PS3 startup issues - continued...

What version of TIU are you using??

4.2.  I have been reluctant to upgrade because everything had been working so well, including my new PS3 engines.  This is the first brain teaser I've had for a good while...

 

I did notice that there is a new "MSV" feature that allws for voltage definition on the TIU channels, and that there are parameters specifically for "Z4K" Z-4000 transformers.  Could this have something to do with it?  

 

Again, no issues with my other PS3 engines;  the E-8 is by far the newest, just out of the oven, but my ES44AC is only a couple of months...

 

 

First, if I had to guess, my thought would be that the culprit is a passenger car, caboose or other lighted car, or a TMCC/Legacy engine, that's degrading the DCS signal.

 

However, if you removed all of those type of suspects and the problem remained, that's not it.

 

Second, do you have a TMCC or Legacy command base connected? I so, unplug it from the wall (AC) power and see if anything changes.

 

Third, I'd suggest using different TIU channels or a different TIU altogether.

That's my guess as well - a proximity thing where once you get to a certain number of lighted cars combined with engines, there's a electrical noise level that this particular E-8 is sensitive to.  Oddly enough, I'm guessing most people would probably not encounter this - I think I just have the right (wrong?) combination of "stuff" cobbled together in just the right (wrong?) way.  

 

There is no TMCC involved.  I may try to re-channel just for fun, and I do have another TIU barring that...

 

 

Just remove all of the lighted cars from the tracks. If the problem goes away, add them back one at time until you find the culprit.

 

Once you find the offending car(s), there's a tried-and-true, permeant solution to the problem.

I have a similar issue.  For several months my Premier ES44AC, when sitting on a siding that can be turned off, will start immediately when power is applied to all track (power to siding switch was turned on before power to the mainline was applied).

 

Today, with the ES 44AC on the mainline, adjacent to the selectable siding, my new Premier EMD E6 and the ES44AC both started immediately when track power was applied.  In both cases I selected the engine in the remote and pressed the Start-up button.  Both engines continued to run (now under DCS control).  

 

Up to this point the EMD E6 had not exhibited this behavior until the ES44AC was close by and on the adjacent mainline track.

 

The track is a single loop with only this one siding, and all are controlled by Fixed One from the same TIU.  

 

Perhaps the ES44AC is a bad influence.

Rick,

 

You may have two many engines getting power at the same time. When that happens, engines can miss the watchdog signal.

I'm starting to think I have a wonky switch or something.   I can pull the E-8 to the opposite end of the loaded siding (engine and passenger cars) and I don't have the issue - as soon as I enter the part of the track adjacent to the siding, BAM!  Problem occurs.  

 

Of course that still doesn't explain why my other PS2/3 engines don't have the issue - just the E-8.  

 

Odd....

I just ahd a weird thought based on my last post....

 

I ran a new drop to the "outer" part of the siding, but not the inside/main line between the two switches.   When you power sidings, do you do the inside, outside or both???

 

This was the first year I added multiple power drops to my layout with excellent results, but perhaps I missed something????

 

Thoughts?

 

 

As I said previously....

Just remove all of the lighted cars from the tracks. If the problem goes away, add them back one at time until you find the culprit.

 

Once you find the offending car(s), there's a tried-and-true, permanent solution to the problem.

" permeant solution to the problem"  ?

 

My wife went to her hair dresser once, and came back w/ a  permeant ----I hardly recognized her. Looked nothing like when she got a permanent!

The Permanent Solution to DCS interference by lighted passenger cars or TMCC/Legacy locomotives is to install a 22uh RF Choke in the power leads of each offender.  The maximum current draw of the car/engine in question will dictate the exact specifications of the choke.

 

So here's the deal today...

 

It doesn't matter WHAT I have in the siding as long as the E-8 AA is "between the switches" in a parallel position.  One car, two, three, and engine, any combination.  It will start conventionally if ANYTHING is in the siding.  It also doesn't seem to matter what passenger cars are connected to the E-8.  I've tried just about all of them in various combinations.   As an added "bonus" it started conventionally on another part of the layout last night - that hadn't happened before.  

