Rev L TIU issues

I sent my TIU in for no DCS signal on any channel.   just got my TIU back, and the repair is very curious.  They replaced a number of logic chips that should be buried deep in the logic, and not connected anywhere close to inputs or outputs I would imagine.  The repair reads Replaced U340, U341, U440 and added Ziner Diodes.  I'm presuming those are actually Zener diodes, but the chips are octal tri-state logic buffers, they operate at 5V.  I'm wondering how they got toasted when everything else was working. 

I'm curious if other folks with the same failure got the same report, maybe that's a clue to what is killing these.

I wonder if this has anything to do with my topic on losing the ability to run PS2 engines thru VAR1 and 2, yet can still run conventional thru VAR1 and 2 AND FIXED 1 and 2 are still working fine?

What component(s) inside the TIU make it possible to run PS2 engines thru the VAR channels (aka turn VAR channels into FIXED channels)?  Whatever they are has to be where the problem starts.

John, Any chance you could open the case and check and see where the zeners were added? I would think they would be obvious. That might enable some of us to head off the problem. Also we could compare early and late serial numbers to see if this has become a standard engineering mod.

Pete

The diodes weren't on the top of the board, and I didn't feel like loosening all the binding posts to get to the bottom of the board.  I was curious as to where the diodes were installed as well.

The chips that were replaced were all 74ACT244, I'm wondering why these would go bad.  I'm certainly curious why all of them would be bad!

Understood John. Maybe some of the DCS service guys will be able to get this info.

Pete

I'd love to know if they're adding the diodes to protect something from possible voltage spikes, that might be a good mod to do to them before they die.

I have 2 new DCS systems one is Rev L the other is Rev. I3 I want to sell one of the systems. After reading about the problems with Rev. L I am not sure whih to keep and which to sell.

I will be interested in hearing from people that are familiar with and using DCS.

 Thank you,

Some people have wondered if too much heat build up is a potential cause for the Rev L failures and have suggested using fans to cool the units. Does anyone think there is a possible difference in heat build up to a certain vulnerable area of the TIU when it is mounted vertically rather than horizontally?

Ed

The thing that I noticed when I got my repaired unit back was they mentioned adding Zener diodes.  I'd love to know what they're protecting with those, that might be a clue.  I also wonder if they're a worthwhile addition for reliability, could we add them to existing units that have yet to go to TIU heaven?

Got an RA and all four are going back today   They did say there is a known problem with Rev L's that were shipped before Dec 2012

Great, does that mean I should ship my other one back as well?  It was well before December 2012.

I realize I am kind of new to this forum, but doesn't MTH monitor this stuff?  Can't MTH make a definitive statement to end all of this mystery? I would expect there is at least one person who is a member of this forum and has strong contact with MTH.

MTH does test everything in design and engines are tested prior to shipment.

I design Industrial Controls.

Sometimes things don't show up until after many hours of use, this sort of thing can slip through the testing.

It's the reason we have the first unit of our new controller running along in the lab indefinitely. Despite detailed design calculations and observations /corrections made during prototyping and testing customers always seem to find some combination of things we missed.

So far we have a year and a half with no issues noted here. Hope it stays that way but you never know for sure. But a customer found an issue !  And we modified the program to fix it, all units available have been upgraded but they have to be sent in for the fix. So there are a few still floating around out there because the owners have not sent them in.

Russell, are you talking about the Rev. L TIU or something totally different?  There has been no announced update process that anyone here seems aware of.

Totally understand testing, retesting, finding bugs, etc.  That is normal in the system development life cycle.  Can you please tell exactly what the identified problem was and its resolution.  Thanks!

Did they say what the problem was and what it affected?  Is there a general recall or is it left to the individual to find this out and take action?  Who do we contact at MTH?

I'm going to call them tomorrow and ask that very question.  I still have two Rev L TIU's that were shipped before that date that haven't had any repairs.  If there's a fix that will keep them from dying of this issue, I'm all for it!

John, Keep us posted on this. Mine is also an early version but is still working. I am curious if it has to break before they will accept it to be modified.

Pete

I don't know, but when they fixed the one, they replaced a bunch of chips and added the Zener diodes.  I'm just waiting for the shoe to drop for the others!

I purchased mine in September 2011.  Worked great until November 2012 when my 1st VAR1 problem occurred, than a couple of months ago VAR2 died.  They still work in variable mode, but can no longer go thru them in the FIXED mode for running command.  Not the same exact issue as original poster, but maybe the fix is the same.

I have our club TIU here to figure out what it's problem is, and it has that exact problem, VAR1 no longer sends out the DCS signal.  However, it's not a Rev. L, so it may be unrelated.

Has there been any progress on the original issue???

We got our TIU's back  They changed the Power FET's with a different rating   They also added a diode ( ot sure wherr )  So far so good with them

WOW Ben, it took four months to get them back?  I hope they work out after all that waiting!

I have three Rev Ls new in the box, two I bought last year and one early this year.  I want to use all channels in the fixed mode.  After reading all this about its poor design, I'm not sure what to do with them now.  I'm at least a month away before my initial outer loop is complete to test these.  How do I test them (I can set up a small loop of track)?  Can I send them back to MTH to have them modified even though they haven't been used?  Sell them and get older versions off the bay?  I have an old Rev G that I can use if needed; I at least know it works.

I guess I'm just ranting because I'm unhappy with the situation.  By time I start to really use two of them, they will be over a year old and I will be stuck with them.  Hopefully by then someone will come up with all the needed fixes that can be done by all thumbs me at home.

Ron

Ron,

 After reading all this about its poor design

There's no "poor design". Much more likely is a batch of weak components. Also, there is no wide-spead failure of Rev. L TIUs, new or old. I had one failure out of 3 Rev. L's that had been on the layout for more than 2 years. The repaired TIU is behaving just fine after several months of use.

