DCC and voltage around my layout.

[NP2626]

First of all I want to explain that the system I am describing has worked just fine for the last 13 years.

 

I have just purchased a DCC Specialties RRampmeter and have tested track voltage at various places around my layout.  Before I tell you the results, here is some information on how my layout is wired.  The layout was started back in 1988 and was wired for DC Block control with two cabs.  All the wire I used when I wired it up is 22 gauge solid wire. So, the wire runs from the cab to the control panel, through double pole/double throw/center off toggle switches.  Then the wire runs out to the track.  Unless a plastic joiner was used to isolate a block, all track joints are soldered.  I used a common negative wire; but, there are many negative feeders along the common rail.

 

When I installed my Digitrax Zephyr, I simply removed one of my DC cabs and installed the Zephyr in it's place.

 

I never installed the suggested 12 gauge buss wires for DCC operation, as the system seemed to work just fine.  I do not see the locomotives slow down at any place along the layout.

 

Now to the voltage checks I have just done and what they’ve revealed.  At a location where the shortest length of feeder wire connects to the track, around 1 foot in length, my RRampmeter reads 13.5 volts.    At the farthest distance from the Zephyr, with feeder wires around 15 feet in length, the meter reads 13.4 volts.

I have on many occasions run two locos at a time; however, they are mostly in different blocks.

Am I doing damage to anything with my set-up? 

[richg1998]

Not a problem.  Our club had 14 blocks with four throttles at one time with same wire. We had to run #14 buss because we could have as many as 10 sound locos running at one time.

Our NCE 5 amp system measures around 13.6 VAC. Three different multimeters agree.

Do you have a load, loco, on these sections when measuring voltage?

Rich

[NP2626]

No, no load; or, loco.

[RR_Mel]

You do need a load to check for voltage loss.  It doesn’t need to be a locomotive, you can use an automotive taillight bulb.  A single filament bulb (1073/1156) normally draws close to 2 amps.  

 

You have almost described the way I wired my layout and I don’t and never have had any problems.  I did run both wires to each block using #20 solid bell wire. I got rattled after all the hype of using #14 wire and rewired my mainline. That didn’t make any difference one way or the other with the DCC operation but it dinged my CTC system occupancy detectors.  I pulled out the #14 gauge wire and reconnected the block feeders and everything is happy.

 

I run dual mode, using either DC or the disliked by many MRC DCC Advanced Squared 2 controller and all makes of decoders without any problems.  A voltage drop of less than a half volt shouldn’t give you any problems.  My longest #20 gauge run is about 15 foot and using a pair of E7s and full lighting on 9 passenger cars up my 3½% grade the voltage drop is .63 volts with the 2.1 amp load.  I’ve never experienced a problem.  I use two identical fluke multi-meters, one at the controller and one at the block so the measurement is accurate.

 

Mel

[richg1998]

Yes, a load is necessary to see if there are any defective track joints, feeders with high resistance. No load, you should see the same voltage as your booster output unless a connection completely broke.

Just put a voltmeter on AC at the booster output. Even is you are concerned about the meter accuracy, all you are looking for is a trend. Same or lower voltage. Standard power line trouble shooting.

Use that voltage as a reference and check different spots with a loco or light bulb as a load.

Auto parts sell a tool that looks like and ice pick with a wire and clip on it. Works on a layout for a quick check. I keep one handy. A dimmer light than at the booster, a defective connection and makes a suitable load. Old school auto trouble shooting.

Between the Harbor Freight meters and the ice pack, never have troubleshooting issues.

Rich

[rrinker]

A .1 volt difference is a non-issue. More importantly than the voltage though is the ability of the wrign to handle the current. This can be tested with an ordinary quarter (other countries, substitute a similar size coin). You just lay the quarter across the rails at various lcoations and verify that the circuit breaker trips and cuts off the power. If it doesn't, that means there is enough resistence in the wiring to provide a load of less than the Zephyr is cable of supplying. It's less of a deal than if you were using a 5 amp or higher booster, but still, 2+ amps at 13.5V is 27+ watts which can be a significant amount of heat. The only way to fix that is to beef up the power feeds to the track, more feeder and/or heavier wire for the feeders.

