track signal question

Clean your rollers and wheels and try again

How many feeds to each block and how far apart are the feeds?

The drops in question are a 30 ft 14g speaker wire which cover a 12 to 15 ft block. With the atlas track and all those joints maybe the blocks need to be shorter  ? i keep the track rather clean maybe the rollers are a bit dirty but they dont appear so.....thanks Marty

John,

  I know you do not post that much, however I do believe this post should have been made in the DCS section of the OGR Forum.  Rich's Rant asks us to be more

aware of posting so he does not have to move stuff around so much.

PCRR/Dave

is this in the correct section??

Of course it is. It relates to DCS signal strength.

Barry,

   John's post was initially posted in the O/O27 section, not here in DCS.

PCRR/Dave

Signal strength as reported by the loco varies considerably as loco moves.  A roller on a track is a lousy contact, a mere point on tubular track and hopefully a line on perfectly flat-topped track.  So as it roll, there is variation.  If you've ever run old Lionel, you'll see constant sparkes at the rollers.

John, you wired it like I would.  I had one Atlas layout that was kicking me and I ran a ground buss wire around and hit it buss style every 15-20 feet splitting the star wire taps.  It was 10 signals from then to now.  I know you wired both outside ground wires together.  Every layout has a twist.

 

I hear guys who are happy with eight and nine signal strength.  That works very well with DCS.  I do not settle for that. Ten is my goal.  Any large layout I have worked on must have that.

Marty..i was out of town and read your last post on my phone....you kinda lost me on the terminology....i get the buss and drop idea...one buss to reach way out then a starburst to several blocks? i thought just many many drops from the terminal block to the track was the current way of thinking. maybe a combination of both? i agree with every layout has a twist. especially with atlas track. 10s all the way around? my hero!! conrail john. thanks

John;

If I read it right Marty is Buss wiring the Return outer rail wire interspersed with doing the star drops. This would give a lower resistance return and would certainly help Legacy or TMCC setup.

As Marty says, Every layout has it's oddities.

On my little den layout one power drop with splits in 2 places keeps it from looping the signal.

For your point on the joints in the atlas, As long as you have 12 or less joints per block you should be OK. Bad connections at joints needs to be fixed. Use a Voltmeter with an engine sitting powered up at the end of the block.

Put one lead on the center rail and one on the outer rail and slide down the track. Voltage will drop gradually as you move down the track, this is normal, look for sudden drops as you pass a joint. If the voltage drops as you pass a joint, Fix it. It may be center or outer rail, test by moving one lead at a time past the joint.

I run a number 12 wire under the layout and connect the outside rails to it.  It is a complete circle of wire under all the main lines.  The rest of the layout can be wired star wired to each block as Barry has in his book with paired wires.  This extra ground will clean up any signal.  I have used this on some very big layouts with total success. 

 

Follow the outline Barry has in his book and you will be happy.  This may not be needed at all if you already have 10 signal.  If you have a properly wired layout and just can not reach those perfect 10 signals, give this a try.   

 

Is a perfect 10 needed to properly run DCS, not at all.  I have a local friend with 4 loops of Atlas track and each loop is about 100 feet around.  He split each loop into four feeds, tied the outside rails together and gets perfect 10 signal.  Someday this signal issue will be a memory.

From Marty quote.

 

I run a number 12 wire under the layout and connect the outside rails to it.  It is a complete circle of wire under all the main lines.  The rest of the layout can be wired star wired to each block as Barry has in his book with paired wires.  This extra ground will clean up any signal.  I have used this on some very big layouts with total success.

 

 Yes I believe this is an excellent idea. Not for everyone if running different scales on the same tiu.  This is the first time I've head this suggestion and  think it's  good one. There would no shortage of routes the return data packs could take  to the tiu if other lines are busy. (I don't even know if that  makes sense)

 

 

I also use the heavy duty buss for ground, both track and all lighting and accessories.  This means it can carry more amperage than the sum of all TIU channels' red circuits,  For this reason, it connects directly to the transformers' U terminals, rather than going through TIU blacks.  There is a drop from all black outputs to the buss.

If I had had to rewire the layout, I'd still be running conventional, using the DCS eqpt as a walkaround control.  I don't have star wiring either.  Wire goes from each TIU red output (I use 6 channels) to a section of the control panel, where each channel feeds 10-15 toggle switches, each feeding a block through 14-gauge wire.  It works.

RJR and it should work good.  Basically my layout stayed common ground buss with wires from the TIU terminal block overlayed.  Both hot and ground were overlayed.

Marty, just what do you mean by "overlayed?"

Robert,

I don't have star wiring either.  Wire goes from each TIU red output (I use 6 channels) to a section of the control panel, where each channel feeds 10-15 toggle switches, each feeding a block through 14-gauge wire.  It works.

