DCS Acting Wierd

If I read your troubleshooting correctly, it's the loop at fault then. Does this loop have multiple feeds (blocks)? Remove them one by one?

Is there something on the loop or connected that could draw down the signal? Like switch motors, lighted cars or buildings, or TMCC equipment?

Did this loop work originally? Did you do anything to change this loop? Have you removed everything from the tracks for a test?

If they don't receive any signal, the engines would start right up when they see power. If they don't and stay quiet when the rails have power, they are receiving the watchdog signal at least.

maxmn_98 posted:

 The TIU version is Rev. L but does not have the USB port.

this confuses me.

I thought all version "L"s had usb ports. Are you sure it's not an I?

If it is you may have to add bulbs or filters to the outputs and furthest points of the track.

You would have to go deeper and explain what your power is? what your track type and connections are? wiring size? why the voltage is varying so much?

The bottom of the TIU has the letter L on it definitely. This may have been an early one. The power supply to the bottom loops is a MTH Z-4000 and the upper an old ZW. There are no blocks on the  layout. The upper loops go through the variable voltage feeds on the TIU and the Z-4000 the fixed voltage. The MTH engines work okay on the upper loop. The bottom loops they do not react to any command and I get the engine not on track.  Once in great while when I get the engine not on track the engine will react bypassing the startup command and moving. The short setup track is powered when I use it by one of the fixed voltage outputs from the TIU and the Engines react as they should. I do not get Engine not on track signal on the Remote.  Everything was working okay on Saturday but Sunday the bottom loops stopped working in the DCS mode. The loops still work with the Lionel command system.  The track is a mixture of Atlas and Gargraves. The switches are Atlas or Ross and the Ross use DZ2500 switch machines.  Tracks are all powered.

First off, if the TIU doesn't have the USB port, it is NOT the Rev. L.  I don't care what the label on the bottom says, the defining external feature of the Rev. L was the USB port.  Every Rev. L TIU shipped had the USB port, it was in the original design.

I would swap the variable and fixed voltage feeds (variable to bottom loops and fixed to upper) and see what happens. That should give you a clue as to what may be wrong.

maxmn_98 posted:

The bottom of the TIU has the letter L on it definitely. This may have been an early one.

Image? Is it the same engines that have issues even after you swap the wiring? The power goes thru the TIU to these loops?

One thing to check.. if they are PS3 engines, the power feed MUST be to the center rail with ground feed to the outside rail. PS2 engines aren’t as sensitive but work better when the power feed is to the center rail.

I swapped the leads around and the fix voltage output works on the upper loops. The lower loops do not work with the variable outputs but the H-8 does light up but just sits there. It does not respond to commands. The test engine a PS2 doesn't react at all but it worked on the upper loops responding to all commands. Any idea what I should look for with the lower loops. Based on the test the TIU appears to be working properly. My setup track uses one of the fixed outputs and the engines react as they should. Note: the engines are PS2's.

DG posted:

One thing to check.. if they are PS3 engines, the power feed MUST be to the center rail with ground feed to the outside rail. PS2 engines aren’t as sensitive but work better when the power feed is to the center rail.

RU you sure about this? Sounds backwards to me.

maxmn_98 posted:

I swapped the leads around and the fix voltage output works on the upper loops. The lower loops do not work with the variable outputs but the H-8 does light up but just sits there. It does not respond to commands. The test engine a PS2 doesn't react at all but it worked on the upper loops responding to all commands. Any idea what I should look for with the lower loops. Based on the test the TIU appears to be working properly. My setup track uses one of the fixed outputs and the engines react as they should. Note: the engines are PS2's.

Do you have each loop of track broken in at least one spot? Cut the center rail there. It's so the signal doesn't double back on itself.

maxmn_98 posted:

I swapped the leads around and the fix voltage output works on the upper loops. The lower loops do not work with the variable outputs but the H-8 does light up but just sits there. It does not respond to commands. The test engine a PS2 doesn't react at all but it worked on the upper loops responding to all commands. Any idea what I should look for with the lower loops. Based on the test the TIU appears to be working properly. My setup track uses one of the fixed outputs and the engines react as they should. Note: the engines are PS2's.

Since the upper loops work off of both the fixed and variable TIU outputs, but the lower loops do not work with either the fixed or variable TIU outputs, it sure seems like the problem lies within the lower loops themselves and not the TIU.

Are the loops electrically separated ?

Have you tested the DCS signal strength on both upper and lower loops ?

There are two lift sections where there are two tracks (4 breaks) on each but there is power on the sections.  If there is a problem with the breaks on the lower loops shouldn't I have similar problems on the upper loops. I've tested each track and I have power in each section. Are you saying that something is wrong with the lower loop section that has the breaks.

