Voltage issue

[philnymark]

I have recently revived my N scale track after a few years of non use. Most of my engines are Kato and I also have two Atlas SD units and a beautiful Broadway Limited E6 A/B unit. I have had trouble with some of these engines. They all worked okay a few years ago but many of them now have performance problems. Of my Kato units, some are running fine, others not so much, and some don't run at all. My Atlases don't run but have always been problematic and are low on my priority list. It's my Broadway Limited that I am more concerned with. It is a great engine and was running pretty well a few years ago. I think they are a bit finicky. This A/B unit would sometimes run and sometimes not. The A unit was getting very hot so I sent them in to BLI. Their service is great and I got them back. After trying to adjust some CV's, I was having no luck with them and the B unit quit running. I took the shell off of the A unit and the wires had melted a bit even though it ran. I sent both back to BLI and they think I may have a voltage issue, like a surge maybe (?). I use DCC and run my track with an NCE Powercab. However, I power my Kato turnouts with a transformer. Could there be some kind of carry over from the transformer that affects the DCC? I also have a number of reverse units and am wondering if they could be causing short circuits. I'm pretty good with all aspects of the hobby but not too smart when it comes to electrical stuff even though I manage to get things done like power reverse units and the like. I have some of my Katos running really well though, so the problem is a bit confusing. Any suggestions or past experience with such a problem would be very appreciated. Thanks!

 

[Smudge617]

Sound like some of your loco's need a good clean and service if they haven't been run for a few years, also all your track and track connections need checking for continuity.

As for a carryover between your turnout power supply, I think you would have known about it when something went bang, a 12v power supply is usually DC, DCC systems are AC, and the two don't play well together.

I not sure what to suggest regarding this alleged surge, I doubt it's your Powercab but it's possible, all I can suggest is to use a multimeter on your track and see if you get any spikes while dialling the power up and down, it could also be that due to the poor running of your locos they're drawing excessive power from the track which would explain the melting cables, but I'm guessing here.

 

[philnymark]

Sound like some of your loco's need a good clean and service if they haven't been run for a few years, also all your track and track connections need checking for continuity.

As for a carryover between your turnout power supply, I think you would have known about it when something went bang, a 12v power supply is usually DC, DCC systems are AC, and the two don't play well together.

I not sure what to suggest regarding this alleged surge, I doubt it's your Powercab but it's possible, all I can suggest is to use a multimeter on your track and see if you get any spikes while dialling the power up and down, it could also be that due to the poor running of your locos they're drawing excessive power from the track which would explain the melting cables, but I'm guessing here.

Hi Smudge617, Thanks for this advice. I am thinking that I need to get under the layout, re-do the electrical and go over each engine as you suggested. Also, getting a multi meter and looking for spikes in power is something I should do too. Letting everything sit for so long wasn't a good idea. I told my wife that I'm not going to let these things sit for long anymore. I packed up my G scale engines a month or so ago but with better weather, I need to get those out too. Much appreciated Smudge!

 

[Iron Horseman]

The A unit was getting very hot so I sent them in to BLI. Their service is great and I got them back. After trying to adjust some CV's, I was having no luck with them and the B unit quit running. I took the shell off of the A unit and the wires had melted a bit even though it ran. I sent both back to BLI and they think I may have a voltage issue, like a surge maybe (?). I use DCC and run my track with an NCE Powercab.

The Powercab is one system I do not own, but I understand it does not have a voltage control. It is supposed to be 13.5V.

I do not have any good advice, only some random thoughts.

I do not think that "surges" or "spikes" would overheat a unit unless the surges are more the norm than the exception. Can you tell if the heat is from the motor or from the circuit board? I would guess the later as only the A is getting hot.

In addition to the good general cleaning as suggested above, I would look to see if one of the auto-reverse units are not flipping fast enough?

If it is a motor one would think it has an internal short in the windings and drawing more power than it should. But you said they had already been back (twice) to BLI and that is something they surely have checked.

So I guess maybe I do have some advice. Get a good DCC voltage and current meter and start diagnosing one thing at a time from the power-out to the track, to the loco. I have an original RRampMeter. Unfortunately, it does not have a retention circuit to detect "spike" voltages. With it one would have to visually detect the spike on the gauge.

 

[philnymark]

The Powercab is one system I do not own, but I understand it does not have a voltage control. It is supposed to be 13.5V.

I do not have any good advice, only some random thoughts.

