Jump to content

A minor pitfall - speed control


NXCALE

Recommended Posts

Takahama Trainwatcher

I have a spreadsheet (for setting DCC decoders to give appropriate maximum train speeds) that needs further testing before I share it - it suggests the CV values to set top speeds. I appreciate having the speeds realistic. The original spreadsheet also provided for setting the acceleration and braking values to realistic (scaled) numbers. I discovered that I could scrap this part of the spreadsheet, because even at the slowest acceleration and deceleration settings, the train acceleration and braking rates were too high.

So, with DCC we can set the speeds to be realistic, but I would posit that we cannot quite set the rates of change of speed to be realistic.

Link to comment

Many folks have fun running their trains at the proper speeds (sounds like you don't, but that's your pleasure). We use to have a digital speedometer on the club layout and it was great for public display as well as keeping to speed, but unfortunately it broke. Now it's a simple task to time a lap to check our running speed. We have a speed/lap time/model list next to the throttle. After a while you can get quite good judging by eye! N scale trains are usually run by folks at 2x or greater than the prototype trains... Really looks great when the shinkansens, express and freight trains are all running a their proper speeds at the same time.

 

Jeff

Link to comment

 

I discovered that I could scrap this part of the spreadsheet, because even at the slowest acceleration and deceleration settings, the train acceleration and braking rates were too high.

This depends on the decoder and train in question. Recently at a public event i discovered that a ludmilla diesel locomotive is capable of much greater acceleration that i was aware of and programmed into the decoder of my model. For japanese trains, especially modern commuters and shinkansens, the DCC accelration rate range convers most types and is usable for even the old and sluggish JNR era DMU-s.

Link to comment

For realistic running on smaller loop layouts, it's possible to treat the distance between station A (your only station) and station B (in reality the same old station) as more than a single loop. So this means you start accelarating, speed up after 2-3 laps, then coast, finally decelerate under 2-3 laps. If you set the cab controls to sluggish, it will be a real challenge to hit the stop mark. This works well for even a 2 by 4 loop. The nice part is that the 'coasting' or speed holding phase can be as large as you want to be, so if you like to watch the train run, you can add a longer strech of virtual track between the two stations. (like as many loops as you want)

Link to comment

There are electronic speedometers that really are not too expensive ($40 or so). These are just tow photo sensors mounted in the roadbed that time the train hitting each. They work well but require light,mso in a tunnel you need a couple of LEDs overhead and they won't work with night running.

 

Speed control is more fun when you run with a train cam and drive from the screen,mthen you get the challenge of estimating distances, watching for markers, etc!

 

I've always fancied getting one of the big tomix controllers to play with, but now the new one looks like a good compromise and price!

 

Jeff

Link to comment
Takahama Trainwatcher

I will enjoy watching my volunteer drivers trying to stop the train at the right mark with the controls set to the most sluggish (call me sadistic). For me, it adds to the fun (each to their own as everyone has suggested :) ).

 

 

Now, let’s take an E231 running at 80Km/h with a brake rate of 4Km/h/s.  In order to stop this E231 at the station platform, we will need the following distance (i.e. stopping distance). In order to simplify this, we could consider that there is not brake delay and the grade is zero (I know that I did not show the grade equation but it is irrelevant for this example).

 

So, we have:

 

a= 4km/h/s = 1.11

u=80Km/h = 22.22m/s

 

Going back to the formula:

 

B=(22.22^2/(2 x 1.11)) = 222.4m

I concur. The other thing to note from this scenario is that a train at 80 km/h will take 20 seconds to stop if decelerating at 4 km/h/s. This was how I picked up that that, when my train was set to the correct top speed, the slowest DCC braking setting was stopping the train in less than 20 seconds.

(That being said, can you imagine driving your model train and having to start its braking 20 seconds before arriving at your model station?)