Your layout works, in large part, because you've added a great many light bulbs here, there and everywhere, as you've stated in the past. Although that may work fine for you it's not one that I'd recommend to anyone else.

 

I think that you might do things differently if you had it to do over again.

Not quite corrrect, Barry.  Bulbs were added early, when layout was powered by 2 Rev G's.  I found that when several sidings were toggled on, signal dropped on other blocks.  This only happened on variable channels.  Bulbs on those sidings only cured it.  I later added bulbs on each TIU output, which you were pushing at the time, which also served to indicate power was on.

I have another something....i am using the two fixed outputs with my z4000 and a variable channel set to fixed with a 180 lionel brick. The 2 fixed with the z have a much better output signal wise and performance wise then the 180 brick thru the variable set to fixed channel, i have tested this by running the same section of railroad connected to the z first and the 180 second. There is a marked difference. let me know if this is possible or maybe magic....then i will fade back into forum obscurity....big word for me!!  i have a train trip to Buffalo today so i will reply when i return...thanks in advance....conrail john

Robert,

Not quite correct

What's not correct? You just stated:

I found that when several sidings were toggled on, signal dropped on other blocks.  This only happened on variable channels.  Bulbs on those sidings only cured it.  

Like I said...  

John,

    Although I love the Z4K and would like to tell you it allows for better DCS signal, it is no the case.  I suggest you actually test your real DCS signal strength, and not your voltage, voltage from DCS to track should be 18 Volts, no matter the transformer.  You have a problem some place because you varied something, make sure both the 180 Brick and your Z4K are both delivering 18 volts to the track at different times, when you test your DCS signal, then if all connections from the DCS are identical, the signal from the DCS should be identical also.  18 Volts is 18 Volts no matter the transformer, when both transformers are working correctly.  Unless you are using a transformer that is not compatible with DCS.  However the 180 Brick is on the Compatible list.  The only way the DCS signal strength can be different, is if you are not set up in an identical manner, when using the 2 different transformers @ 18 Volts.   No Magic, just engineering.

PCRR/Dave

 

It's not correct to say I have bulbs all over.  The bulbs on TIU outputs added nothing.  I only left them there to indicate power on.  Basically, there were 5 that had effect, all on the ends of sidings on variable circuits.

Robert,

It's not correct to say I have bulbs all over.

Let me re-phrase...

 

You have several bulbs that were used to augment DCS signal strength that were placed "strategically about the layout".

 

Better?  

RJR,

   Actually the Magic bulbs at the head of each terminal block do enhance the DCS signal pick up, your set up may have been strong enough at the time, due to your other lights, so you could not tell the difference however.  They helped me run with all 10's thru out my multi level layout.   When I took them off to test, the signal dropped to about 7-8, now on the 1st level, where I use the 36" track pieces and 3/4 of the joins were eliminated, it did not matter whether the magic light at the head of the terminal block was used or not, the 10 signal strength was still present.  Those lights did matter, you just could not see it, cause you had the 10 signal strength being delieverd already.

PCRR/Dave

Dave, I don't have any terminal blocks.  Also, this is not the first time I have heard statements that signal strength through fixed channels was better than through variables.

 

Barry, not only better, but a perfect statement.  This was in accord with all advice, past & current, concerning enhancing signal strength on Rev G & H TIUs.  Was 5 dead end sidings that, when any of them was toggled on, pulled down signal on other blocks on that output.  That was the strategy

 

 

RJR,

   I get what you are saying, however it was the addition of all the track joins that varied the signal, you used the magic lights to eliminate the track join problem.

The magic lights no matter where did help, it was not the Var or Fixed channel that made the difference.   This was the reason I started using the long track section and eliminating many joins on my layouts, most people are unaware that track joins degrade the DCS signal.  I proved this by simply switching the Fixed and Var channels on my layout, at different times to see what would happen.  Barry is right your magic lights usage was the difference in signal portrial, along with the toggle for adding/subtracting the track joins.

 

PCRR/Dave

I don't know what you mean by "track join problem" or "toggle to adding/subtracting..."

 

All I said was that people on the forum some time ago had commented that signal strength on fixed seemed to be better than on variable.

 

As to lights, putting lights on the ends of certain dead-end sidings eliminated low signals elsewhere.  This is something I did some 10+ years ago.  These dead end sidings at the most had 2 track joints.

