I have been interested in making tight (read low resistance) connections between track sections for some time. I have tried using an ohmmeter with less than accurate or quick results. A few days ago, I rigged up a Kelvin connection for measuring resistance. It will measure as low as 20 micro-ohms. With this I have made the following measurements:
A good connection between track sections is 1.5 milli-ohms. This is with clean pins and clean openings in the two sections. Also, it is necessary to bend the ends of the section where the pins are installed to increase the contact pressure between the pins and the openings in the adjacent section. With the pins of a section pointing away from me, I bend the left rail where the pin is installed to the left and bend the center rail to the right. I also use pliers to form the ends of the open rail back to where they should be if they are bent out of shape.
Resistance of a Lionel O-31 curved section = 0.0035 ohmsResistance of a good joint between sections = 0.0015 ohmsResistance of a Lionel O-31 straight section = 0.0032 ohmsResistance of a Lionel prewar 072 curved section = 0.0048 ohmsResistance of a Lionel prewar 072 straight section = 0.0058Resistance of a prewar American Flyer O-40 curved section = 0.0040Resistance of a prewar American Flyer O-40 straight section = 0.0040Resistance of a Lionel O-42 curved section = 0.0039 ohmsResistance of a Lionel prewar 072 (711) switch = 0.0100
About half of the resistance of the 711 switch is the wire that connects from the rivet holding the fat rail to either of the other two center rails. This wire is apparently steel and not copper. Replacing this steel wire with a copper wire would reduce the resistance of the switch by about 0.003 ohms. With the steel wire, the resistance is 0.0047 ohms.
The four 711 switches that I tested had a significant amount of resistance between the rivet that holds the fat rail and the rivet itself. I soldered the rivet to the fat rail.
Making a Kelvin connection requires a constant current supply and a voltmeter than can read down to 10 micro volts. I am using a current of 100 milliamps and reading the voltage with a meter with a resolution of 10 micro volts. I multiply the voltage reading by 10 to get the resistance.
Is your "section" resistance for the entire circuit, that is, outside rails in parallel and that in series with the center rail, or just one rail?
Bob Nelson
It is just the center rail. Measuring the outer rails is a bit difficult since they are in parallel.
I just wanted to verify that one should increase the resistance that you measured by 50 percent, to 4.8 milliohms per section, to account for the entire circuit. I figure that that is approximately equivalent to 18 AWG copper wire. I have advised using no smaller than 14 AWG for feeders, on the basis of CTT's similar published measurements, that I estimated to be equivalent to 16 AWG. The feeder resistance of course should be lower than the track resistance in order to improve the overall circuit conductance by a significant amount.
Based on my measurements of the joint resistance with and without the rails bent where the pins are, the joint resistance is the problem. It can become quite large if the pins are not clean, the hole in the end of the rail, and there is no side pressure on the pin (rail not bent where the pin is installed). Bob. I know that you solder a jumper between track sections, but with a floor layout I don't have that option. So I am modifying the track sections to get the joint resistance down very low.
Most of the track that I have is old. Some is 072 prewar, some is AF prewar, and some is O-31 post war but may be 50-60 years old. I am modifying and cleaning the track so I can put it together and not have a lot of poor connections. Adding a feeder is an option I am trying to avoid.
servoguy I have been interested in making tight (read low resistance) connections between track sections for some time. I have tried using an ohmmeter with less than accurate or quick results. A few days ago, I rigged up a Kelvin connection for measuring resistance. It will measure as low as 20 micro-ohms. With this I have made the following measurements: A good connection between track sections is 1.5 milli-ohms. This is with clean pins and clean openings in the two sections. Also, it is necessary to bend the ends of the section where the pins are installed to increase the contact pressure between the pins and the openings in the adjacent section. With the pins of a section pointing away from me, I bend the left rail where the pin is installed to the left and bend the center rail to the right. I also use pliers to form the ends of the open rail back to where they should be if they are bent out of shape. Resistance of a Lionel O-31 curved section = 0.0035 ohmsResistance of a good joint between sections = 0.0015 ohmsResistance of a Lionel O-31 straight section = 0.0032 ohmsResistance of a Lionel prewar 072 curved section = 0.0048 ohmsResistance of a Lionel prewar 072 straight section = 0.0058Resistance of a prewar American Flyer O-40 curved section = 0.0040Resistance of a prewar American Flyer O-40 straight section = 0.0040Resistance of a Lionel O-42 curved section = 0.0039 ohmsResistance of a Lionel prewar 072 (711) switch = 0.0100 About half of the resistance of the 711 switch is the wire that connects from the rivet holding the fat rail to either of the other two center rails. This wire is apparently steel and not copper. Replacing this steel wire with a copper wire would reduce the resistance of the switch by about 0.003 ohms. With the steel wire, the resistance is 0.0047 ohms. The four 711 switches that I tested had a significant amount of resistance between the rivet that holds the fat rail and the rivet itself. I soldered the rivet to the fat rail. Making a Kelvin connection requires a constant current supply and a voltmeter than can read down to 10 micro volts. I am using a current of 100 milliamps and reading the voltage with a meter with a resolution of 10 micro volts. I multiply the voltage reading by 10 to get the resistance.