 

This depends on the decoder and train in question. Recently at a public event i discovered that a ludmilla diesel locomotive is capable of much greater acceleration that i was aware of and programmed into the decoder of my model. For japanese trains, especially modern commuters and shinkansens, the DCC accelration rate range convers most types and is usable for even the old and sluggish JNR era DMU-s.

Hopefully I will be able to put this to the test (on multiple trains) in the next month or so. I have Digitrax/Kato decoders only in freight locomotives and EMUs. So far, the slowest acceleration I have been able to achieve has been too fast for a rapid commuter train. But... it is not a hobby of perfect scaling, as we know.

Link to comment

I have a few digitrax and fleischmann decoders and was able to set an acceleration rate of 40 seconds while using a braking time of 20 seconds. It's only usable on loop layouts and on the occasional exhibition one as you have an average of 4 minutes between stations.

Edited by kvp
Link to comment

It's worth giving models a good thrash at high speed occasionally, as it helps clean the commutator on the motor. I've bought a few older items over the years which improved hugely after being left circulating for an hour or so just below the speed at which they'd fly off.

  • Like 1
Link to comment

Hello,

 

This is very interesting.  Could you provide me with a reference for the stopping distance formula?  You mentioned that the original formula factored in the grade.  Does it allow for train mass, or is that accounted in the brake rate?

Link to comment

Brake delay happens due to the use of air pressure to active the brakes of the train. This delay could be 6 or 7 seconds in an EMU or around 30seconds in a loaded freight train. Air pressure has to “travel” from one side to the other side of the train to activate the brakes. If the train is long, this takes longer.

I don't think this is correct. Pressure is kept to a maximum or close to maximum at all times in tanks.  Those delays exampled are way too big.

Edited by katoftw
Link to comment

There are three common pneumatic braking strategies:

-straight air with no separate charge pipe (used for slow freight trains)

-straight air with separate charge pipe (used for many loco hauled passenger trains)

-electropneumatic with two pipes and cross train brake cables (used for fast passenger trains and most emu-s/dmu-s)

 

The single pipe variant is slow to brake and even slower to release as the main pipe is used for charging back whole train. The second is still slow to react, but at lesat can be released at same speed as the braking. Electropneumatic uses the through brake pipe as a failsafe only and pressure is released and reapplied at each car by a solenoid valve controlled from the cab though the train bus cables. Most japanese trains use this system and even many JNR era ones used this. European and american practice lags behind with only the 2 pipe system in wide use on most freights, while for some reason the soviets installed the electric cables on several conventional freight cars to increase maximal train length and make them more resilient against freezing (electric lines are not affected by cold as much as air hoses). In american practice, the radio controlled booster units mid train also act as brake boosters as they receive the brake signals faster through the MU link than through the air pipe. This can cause the pressure drop to emit from all MU-ed locomotives in the consist at the same time, regardless of their position. Some fred-s with bidirectional communication also allow braking from the rear too without a locomotive there.

Link to comment

The air brake system will not run out of air with one single brake application. The air has on where to go, as it is a sealed system. It only vents and wastes air upon multiple brake applications. Which is when the air runs out and needs replacing.

Link to comment

Lucky for the passengers, most trains in the modern world don't use suck simples systems as in those diagrams.  Lots of bypass in the newer systems means air will always be available.  And if not, the secondary system to getting air to the brakes will be there.

Link to comment

Air brake systems can leak and this is one of the reasons why a constant charge pipe is used even on freight trains in the US. This allows the replenishment of the air supply while the main pipe is wented (brakes are applied). Several runaways, (last time in Canada were) caused by the crews turning off the compressors whithout applying enough handbrakes. The brakes will hold for a time (around 48 hours is mandatory), but many times they run out of air faster due to leakage on old and/or badly maintained brake systems.

 

For keeping the cars safe without air, many european countries mandate spring force storage parking brakes. In this case air is used to unlock the brake and loss of tank pressure releases them. In most systems these springs are parallel to the main brake cylinders and an emergency application activates both causing a bigger than service maximum brake application. So by venting the charge pipe it's possible to automatically lock the parking brakes on all cars.

