N Scale Track Speeodmeter...

[Sumner]

The 'Curved SpeedTrap' speedometer is done and the print files are up on Thingiverse.com. The 'Straight SpeedTrap' speedometer should be done in a day or so and those print files should also then be up.



SpeedTrap software is finished and seems to work fine.

The print files are up on thingiverse.com ( HERE ).....

https://www.thingiverse.com/thing:6697804

For the whole build of this speedometer go ( HERE )....

https://1fatgmc.com/RailRoad/DCC/page-30.html


Sumner

[brill27mcb]

I would think that you may get different speeds running and measuring on a straight track than you would on a 12.5" curve. I would put more faith in the straight-line speeds, since different locomotive styles and different model manufacturers' chassis designs will tend to slow differently when rounding a curve. It would be interesting to see if you can quantify a difference when you have the straight-track version working.

Rich K.

[Sumner]

..... I would put more faith in the straight-line speeds, since different locomotive styles and different model manufacturers' chassis designs will tend to slow differently when rounding a curve. It would be interesting to see if you can quantify a difference when you have the straight-track version working.

Rich K.

Thanks for the input Rich.  I think they will be the same but should find out maybe day after tomorrow.  Need to print one more part and then assembly it.

Whatever part of the loco that trips the first beam should also trip the second beam so I can't see where the shape of the loco would play into it.  I tried to make the distance on both 6 inches on the centerline so they should read the same but since the one is on a curve it probably will read slightly faster.  It is easy to calibrate either by changing one number in the software.  If you watched the video you will see that there was only about 1-2 mph difference over the loco's slowest speed (4 mph) to its highest speed (65 mph) compared to the throttle's speed.

I speed matched the loco to the throttle speed steps with a stop watch timing a lap on the layout where I have the distance 'fairly well figured out'.  It might be wrong and I'm not sure how much time I spent on that loco for lowest speed, mid speed and top speed.  Still the SpeedTrap speedo compared to the throttle speed and only being 1-2 mph off over the entire speed range is pretty good.  A lot of older car speedometers were off more than that (not sure about newer ones).

When I feel a loco is accurate to the speed vs. a timed lap on the throttle I'll adjust the 'curved' and 'straight' to read the same speed as the throttle.  An accuracy of 1-2 mph will be good enough for me and seems to work fine in the few consists I've put together.  Anyone who builds one of these will be able to adjust them also by changing that one number in the software's config file.

Sumner

[reinhardtjh]

Going around a curve adds drag which can affect a loco's speed.  And a 12 1/2" radius curve is one of the sharper ones in N-Scale so might affect speed more.  Not so much due to the loco's shape, but how the wheels and flanges react to the curve.  Look at the recent threads about steam locomotive pulling capacity and how it is different between straight track and curves.

Steam Pulling Power observations

[Sumner]

Going around a curve adds drag which can affect a loco's speed. ...

OK, thanks for the clarification.  So I guess you are suggesting that even though a loco is on the same speed step for the decoder that it could be slower going around a curve vs. at the same speed step on a straight section of track?

The straight one should be up in a day or so with that being a possibility for anyone that has a concern.  I really don't expect the curved one to fit many people's needs.  For my needs the curved one was required for my test track as that is where I do all my programming and speed matching.  Once I get the straight one done I'll try and test them if possible back to back on the test layout.  I'll have to put the straight one over the top of a turnout but should work there.  I also can't do any steam testing as I'm all diesel.

I do have locos from small switches to DDA40X's.   The DDA40X might not clear inside the SpeedTrap on a curve but a slightly shorted DD35A did so I can compare some of those options when I find the time if that happens 

Sumner

[reinhardtjh]

OK, thanks for the clarification.  So I guess you are suggesting that even though a loco is on the same speed step for the decoder that it could be slower going around a curve vs. at the same speed step on a straight section of track?

Exactly.  Probably not a lot but it might work out to a Mph or two (or three) - just not sure.  It would be an interesting experiment.

[Sumner]

--- N Scale SpeedTrap Speedometer – Pt. 5 --- Loading the Software/Sketch for the esp32 ---
=========================================



Loading the code (if the first time for this take your time and follow the steps slowly):

1. Connect the SpeedTrap speedometer to your computer and Download the Arduino IDE.
…...Available from https://support.arduino.cc/hc/en-us/articles/360019833020-Download-and-install-Arduino-IDE

2. Open the Arduino IDE.

3. Download the esp32 boards in the Arduino IDE with the following steps. ( Use ESP32 by Espressif Systems Ver 3.0.0 or later).