I do not think that "surges" or "spikes" would overheat a unit unless the surges are more the norm than the exception. Can you tell if the heat is from the motor or from the circuit board? I would guess the later as only the A is getting hot.

In addition to the good general cleaning as suggested above, I would look to see if one of the auto-reverse units are not flipping fast enough?

If it is a motor one would think it has an internal short in the windings and drawing more power than it should. But you said they had already been back (twice) to BLI and that is something they surely have checked.

So I guess maybe I do have some advice. Get a good DCC voltage and current meter and start diagnosing one thing at a time from the power-out to the track, to the loco. I have an original RRampMeter. Unfortunately, it does not have a retention circuit to detect "spike" voltages. With it one would have to visually detect the spike on the gauge.

Thank you for this advice. I think getting a good DCC voltage and current reader is something I need to do in addition to making sure my decoders in each engine are contacting properly. The heat does seem to be coming from the circuit board. Track and wheels are all very clean but my wiring needs to be gone over thoroughly. I really should never have left things sitting for so long. Thanks for all the knowledge and ideas!

 

[ctclibby]

Just my $0.02 worth. Forget about the stand alone DCC current/voltage reader. Get an oscilloscope or DSO as they are more ...um... usable. I have 2 Hantek's that I use all of the time. Ya, also have an analog 200Mhz scope too. The freq's used on MRR's are pretty low so you don't really need a 100MHz DSO. The Hantek's have grown in price, but there are a ton of others out there for under $100, find one that fits your fancy.

One of the things that fall out using a DSO is that you can see the pulse train of the DCC signal. It is NOT an AC signal, it is a modulated signal using 2 different pulse widths for binary 0 and 1 ( yes/no, up/down, left/right .. whatever )

Have fun!

 

[philnymark]

Just my $0.02 worth. Forget about the stand alone DCC current/voltage reader. Get an oscilloscope or DSO as they are more ...um... usable. I have 2 Hantek's that I use all of the time. Ya, also have an analog 200Mhz scope too. The freq's used on MRR's are pretty low so you don't really need a 100MHz DSO. The Hantek's have grown in price, but there are a ton of others out there for under $100, find one that fits your fancy.

One of the things that fall out using a DSO is that you can see the pulse train of the DCC signal. It is NOT an AC signal, it is a modulated signal using 2 different pulse widths for binary 0 and 1 ( yes/no, up/down, left/right .. whatever )

Have fun!

Thanks for this advice. Model RRing has helped me increase my electrical knowledge and tools like this can only help that much more. Much appreciated!

 

[Selector]

In my opinion, a $10 digital meter is all you need. You're not looking for voltage spikes so much as amperage draw. It's the voltage's capacity to carry amperage that makes electrical power dangerous. It isn't voltage that will kill you...it's amperage. Similarly, voltage won't damage electrical components so much as the heat generated from over-amping the components.

A cheap digital meter, which is all I have ever used (same one purchased 20 years ago), will give you a close approximation of the actual voltage to the rails IF you use the correct setting, that being the one for 20 volts AC. Remember, DCC is an alternating current. So, set your El Cheapo digital meter to the correct voltage range (to keep the reading sensitive to the true voltage), touch the leads to the rails, and you should get a reading between 14 and 16 volts, typically. If it shows that, that's all you need to know...your rails are getting the typical DCC voltage. Your heat problem is due to resistance, and that is endemic to wiring or faulty/worn/dirty components typically. The motor, for example, has to drive a sticky gear tower, mebbe, and get the whole train moving, not to mention the rest of the locomotive's bulk. It will draw amperage to make that happen, and if the amperage is high enough to generate heat, guess what.....your locomotive will soon be hot to the touch.

 

[ctclibby]

Crandell ( and others ). DCC is not AC or Alternating Current or a form of Alternating Current. Please see the wiki:

DCC Wiki

Without reading the wiki, basically there are pulse width modulation waveforms on both rails, identical except one of the waveforms is inverted from the other. In other words, 1 wave is +10VDC while the other is -10VDC at the same instant, both referenced to 0VDC. Modulation for DCC uses pulse widths ( 58uS and 100uS ) if I remember correctly; where the 58uS is a binary 1, 100uS is binary 0.

Unlike analog DC, DCC signal polarity does not contain direction. and the DCC voltage does not contain speed. Both of those are handled via the engine's ( or other device ) decoder.