RJR,

  The people who say there is a difference in Var and Fixed DCS signals IMO are mistaken.  As for the Toggle and Track Joins let me explain, long ago when I started using the magic lights as Per Marty F's instructions, I ran across a different problem.  If I used many different tracks, I had to use a magic light about every 13 track sections, to maintain the DCS 10 signal strength.  When I use the 36" tracks and eliminated many of the track joins, the DCS signal remained at the 10 level.  This means your siding with lots of track joins when added to the main line DCS signal, makes a big difference (loss) in DCS track signal portrial, when the toggle activates each siding.  Hence your need for the magic lights at each toggle, to boost the DCS signal, over riding the DCS signal loss caused by each track join, on that siding.  The more toggles activating the track sidings, the greater loss from the numerous track joins.  When the toggles are closed, the DCS signal was not effected, the track joins were eliminated from the DCS main line operating signal, strong DCS signal again.  Switching from Var to Fixed made no difference in the DCS Signal.

PCRR/Dave 

Robert,

that people on the forum some time ago had commented that signal strength on fixed seemed to be better than on variable.

That's a DCS "urban myth".

It may well be an urban legend, just as were the biblical pronouncements that both hot and ground wires had to be the precise same length.  But who's to say what situation exists on someone else's layout?  I think one should not conclude that, because something is such and such on their layout, that the same holds true on all.  No 2 layouts are electrically identical.

 

Dave, I have to disagree on the effect of turning on sidings.  When there are 2 sidings side-by-side, both of the same length and same number of track joints, why would turning on one have an effect but not turning on the other?  Out of 6 sidings on that circuit, only 2 or 3 needed bulbs, although I added bulbs to all bumpers for uniformity of appearance.  My experience was also that once one was turned on that affected the signal, there was no further effect from turning more on.  And on none of the sidings was there ever a low signal. Why would there be a loss resulting from track joints in an area where there were no locos, no nothing?  Also to further muddy things, when the sidings were on and signal low, I found that if I took a long wire with alligator clips, placed one end on the TIU and the other on the center rail of the track block whose signal was down---thereby having 2 feeds to that block---the signal immediately jumped to 10.  I would also note that other areas of the layout, where there were more track joints (fed from fixed 1 & 2), had solid 10's from day 1. 

 

My point in all this is merely to note that no one knows with absolute certainty what happens to the DCS signal.   Trial & error reveals what is best for a particular layout.  Not having a Rev L, I won't guess as to what effect the different signal generators might have if Rev L had been available back in 2002.

Robert,

No 2 layouts are electrically identical.

True, however, the laws of physics are the same for all layouts.

RJR,

   I agree with the individual trial and error, however using the 12 track join discipline, my layouts always stayed in the 10 signal range, thru out the entire layout.  Violating the rule, the track signal was degraded, and additional magic lights were needed to correct the problem.  Its not the engine on one particular track siding, its the track joins, that effects the entire layout signal strength.  When you add or subtract the siding, with multi track joins, by the use of the toggle switch, it changes the over all signal strength of the entire layout, when you have no magic light.  Adding the magic lights completely eliminates the drop in signal strength. Further I have found that the toggle switch itself, helps to degrade the track signal also.

PCRR/Dave  

Did not the 12-track "rule" apply to a block rather than an entire TIU circuit?

 

See p 55 of Barry's book.  I'd copy and paste the text but he has it protected.  Attempt to limit number of blocks to 12-15 per channel.  Limit blocks to 11-12 sections.

 

Since my layout was built to run many conventional locos at once with short blocks individually fed, and uses Gargraves 36" track, nowhere do I even come close to this number of section per block. 

Robert,

Did not the 12-track "rule" apply to a block rather than an entire TIU circuit?

Yes, however, another guideline (pre - Rev.L) was no more than 12-13 blocks per TIU channel.

RJR,

   It actually applies to both, each block can have no more than 12 joins before the signal starts to degrade, if you want your DCS signal to remain at about 10 thru out each block on the complete DCS layout.  This is why it's best to place magic lights at the head of each terminal block and if you use toggle switches for each siding, there also, and as Barry points out only so many per TIU, I do cheat a little on this and get away with it by using a few more magic lights.

PCRR/Dave

  

Barry & Dav, my edit crossed your posts.  I am well within Barry's guidelines.

I think John's problem might be engine related if one engine gets a much higher signal than the other.  I had a PS3 in the shop last week that got a very week signal if you were lucky.  It left at 10s.

 

Ground:

remove black from ground screw

remove black from wheels

 

Board:

re-touch solder joints on signal generator (thank you Marty!)

re-touch solder on caps

 

Optional:

remove 2/3 switch

remove DCS/DCC switch

 

Dave

 

Barry,

   Now with Rev L I am still building in the same manner, however I am needing to use a lot less magic lights, unless they are absolutely required.   In fact this years multi level DCS/Legacy Christmas layout needed no magic lights what so ever.  Of course the layout is only a 6x8 type living room layout.  Are the Rev L TIU's that much better signal providers, or is it, that I am a much more disciplined DCS layout builder, that no magic lights were needed at all. 

PCRR/Dave