Voltage drop

[AndyID]

In my (French, admittedly!) club we make a lot of use of choc blocks, but we always use ferrules on the end of flexi cable. We have one permanent (but transportable once or twice a year) layout that is about 90ft long including the back return loops. 

 

This sort of thing

 

http://www.tme.eu/en/katalog/bootlace-ferrules_100041/

 

That solves the problem. The ferrules provide enough mechanical resistance to keep the screw and the connection in compression. You get the same effect with heavy solid copper wire, but multi-stranded wire deforms too easily under the pressure exerted by the screw. The screw also tends to break some of the strands.

[cliff park]

Be very wary of soldering the multicore before inserting into screw type connectors. They will break off where the solder ends at the slightest flexing. 

[tigerburnie]

I have some bootlace ferrules, easy job to fit, so I will do that.

[AndyID]

I have some bootlace ferrules, easy job to fit, so I will do that.

 

Except that.....Oh, never mind.

[DCB]

 Well worth using ferrules in screw connectors, I have used brass tube cut down with a stanley knife. Even solid conductors benefit as I have seen conductors almost severed by overtightened burred screws which subsequently snap off even on mains cabling.  Soldering the ends of multicore flex where they enter the choc block is another quicker not quite as good ploy.

 

 

 

That's just muddled thinking. Apply ohms law and you will understand what is going on.

 

Ohms law states  V= I R  and also  I = V/R.  

 The devices Crosland describes rely on the Controller setting and the resistance of the circuit including the loco motor to regulate the voltage and speed, V = IR and are greatly affected by track and wiring resistance.   [a]

 

The ones I describe are I = V/R where the controller controls the Voltage so the current changes with resistance and the voltage and thus speed remains pretty much constant.

 

The Diode controller can drop voltage by up to 14 volts in 1.4 volt steps assuming the oft quoted diode drop actually is 0.7 volts virtually irrespective of the current..  The OnTrack varies the voltage from + 12 volts to -12 volts seamlessly with no dead spot, the H&M is a variable transformer, possibly an autotransformer, which delivers 2.5 to 16 volts.   Very little difference in off load to on load voltage on any of them.

 

Going back to the original topic, you will get less trouble with "Voltage drop" if your controller controls the Off Load voltage.

[AndyID]

 

 

Ohms law states  V= I R  and also  I = V/R.  

 The devices Crosland describes rely on the Controller setting and the resistance of the circuit including the loco motor to regulate the voltage and speed, V = IR and are greatly affected by track and wiring resistance.   [a]

 

The ones I describe are I = V/R where the controller controls the Voltage so the current changes with resistance and the voltage and thus speed remains pretty much constant.

 

The Diode controller can drop voltage by up to 14 volts in 1.4 volt steps assuming the oft quoted diode drop actually is 0.7 volts virtually irrespective of the current..  The OnTrack varies the voltage from + 12 volts to -12 volts seamlessly with no dead spot, the H&M is a variable transformer, possibly an autotransformer, which delivers 2.5 to 16 volts.   Very little difference in off load to on load voltage on any of them.

 

Going back to the original topic, you will get less trouble with "Voltage drop" if your controller controls the Off Load voltage.

 

Oh dear!

 

David, I do not mean to be disrespectful, but this is just not correct. As you seem to have your own ideas about how this stuff works I don't think there is any point in trying to explain what's wrong with it. What does worry me is that others might find it very misleading.

 

Andy

[SHMD]

I'm just loving this thread!

 

David has a wealth of practical experience - which I admire and is useful - but his theories on how electronics works should not be, shall we say, heeded/used/applied/remembered!

(Except for personal amusement purposes!)

 

 

Kev.

[raymw]

Making some broad assumptions - if you have a voltage controller, as the loco moves further away, the increased track resistance, bad joints, etc, will reduce the voltage available to the motor, so for the same voltage output from the controller the motor will slow down. If there is such a thing as a constant current controller, then the length of track will make no difference to the speed of the motor, since the track and motor are in effect series resistors. I believe that is what David was saying, but in a different manner.

[AndyID]

Making some broad assumptions - if you have a voltage controller, as the loco moves further away, the increased track resistance, bad joints, etc, will reduce the voltage available to the motor, so for the same voltage output from the controller the motor will slow down. If there is such a thing as a constant current controller, then the length of track will make no difference to the speed of the motor, since the track and motor are in effect series resistors. I believe that is what David was saying, but in a different manner.

 

That's not the impression I got, but you could be right

 

Funnily enough a few years ago I cooked up a design for a controller that supplies a controlled current rather than a controlled voltage to the track. That's fairly easy to do. The not so easy bit is putting the electronics in the locomotive to convert the current into a voltage that is fed to the motor where the current fed to the track represents the speed that the motor is supposed to run at. The electronics in the locomotive is actually a shunt regulator that "spills" unwanted current past the motor. It all became rather complicated and there are problems with heat dissipation so I gave the whole idea up as a bad job.

 

IIRC someone at MERG has actually tested a controlled current source controller. I seem to remember it was extremely difficult to control the speed of the train. That would make sense because our DC motors produce a back EMF that is directly proportional to the speed of the motor. If the motors had no internal resistance they would maintain constant speed for an applied voltage regardless of the load on the motor.

