3V had more memory so several flash codes could be combined.  PS-3 still needs those codes.  That is why I pushed MTH(Jason) to get those PS-32 flash codes out.  GG-1, small smoke, trolley, COORS etc...

I was doing so many PS-2 5V and 3V repairs, but needed those special features that MTH had to get them on the fast track.

As far as the newer boards, MTH has seen processor boards get corrupted.  If the FPGA (john correct me if I get it wrong) has to store memory items in the correct spot.  If it doesn't the board can no longer function correctly.  It does not know how to start up right.  I have had a few PS-3 do that with just a normal load early on.  Had to send board back to MTH so they could reflash the board.   Remember the document about Morse code type after a reflash?  Certain Morse code meant a corrupted memory.  G

G,

I've known about the flashable boards for those things but never knew there were ones for the smoke units.

In my experience with creating custom sound files, I've found the the memory locations for the Horn, whistle,bell sounds are located at the same address in different files.  You would almost have to replace PS2 control features with the same features from a corrupted file.  This still doesn't make sense to me how changing the sounds could corrupt the board.  The memory addresses are different in the sound file.

I could see if the control commands are replaced and the board cannot see the DCS signal...maybe.

Just trying to wrap my head around this.......

@gunrunnerjohn posted:

I just lived through a PS/3 steam issue with a sound file that got corrupted with a bad sound file.  Another tech came down and we were trying to sort out the issue, and even getting the sound file again from MTH still corrupted the boards.  Finally had to send all of them back to MTH for them to beat them into submission. 

It's a bad feeling when you load a sound file and the board craps out, then you load another one and it does it again!

I attached the code audio for good and bad flash.  Luckily, I normally hear the good one.

John,

I have had quite a few issues with PS3 steam boards not taking a flash and getting bricked.  I even built a simple loader harness and didn't use the PS3 test fixture thinking that there might be an issue with it.

That is a sick feeling!  Especially when you only have a couple left.

@David Minarik posted:

That is a sick feeling!  Especially when you only have a couple left.

And no more to be had it appears...

@David Minarik posted:

I've known about the flashable boards for those things but never knew there were ones for the smoke units.

 The PS32 boards have steam and diesel loads for standard smoke units and also another load for the single resistor small unit.

I’ve flashed a couple of boards and never run into this.  One was an older 5v board in a 4-4-2 E6s in which I modified the sound file quite a bit as the sounds were pretty weak sauce.  I probably flashed it 10 times adjusting the sounds to make it the right volume and tweaking the reverb. The other was a PS2 3v board in 2-8-0 from a RTR.  In both cases it worked like a charm.  To hear about these other issues is concerning.

I probably won’t be flashing any of my other locos in the near future, but I have at least one PS2 5v loco I’d like to upgrade at some point.  There are also numerous PS-1 locos I’ve considered buying with the hope I could eventually upgrade and custom program the sound files.  Obviously who knows what will happen with MTH/DCS and board availability, but I figured I could Frankenstein some later with boards from locos that I either like less or are from RTR locos.

So this leads me to a couple questions.  First, what is meant by a “PS3-2” board? Is this a PS2 3v board, a PS3 board, or a PS3 upgrade board? If one of the latter, what does the -2 indicate?

Second, based on the information about the small smoke unit boards, if I bought a run of the mill PS3 diesel upgrade board on the MTH website (assuming they were available) would it work in a small diesel like the ALCo S-2 that has the little smoke unit?

The PS32 board is the two board stack that is in the steam upgrade kits and also is used to replaced failed PS/2 boards.  It consists of a special PS/3 board (different layout), and an interface board that makes it look like a PS/2 board in regard to interfaces.  Specifically, instead of LED's, you use 6V bulbs, and instead of the PS/3 tach reader, you use the older PS/2 tach reader.  Here's a thread by Marty F that has lots of pictures of the PS32 board.

PS3 upgrade information

@gunrunnerjohn posted:

The PS32 board is the two board stack that is in the steam upgrade kits and also is used to replaced failed PS/2 boards.  It consists of a special PS/3 board (different layout), and an interface board that makes it look like a PS/2 board in regard to interfaces.  Specifically, instead of LED's, you use 6V bulbs, and instead of the PS/3 tach reader, you use the older PS/2 tach reader.  Here's a thread by Marty F that has lots of pictures of the PS32 board.

PS3 upgrade information

Thanks!

I'm sorry to bump a relatively older post, but I'm in the exact same position as the OP.  Have done numerous flawless sound file changes with other engines but suddenly seem to have corrupted the board on a Premier PS3 GN S-2 4-8-4 from 2013/14.

Has the OP found any resolution for this issue?  Trying to get this fixed asap knowing MTH is going away soon.

If you've bricked the board, only MTH has the capability to restore it.

Exactly what are the symptoms?  There are issues with loading chain files that don't match the board.  MTH went to the EMI boards after your locomotive was released, you may be loading incompatible code.

It took almost three hours to load the sound file.  Kept running down to the "trying again slowest" level and often throwing the timeout error.  Pushed through and it finally finished.  Engine would then add to the remote, run, and play sounds, but the name and engine-specific data wasn't showing up in the remote, just showing blank.

Tried reloading the original sound file and after several attempts, It won't go past step 0/32.  Tried another laptop, but no luck.  I was thinking it was a signal issue, but even clip leading directly to the engine's wheels it won't do any better.  So unless there is a trick I haven't tried, I'm think it's board related.