 

Around a 150 years ago this was the main brake on many trains in Hungary and the single pipe had to be charged to release the brakes. This made the air tanks and hand brake crews unnecessery. Then most railways switched to the westinghouse single and double pipe system (Gz/Pz modes) which could apply greater forces than the springs and these solutions became optional or a backup. (nowdays mandatory on double pipe equipped cars including trams)

 

One interesting variant was the electromagnetic brake used on all electric trams and subway cars, where the spring force storage brake was connected to a series electromagnet unlocking on power application and locking on power loss, pretty much like a lift motor. Main service brakes were elecromagnetic track brakes fed from the feedback of the traction motors. Loss of traction power behind a red signal automatically applied the brakes without driver involvement. The last tram that used this system was retired around 2009. (built in 1918, rebuilt in 1956) In case of a breakdown or for shop movements a handheld battery had to be connected to an external socket to unlock the brakes and make the cars movable.

Edited by kvp
  • Like 1
Link to comment
Takahama Trainwatcher

The spread sheet (Excel) for assisting in setting CV05 and CV06 in the DCC decoder for desired maximum (scaled correctly) train speed seems to be working okay, having used it in the last couple of weeks for a few trains. It is available at the link below (I hope):

 

https://drive.google.com/file/d/0Bz1dMRnAuqWLSnY5NkZjU2xzNGs/view?usp=sharing

 

Background:

I manually set an E531 to have the correct top speed of 130 km/h, but found the process too tedious to repeat for another train (an E231 or E233) on another occasion. I took a shortcut and assumed that the decoder and motor behaviour for this second train would be the same as for the E531, and set CV05 to have a correspondingly slightly smaller CV05 value appropriate to give a top speed of 120 km/h.

I thought this was fine until a later occasion when I set up a temporary layout and ran the 2 trains together, noting, to my chagrin, that the E531 of top speed 130 km/h was being outpaced by the slower train! There came the realisation that I would have to set up each train by measuring each train's speeds individually. Therefore, I started on the spread sheet to automate the process as best as I could. 

 

Release notes:

You put in the length of your test track, the scale, and the desired top speed in km/h.

The spread sheet tells you the target time for your train to traverse this length of track.

You set the decoder CV05 value to something and time your train. You can do multiple times if you want, and the spread sheet will average them.

You adjust the CV05 value up or down according to whether your train is too slow or too fast, and again record the time in the spread sheet.

After recording 2 different CV05 values with different times, the spread sheet estimates what CV05 you should use to get the desired speed (by assuming a straight line relationship between CV05 and speed). This enables you to get the correct CV05 value quicker. Additional values entered give a more accurate estimate of the correct CV05 value.

The process can be repeated for CV06 (50% throttle speed). Importantly, you should have a small CV06 value set (not 0) in the decoder before starting to set CV05.

  • Like 1
Link to comment

I went through this exercise also. I bought the scale speed tunnel and after it was said and done everything was thrown out the window. There is no formula that one can use to setup braking/stopping distance and speed setting because of the layout variables. One must rely upon the visual aspect of the train motion.

 

When you try to calculate and create a formula with actual values the data cannot be used. It is the scaled nature of the layout which creates an illusion of distance traveled. A basic layout represents a number of stations/cities less than a Kilometer.   1.2  N scale miles (2 kilometers) is about 36 feet (10 meters). That is almost 3 layout end to end and that still won't represent the true essence of the real world. 

 

Because we are creating an illusion of travel time, distance and speed we can cannot use a standard formula. A realistic presentation of the train's starting and stopping motion needs to be orchestrated with the visual aspect of the layout.

 

When the train is in full motion top speed can be referenced because the eye will register speed not distant traveled.

When the trains stops the eye registers that motion by deceleration time to stop not speed.     

 

Inobu

  . 

Edited by inobu
Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...