Double commands

This can happen if you have more than one connection from the TIU to a track block.

This and a whole lot more is all in "The DCS O Gauge Companion 2nd Edition", now available for purchase as an eBook or a printed book from MTH's web store site! Click on the link below to go to MTH's web page for the book!

Appreciate The Link Barry, but I do have your book. From the parts I have read, it is excellent.  

   Anyways, I do run Two loops off of Variable 1. Would you perhaps suggest using variable two as well? I do not have intercepting blocks.

I do run Two loops off of Variable 1. Would you perhaps suggest using variable two as well? I do not have intercepting blocks.

Are you using isolated blocks, with one TIU connection per block? If not, that could be the source of your problem.

Whether you use one channel or two is not the issue.

Yes, the blocks are isolated from one another. It is just strange. it only seems to happen with that command, and completely random at that. Some days It happens every now and then, some days not at all. 

I had that problem along with some others on the forum. Barry thought maybe I had too many blocks isolated on mine. They were probably 8' long. When you run a 8 or 9 car passenger train being pulled by a ABA set of Alco's your talking about 18'. So it was possible the train could be covering 3 blocks at one time. I lengthen the blocks to 16 feet and it improved. I also found that running consists of freight lessened the chance of these problem. I guess it has something to do with the lights in passenger cars.  I'm just guessing as I'm no DCS expert But this worked for me.

I experience that same exact thing at times. I have blocks that are each fed with 1 wire and still get the double command. Especially with the forward horn (2 quick blasts). I got it before I had blocks, so I broke it up with insulated gaps to eliminate it, but it didn't.

I get the same thing over at my Dad's house on the big layout, and we wired that one with blocks from the get go. Forward signal (2 horns) is repeated even when the engine is in a block all by itself... with only one feeder.

If you activate the Crossing sound, and then quickly activate it again while the first one is still playing, does it repeat in the same way you describe?

If you only get the one, then ignore this post.  But if you get two, then the command is one that can be queued in the engine.  In which case the likely explanation is the engine is receiving the command twice and has queued up the 2nd request. This can happen on a single block as witnessed in the last post.  It's a little involved but here's how it works.  The TIU sends commands to an engine and waits for an acknowledgement saying the engine got it.  If the TIU does not get an acknowledgement, it sends the command again. This re-trying continues for some interval after which you get an error message like "Check Track" or whatever.

So suppose the engine indeed receives the Crossing command on the TIU's first attempt and it sends the Acknowledge to the TIU.  The engine starts playing the sound.  But for whatever reason, the acknowledge is missed by the TIU.  The TIU then sends the command again thinking the engine missed the first one.  So the engine now receives the command again and because it's a queue-able command, it saves it and plays it when the first one is done.

I understand how the two way commands work, but I want to know WHY the engine is acknowledging the first command, but another command is still sent out.

I understand how the two way commands work, but I want to know WHY the engine is acknowledging the first command, but another command is still sent out.

I would imagine that if the engine had it's front pickup roller on one block and the rear pickup roller on another when the command is given, technically linking 2 blocks together, this could cause two commands being received by the engine.  Even cars with 2 rollers could also cause the problem by bridging the blocks. That's one reason to use as few blocks as possible in my opinion.

The thing is I only have one set of feeder wires on this loop

You might want to try breaking the loop into two halves with one pair of wires to each half.

Do you have the center rail gapped on the side of the loop opposite the feeders?  If not, you're probably sending two DCS signals around the loop, one clockwise and one counterclockwise.

The TIU may be missing the acknowledge.  All 10s does not mean 100.00% absolutely positively error-free transmissions.

Stan,

The TIU may be missing the acknowledge.  All 10s does not mean 100.00% absolutely positively error-free transmissions.

Not exactly.

From The DCS O Gauge Companion 2nd Edition, page 55:

DCS Signal Strength

DCS signal strength is a number, from 1-10, that's calculated based on the number of data packets that the TIU sends and receives back from a PS2 engine that's performing a DCS signal strength test.