Hobby folks that are not familiar with scopes, just setup one probe, set it to AC ( capacitively coupled ) and assume that the signal is AC. To really view a DCC signal ( DON'T do this at home unless you really know what you are doing ), float the scope from the AC Mains then connect each probe to a rail, with probe GND connected at the Command Station GND. You now are looking at DCC. Since you know that the 2 signals are the same but one inverted from the other, unhook all of the scope stuff and go back to a normal AC Mains with Ground to protect yourself and use the DC setting on the scope. On second thought, don't do any of the scope floating above, just trust me!

As far as using a cheap meter - do what ever floats your boat as long as you realize that the reading is not correct. Good for generalities, but for trying to pin down power leaks ( heat ) it may or may not work as expected.

To get a closer reading, put a 1 Ohm 1 Watt 1% resistor in series with the N/HO load ( engine ) and connect the 2 scope probes; one on each end of the resistor and difference the values. The voltage reading you see is darn close to the current flowing in the circuit. Yes, it could be somewhat wavy. Depending on your scale, that 1 Watt resistor may need to be upwards of 5 Watts. This is better than that cheapo meter as you can see spikes on the scope somewhere in the middle of the wave, whereas the meter will probably miss them altogether if indeed there is some sort of current issue abet only for a few micro seconds each pulse.

Of course, this all depends on the clarity of your DCC bus. If it is ringing big time ( over/undershoot on rising/falling edge ) you have other problems, start there first to get stuff settled down - google snubbers for DCC or the like.

Have fun!

 

[philnymark]

In my opinion, a $10 digital meter is all you need. You're not looking for voltage spikes so much as amperage draw. It's the voltage's capacity to carry amperage that makes electrical power dangerous. It isn't voltage that will kill you...it's amperage. Similarly, voltage won't damage electrical components so much as the heat generated from over-amping the components.

A cheap digital meter, which is all I have ever used (same one purchased 20 years ago), will give you a close approximation of the actual voltage to the rails IF you use the correct setting, that being the one for 20 volts AC. Remember, DCC is an alternating current. So, set your El Cheapo digital meter to the correct voltage range (to keep the reading sensitive to the true voltage), touch the leads to the rails, and you should get a reading between 14 and 16 volts, typically. If it shows that, that's all you need to know...your rails are getting the typical DCC voltage. Your heat problem is due to resistance, and that is endemic to wiring or faulty/worn/dirty components typically. The motor, for example, has to drive a sticky gear tower, mebbe, and get the whole train moving, not to mention the rest of the locomotive's bulk. It will draw amperage to make that happen, and if the amperage is high enough to generate heat, guess what.....your locomotive will soon be hot to the touch.

This is great information. I'm going to be working on my track over the next couple of days and now have a better idea of what to work on. Thanks so much Selector!

 

[Smudge617]

Without reading the wiki, basically there are pulse width modulation waveforms on both rails, identical except one of the waveforms is inverted from the other. In other words, 1 wave is +10VDC while the other is -10VDC at the same instant, both referenced to 0VDC. Modulation for DCC uses pulse widths ( 58uS and 100uS ) if I remember correctly; where the 58uS is a binary 1, 100uS is binary 0.

Unlike analog DC, DCC signal polarity does not contain direction. and the DCC voltage does not contain speed. Both of those are handled via the engine's ( or other device ) decoder.

For us more electrically challenged shall we say, although describing DCC as an AC signal is not correct, it is an excellent explanation that we, more technically challenged of us, can get our heads around.

 

[Selector]

Crandell ( and others ). DCC is not AC or Alternating Current or a form of Alternating Current. Please see the wiki:

DCC Wiki

Without reading the wiki, basically there are pulse width modulation waveforms on both rails, identical except one of the waveforms is inverted from the other. In other words, 1 wave is +10VDC while the other is -10VDC at the same instant, both referenced to 0VDC. Modulation for DCC uses pulse widths ( 58uS and 100uS ) if I remember correctly; where the 58uS is a binary 1, 100uS is binary 0.

Unlike analog DC, DCC signal polarity does not contain direction. and the DCC voltage does not contain speed. Both of those are handled via the engine's ( or other device ) decoder.

Hobby folks that are not familiar with scopes, just setup one probe, set it to AC ( capacitively coupled ) and assume that the signal is AC. To really view a DCC signal ( DON'T do this at home unless you really know what you are doing ), float the scope from the AC Mains then connect each probe to a rail, with probe GND connected at the Command Station GND. You now are looking at DCC. Since you know that the 2 signals are the same but one inverted from the other, unhook all of the scope stuff and go back to a normal AC Mains with Ground to protect yourself and use the DC setting on the scope. On second thought, don't do any of the scope floating above, just trust me!

