Overall Objectives

After much thought and a lot of observation on the MRH Forum, I decided that I would build a five track test “layout”.  This will give me the opportunity to do a number of different tests as the need arises.

Test items that are currently being developed are:                         

1. Loco performance
2. Track additives to eliminate or minimize dirty track
3. Alternate track material or treatments for performance and reliability
4. Loco Performance Variation
5. Loco Performance deteriation
6. Loco tune up- what makes the most difference to restore performance
7. Can vs open frame motors- what are the performance parameters

Other items will be added as time and interest allow.  Suggestions are always welcome. 

Just to comment on the previous Higgins, et al model locomotive tests, they were done with a great deal of care and attention to detail.  In my research, I found the test process defined in two articles:  http://www.trainlife.com/magazines/pages/551/39251/spring-1981-page-41  & http://www.trainlife.com/magazines/pages/15/1056/march-1990-page-24. 

In the later test result articles, a brief summary of all of their tests and magazine issue of the test article is included.  (Unfortunately, I have not been able to get these links to work for the last few weeks. They are included with the hope that the problem will be fixed at some point in the future.)

The problem with the previous testing came from the limitations imposed by a print article.  Most of the data that was taken was not published in the articles.  And in some cases, the data published was improperly edited.   I suspect that was not the fault of the authors, but in the publishing process. 

For the current test objectives, the givens and druthers for the test layout are as follows:

1. The primary focus will be on HO model locos, but the space needs to be able to accept track of other scales if required.
2. The test track needs to be conveniently located near my workshop.  The main motivation for building it is to have a place to quantify my tune up impact of the various HO locos that I work on.
3. It needs to be in a spot that will be undisturbed for long periods of time.  The data from these tests easily can be compromised by the variations that will occur from occasional bumps, as well as rebuilds.  
4. It needs to be built to a significant attention to detail.  The grade needs to be zero in the level portion.  Certainly within a tenth of a percent of grade.  The tangential plane needs to be held to the same tolerance. 
5. The movable portion will require the same concern for the plane slopes.  When the slope is set a three percent, the whole movable part needs to be at that slope.
6. There will be a transition section to the Layout.  This section will smoothly change the grade where required.  It needs to be flat when required and smoothly concave when required.  The will likely require at least two transition modules. 
7. The length of the active sections needs to be long enough to ensure that the acceleration of the test unit has stabilized to zero.
8. The track sections shall be wired using DCC best practices.  The initial tests will all be DC, later testing will also include DCC testing.
9. The track shall be fixed but removable.  No ballast or hard adhesive bonding.

After considering the various aspects from the Higgins tests, I decided I would focus on HO loco performance on a level track and on a level track transitioning to as much as six percent grade.  In the previous testing, the authors focused on defining the maximum grade, (approaching 30 percent) that the loco could climb by itself.  From a practical point of view, those levels tend to be extreme and add little to the understanding of how the loco will perform for the modeler.  Instead, I will be measuring the loco scale velocity at the predetermined grade with a resistance provided from zero to 100 NMRA weighted cars.  How I get the equivalent of 100 cars in a reasonable space is the subject of a subsequent Blog Topic “Design and construction of the loco performance test cars”.  The cars have been completed, but the blog write up is in process.

The issue of space is actually the dominate consideration for this test track layout. 

The A sketch of the layout side view is shown in the following picture:  

Test layout location and length

A critical factor in the selection is the length required to ensure the acceleration is zero in the measuring section.  Depending on how the throttle is applied and the resistance of the particular test, the unit will accelerate past the sustainable velocity.  Then it will oscillate around the sustainable level until the velocity is steady.  This steady region is where the speed measurements need to be taken.  This point will vary from engine to engine as well as for different conditions. 

Without actually getting into the calculus; the longest section, of unobstructed available space available edicts the length of the test layout.  Fortunately, for me, this is a wall that is in the room where my work bench resides.  This wall is shown in the next picture.  This wall is in the passage way from the stairs out of the garage into the finished portion of our walk out basement.

This wall is not totally unobstructed.  On the left end of the wall is a door going into the finished portion of the house and on the right end there is a stair well up to the garage.    When I designed the house, all ceilings were specified to be 10 feet or more.  So the test track can comfortably be installed above the door on the left.  The right end can be designed to minimize the potential interference with the stair well.  This wall is twenty feet long.  That will have to be adequate. 

To be sure the steady velocity has been achieved; speed measurements will need to be made at several locations.  Where the velocity does not stabilize, the data will be so noted. 

Test Layout Design

My initial concept was to allow the grade to be adjustable up to 40 percent.  This required some length tailoring, because of a overhead obstruction near the left end of the room.  The basic design remains the same even though I have decided to limit the grade testing to six percent.  One concern is the number of transition sections required to get to the extreme grades for only curiosity.   The space will allow testing of a larger grade at some later time. 

By imposing a restriction on the grade, the transition section will only have to be changed between the level and the grade tests once.  Defining the maximum grade and then incrementally testing grades below that may have required three or four transition sections and that many changes per test series.  A testing nightmare that does not add much useful data.  Plus the number of changes would adversely impact the overall quality of the grade testing. 

As shown, in the track plan, the test section is a foot wide.  There will be five separate test track runs.  These will be on two inch centers. 

From the outside to the wall, the second track in will be the primary engine test track.  This track will be code 100 Atlas Nickel Silver flex track.  The plan is to clean this track between every test. 

