Skip to main content

In a conversation this morning with MTH R&D, we discussed a technical issue that has lately been a topic of conversation on the forum. The subject was regarding updating sound files in DCS engines. I asked if I could get a better understanding of the issue and the reasons behind it. thsi thread is intended to pass what I learned along to any of you who may be interested.

 

Several forum members have reported receiving advice from MTH that, when transferring sound files via the DCS Loader Program, certain techniques should be utilized so as to decrease the occurrence of errors and to reduce the time necessary for the transfer. The two techniques are:

  • Use DC voltage for track power as opposed to AC voltage
  • Ensure that at least 20 feet of wire is between the TIU channel output and the track upon which the target DCS engine resides.

However, there are two caveats that are sometimes not stated:

  • These are only applicable when the TIU is a Rev. L and the engine is a PS3
  • If DC voltage is used, the 20 feet of wire is not necessary.

Further, the reason for these techniques also are often not mentioned.

 

The Rev. L TIU was designed to be more effective on larger layouts than previous TIUs. It's equipped with an FPGA (Field-programmable gate array) chip for its digital sound processor (DSP) as opposed to the ASIC (application-specific integrated circuit) used in earlier TIUs. The FPGA allows the Rev. L TIU to communicate more effectively with all DCS engines than could previous TIUs.

 

Further, PS3 engines also have FPGA chips instead of the ASIC chips that are in PS2 engines (both 5 volt and 3 volt varieties). The combination of both Rev. L TIUs and PS3 engines having FPGA chips is what allows for improved communications between the Rev. L TIU and PS3 engines, which also leads to improved PS3 engine signal strength.

 

However, one drawback of this scenario is that at times  complex communication with a PS3 engine (such as what occurs during a sound fike transfer) may actually degrade because the TIU is "shouting" at the engine and the engine is "shouting" back. The two things that can correct this "shouting" match are to either quiet any electrical background noise by using DC voltage rather than AC voltage, or by separating the TIU from the engine with 20 feet of wire. Either one should suffice to eliminate the problem.

 

Again, this scenario can only occur when using a Rev.L TIU to transfer sound files with a PS3 engine.

 

As an experiment, I intend to try something I never would have attempted before. That is, transferring a sound file between a PS3 engine and the PC, while the engine is on the layout.  If someone beats me to it, please post your results on this thread. I'll do the same.

Original Post

Replies sorted oldest to newest

Barry,

 

I know I'm not using 20 feet of cable (maybe only 2 feet).  What kind of problems are people seeing or that I can expect?  Will the file still download successfully only to be found corrupted when I run the engine?  Again, what indications of a corrupted file can we expect?

 

Admiral Grace Hopper was credited with saying something like "If you want to speed up the process...shorten the wiring".  Guess she never saw command control trains in operation

Bob,

 

If there are any issues, they will manifest as error messages and longer than normal transfer times.

 

If it gets to the end of the load process, it should be fine. 

 

If you're nervous, do thee file transfer while the engine is on the layout, substituting 20' of track for 20' of wire, That should do it. Just ensure that no other engines are getting power during the transfer.

Barry, Can you clarify this with R&D?  What I have been told is the 20' of wire is specific to PS-3, but the use of DC is also beneficial for PS-2 boards.  MTH Service has stated use of DC with loading PS-2 boards with any TIU increases speed and reduces errors.  I agree from my experience.

 

So the use of DC power, is not a PS-3 specific issue.  Just the 20' of wire.   G

George,

 

This is what I received as a follow-up from MTH this morning...

DC power does, overall, have the tendency to provide better signaling. This assumes a clean, filtered DC supply with little ripple. This applies to all TIU revisions.

 

That statement (and it’s a paraphrase of one I made originally) is based on DCS signal theory and anecdotal evidence from personal experience but is not exactly scientific.


Across all TIU revisions DC power CAN provide the potential for better signaling which would show in reduced retries when loading data across the rails.