Why worry about it? Those values are so insignificant that it's not worth it.
Personally, if I was worrying about something like this, I'd be looking for a new hobby. I have a Bachelor of Science degree in Electrical Engineering Technology as well.
Jim
RedfireS197 servoguy I have been interested in making tight (read low resistance) connections between track sections for some time. I have tried using an ohmmeter with less than accurate or quick results. A few days ago, I rigged up a Kelvin connection for measuring resistance. It will measure as low as 20 micro-ohms. With this I have made the following measurements: A good connection between track sections is 1.5 milli-ohms. This is with clean pins and clean openings in the two sections. Also, it is necessary to bend the ends of the section where the pins are installed to increase the contact pressure between the pins and the openings in the adjacent section. With the pins of a section pointing away from me, I bend the left rail where the pin is installed to the left and bend the center rail to the right. I also use pliers to form the ends of the open rail back to where they should be if they are bent out of shape. Resistance of a Lionel O-31 curved section = 0.0035 ohmsResistance of a good joint between sections = 0.0015 ohmsResistance of a Lionel O-31 straight section = 0.0032 ohmsResistance of a Lionel prewar 072 curved section = 0.0048 ohmsResistance of a Lionel prewar 072 straight section = 0.0058Resistance of a prewar American Flyer O-40 curved section = 0.0040Resistance of a prewar American Flyer O-40 straight section = 0.0040Resistance of a Lionel O-42 curved section = 0.0039 ohmsResistance of a Lionel prewar 072 (711) switch = 0.0100 About half of the resistance of the 711 switch is the wire that connects from the rivet holding the fat rail to either of the other two center rails. This wire is apparently steel and not copper. Replacing this steel wire with a copper wire would reduce the resistance of the switch by about 0.003 ohms. With the steel wire, the resistance is 0.0047 ohms. The four 711 switches that I tested had a significant amount of resistance between the rivet that holds the fat rail and the rivet itself. I soldered the rivet to the fat rail. Making a Kelvin connection requires a constant current supply and a voltmeter than can read down to 10 micro volts. I am using a current of 100 milliamps and reading the voltage with a meter with a resolution of 10 micro volts. I multiply the voltage reading by 10 to get the resistance. Why worry about it? Those values are so insignificant that it's not worth it. Personally, if I was worrying about something like this, I'd be looking for a new hobby. I have a Bachelor of Science degree in Electrical Engineering Technology as well.
Life's hard, even harder if your stupid John Wayne
http://rtssite.shutterfly.com/
The numbers I published for the joints are for the best joints I have measured. Joint resistance can be a problem and is evidenced by the number of posts where someone is complaining about their train slowing down at some places in the layout. I have seen joints which were not clean with resistances as high as 80 milliohms. If you think this is not a problem, calculate the total resistance for a loop of track with 50 sections and then calculate the voltage drop for loco that pulls 3 amps. You will be amazed how high the voltage drop is.
For the most part you and Nelson are really the only ones worrying about it enough to pull a meter most of us just put drops down every so often and were good with what we get. Its a hobby not a perfection operation where everything must be perfect it's suppose to be where one comes to relax. Not to cause more headaches. Sorry I just can't go there.
No one's asking you to go anywhere. I don't understand why anyone would think it his place to tell others what they may or may not enjoy about the hobby. Can't you understand that one man's "headache" may be another man's relaxation?
I find the discussion about track and track joint resistance quite interesting. I have also experienced voltage drop due to poor electrical connectivity between track sections and running feeder wires to every section is not a viable option every time. Keep up the good work "servoguy" and "lionelsoni"!