                        --Randy

[NP2626]

rrinker

A .1 volt difference is a non-issue. More importantly than the voltage though is the ability of the wrign to handle the current. 

                        --Randy

 

Randy,

I'm uncertian about what you meant by the word "wrign" in the above post, can you clarify? 

[richhotrain]

I'm sure he meant to type 'wiring'.

Rich

[rrinker]

Yes, wiring.

        --Randy

[NP2626]

That makes sense.  However, Randy being the electronics Wize Kid that he is, I thought maybe "wrign" was some type of test equipment; or, other!

[richhotrain]

The most recent issue of Model Railroader magazine has an interesting article on this very subject of wiring, resistance, etc.

I wonder about two issues raised in this thread by the OP.

One, is 22 gauge wire too small for a 5 amp booster, raising concerns about heat and resistance?

Two, are 15 foot feeder wires too long, raising concerns about voltage drop?

Rich

[carl425]

NP2626

However, Randy being the electronics Wize Kid that he is, I thought maybe "wrign" was some type of test equipment; or, other!

As a follower of Randy's posts, I've noticed that his talent for electronics is not matched by his typing skills.

[peahrens]

You might find the section "the great feeder experiment" in the attached webpage interesting:

http://www.wiringfordcc.com/track_2.htm#a26

If I interpret it correctly, with a 5A system, one would not want to use 22AWG feeders of 6' or more, which fail (3' passed and 4' is "marginal") A. Gartner's quarter test.

With a lower amp system, longer small wire could be used.  And from the standpoint of heat generation (which could melt ties) and long small wire runs, some of the voltage drop / heat would be lost in the wire vs. the rails, so not sure 2A could melt the ties.  I think the bottom line is that if a quarter test works consistently everywhere on your layout, the system will trip when it needs to, which is what matters as long as you're getting the voltage the locos need around the layout.  So IMHO definitely don't skip the quarter test.   

[NP2626]

Every place that I placed a quarter, the Zephyr showed a short (the LED loco address window flashes).   

Under load (12 volt auto tail light) I get 11.6 volts and .04 amps.  With a locomotive the voltage only dropped to 12.5 volts. 

I left the 12 volt light on the track for 5 minutes and could detect no heat build up.

[selector]

I agree that feeders, as that term is commonly understood in this hobby, should not be longer than 3' maximum.  I had several 22 gauge feeders on my last layout, the longest being just short of 3'.  I never had problems, and I had the odd short where my 5 amp supply could have caused a hot wire, not to mention a fried decoder had it not been for the detection and cut-off circuit in the DB150 base station.

Looking at readily available resistance tables on the www, 22 gauge is suitable only for short runs AS FEEDERS........not as a bus taking 5 or more amps for reaches of ten or more feet.

[jrbernier]

Interesting,

  We ran into a problem on our club layout.  The layout fills a 27' by 27' room and has 2 levels.  There are three 8 amp Digitrax boosters, feeding 16 PM42's with 46 power districts.  The main bus wires are either #10 or #12 wire, with #18 feeders to the track.  3M Scotchlok or PosiTap connecters attach the feeders to the bus. Everything has worked great for 10 years.

  A few weeks ago, we started getting some of the PM42's to trip - We thought that it was due to overloads or shorts/derailments, but they kept happening more often.  And in only certain areas.  A check of one of the suspect PM42's showed that it was set to the configuration we put in it 10 years ago(4.5 amps/quick trip).  The problem kept showing up, so one of the guys built a harness with alligator clips and a 1156 lamp.  We check the Rrampmeter voltage under no load(13.8v) for every section of track.  We then added the 2.2 amp load from the 1156 lamp, and tested again - 12.2 volts or less on several of the power districts!  Cross referencing the power district table, it appeared that the problems were limited to three of the PM42's.  We checked each of them, and the configuration was correct.  Swapping them out showed that the problem followed the PM42's.  At least our wiring was 'Bullet Proof'!

  So it appears that those PM42's will be visiting Digitrax in Florida for a vacation($20/unit, plus shipping)

Jim

 

[carl425]

NP2626

Every place that I placed a quarter, the Zephyr showed a short (the LED loco address window flashes). Under load (12 volt auto tail light) I get 11.6 volts and .04 amps. With a locomotive the voltage only dropped to 12.5 volts.