During the test, the TIU sends a continuous series of data packets to the PS2 engine. The PS2 engine is expected to reply to every packet that it receives, however, some packets are either not received by the engine or the response is not received by the TIU. Regardless, the TIU calculates DCS signal strength by counting how many response packets the TIU receives from the PS2 engine out of each hundred data packets the TIU sent, and then looks at a sliding scale to determine the signal strength that should be reported. The scale is not linear, rather, it is such that 87-100 packets equates to a DCS signal strength of “10”, 80-86 equates to a “9”, and so on.

This and a whole lot more is all in "The DCS O Gauge Companion 2nd Edition", now available for purchase as an eBook or a printed book from MTH's web store site! Click on the link below to go to MTH's web page for the book!

Now it's my turn to say Huh? 

Barry, you say 87-100 will report as a signal strength of "10".  That's exactly my point.  Just because he reads 10s all around does not mean he's getting 100% error-free transmission.  As you say his actual performance may be only 87.

Or maybe you're agreeing with me and I'm misinterpreting your terse response.

 Obviously, I have to bow to you guy's experience.

I remember a post from days back when it was mentioned to play music thru the engine. Although you may get tens all around, the music playing uses more of the system's capabilities and will distort if there's errors. So what do you think??

 I have noticed myself that tens all around doesn't mean there won't be problem areas when conditions arise. The right combination of engines or equipment seems to effect the signal in areas of weakness that were believed to be solid.

 All that said, following Barry's rules exactly and experimenting just a little with moving connection points, .......and a good step back to view the rules are applied correctly, my RR is pretty perfect and I can run trouble free.

 Another disclaimer: I'm running 2 rail!! (which is tougher to perfect)

Stan,

I'm just attempting to clear the air and put signal strength in context.

Typically, double-execution of commands is not a signal strength issue per se, rather, it's a doubling of command packets coming at the engine from different directions.

Regardless, 87 of 100 packets is plenty good enough for error-message-free DCS operation.

Rewiring is not exactly Christmas, but ya do what you have to do.

I will split the loop and report my findings. It could be sometime because as I've said, it is completely random. 

Answering my question might be simpler than rewiring. 

Don't worry Gary, I had your proposition stuck on my forehead. That was going to be my first attempt before rewiring anything.

I cut the center rail as Gary suggested, and Ran the engines on that loop this morning. I gave out the crossing signal command about thirty times, and it only repeated once.  Still repeated once though. 

I'd consider one out of thirty close enough.  On the one time was the engine by any chance astraddle the gap?  Under any conditions it's possible to get an occasional "echo".  Add a light bulb or a filter and it may go to one out of fifty or a hundred.

I have this "problem" occasionally.  I have one feed per block, multiple blocks on each "loop,"  train does not extend out of block, and signal is 9 or 10.  But since operations aren't affected, I don't spend any thougt trying to figure the cause---I have higher priority items.

Barry,

Are you suggesting that a single command transmission from a single TIU output is being received by the engine twice - one going CW and the other going CCW around a loop?  That is, the time delay from going around the longer direction is large enough so that a TIU transmission is received twice at the engine?

Stan,

Are you suggesting that a single command transmission from a single TIU output is being received by the engine twice - one going CW and the other going CCW around a loop?  That is, the time delay from going around the longer direction is large enough so that a TIU transmission is received twice at the engine?

Pretty much, however, I'm open to "alternate theories of the crime."  

Well, I'm standing by the alternate theory that the engine is acknowledging the Crossing command but the TIU is missing it...so the TIU re-transmits an additional command which is queued and then played.  The missing acknowledge could be the result of a small section of dirty track, momentary wheel/roller bounce, random noise from wherever, etc.  So even if the signal strength reads 10s all around, this doesn't mean you have error-free transmissions or acknowledges - though as Barry points out, 10s usually means error-message-free transmissions.  The OP's latest experiment of getting a 1 out of 30 repeats essentially falls in the range of a 10 since this would be in the high 90's.

To dig deeper I suggest doing what Enginear-Joe says.  Run the audio test. It is not quantitative but it might identify areas of the track that report 10s but may be getting dirty or have an unsmooth joint causing roller/wheel bounce or whatever. 

IMO, the two-directions theory should be laid to rest.  If you look at a TIU transmission on an oscilloscope, and then calculate the speed at which the electrical signal travels down the rails, you'd need well over 100,000 feet (tens of miles) of track for the signal to be delayed enough to appear as a duplicate from the long way around.  The loss in signal strength over such a distance would also be substantial.