 

I noticed this morning that if I remove all cars/trains from the siding, I can "beat" the startup if I get to 17/18 volts quickly.  I don't really understand this, but it does work.  The problem is that I'm not that accurate with the handles and I tend to jump to max power (or close to it) before backing down.  I can't believe that's good for the lighting in the cars.  

 

My wife is home now, so I can't test until later, but I'm thinking about hooking a Z-1000 to the loop to see how that affects the process.   Those go to 18 the moment you turn them on from what I understand.  Will be interesting to see if this changes the outcome. 

 

If not, I'm left to assume that the E-8 is just extremely sensitive.  Again, don't have this issue with any of my other engines/combinations of engines.  This one just has me flustered....

 

Thoughts?

 

-Eric

OK, I can beat it.  With any combinations of any cars on the track, I can 'beat' the conventional startup every time with a quick throttle up to max and then back down to 17/18 volts.  Two questions:

 

1) Is there any potential harm to the passenger cars or engines by jacking up the throttle to 23V and then back down???  I can get it back down pretty quickly - it's only at 22/23V for literally a second; maybe less.   

 

2)  What does this say in terms of diagnosis?  When I slowly, methodically apply power, it would appear that the TIU is sending the watchdog before the E-8 is ready for it??? But that doesn't make sense to me because I was under the impression that the TIU would only send the watchdog when 18Vin hit the channel.  At 18Vin, the E-8 AA should be ready to respond.  Perhaps the E-8 is jumping the gun BEFORE seeing the watchdog?  This might make more sense because when it starts conventionally, the headlight comes on pretty quickly and I hear a relay click as soon as I start applying power.  In this case, it never gets a chance to hear/process the watchdog because as soon as it sees power under 18V (or somewhere between 0-18V) it jumps the gun an initiates conventional start.  Does this make any sense?

 

3) Would this be seen as "defective", and would it be hardware or software.  My gut tells me that it would be a code update to the engine...

 

4)  Should I call/contact MTH about this?  Anyone have a contact there that might be especially interested?  

 

Thanks in advance (as always...) !!!

 

-Eric

Jacking the voltage up over 18-19 volts will not do anything positive, but it has the potential to do something very negative!

 

The WD signal comes out as soon as the TIU senses power on the channel input.  It's present for a few seconds, so normal application of power shouldn't cause you to miss the WD.

Yet it does...  consistently on this engine.  That's the frustrating part.  Seeing that it has now happened on other parts of the layout, I'm thinking this one may have to go back to it's maker.   The last thing I can test from the timing perspective is to try the Z-1000 brick.  If it works properly, I'm not sure what else I can do unless there is a way to "lock" the Z-4000 at 18V on startup.  I'm not sure that's possible because I believe it to be designed to avoid such a thing purposely.

 

Going to get the Z-1000 now.... more to come....

 

 

FWIW, Bill Henning has the MTH A-B-A E8, and his has lots of problems on our club layout as well.  I can run other MTH stuff around fine.

Works perfectly EVERY time with the Z-1000!

 

Leads me to believe this PS3 implementation on this engine is a dud...

Called MTH and talked to a tech.   He is concerned that there is indeed a problem.  They have had some problems with pins not being fully inserted into sockets; somehow this creates intermittent issues with the engine being confused as to whether it is in DCC or DCS mode.  

 

I personally don't think that is the issue - it appears to be certain of what mode it is in.  It clearly speaks DCS and it's not "intermittent".  Ive got it to the point where I can reproduce it at will under any conditions and I know what fixes it EVERY time - problem is, it requires not using my Z-4000 and switching to Z-1000s (not going to happen) or hitting the TIU with 18V or higher immediately upon power.  Also not going to happen - it's the "or higher..." part that concerns me in this case.  

