Track voltage too low

[Anonymous]

I just measured the output of my Digitrax DCS50 Zypher using a voltmeter set on AC and the voltage was 9.5.  This is with the unit completely disconnected from the track.  Connected to the track shows approximately the same reading.  The transformer which came with the set shows voltage at 17 volts with no load.  Is it safe to say there is something wrong with this unit?  The owner's manual says the output should be approx 12.8 volts.  I understand that an AC voltmeter is not the best way to measure, but a reading of 9.4 volts is much lower than what I see others are reading with their systems.  The trains operates OK, but the track has to be spotlessly clean or I have problems.  Perhaps this is due to the low voltage.

[selector]

I have no information on the Zephyr, but most DCC units have settings for track voltage commensurate with the demands of the various scales.  Your setting may be for N-scale, which you can verify by reading the instructions and looking for a switch and indicator.

Sorry that I can offer nothing more.  Hopefully, it will be something as simple as setting the voltage for your scale.

[Soo Line fan]

I am still using DC so I am hardly an expert but a couple things come to mind. Is your DMM a “true RMS” meter? I have 3 DMMs and only my Fluke 87 is an RMS meter. If not you may be getting an inaccurate reading.

For example: An automotive Hall effect switch measured with a DMM will read ½ of system voltage due to the 50% duty cycle. The meter averages the voltage signal. That is why many meters have a duty cycle setting on them; using voltage on a Hall effect is misleading.

One work around worth trying is to remove the shell and check the voltage after the decoder converts it. Then a standard DMM will be able to measure it. I have never heard a good explanation of exactly what a DCC signal is. A lab scope trace would be nice to see.

 

Jim

[Stevert]

I am still using DC so I am hardly an expert but a couple things come to mind. Is your DMM a “true RMS” meter? I have 3 DMMs and only my Fluke 87 is an RMS meter. If not you may be getting an inaccurate reading.

Jim

  The DCC track voltage is square wave, so even a true RMS meter will not give an accurate reading.  That said, the OP's voltage does seem low even taking into account it was measured with a "normal" ac voltmeter.  If there isn't an N/HO setting that can be changed (or possibly an OPSW?), I'd contact the Digitrax folks.

HTH,
Steve

[rrinker]

  The Zephyr has no voltage adjustment, it is set to about 12.5 volts. I have an el-cheapo meter ($5 special) that reads about 11 volts, and an very expensive older Fluke benchtop true RMS unit (that hasn't been recalibrated in over 10 years - could be an issue here) that shows 13.8 volts.
 One thing that can throw off the reading is use of address 0. Since running a DC loco is accomplished by stretching the positive or negative side of the square wave, this willt hrow off the meter reading, especially in a non-RMS meter. Select address 00 and make sure the speed control is set to stop and check again.


                                       --Randy

[Isambard]

 I have never heard a good explanation of exactly what a DCC signal is. A lab scope trace would be nice to see.

 

Jim

As noted in other posts, the DCC signal, which acts both to provide power and commands, is essentially a series of square waves. The Digitrax Big Book of DCC describes the DCC signal as a bipolar square wave with the amplitude of the full track voltage. A binary "one" half cycle is between 52 and 64 microseconds duration and a binary "zero" half cycle is at least 90 microsecond in duration . The Digitrax Big Book (currently under revision) is a good source of information, including scope traces. MR's "DCC Made Easy" by Lionel Strang is also a useful reference.

[Stevert]

I have never heard a good explanation of exactly what a DCC signal is. A lab scope trace would be nice to see.

 

Jim

  Sorry, I missed this the first time around.  Isambard gave you the good explanation, and the scope traces come courtesy of Allan Gartner's excellent DCC Web site.  Note that this series of traces are in illustration of a specific DCC wiring problem and fix, so while the first image shows a "normal" trace, the others show various levels of distortion:    http://www.wiringfordcc.com/dcc_waveforms.htm

Steve

[jeffrey-wimberly]

This thread made me curious as to what voltage my Bachmann EZ Command system is putting out, so I checked it. It's 16.5 no load and 16.2 under load load. Not bad.

[TomDiehl]

I have never heard a good explanation of exactly what a DCC signal is. A lab scope trace would be nice to see.

 

Jim

  Sorry, I missed this the first time around.  Isambard gave you the good explanation, and the scope traces come courtesy of Allan Gartner's excellent DCC Web site.  Note that this series of traces are in illustration of a specific DCC wiring problem and fix, so while the first image shows a "normal" trace, the others show various levels of distortion:    http://www.wiringfordcc.com/dcc_waveforms.htm

Steve

Looking at the waveform in the link, the signal is a pulse train, almost impossible to read accurately with anything but an oscilloscope. Note that the positive and negative portions of the wave are not equal in duration (width on the scope screen). A true square wave can be measured somewhat accurately with a voltmeter, as long as you know you're dealing with a square wave and not a sine wave. The variation you see on the voltmeter can simply be the throttle setting depending on what the setting does to each pulse.

[Vail and Southwestern RR]

I have never heard a good explanation of exactly what a DCC signal is. A lab scope trace would be nice to see.

 

Jim

  Sorry, I missed this the first time around.  Isambard gave you the good explanation, and the scope traces come courtesy of Allan Gartner's excellent DCC Web site.  Note that this series of traces are in illustration of a specific DCC wiring problem and fix, so while the first image shows a "normal" trace, the others show various levels of distortion:    http://www.wiringfordcc.com/dcc_waveforms.htm

Steve

Looking at the waveform in the link, the signal is a pulse train, almost impossible to read accurately with anything but an oscilloscope. Note that the positive and negative portions of the wave are not equal in duration (width on the scope screen). A true square wave can be measured somewhat accurately with a voltmeter, as long as you know you're dealing with a square wave and not a sine wave. The variation you see on the voltmeter can simply be the throttle setting depending on what the setting does to each pulse.

In theory the high and low side of each "wave" are of equal duration, but the zeros are long, and the ones are short, so it is only a square wave during a stream of ones or zeros.  If you are using address zero to run a DC loco the high or low side of the zero is stretched to change the average voltage from zero to a positive or negative voltage to drive the DC loco.  I think from reading the spec that it would be allowable to have an average of other than zero, but then channel zero would not be available.

 

[Soo Line fan]

Gentlemen: Thanks for the explanation. The link and  traces certainly made it clearer.

 

Jim