I know one guy who soldered all his track together? Who was that - it was someone who posted on here from time to time back a few years ago....
If the train slows down enough to notice - add a wire. That's all you need to do. If you do all these calculations and figure it out to these super tolerances, what will you do when you change the plan a little or add another switch etc - calculate the whole thing over again? Just slap on a feeder where needed and run those trains flat out as fast as you can :)
RIP Chewy - best dog I ever had.
Yes there are voltage drops but simple wiring will take care of that issue without knowing the resistance value of each track section.
If that's the kind of thing you like to do have it. but it sure isn't necessary..
"IT's GOOD TO BE THE KING",by Mel Brooks
Charter Member- Tardis Train Crew (TTC) - Detroit3railers- Detroit Historical society Glancy Modular trains- Charter member BTTS
I would love to chime in here,,,,,,,,,,,,,,,,,but wait, who cares? As long as I am here, well here it is. JMHO.... As long as they run is good enough for me. I tend to agree with JimmyT's thoughts, Find a hobby where you can use all those smarts you have that no one cares about anymore now that you are retired. I did. Toy Trains are what I do.
Move on guys!.
Joined 1-21-2011 TCA 13-68614
Kev, From The North Bluff Above Marseilles IL.
servoguyThe numbers I published for the joints are for the best joints I have measured. Joint resistance can be a problem and is evidenced by the number of posts where someone is complaining about their train slowing down at some places in the layout. I have seen joints which were not clean with resistances as high as 80 milliohms. If you think this is not a problem, calculate the total resistance for a loop of track with 50 sections and then calculate the voltage drop for loco that pulls 3 amps. You will be amazed how high the voltage drop is.
..........Wayne..........
Does the 657 have a Kelvin connection? I just looked at the operator manual here:
http://www.simpson260.com/downloads/simpson_657_user_manual.pdf
It doesn't look like a Kelvin connection. It also doesn't appear that it gets to a milli ohm or 100 micro ohms. My setup will read down to 20 micro ohms. My setup also gives me instantaneous readings and is not dependent on the connection resistances. If the constant current source has resistance in the clip leads that attach it to the rails, the voltage just goes up until the current is at the set value. The voltmeter has a high impedance and so it doesn't matter if there is 100 ohms or more in series with the voltmeter. This is the advantage of a Kelvin connection.
The Simpson 657 Milliohm Adapter uses a four wire measurement technique. The leads to the alligator clips look like lamp cord (two wires) and each wire connects to one jaw of special alligator clips (special because each jaw is electrically isolated from the other jaw.) The wire, aligator clip and contact resistance of the aligator clips introduce no error because the voltage measurement is a separat circuit from the current source circuit.
One jaw of each alligator clip is supplied by the current source, the other jaw is connected to a differential amplifier to read the voltage developed across the resistor (or connector) being measured. It is not exactly a constant current source, nor is the differential amplifier the most stable in the world. The adapter has zero and full scale adjustments to compensate for battery and amplifier drift. After these are adjusted the adapter is set to [READ] and the resistance read on the meter's DC scales.
The Simpson 260 0-10 DC scale has a resolution of 0.2 across the scale. With the adapter, set to 0.1 ohm range, full scale becomes 0.1 ohm on the DC volts 0 to 10 scale and resolution anywhere on the scale is 0.002 ohm (2 milliohm.) Minimum reading is 0 ohm with the same resolution (2 milliohm.) One should be able to measure 0.080 ohms (80 milliohms) with this meter and adapter (reading 8 on the 0-10 scale). Even 0.0015 ohm (1.5 milliohms) is doable, although with 0.002 ohm (2 milliohm) resolution you would say the meter was reading 0.002 ohm (2 milliohm.) Close enough for toy train work in my opinion.
Did anyone take up Servoguy's challenge? I got 4.5 volts for an answer. Anyone else do the math?
To complete the picture of old track and joint resistance, I put two sections of unmodified track together and measured the resistance of the joint of the center rail. It was open circuit. You can't fix that by adding feed wires. This track is pretty bad, and the only reason I am using it is I want some 072 track and I don't want to pay a lot for it. I have about 40 sections of 072 track and AF O-40 track modified, and I have more to go. The joint resistance of the center rail for each of these sections is less than 0.003 ohms.
Our community is FREE to join. To participate you must either login or register for an account.
Get the Classic Toy Trains newsletter delivered to your inbox twice a month