 

I still feel like it's a timing glitch related to input voltage, or improper interpretation of/reaction to DCS signal at voltages below 18V.  

 

I'm probably wrong.... but its back to the mother ship tomorrow regardless.   

 

Bummer... 

 

 

Well a layout is supposed to be powered up fast when using DCS. Lots of folks have all power supplies plugged into a power strip...throw the on switch and everything is full power now. Bringing the power up slowly will cause PS2/ 3 equipped engines to miss the watchdog signal.

 

Jacking the voltage up over 18-19 volts will not do anything positive, but it has the potential to do something very negative! 

 

What's that going to harm? My target track voltage is 24V DC with the capability of going up to 30V.

I suspect that at the very least, the passenger car lights will cook off prematurely.  But I'm literally talking about 1 second at startup, returning to 18V  steady state very quickly.

 

It is frustrating - If I could get the Z-4000 to power up at 18V automatically, I might be willing to overlook the 'glitch' that I'm seeing.  Of course then I wouldn't be seeing it...

 

I suspect this gets back to MTH and they struggle with it a bit.  They seem to think it's a lose wire or pin, and perhaps they are right, but I think it's going to require a code rev or a new board on the off-chance that mine is a dud.  

 

I just keep coming back to the "works perfectly with the Z-1000" thing;  it just seems so trivial yet remains vexing in it's solution.  It's the one thing the tech couldn't explain away.....

 If I could get the Z-4000 to power up at 18V automatically, I might be willing to overlook the 'glitch' that I'm seeing.

There are two ways to do this:

• The first is to use variable channels with TIU tracks, and set the initial voltage to 18 volts. One thumbwheel click gets you right to 18 volts.
• The second way is to use a Z4000 Remote Commander receiver with Z4K Tracks on fixed channels. It works the same way as the variable channels work.

This and a whole lot more is all in MTH’s “The DCS O Gauge Companion 2nd Edition", available for purchase as an eBook or a printed book at MTH's web store!

I did think about this, and did some research on it.   It would  work, but introduces another step in the "turn on, run trains" cycle that I like to keep as simple as possible.  Since none of my other engines appear to require it, I am of the stubborn mindset that this one shouldn't either 

 

In your opinion Barry, what is the potential issue with slamming the power up to max on startup and then quickly back down?  There seems to be differing opinions on this and I'd like to hear yours.  I won't hold you responsible 

 

Thanks for all the help so far, to Barry, GRJ, and all others!

 

 

what is the potential issue with slamming the power up to max on startup and then quickly back down?

Just about none.

 

However, I don't see why you can't simply raise the voltage to approx. 18 volts in about 1 second or so.

 

Regardless, how about just pressing Startup or Shutdown when you access the engine?

 

Personally, I think you're making a mountain out of a mole hill.

I've tried, but I don't have the accuracy with the handles;  I need to get to 18V almost instantaneously: approximately doesn't work.  If I don't hit 18 or better virtually immediately, it trips.   I've tried this about 50 times - the only way to catch it is to jack it up fast and then ease it back down.  I don't get a whole second to make this happen.  Your solution with the TIU tracks IS a good one, but I'd prefer consistent operation across all my engines and I don't think that is too much to ask.  

 

Perhaps I am making more of this than I should - it might be down to a fundamental misunderstanding of DCS.  When I let it start conventionally, it comes up full blast, full smoke, full everything - I don't typically run my engines this way.  I can shut it down after it comes up, sure, but when I restart it, it retains the "full blast" settings from when it came up conventionally.   Then I wind up having to reset the parameters I want.  I suppose I could dial down the volume and smoke pots under the engine manually, but there are times I do like to run full smoke and volume (when wife is out of the house...) and would prefer not to have to hoist the engine off the layout to reset them.  Am I missing something with regard to the conventional "full blast" startup?  My other DCS engines exhibit this as well - I am unaware of any way to change this particular behavior.