…...a.) In the Arduino IDE, go to 'File' > 'Preferences'
…...b.) Enter the following into the 'Additional Board Manager URLs' field (bottom of page): …..…....…... https://dl.espressif.com/dl/package_esp32_index.json
…...c.) Then Use the 'Boards Manager' in the 'Arduino IDE' to install the esp32 board support: …...…...…... Go to 'Tools' > 'Board' > 'Boards Manager' …...…...…... Put 'esp32' in the search window. …...…...…... "esp32" by Expressive Systems should show as an option. …...…...…... Install it.
…...d.) Next you need to install LiquidCrystal I2C library by Frank de Brabander.

.... enough of the software instructions the rest are here....

https://1fatgmc.com/RailRoad/DCC/page-34.html

Everything is up now to build this at...

https://1fatgmc.com/RailRoad/DCC/SpeedTrap-Index.html

Not a valid youtube URL

Sumner

[Sumner]

The 'straight & curved' SpeedTrap speedometers are done....

and the print files are up on Thingiverse.com. The whole build for both is ( HERE ).

SpeedTrap software is finished and seems to work fine. Click the image above or ( HERE ) to see it in action and.....

…..above or ( HERE ).

The 'straight' SpeedTrap print files are up on thingiverse.com ( HERE ).

The 'curved' SpeedTrap print files are up on thingiverse.com ( HERE ).

For the whole build of this speedometer go ( HERE ).

Sumner

[Sumner]

I had a request for KPH readings on the SpeedTrap speedometers in addition to the MPH readings that are there.  I spent some time today and added that feature.  When you load the software you have a choice now....

...…... Open the internet link shown next that is for the software/sketch you want to load: …...
…...…........…... Sketch for 'straight' SpeedTrap showing MPH........ ( HERE )
…...…...…...…... Sketch for 'curved' SpeedTrap showing MPH......... ( HERE )
…...…...…...…... Sketch for 'straight' SpeedTrap showing KPH........( HERE )
…...…...…...…... Sketch for 'curved' SpeedTrap showing KPH......... ( HERE )

I tested the code out in the 'curved' & 'straight' SpeedTraps and ....

... it seems to be working fine.

  For the whole build of the SpeedTrap speedometers go ( HERE ).

Sumner

[John]

Im going to give this a try when I get back home

[Sumner]

Im going to give this a try when I get back home

I put the second one together (after the parts were printed) in about 4 hours total.  Let me know if the instructions need clarifying and any other build observations would be helpful.  Be glad to help where/when I can.

Sumner

[TravelingCarpenter]

Pretty cool!
Might do one as a backup to my accutrak 2.
For the past three days I have been running speed tables and matching for my ESU lok sound chips for the E24 n scale event in Evanston Wyoming. I use a 13 3/4” radius track. To those that say it needs to straight track,, I say 🧐. Does it matter?,as long as you are consistent? But to each their own
     
     As I do not like the driving and motor settings that come from the factory. So I adjust the start/stop momentum and minimum/ maximum speeds. Step 1 has me at 1.5 smph, then step 16 has me at 18smph. 
     I have 7 three unit lash ups that all run the same speed on the same speed step. So I can interchange them. The plan is to make all my fleet exactly the same, within.1 smph. The speed step are also coordinated with throttle notching on my Proto throttle.
     The point in my rambling is a speedometer is most needed in speed matching full stop!

[Sumner]

....Might do one as a backup to my accutrak 2.....

Let me know how the speed compares to the accutrak.  I used a formula that should be correct.  I speed match to the speed steps on the throttle.  Each step is a N scale mph.  I did that as close as I could to the E9 that is in the videos timing a lap of known distance on my test track (oval).  I got the speeds from 4 to 65, first and last speed stop on my HandCab throttle to match within a second timing a lap.  Build a spreadsheet for that, 65 mph = 13.0 sec.

When I ran the SpeedTrap against the throttle speed steps I had speed matched the E9 to almost all were within 1 mph (there are some rounding in the software).  With that in mind I'm happy with using the speeds I'm seeing.  I had to adjust the numbers for the 'curved' SpeedTrap some since it is a slightly shorter distance between the IR sensors. 

If one wants to adjust the speed it is easy to do.

 constantMph = 57000;     // Use for Straight SpeedTrap 
  //   constantMph = 55500;  // Use for Curved SpeedTrap

Use the top line for the 'straight' SpeedTrap.  Leave the bottom line above commented out (//).  To adjust the speed you Increase the 57000 to make mph higher -- Decrease to make mph lower.

Sumner