As far as using a cheap meter - do what ever floats your boat as long as you realize that the reading is not correct. Good for generalities, but for trying to pin down power leaks ( heat ) it may or may not work as expected.

To get a closer reading, put a 1 Ohm 1 Watt 1% resistor in series with the N/HO load ( engine ) and connect the 2 scope probes; one on each end of the resistor and difference the values. The voltage reading you see is darn close to the current flowing in the circuit. Yes, it could be somewhat wavy. Depending on your scale, that 1 Watt resistor may need to be upwards of 5 Watts. This is better than that cheapo meter as you can see spikes on the scope somewhere in the middle of the wave, whereas the meter will probably miss them altogether if indeed there is some sort of current issue abet only for a few micro seconds each pulse.

Of course, this all depends on the clarity of your DCC bus. If it is ringing big time ( over/undershoot on rising/falling edge ) you have other problems, start there first to get stuff settled down - google snubbers for DCC or the like.

Have fun!

From the NMRA chapter on DCC:

A: Technique for Encoding Bits 10 The NMRA baseline digital command control signal consists of a stream of transitions between two equal voltage levels that have opposite polarity.1 Alternate transitions separate one bit from the next. The remaining transitions divide each bit into a first part and a last part. Digital Command Stations shall encode bits within this digital command control stream of transitions by varying the duration of the parts of the bits, or frequency of the transitions. In a “1” bit, the first and last part of a bit shall have the same nominal duration, and that duration shall be t12 15 , giving the bit a nominal total duration of (2x t1). Digital Command Station components shall transmit “1” bits with the first and last parts each having a duration within the t1 range....

...

2.3 Power Transmission and Voltage Limits for Transmitting Power through the RailsThe baseline method for providing the power to operate locomotives and accessories, which shall be supported by all Digital Command Stations and Digital Decoders, is by full-wave rectification of the bipolar NMRA digital signal within the Digital Decoder6. In order to maintain power to the 75 Digital Decoders, gaps in bit transmission are only allowed at specified times (see S-9.2, Section C).

What is described in bold is AC current.


SaveShare
Reply Quote

 

[ctclibby]

For us more electrically challenged shall we say, although describing DCC as an AC signal is not correct, it is an excellent explanation that we, more technically challenged of us, can get our heads around.

Smudge: Maybe you are trying to make more out of it than it is? Just think of it as a signal with encodings that go to all DCC devices.

DCC signal - you can call it what ever strikes your fancy: AC, Pulse Modulated DC, Apples, Oranges or Grapes. No matter what you call it, it is still just a DCC signal.

Later

 

[ctclibby]

From the NMRA chapter on DCC:

A: Technique for Encoding Bits 10 The NMRA baseline digital command control signal consists of a stream of transitions between two equal voltage levels that have opposite polarity.1 Alternate transitions separate one bit from the next. The remaining transitions divide each bit into a first part and a last part. Digital Command Stations shall encode bits within this digital command control stream of transitions by varying the duration of the parts of the bits, or frequency of the transitions. In a “1” bit, the first and last part of a bit shall have the same nominal duration, and that duration shall be t12 15 , giving the bit a nominal total duration of (2x t1). Digital Command Station components shall transmit “1” bits with the first and last parts each having a duration within the t1 range....

...

2.3 Power Transmission and Voltage Limits for Transmitting Power through the RailsThe baseline method for providing the power to operate locomotives and accessories, which shall be supported by all Digital Command Stations and Digital Decoders, is by full-wave rectification of the bipolar NMRA digital signal within the Digital Decoder6. In order to maintain power to the 75 Digital Decoders, gaps in bit transmission are only allowed at specified times (see S-9.2, Section C).

What is described in bold is AC current.


SaveShare
Reply Quote

Exactly - 'two equal voltage levels with opposite polarity' - as described in my post above. Note that two equal but opposite voltage levels can NOT exist in an AC waveform - the net result would be 0 ( zero ).

The 'full-wave rectification' is a bad use of words on the NMRA's part. All that rectification does is to set the correct polarity for the internal decoder mechanism. In other words, it don't matter what way you put the engine on the track.

Later

 

[Selector]

Except that the equal and opposite voltages are not concurrent. They alternate, as per the diagram. If they didn't alternate, then the voltage would be zero and they'd all cancel out...meaning no amperage to drive the mechanism. The zero stretching is what gives the signal for forward or reverse motion, not the voltage.