Skip to main content

choochoopaul posted:

I realize that the functions of control with a handheld are lost with the addition of the board into the DCS-RC, but with a longer timed reset cycle could you send a data string to a engine between resets?


Yes you could.  Obviously the factory-stock DCS-RC without the PBW (perpetual barking watchdog) board has a much longer reset cycle time of forever (it never resets unless you remove power)...and during that longer cycle is when data strings are sent to the engine under handheld control.

When you get your boards back, if you want to experiment with a longer reset cycle, probably the easiest is to put a capacitor in parallel with the existing capacitor.  That is, it's arguably easier to tack a capacitor in parallel with an existing one (to increase it's value and hence the time delay) than to add a resistor in series with an existing one.  It's the product of the R and C that sets the time delay so a 10% increase in the capacitance would increase the delay by 10%.

But, to repeat, I think you need to confirm your unmodified DCS-RC can start your PS engine silently in command mode in the first place.  Again, I realize the TIU sends its watchdog for up to 45 seconds (depending on TIU Rev) but the fact is the DCS-RC sends that single 1/10th second burst just after it receives power.  And that seems to be all that's needed to force a PS2/3 engine in silent command mode on power-up. 

Another thought.  I realize you'd need to factory-reset your engine to the default DCS address but did you confirm that your particular unmodified DCS-RC can in fact control an engine using the handheld remote?



Yes to all of the a fore mentioned questions.  This morning with first one, then two, then three MTH engines all reset to address (0) sitting the longest distance from the power source on the layout.  With the DCS-RC connected passively to the track and applying power, all three remained dark and silent.  When the startup command was sent out all three woke up, and when commanded to blow their whistles they did, when commanded to move they all moved in a very bad looking consist.  When commanded to shut down, they all did.  I would say at this point I do have a working test bed with both the DCS-RC and the full DCS-TIU systems.

I even checked that with the very bad consist running around the loop of track, I could also fire up a Legacy TMCC engine on a siding connected to the same loop.

How much time does it take from the application of power to the DCS-RC to the start of the watchdog burst?

I am ready for the boards to return from their vacation out east.

choochoopaul posted:

How much time does it take from the application of power to the DCS-RC to the start of the watchdog burst?

I am ready for the boards to return from their vacation out east.

It would depend on the track supply voltage and if AC or DC since it is charging up some capacitors which then becomes regulated voltage to run the microprocessor.  I have not measured this but I suspect it's less than 1 sec.  Once the microprocessor starts running you would have the same 3/4 sec delay. 

But didn't you report yesterday you powered up an unmodified DCS-RC on a short test track and the engine started up in conventional?

Has anyone besides Paul tried the watchdog boards?  He had issues getting them to keep the engine silent and sent them back for me to test.  I couldn't find any issue, and they worked here, but we may have overlooked something for other environments in our little science experiment.  If you're tried them and they worked, or not worked, let's hear your story.  I'm curious why we get different results from seemingly similar test setups.


Finally got to trying out one of the WD timers PBW's you made and it works fine here. I tried it on one spur track and it worked. While under the layout (scratching my head, looking at all the wires I have added and not documented ) I thought why not just connect it to the MTH terminal block I have feeding loop 2 with all the spurs and sidings on it. I have only been fiddling with one train and a PS3 RK RS3 engine, but that seems to work fine too. It's one of the newer ones that also starts with the thumb wheel, tried starting it both ways. I will get the rest of the trains back on the tracks and see if all is still well, but I see no reason they should not be just fine.  

Forgot to add this earlier: The DCS-RC's green LED came on green when powered up and had a blip in it at just about 1 sec as near as I could time it with my watch. Pretty close anyway.

FWIW, if it helps with problem solving, here is my setup right now. Loop 2 (the one tested) is a center rail isolated loop using Atlas O-54 track. It has 2 inner sidings and 4 inner spurs. There are 4 blocks on the main and each siding/spur is isolated coming off the switch that feed it. The sidings and spurs are switched with LED rocker switches at a small panel. Each switch powers a relay that actually switches the track power, center rail only (small glitch with the switch's LED commons). All the power comes from an MTH terminal block and that is where I now have the DCS-RC PBW connected in passive mode. It feeds the entire loop, all sidings and spurs. Power source is PH-180 to PSX-AC to TIU to MTH terminal block, Legacy connected to TIU commons.  

I will try the rest of the timers PBW's tomorrow with the same setup above. Will let you know on that too. 

These are really pretty cool little devices, work great so far and no more watchdog problems. Just flip the power switch and go when ever you are ready! Also finally got my gantry crane track, found it on the front porch a couple days ago. 

(Oops, was re-reading the problems above and noticed the proper name - Perpetual Barking Watchdog (PBW), post edited). 

Last edited by rtr12

Paul had issues, and I don't know if he's made any progress.  He sent all his modules and a DCS-RC base back to me and I tested all of them and they worked for me.  That's why I'm curious as to other people's experience, there has to be an environmental issue that we're missing I would imagine.  I actually got a couple more boards and built up a module so I could have one to test if more questions came up, I had previously sent them all out.


