Sunday, September 27, 2015

September Update


September has seen me sick for a bit over two weeks but I did spend a little time in the train room for live testing after programming decoders. I "killed" my SprogII, learnt some decoder programming lessons, installed some decoders, established my new freight loco speed curve and speed matched 5 locos. I finished the Control Panels for Chris's layout whilst he was holidaying in England, Scotland and Ireland.    There was no work on my upper level signal control stuff this month.


Decoder Programming Lessons

I use a Sprog to read my decoder CV's as my Roco command station - Lenz LI-USB interface does not work as well as I would like.  The first time successful "writes' to high current draw decoders is not totally reliable either.  However the Programming on the Main (POM) works like a charm.

To ensure test track programming speeds you MUST use the system used on the layout as decoders do things "proportionately". So if the test track output voltage is 15V and the layout 18V then things will run faster on the layout.

Whilst the Sprog is very reliable with my test track setup the output voltage is 3V lower from the Sprog when compared to the Roco system.


Sprog happenings

I tried to program my new Tsunami decoders but the Sprog II but it would not do the job.  I researched the Sprog DCC website and found info that the firmware needed to be upgraded to address the issue. Well I tried to upgrade the firmware in my Sprog II and managed to kill it.....  I picked up a Sprog 3 and it works like a dream.

Sprog II


Picture shows a v3 mine is a v2.5

I used DecoderPro to do the firmware update BUT I had downloaded the wrong firmware file from the Sprog Website.... The process did not complete due to my error and the Sprog II never came back to life.  I emailed the manufacturer who responded with a series of questions which determined that I had "stuffed it up".  He offered to fix for no charge if I returned to product to him in the UK.  It will return to me in due course and I will then have a spare.

Sprog 3



I loaded the device drivers and it worked first time in DecoderPro and handled the Tsumani's just fine.


ATSF 4-8-4 TCS Wow install

A few years ago I acquired a secondhand Paragon 4-8-4 Niagra.  It had a QSI v6 decoder which regularly "lost its settings" requiring refreshing the CV's, further it ran nicely for about 5 minutes then failed - this was repeatable and caused me to remove the loco from the layout.

I decided to try the new TCS Wow Steam decoder so I removed the QSI deocder from the tender and hard wired the new one.


I sealed the speakers in place with Kystal Klear and retained the backup light PCB.


I fitted some brass strip to mount the new decoder on.



Decoder and Keep-Alive  mounted in the tender.

There is a problem with the headlight in the factory wiring as the decoder output is functional as is the led on the PCB light board in the smoke box.  The loco awaits stripping to find the faulty connection in the wiring harness - I think it is at the plug under the cab.

I have also done the speed matching to align with the Cab Forward baseline detailed below.


Cab Forward Modifications

A while ago my engine "lost its chuff" and after contacting BLI they sent me a replacement reed switch but did not give any details on how to fit it.  Reading posts on various forums it was evident that this was a common problem.  I had previous experience with an Athearn Genesis Big Boy that had a Tsunami decoder and the sound quality and motor control was much better than what was being achieved with the factory decoder and second motor control decoder in the Cab Forward.

I decided to replace the factory decoder and chose a SoundTraxx 826105 Tsunami Decoder.  I removed the existing decoder leaving only the wiring harness.  I used the existing speaker and  hard wired the new decoder after gluing with Krystal Klear to the metal base of the tender.

Thanks for the theory lesson must go to Mark Gurries, Bruce Petrarca (AKA MrDCC) and Detlef for the Motor Decoder Tuning - SoundTraxx information to set the BEMF up properly.  My goal is to have a top speed of around 40 MPH for my freight locos.  After watching Programming Tsunami Decoders for realistic operation.Parts 1, 2 & 3 by FishplateFilms. I decided the programming order was BEMF, Top Speed, Speed Steps 7, 14, 21 & 28 then adjustment of remaining speed steps followed by acceleration & braking.

The big lesson apart from the BEMF was using CV66 (Forward Trim) & CV95 (Reverse Trim) for top speed adjustment.  This method allows finer control using the speed step controls.

I used DecoderPro and the Bacrus Speedometer to set the desired speed curve - see the video below.


Well the final test was a 10 car load firstly on the flat with a cold motor and it failed to start so CV 210 Ki Coefficient (BEMF) was increased (after setting CV 209 to 0) and a satisfactory result was achieved and CV 209 reset.

The next test was starting on a fair grade with a 10 car load - again it failed, the sounds commenced before wheels turned, so a further increase to CV 210 resulted in a satisfactory outcome and I again reset CV 209.

I am finally happy with the result which will become the new benchmark and have recorded the procedure for future use.




More Decoder Programming

After deciding on 40 MPH top speed and a speed curve for reference I have been trying to match locos with different drives and decoders which means the acceleration, braking and BEMF performance requires different settings for each combination.

Blueline RSD-15

I removed the factory decoder and hard wired a new Tsunami decoder in.  Being a freight service loco I used the speed curve previously determined to set the speed steps using the same process as set out above for the Cab Forward.


Athearn U-50

After my leanings this month I decided that all  y Tsunami decoders would be reprogrammed and as I just did this one last month it was an early candidate.  I tested the results with a command station consist between the RSD-15 and on parallel tracks to check the acceleration and braking rates and they match perfectly!

Atlas RSD-5

This is an Atlas Gold loco which is fitted with a ESU Lok Sound Version 4 Select decoder.  I had to read up as this decoder is significantly different to the non sound Lok Pilot 3.0 DCC I have previous experience with.  Well after the first attempt followed by a factory reset I finally got it in the right paddock.  A bit of fine tuning on the layout and I think I have it performing quite close to the Tsunami's..


Next Steps

Before I launch into reprogramming 50 odd locos I want to invite a few individuals around to "try out" the revised settings to make sure my perceptions and those of modelers I respect and some of my operators are in a "similar" space.   Don M, Rossco, Grahame H and Wayne L wait for the email/viber/phone call if you have not already heard from me!


Chris's Control Panels

Last month I mounted the switches and this month I have completed the wiring of the panels ready for installation.




Two yards have duplicate control panels to enable control entry to each end and switching.  Experience has shown that each solenoid on each point motor needs to be isolated from each panel so conflicting control switch settings for a particular solenoid does not cause a shorts circuit - this is achieved through fitting diodes in each feed to the solenoid from both panels.


The diodes are fitted between terminal strips which are in turn fitted between the switch and the solenoid.




First panel fitted one toggle switch not working and will need replacement during the next work session. ...


Thanks for reading.

Till the end of October 2015.





-ooOOOoo-