The first, third and fifth set of track sections are intended to be used to examine the impact of various track treatment options. 

The fourth set will be used to test alternate track materials or unique features.  To begin with this will be brass track.

These last four sets of track are not part of the primary testing activity.  They will be the subject of additional blog topics.

The tracks will run to the end of the layout, but the last Flex section piece on either end of each track run will not be powered.  Each track will also have a bumper plus an additional stop device to prevent locos and cars from rolling off the end.

Construction

The building process is in progress and what has been done is shown in the next few figures.

The test track is made up of four major modules; the Stationary module on the right, the movable module on the left and two transition modules.  The transition modules are unique to level surface testing and grade testing.  Because it is thought that most of the testing will be on the level surfaces, the level transition section is not meant to flex in any way.  The grade transition section will provide the transition from the horizontal surface to the grade surface.  Thus depending on the grade being tested, the transition piece will vary in curvature.  For small grade changes, it is felt that this continuous flexing will have a minimal impact. 

Initial segment checkout

For a number of reasons, a test checkout series is needed before the layout is totally completed.  The purpose of this is as follows:

1. Establish the accuracy of the data that can be taken. 
2. Is the proposed test series possible and does it make sense.
3. Generally establish a procedure for the test sequence.
4. Are the test lengths adequate?
5. Are the stopping segments sufficient
6. Provide a set of data to be used in the data reduction computer program
7. Are the voltage losses from the power supply acceptable? 
8. Are the track segments clean?

These last two items will be repeated as the construction evolves.

This testing of the test setup has begun using the deck of the fixed segment.  As shown in the following figures:

Understand, this work will be finalized as construction continues.

The initial set of testing has begun using an older Athearn Blue Box SD9 loco.  This unit has a round grey motor and metal side frames.  These tests were performed with the loco only, no “car” resistance on a zero grade plane. 

The measured speed variation ranged from 2 to 180 SMPH.  The voltage control is very good.

The areas of problem are pull force determination and current level and stopping distance above 50 percent power.  As expected, the current readings fluctuate after initial start.  They will eventually settle out to a steady value, but with this unit that may be 3 or four lengths of test section.  I have observed this in the past, but have not associated it with a travel distance.

The speed for this unit at 50 percent power was around 110 SMPH.  At that speed the vehicle and FW momentum is too large for the 2 foot stopping distance I have allowed. 

I have another device and a set of Weights That will help with the draw bar pull.  I will be repeating this checkout testing on other loco units before deciding about the current and stopping issues. 

For me this unit is one of a type that I will be doing tests on, but in those cases, I will not be looking for a full operating range characteristic.  This info is the reason for doing this pre completion testing.

The initial data taken from

The initial data taken from the trial testing of the test track is being sanity checked and the test plans are being modified accordingly.  For this test testing three different locos were tested with the track in a temporary set up.  Because of chance, each of these locos were older Athearn Blue box units with metal side frames (outside frame wheels) and the round grey motor. 

The first of these was an SD9 unit.  For this unit there were four different tests run after a series of tune up steps.  These steps are as follows:

1. The first was in the as received condition.  This unit has been well used, so the as received condition could have been really bad.  In this case it was not that way at all.  The unit ran over the speed range on a level surface without any problems.  One might have said it did not need a tune up watching it perform visually. 
2. For the second run the wheels were “polished” with a Standard Kadee Speedi Driver Cleaner.  Each wheel set was cleaned for 10 seconds with the brush. 
3. The third setup had the motor and truck towers tuned to a small degree.   In this case the motor and truck towers were lubed with the appropriate grease or oil.  The communicator was also cleaned. 
4. The fourth set had the motor isolated, the mounts changed to the newer screw type, the five wire electrical wiring was installed and the wheels were up graded to a set of NWSL replacements. 

The second unit tested was a super powered F unit that someone had changed the truck gears to reduce the speed at power.  For this unit, three test steps were performed:

1. As before, the first was in the as received condition.  This unit has also been well used.  The condition was average or below.
2. For the second run the wheels were “polished” with a Standard Kadee Speedi Driver Cleaner. 
3. The third setup had the motor and truck towers tuned to a small degree, as in the fisrt set of tests. 

This unit did not get the five wire or wheel up grades.

The third unit tested was a no fly wheel F unit.  This unit had the tune up features of the step four from the SD9 tests already in place.  That is the only set of data recorded for this unit.

For each of these tests, at each test point the data was taken by resetting the power supply voltage and measuring the corresponding current and velocity.  Each point was repeated three times.  There is some variation at any point, but generally small.

Because of the variations previously observed in the current measurements, a set of loco test rollers were used to measure the current levels as a back up to the readings while the unit is running on the track.  This allowed the current measurement to settle to the steady state level.  The question is that an measurement problem or is the motor current draw varying in this manner?  The rollers introduce two unwanted effects on the measurements.  The rollers introduce additional voltage drop between the power supply and the motor and the rollers tend to be free rolling so the motor load is not the same as running directly on the track. 

At low speeds the difference was small.  At high power the difference was much larger, indicating a measurement problem in that region.   The length of the designed track section will help, but there will likely still be uncertainty in the current measurements over 100 SMPH.

This data shows that the testing will sense changes in the loco condition when various tune-up features are done.  Because of the general uncertainties in the data at this point, I’m not going to make any conclusions.  I thought I would share some interesting charts. 

I’m encouraged to go on with the next phase of construction on this test track.

Having fun with trains,

Larry