 

Most people, however, fall into your camp where they’ve not had any noticeable issues and their data files move across the rails to the engine without issues; typically using AC power.

Everyone should feel free to draw their own conclusions based on the above. Personally, I wouldn't bother with DC power, other than in a Rev. L and PS3 scenario, unless I was having issues, which I most certainly am not.

Last edited by Barry Broskowitz

Further, PS3 engines also have FPGA chips instead of the ASIC chips that are in PS2 engines (both 5 volt and 3 volt varieties). The combination of both Rev. L TIUs and PS3 engines having FPGA chips is what allows for improved communications between the Rev. L TIU and PS3 engines, which also leads to improved PS3 engine signal strength.

 

That's what I always thought was the RevL TIU was made for PS3 engines though nobody would admit it. Infact I was chastized for even thinking like this when the RevL was introduced.

 

 

Chuck,

That's what I always thought was the RevL TIU was made for PS3 engines though nobody would admit it. Infact I was chastized for even thinking like this when the RevL was introduced.

Actually, that's correct - they were made to complement each other. I made this point in the 2nd editionof my book.

 

Further, TIU and engine electronics have a history of evolving together. The Rev. I TIU was the first TIU with a 3 volt processor, just like the 3 volt processor in the PS2 3 volt engines.

 This is interesting. I fall into not seeing to much diff between AC/DC camp. In fact, I was advised to switch my whole AC G scale layout to DC. After great expense buying a huge powersupply, it's not much different. The DC power supply I got from "Tank" actually is working quite well after all.

 The rev "L" is the biggest change I've seen. I can make my older one just as good. It just takes some work with bulbs/filters.

 Now I have not changed any PS3 engine's sound files yet. I am a little dissapointed that my PS3 dash 8's volume is lower. Enough to where I have to adjust it in a consist to hear the horn at all. I was hoping they would match. The sounds are very good.

 Now if I could figure out why I just lost another engine from the remote this week????

Originally Posted by Barry Broskowitz:

Ed,

How do you apply DC power?

Just as if it was AC, or am I not understanding you correctly?

Berry.....oops, sorry Barry,

 

It sounds like he is wondering (since I have never done any updates/transfers, I have no idea) where the AC power is input---I think it goes to the TIU, but I could be confused.

 

I am thinking he might be confused, and think that the AC that goes to the engine is what is being referred to as using DC as a sub. Reading the thread, if one is a neophyte, it is confusing to them. (me)

 

I could be wrong, but just a guess.

 

Thanks, Greg

Greg,

and think that the AC that goes to the engine is what is being referred to as using DC

He would be correct!

 

As I stated in my initial post, "Use DC voltage for track power as opposed to AC voltage".

 

The issue is not how the TIU is powered, rather, how the PS3 engine is powered. What is being discussed is using DC voltage as input to the TIU channel (must be a fixed channel) rather than AC voltage.

 

As an FYI, one of my close friends who is also a DCS beta tester and MTH-trained tech called me one evening to discuss problems he was having transferring a PS3 sound file while using DC voltage.

 

The engine kept coming up in conventional mode as soon as track power was applied and the DCS Loader Program couldn't find it.

 

As it turned out, the problem was that the TIU itself was being powered by AC power through its Aux. Power port. When that power source was removed or also changed to DC, the transfer worked without a hitch.

 

Unfortunately, we never did any further research into the issue. However, as I recall, he was using an older TIU model rather than a Rev. L.

Hey, thanks for keeping everyone on track.

I seems I posted too early, sorry. Or maybe I kept Ed straight by him getting your answer!

I went back to the top and re-read. In the first post, it was clear:

 

"Use DC voltage for track power as opposed to AC voltage"


Did not mean to break the train of thought here.

 

Thanks again, Greg

Post
The DCS Forum is sponsored by
×
×
×
×
Link copied to your clipboard.
×
×