Follow Engineer-Joe's advice and cut the center rail. Adding proper blocks (per DCS Companion book) eliminated all my 'engine not found' and all similar DCS errors. I haven't had an error caused by my DCS system since I added the blocks in 2014. Had plenty before doing that.

You need to isolate each TIU channel from all the others, then you need blocks (at least one cut as Joe suggests) in each loop of track.

Yes, the DCS signal from the TIU goes out thru the positive (red) post to the track and the engine sends back to the TIU thru the ground (black) post. Starting with PS3 engines, they were real sensitive to that. It’s even worse I’d you use common ground loops instead of “star” wiring. I am not sure how sensitive 2 rail engines are to that, but 3 rails are sensitive. The DCS signal is a DC current riding on the AC current going to and from the engine on 3 rail engines. The “watch dog” signal is only going out thru the positive feed when power is applied. 

Well, having power (voltage) is not the same thing as having a good DCS signal and I'm still not certain from your answer if the upper and lower loops are electrically separated or not - should be just a yes or no answer. You say the upper is fed by a ZW and the lower by a Z4K - if the loops are electrically connected are the transformers in phase ?

You could try Joe and RTR12's suggestion to cut the lower loop tracks into blocks and see if that helps, but I think you first need to determine if the lower loops are getting a good DCS signal. I would also find unusual that one set of tracks fed from the same TIU would not need blocks and work fine, but the other would. If that's the case then there might be something interfering with the DCS signal on the lower loops.

I agree with Richie about the strength of the DCS signal, one thing that eats the DCS signal alive is operating cars such as searchlight cars and operating accessories that are running off of track power. Some switch machines that are running off of track power also eat the DCS up.

DG posted:

 ...  It’s even worse I’d you use common ground loops instead of “star” wiring. 

Maybe I am mis-interpreting this, but my layout is wired just as you describe and it's been working quite well for over 6 years now. The signal wire from my Legacy system is connected (jumpered) to all four common outputs on my TIU.  All I have on my layout are PS3 and Legacy engines (I have no PS2s).  Haven't had a DCS error since 2014. That's when I got this layout wired up. It has blocks (star wired, per DCS Companion book) isolating only the center rails, all common rails are connected as well as jumpered at the TIU. Both systems work very well here, with each other and at the same time. 

My layout is also wired with a very large common ground system and does not use the “star” wiring system and I also have no problems. The entire layout is 25 by 40. The are many isolated blocks, each with its own power feed running back to a power strip near the TIUs. The star system was recommended by MTH because alot of people were using a common ground with only one or two power connections and one or two ground connections with only one block on a medium to large layout, thus they saw signal loss problems. If you have isolated block of 20 to 30 feet with one power drop and  ground drops for each block,  you are just fine. Not sure how big the OP layout is and how it is blocked. The “star” system tends to improve the signal but requires isolated blocks also. MTH says the power feed and the ground feed should be the same length for best results, mine probably aren’t but I get good results. MTH definition of “star” wiring is a paired wire with one carrying positive and the other carrying ground. You and I are calling “star” wiring for only the positive feeds.

I think I found the problem.  I have a TMCC signal booster connected to one of the lower loops to boost the signal to the whole layout.  Reading through some of the posts in the forum, there was some discussion about the TMCC signal interfering with the DCS signal. I decided to take the booster out of circuit and low and behold DCS is working on the lower loops again.  Now can someone tell me how to to put the booster back into the circuit so it doesn't interfere with the DCS signal. The booster has been connected for a number of months and did not seem to affect the DCS system and just this week it killed DCS on the lower loops.

My guess is if you check the signal strength of the DCS signal out of that TIU channel, you're likely to find it's lost a little of it's "punch".  The Rev. L has a habit of slowly losing DCS signal strength on a channel as gates in the 74ACT244 DCS driver chip die due to transients.

Do you have the amplitude adjustment mod on the TMCC Buffer?  That aids in minimizing the DCS interference from the TMCC signal.

I have a DM TMCC Buffer Version 1.1. I do not see anyway to adjust the signal. Anyway to add a way to adjust the signal externally?

I have a simple kit to add the adjustment pot, send me an email and we'll fix you up.  If you don't want to do it yourself (it does involve work on SMD components on the PCB), send me the buffer and I'll install it.

Since you said it used to work, one quick thing you could try is cleaning the rails, all three (center and two outside). The DCS signal goes out on the center rail and the PS2 board in the engine sends back to the TIU thru the wheels contacting the outside rails and thus a path back to the TIU. Since the TMCC signal booster adds more interference to the ground rails, it makes it even more important to have a strong DCS signal back thru the ground wiring.

I’ve found adjusting the level of the Buffer to be very useful. I very much like the Buffer to monitor the health of the signal coming out of the base as well as boosting it a bit. At one point I was missing some sensor track triggers and fixed that by adjusting the Buffer’s output down just a tiny bit.