Woody,

OP here. No found resolution to the issue. I'm a bit weary of sending it to MTH given the current state of everything (The pandemic, closing up shop or at least to our knowledge). Your situation is probably the same as mine, and would most likely require a re-flash done by MTH as GRJ posted above. Nice to know I'm not the only one with this issue, but sad to see another locomotive fall susceptible to this issue. Let me know if you're able to figure it out.

Joe

Take a short piece of track and connect it directly to the TIU channel.  A "trick" here is to use about 30 feet of wire to connect the track to the TIU channel.  Power the channel with DC at around 16-18 volts, obviously you have to use one of the fixed channels.  This gives you the best chance of getting PS/3 files to load.

The fact that you're loading something suggests the board may not be bad.  If only the name doesn't show up, is this the original board?  New boards come with a blank name, and to load a permanent name, you need the dealer loader.

This may be a foolish question but why 30 feet  of wire? 

My only concern that leads me to think it may possibly be signal related when performing the load is that when checking the "track signal" I'm seeing 8-9 out of 10 for signal strength, occasionally going as low as 7 or even 6 - this even when directly wiring the wheels to the TIU output.  Thinking about maybe pulling the board out and directly clipping to the power inputs (want to absolutely make sure I'm connecting to the correct ones of course!).  

I reached out to MTH, but don't know what to do if they won't take out of warranty repairs as I'm hearing.  Hopefully they will have a replacement board in case mine is indeed bad.

@gunrunnerjohn posted:

Take a short piece of track and connect it directly to the TIU channel.  A "trick" here is to use about 30 feet of wire to connect the track to the TIU channel.  Power the channel with DC at around 16-18 volts, obviously you have to use one of the fixed channels.  This gives you the best chance of getting PS/3 files to load.

The fact that you're loading something suggests the board may not be bad.  If only the name doesn't show up, is this the original board?  New boards come with a blank name, and to load a permanent name, you need the dealer loader.

DC?  I have an HO transformer that should be able to do the trick, but that won't hurt anything?

I'll give it a shot.  And this is the original board.

@amtrack5899 posted:

This may be a foolish question but why 30 feet  of wire? 

Why anyone with a ... nevermind.

Just kidding, it isn't like it's something that springs to mind.  It seems that the DCS signal is more likely to get distorted if the output of the TIU to the locomotive being programmed is just a few feet.  So, and especially for PS/3, MTH at their lab had a 30 foot coil of twisted pair feeding the test bench for programming sound files.  They also used pure DC to power the board/locomotive that was under test.  Both of these helped with speedy loads, so I do the same thing on my bench, and it is more reliable and faster than using short wires and AC on the locomotive or board.

Interesting!  Well I just dusted off my old Bachman spectrum DC transformer from my HO days and we’ll give it a shot!  Btw this one only goes to 15.6 vdc, should that be ok?

Also, does polarity matter?

So I just tried it.  Wow that zipped through fast!  Now the engine is working perfectly, but for some reason it’s still not showing a name in the remote!  Not a big deal to me as long as it works, but any idea why?

Now, to try my hand at it.......

@Catonsville Central Railway posted:

So I just tried it.  Wow that zipped through fast!  Now the engine is working perfectly, but for some reason it’s still not showing a name in the remote!  Not a big deal to me as long as it works, but any idea why?

I suspect that for whatever reason, the name was never added.  You can put a name in there, it'll survive until a factory reset.  If you happen to be around a MTH tech, they can put a permanent name into the engine.

Got it.  Odd, because there was always a name in there before.  But I'll just add one for now.

Thanks so much for you help!

@gunrunnerjohn posted:

Why anyone with a ... nevermind.

Just kidding, it isn't like it's something that springs to mind.  It seems that the DCS signal is more likely to get distorted if the output of the TIU to the locomotive being programmed is just a few feet.  So, and especially for PS/3, MTH at their lab had a 30 foot coil of twisted pair feeding the test bench for programming sound files.  They also used pure DC to power the board/locomotive that was under test.  Both of these helped with speedy loads, so I do the same thing on my bench, and it is more reliable and faster than using short wires and AC on the locomotive or board.

Differential signaling and cable capacitance is my CS-major-who-likes-to-solder guess.  

Ok I spoke too soon.  Engine was running fine for a while, then at one point I got the “engine not on track” error, yet I could still control the engine.  Tried restarting and now it can’t find the engine.  However, the plot thickens.  Now even trying other MTH PS3 engines my tiu (Rev L) can’t find any of them!  I tried a reset on the remote and TIU but no luck.  Power gets to the track, the engines don’t start up, so they see they’re in a command environment, the remote communicates fine with the TIU, but the TIU can’t find any engines.  Have never had this problem before.

There was nothing I needed to switch before powering with DC, correct?

It sounds like you might have a TIU issue.  A popular item to verify your installation is this unit.  It allows you to test the DCS signal strength on each of the TIU channels.

Design of a $10-20 DCS-TIU Port Tester Tool?

I’ll look into getting that.  So it looks like TIU issues now oddly enough.  I’m getting “no engines to add” despite trying with three of my PS3 engines.  All engines run normally in conventional.  Trying with the WIU and my phone I still can’t add any engines, so (I think) that rules out the remote.

Is there any hard reset procedure I can use for my TIU before seeking service from somewhere?  Or perhaps would reloading the firmware with the mth loader help?

Whats so odd is that the remote sees the TIU, the engines see the TIU watchdog signal (they don’t turn on), but the TIU can’t see any engines.