Hook up your TIUs, run some trains and check the DCS signal strength. If you have issues, contact MTH.

Just do this before you use up your one-year warranty.

Really!!  I have a list of emails from guys that beg to differ. 

Ben,

Really!!  I have a list of emails from guys that beg to differ. 

Yes, really!!  

When I spoke with MTH a few months ago, when this issue began to "resonate", there were about 75 or so actual TIUs in for repair, for a couple of different reasons. These weren't all Rev. L's, either. As a result, it's possible that MTH has made a small change going forward to better protect the DCS signal generators.

To date, there has been no recall or other cause to lead one to assume a problem of epic proportions. Rather, some quantity of TIUs have failed and some individuals have determined that the sky is falling. I expect that TIUs will regularly fail due to misuse and other customer-inflicted reasons, as well as due to defects in manufacturing. However, there's no way to know with any certainty what is the actual distribution of the causes of failure.

I don't at all dispute that the Hi-Railers have had more than their share of failures, and I most certainly take you at your word that every failed TIU that you've examined and identified as defective is truly so. Further, if you are certain that the failures that you've experienced were not the result of accident or abuse, I certainly take you at your word.

However, that's specific to your knowledge and experience. In my experience, I had one Rev. L fail totally, out of my three, all of which went in-service more than two years or so ago. I really have no idea as regards what event may have caused the failure.

Regardless, it would be a mistake to assume a general case from yours or my specific cases.

 Geez Barry, you got me laughing. That was one of the best, well written responses I have seen on this forum. You're getting better with age!!

 I can't help but agree that for some reason, too many people are ready to trash about anything for any reason. Maybe it's the heat setting in??

 Reading this about Ben's troubles is alarming even me to a degree. I keep going down to run trains and expect something to blow up! So I'm guilty as well. My single L version keeps working. It smells bad. The signal is strong. I imagine that there was a batch with certain components that got released and wrecks consumer confidence.

Well, I've had one of three TIU's fail, and the returned unit had bad chips replaced and added diodes for some reason, at least that's what the repair order says.

Barry has had one of three fail as well.

I don't know about others, but a 33% failure rate is not a record to be envied.

I've often wondered about the burnt smell coming from all of my TIU's, it's stronger than anything else I've encountered with other electronic products.

John and Ben, didn't your failures involve using at least one of the variable channels in fixed mode? I suspect someone with a small layout, like me, who only uses a single fixed channel might never see this problem. 

Pete

Mine was all the channels stopped sending out the DCS signal, the remote connected with the TIU and I could configure it, even update the software, but no DCS operation.  I have two variable channels connected to the tracks, as that allows me to work on and test conventional stuff.  I normally just run them in fixed mode.

 Come on John, your math is used to illustrate your point well taken. However how about you factor in every TIU produced then?? Would that make it like .0000001%? How's my pie chart coming???

 I believe we heard that it was a production problem within a certain time frame that the components are failing in. It appears you got one of those. Should we poll every one who ever bought a tiu??

Very true Joe, I have no idea what the true failure rate is.  Also, since the TIU is pretty much the only game in town if I want to run my MTH PS/2 locomotives in command mode, it's somewhat of a moot point.

I do know that after taking the boards out of all of the TIU's, there are at least two quite different layouts of the Rev. L TIU's, so there had to be some reason for the changes.

If you have a strong odor then suspect some component is being overdriven or the engineering did not account for heatsink area to dissipate the heat rise above ambient.  Or MTH used a components that was "good as or exceeds" the hard to get part ("don't worry this part will work just fine" (as a retired avionics EE engineer, ask me how many times I heard that line... )

Should we poll every one who ever bought a tiu??

No, but MTH should and put out a recall to fix them.  Why should anyone have to send in their TIU and pay to get it fixed when MTH has discovered that faulty components were installed?  We'd ask for the same if it was a $300 engine or a $300 piece of electronics for the home with a widely known problem.

So, how does a person determine if their TIU is in this group?  Is there a sticker on/in it that shows the date built?

All MTH would have to do is place a notice on their website telling of the problem (if one has really been identified), how to identify if your TIU is affected, and who to contact for a RA number or repair shop to take it to for the fix.  Easy Peasy, Lemon Squeezy!

John,

Barry has had one of three fail as well.

All 3 of mine were from the original batch of Rev. L's and all 3 were in-service for 2+ years. That's an indication of either a catastrophic event or a weak component.

Since all 4 signal generators failed after 2+ years and all at the same time, I'm thinking catastrophic event.

They had a strong odor right out of the box brand new.  It's a common complaint, and I think it's not an overheated component, but I'm not sure what it is.

That's coming from another retired avionics EE.

 Just a guess here guys and I don't work for them!

I would think a recall means that all the TIUs must be bad or sure to fail in a given period. I'm guessing again, but I thought they all have not failed?

 It would be good for MTH to honor the warranty for anything made with known faulty components I agree. I don't think this forum scattered responses shows all the TIUs in use around America so maybe they have a better picture of what's happening overall??

 MTH has fixed things for me in the past that were suspect. I expect them to do it in the future when applicable. If mine goes bad, I will contact them first. I don't expect them to share all the design details with a regular consumer like myself.

 Even at my level, I'm surprised still, by the strong odor that tells me something is not right. Until it fails, I'll use it. I could only guess that something is overdriven or has some inferior coating on it?? It's still working perfectly?

 Reminds me of my Chev truck. It has a noise emanating that sounds just like a bad water (pump from the old days). There was a lawsuit claiming it hurt the resale value of them as they all do it. No recall ever came about. 12 years later, mine still runs fine.