The quarter test is good news, but your measurement results are strange.  Based on a couple of different voltage drop calculators I tried, your drop if far more than they predict.  I'm curious to see what kind of responses you get with possible explanations.

[rrinker]

 Depends what tail light bulb was being used. An 1156 is about 2.2 amps, which is pushing close to the Zephyr's limit of 2.5 amps. Loading plus the drop in the wiring could explain this. Remember to double the distance for the wiring loss calculation - if it's 15 feet from the source to the test point, that's 30 feet of wire for calculating drop based on resistence.

 I never claimed to be a typist. It's actually BETTER since I got a mechanical 'click' keyboard on my desktop. But then those laptop keyboards...

                      --Randy

[NP2626]

carl425

 

 

NP2626

Every place that I placed a quarter, the Zephyr showed a short (the LED loco address window flashes). Under load (12 volt auto tail light) I get 11.6 volts and .04 amps. With a locomotive the voltage only dropped to 12.5 volts.

 

 

The quarter test is good news, but your measurement results are strange.  Based on a couple of different voltage drop calculators I tried, your drop if far more than they predict.  I'm curious to see what kind of responses you get with possible explanations.

 

 

I need to do a bit more thorough job of checking volts and amps and document my findings.  I may have screwed up something in some of my findings.

 

I don’t know what the specs. on the 12 volt bulb and after a looking at it I realized it is an interior light from an R.V.

 

However, the quarter trick was right and the loco amps and volts were what I found.

[peahrens]

This all is very interesting!  While researching the literature, including the websites, would perhaps lead a person to the right setup for his/her situation, it seems a "decision tree" would be a teriffic tool to assist people to the right approach for their situation.  No, I'm not volunteering.

Another, easier, approach is to just install conservative stuff, if at that stage.

[NP2626]

It appears to me that some of you are assuming that I'm running into problems, I’m not. When I first installed my Digitrax Zephyr, I assumed I would need to install a #12-#14 buss and cut my #22 block leads short and connect them to the buss.  Once I installed the Zephyr and hooked it up to my block control system, I found that the system functioned well without installing a buss.  This was 14 years ago and the system has worked flawlessly since.

 

I don’t plan to run a buss unless I can determine I have a problem.  The reason I started this thread is because in describing my layout’s DCC system to a friend here, he thought it was interesting and suggested I start a thread on it.

 

I understand what the manufacturers recommend and what the general consensus is, here on the MRForum about how a DCC system should be installed. 

 

I tested the amperage draw of the 12 volt light I have used as a load and found it draws just under one amp.

 

With this bulb attached to the track and my RRampmeter, I checked at six locations around my layout to see what my readings were.  My voltage ranged 12.3 to 11.4 volts and the amperage was between .92 to .96 amps.

 

So, I ran a sound equipped steam locomotive with as long a train as I generally run on my layout.   On the flat; or, down grade the voltage ran between 13.3 to 13.4 and the amps from .04 to .05.  On my 2.5% upgrade on a spiral curve, the loco volts were around 13.2 and the amps were the same as the level ground.

If you feel I should conduct more/different test, I would be willing to do so, if it isn’t too expnsive.

[carl425]

NP2626

If you feel I should conduct more/different test, I would be willing to do so, if it isn’t too expnsive.

You could test it with 12ga wires. just kidding

The wire guys say you don't want more than a 4% drop in voltage.  The DCC guys say you don't want more than a 1 volt drop.  If you only run a train or two with only single engines, I think you are far from drawing enough current to have a problem.  I think there is a reasonable possibility that your Zephr won't supply enough current to create a problem. I'd have been worried if the quarter test failed, but since it didn't it sounds to me like you have reached the "if it works, don't fix it" point.

Just keep in mind that when you get that 8-amp system and have 5-6 operators running 4 unit trains with 50 cars up that 2.5% grade, you'll need to upgrade your wiring.

[NP2626]

I doubt that I will ever need to add even a 5 amp booster to my layout.  However, I understand the "Never say Never" premise and if at some point I decide I need to do a “Tim Allen, MORE POWER!!!” on my layout, a re-wire with a #12 buss would be forgone conclusion!