 

I respect your opinion Barry - I truly do.  Your contributions here are epic.  In my defense, however, I mostly run Railking engines;  on the rare occasion I spring for a Premier, I want it to be at least as smooth and responsive as my RK stuff across all manners of operation.   If I someone bought a new car that would only start up with a "trick" move or else the radio came on full blast, the windshield wipers came on, and the heat blew full bore, until they turned it off, back on, and reset everything, and this happened every time they missed the "trick" startup...  I'm guessing most people would take it back to the dealer and have them fix it.  That is currently my plan.  I acknowledge that a new toy train isn't of the same magnitude as a new car, but it's still a decent expense, especially for a "hobby".   

, but when I restart it, it retains the "full blast" settings from when it came up conventionally.   Then I wind up having to reset the parameters I want.

 

That sounds like a weak battery that's not holding settings from startup-to-startup. PS2 engines need a strong battery to hold any changes that you make to their settings under DCS.

I suppose I could dial down the volume and smoke pots under the engine manually

That would only work when in conventional mode.

My other DCS engines exhibit this as well - I am unaware of any way to change this particular behavior.

All DCS engines will come up in conventional mode with sounds and smoke as per the pots under the engine. If all of your engines don't hold their DCS settings form startup-to-startup, I'd suspect that their batteries also are not sufficiently charged.

As long as they don't come up conventionally, they always hold their settings.  Save for two PS2 engines, they are all PS3, so there's no battery to be concerned with anyway.  The tech I talked to today did mumble something about the "super cap" and that they have had some recent issues, but then he got wrapped around "loose wires" and unseated pins/sockets.  He might be right, but I think it's something else.  I used to troubleshoot a lot of high end components, and this "feels" like a firmware thing to me.  Maybe a bad board, but I don't know... I just have that vibe.   

 

The thing is.. I know I can work around it at this point, but I want to SOLVE it;  I'm just goofy that way.   I like to tinker and I like to understand.  I wish I had a more solid EE background sometimes.   One thing I thought about was that the E-8's (and the F units) are kind of unique in that they have two engines but are not technically a lash-up.  I have a lash-up that runs great, but it is a true lash-up.  Could this be a contributing factor with the timing?  Heck, now Im just grasping at straws

 

I need to chill out and see what MTH says.   Hopefully I get a report back on what the issue actually was - I really want to know.

I startup PS2 engines all the time on my test bench with a standard transformer.  I don't bring the power up at lightning speed, I just reach over and move the throttle up to max.  I've never had a properly working PS2 engine come up in conventional with that process.  The transformer is the input to the TIU channel powering the test track.

 

 

I've offered to you multiple solutions to your issues. The ball's in your court.

 

Good luck.

I agree with John and Barry. I also use a Z-4000 on my work bench and don't throw the handle right to 18v. There's no reason to try to hit that from the get go. Even with the new PS3 engines. It does sound like something is wrong with your engines and the techs are pretty darn good at MTH.

Ralph

So I just got a reply from MTH Tech Support.  I spoke with them a few days ago, but the tech seemed distracted and I don't think he really 'got' what I was describing, so it filled out the online ticket form yesterday morning and got this reply back this evening:

 

"Hi Eric,

 

 

I have showed your message to a technician and he said this is something we are aware of and the best known solution as of now is to start things as you normally would and when the E-8 starts up, select it, press start up to take control and then shutdown if you don't want to run it.

 

 

 

I hope this helps,

Diana@MTH"

 

So this is a problem that they are aware of, but apparently don't have a fix for just yet.  I'm glad I didn't send the engine in just for it to sit or come back unfixed.   I like Barry's work-arounds better than theirs because if I let it start up conventionally, it loses its settings, comes up full blast,  and I have to reset everything on the subsequent restart - not fun.

 

Just wanted to add this in case anyone else stumbles across it.  Will post the fix when it becomes available.

 

-Eric