Please pardon the delay, I work on the layout catch as catch can.

I was down there tonight attempting to electrically isolate the yard to allow me to completely turn off the individual tracks.  I wanted to do that to see if I could first create the need for the watchdog signal thinking that way I could accurately test it with and without the chip.

Most of the time I was finding the things I put away for the project, i.e.  rotary switch and electrical box so that it will be easier to deal with.   I hope to work on it next weekend.

Still re-reading the entire thread here. Probably should have mentioned that in the last testing I had the main connected to the PBW as well. Main had a Legacy engine and another PS3 on it, no problems there either, all seemed to work well.

Would probably be fine as is, but final install will be a re-wire so only the spur and siding tracks are on the PBW full time, main will be separate. 

gunrunnerjohn posted:

The "tuning" is done with a scope, I set the retrigger time between 950ms and 1050ms by adjusting the value of R1 in the schematic.  Values ranged from 1meg to 1.5meg.

I finally got it done and tested.  It works perfectly. 

The R1 tuned at 1.2meg gave me a cycle just a little over 1 sec. 

The WD is clearly visible every second on the scope and, so far, while using the PBW in passive mode, I did not have any issue controlling the trains with the TIU.

Thanks for sharing the design, the Gerber files and the assembly details.

Fichier 17-01-26 22 23 56


Images (1)
  • Fichier 17-01-26 22 23 56
Last edited by Daniel Auger


My homework  was to install the WDS signal and give a report.  I set up my yard and isolated the tracks.

I read everything I could on the forum again and realized that while everyone was saying DCSRC (DCS remote commander) I was thinking DSC remote control.  I actually took apart my remote control and realized that it was not the same board as the images or the discussion.  Thanks to Barry's book no harm no foul.

I then took apart the precursor to the remote commander that I had from a RTR set from 2000.  It does not appear to be the same board everyone else is using.

It has remote commander circuit boardmore components and inputs for proto dispatch and proto cast.

Is there a spot on this board for me to connect your board?

I have included a picture of the board.  I am stuck at this point and need to either find a remote commander board or rig up your circuit a different way. 

My  best excuse is that your watchdog signal ate my homework.  Sorry for the pun.

Thanks for any suggestions.







Images (1)
  • remote commander circuit board

Just got one DCS-RC modified with the PBW and replaced on the layout. Works perfectly!

I found the underside pins did not need any trimming, and there is ample clearance inside the RC cover for the PBW board. Pretty slick.

I tried it on a couple of PS_2 engines, one on a t'table track and another in the adjacent yard. Both sit dark and quiet until started up, and I don't power these tracks until needed; well after the main TIU WD signal has come and gone. The green led on the RC blinks at a frequency of just longer than once per second. All good.

Thanks Stan and grj for a nice little add-on.



Success stories are always good.   I think a few of my projects would have never happened if it weren't for the Internet.  I love the instant feedback and alternate ideas that come up discussing these things, and usually a better design is the end result.

This one was kind of an exception, the design was really all Stan, there was no reason for any improvement, it worked the way it was.  My contribution was to turn it into a PCB that could be produced.

Thank you all for all of the thought that has gone into these concepts and builds.  I am at a point on a new build when I need to zero in on control of turntable whisker tracks as well as several sidings.  I am leaning towards using an AIU with relays for track power and am also considering the  perpetual watchdog signal apparatus.  I’ve just read three separate threads and so I’m still trying to digest it all.  But I do have a question requiring connecting the PBW - since it is connected to both the center and outside rails of each siding, when connected to one siding,  doesn’t it provide a conduit to power all sidings?  Or maybe using it as I mentioned above doesn’t work?  Your guidance is appreciated.

Bob Golfs posted:

... sinceit is connected to both the center and outside rails of each siding, when connected to one siding,  doesn’t it provide a conduit to power all sidings?  Or maybe using it as I mentioned above doesn’t work?  Your guidance is appreciated.

It is connected to both the center and outside rails of each siding that is powered.

Stated differently, the PBW is connected to sidings/whiskers when and only when they are powered via a toggle switch or relay (e.g., AIU ACC ports) that connects the active/powered/hot center rail to a particular siding/whisker.   The PBW itself is NOT directly connected to all the affected sidings/whiskers.



Yes to your question.

But allow me to elaborate.  Some layouts with a roundtables, plus a yards, plus multiple sidings, etc. are so large that when you draw out the wiring diagram with multi-position rotary-power switches, multiple AIU ACC(essory) ports, or whatever block/section powering method you turns out that the wiring is simplified by using 2 or more PBW's.  If you bide your time, you can pick up a DCSRC for next to nothing on ebay or thru the buy-sell OGR forum.  

Add Reply

The DCS Forum is sponsored by

OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
800-980-OGRR (6477)

Link copied to your clipboard.