 

It's actually quite simple to compensate for the motor's internal resistance and effectively cancel it out by adding a voltage that is proportional to the current drawn (it really works too). The snag with that approach is there cannot be much resistance between the controller and the locomotive. If there is the controller "thinks" the motor is running faster than it really is.

[kevinlms]

Making some broad assumptions - if you have a voltage controller, as the loco moves further away, the increased track resistance, bad joints, etc, will reduce the voltage available to the motor, so for the same voltage output from the controller the motor will slow down. If there is such a thing as a constant current controller, then the length of track will make no difference to the speed of the motor, since the track and motor are in effect series resistors. I believe that is what David was saying, but in a different manner.

Rather than attempting to produce 'a constant current controller', which would fail on the 'is the train running up hill or down hill' principle anyway (due to motor Back EMF). Put simply, a motor needs less current running down hill, than going back up. Anyone disputing this principle, has never tried riding a push bike!

 

The secret for more consistent running is to reduce the track/wiring resistance, as much as practical (if you like, ride your bike on flatter surfaces). That topic IIRC, is where this thread started!

[AndyID]

 

The secret for more consistent running is to reduce the track/wiring resistance, as much as practical (if you like, ride your bike on flatter surfaces). That topic IIRC, is where this thread started!

 

Agreed, but I think we pretty much beat that to death on the first page

[raymw]

 

Put simply, a motor needs less current running down hill, than going back up

Bit like a steam loco, then - you'd actually have to drive it.

[AndyID]

I wasn't going to do this but I thought, "what the heck, maybe we'll all learn something from it."

 

This an equivalent circuit for a DC motor. It's pretty basic and it ignores the fact that a lot of what is going on in a DC motor is actually AC, but it's probably close enough for now. It is relevant to the discussion because (I think) it helps to explain why voltage drop along the track is important.

 

 

V is the voltage applied to the motor

I is the current drawn by the motor

E is the back EMF (voltage) produce by the motor while it is turning

R is the internal resistance of the motor (this is the resistance of the wire in the motor windings and the resistance of the brushes

W is the work done by the motor. Some of it is useful and makes the model move and some of it is consumed in overcoming friction in the motor itself

VR is the voltage drop within the motor due to the motor's internal resistance.

 

One interesting thing is that the current has to flow though the internal resistance R. Another interesting thing is that E (the back EMF) is ALWAYS proportional to the speed of the motor. If the motor speeds up or slows down for any reason, there is a corresponding change in the EMF.

 

Also, V always has to equal E plus VR.

 

So, the simple solution is to build a motor with no internal resistance at all so that the motor has to rotate at the speed dictated by the voltage on the track. Unfortunately that would require superconducting wire in the motor windings. While that's not impossible it's not very practical at the moment.

 

Anyway, the point of all this waffling is that we usually want to control the speed of our motors more than anything else and the dominant factor that determines the speed is the voltage applied to the motor. The motor will attempt to maintain that  speed by drawing more or less current depending on the circumstances (gradient, drawbar load, turn radius, etc) and that is why it's important to regulate the voltage applied to the track.

 

You can even do things to reduce the VR effect with various forms of feedback, but that's a secondary issue.

[AndyID]

Bit like a steam loco, then - you'd actually have to drive it.

 

Quite right Ray. And IIRC that's what the guys at MERG found.

 

"Real" trains are driven by controlling the power input to the "motor", be it steam, electric etc. But real trains have enormous amounts of momentum/inertia and relatively small amounts of friction. Our models are the other way around. They have huge amounts of friction and relatively little inertia/momentum. As soon as we alter the power input there is an almost immediate change in speed. That is why it's important to regulate the voltage.

[kevinlms]

Quite right Ray. And IIRC that's what the guys at MERG found.

 

"Real" trains are driven by controlling the power input to the "motor", be it steam, electric etc. But real trains have enormous amounts of momentum/inertia and relatively small amounts of friction. Our models are the other way around. They have huge amounts of friction and relatively little inertia/momentum. As soon as we alter the power input there is an almost immediate change in speed. That is why it's important to regulate the voltage.

Agreed and easily proved. With a resistance controller, no matter how carefully you start a model train, once it starts you have to quickly back off power. That is because the running voltage is less than the starting voltage.

[tigerburnie]

Agreed and easily proved. With a resistance controller, no matter how carefully you start a model train, once it starts you have to quickly back off power. That is because the running voltage is less than the starting voltage.

 

Is this where we introduce FLC into the equation? Can we introduce a cascade starter into the circuit/motor?.............................  

[AndyID]

Is this where we introduce FLC into the equation? Can we introduce a cascade starter into the circuit/motor?.............................

 

It would be simpler to ditch the conducting rails altogether and switch to battery power with radio control (and if battery technology gets a bit better, I might just do that.)

[AndyID]

Agreed and easily proved. With a resistance controller, no matter how carefully you start a model train, once it starts you have to quickly back off power. That is because the running voltage is less than the starting voltage.

 

There might be a simple way to compensate for the voltage drop due to the internal resistance of the motor - stick a NTC (negative temperature coefficient thermistor) in series with the motor. That's assuming you can find one with the right characteristics and you are willing to accept that it might get a bit toasty and melt any plastic in its vicinity. I'll probably get around to trying it myself one day but I'm not too optimistic about it working properly. If it did I think we would have heard about it already.