As a few of you may have read in my other thread, I recently installed my first TCS WowSteam decoder in a Trix 2-8-2 Mike. I'm still getting used to programming some of the CVs using Audio Assist but I'm pleased with results so far.
I had read that programming TCS Wow decoders with JMRI can be problematic for some users, as Decoder Pro has difficultly reading the indexed CVs of the Wow decoders. I tried it the other night using my NCE Power Cab and USB module and - sure enough - I experienced a number of "Programming Error: No locomotive detected (301)" error messages before giving up on the idea.
Today I received 2 sets of roller bearing stands (4 per set) that I ordered from Micro-Mark. I'm presently WOAL (w/o a layout) and only have a programming track. So the roller bearing stands will allow stationary break-in and programming of my locomotives until I build my next layout. Setting up the roller bearing stands was pretty straightforward and they work well as advertised.
This evening I decided to try again and see if I could read & write the CVs that I had programmed/changed in my Trix Mike; this time with it positioned on the rolling bearing stands on my programming track. I fired up Decoder Pro on my Windows tablet, clicked on the CV tab, then clicked on Read Full Sheet. Well, lo and behold - Decoder Pro read ALL the CVs without a single error. Same went for writing the CVS. And, presently, I'm reading my other Wow decoder (installed in a brass H20-44 switcher) with the same success.
[Edit: I made an error. My H20-44 switcher actually has a Loksound Select. My Athearn Genesis F3A is the one with the WowDiesel decoder in it. Read comment in later post.]
So, what is it about a rolling bearing stand that would aid Decoder Pro in reading all the CVs successfully? Is it the larger surface area contact between the base of the roller bearing stand and the rails with the "two-point" contact of the bearings with the drivers of the locomotive? I'm not complaining. I'm just trying to understand why the difference.
Thanks for reading my musings and for any input you might have...
Tom
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
Just finished reading and writing my WowDiesel decoder to Decoder Pro.
The roller bearings and stand are acting as a resistor. If it was made out of copper or gold, it might not provide more resistance. Resistance is an R factor x length / diameter of the wire. In the case of the bus wire, a bigger diameter means lower resistance, because we are usually comparing copper with copper wire. The stand looks like aluminmum, a bigger R factor.
As to why some decoders need more resistance to be read/written....Randy will be here in the morning.
Henry
COB Potomac & Northern
Shenandoah Valley
Thanks, Henry. I know about resistance but had read that using a resistor across the two programming track rails helped with programming certain sound decoders. However, I didn't realize that it might also help Decoder Pro with read/write issues with said decoders.
Yea, we'll see what Randy has to interject when he comes on the forum later this morning.
[Edit: Maybe I need to change the title to Loksound for Randy to peek in? ]
As corrected in my OP, my brass H20-44 switcher actually has a Loksound Select installed in it and my Athearn Genesis is the one with the WowDiesel decoder.
I'm presently reading the F3A on the roller bearing stands. I've had a few initial "Programming Error: No locomotive detected (301)" messages on indexed CVs but only early on. Before that I tried programming w/o the stands and got a number of error messages. I cleaned the programming track and roller bearing stands with alcohol and Decoder Pro is now purring right along reading the decoder.
Okay, Reading full sheet has stopped and I was able to correct the few error messages by manual clicking the read button next to the CV. Along with resistance, perhaps the roller bearing stands help to maintain more consistent electrical contact by keeping the locomotive in a stationary position; thereby reducing the number of reading errors? Otherwise, the locomotive has to be repositioned over and over again on my 27" long programming track because the locomotive moves forward a little or a lot depending on the CV read.
And now Writing full sheet has stopped with no errors or corrections needed. Glad I picked up those roller bearing stands. They make read/writing and chuff syncing so much easier.
Tom, if the roller mechanism has the pick-up tires, say on the drivers, running on curved surfaces, the contact area would theoretically be less than it would be on a flat surface...wouldn't it? Two curves surfaces making contact, instead of a flat and a curved one?
It could just be that your programming track needs to be cleaned!
It's happened to me that Decoder Pro will give me the 'no locomotive detected' error message, and I've slid the engine a 1/2" ahead or back on the programming track, and then - boom - it reads it fine. The programming track is low power compared to regular DCC, so is more likely to be finicky.
Other way around, i think. The roller stands have far better and more reliable contact with the wheels than a piece of track.
That is pretty muich the ONLY problem with sound decoders that have large numbers of indexed CVs - there are SO many reads and writes required that doing it the "NMRA way" needs long stretch of continuous contact to get a clean read on the whole decoder. And since the motor is pulsed for acknowledgement, the wheels turn a little bit. Success with the loco on rails may improve with a programming device that can do a proper direct mode, like SPROG or the PR4, because it needs far fewer queries to read the same number of CVs. I'm surprised that TCS doesn;t have a proprietary method, but then again you can;t load your own sounds so an alternative protocol is not needed like it is with Loksound. When you have to read 500 CVs by asking each and every one "is the value 0? Is the value 1? Is the value 2?" and so on until you get a response, that's a LOT of reads and writes. Add to that having to write the index CV before asking for the value of the CV in question, and that just adds to the total count. There's a good reason DIgitrax released the PR4. As more fetures get added to sound decoder, this situation is only going to get worse with even MORE CVs to read and write. It may be realtively easy to figure out WHAT to write using JMRI, but even if you can easily pick the settings you want, it takes many many CV writes to actually set the decoder. Miss one, and you might just get some unexpected results. Miss the INDEX CV and it can all go downhill - an incorrect index would probably just mean the base CV intended would not get set, but you could think you are setting a volume when because the index CV got missed, the decoder uses the last valid value and you end up setting the whistle type or something.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
selector Tom, if the roller mechanism has the pick-up tires, say on the drivers, running on curved surfaces, the contact area would theoretically be less than it would be on a flat surface...wouldn't it? Two curves surfaces making contact, instead of a flat and a curved one?
If I'm understanding your first sentence correctly, Crandell, the Trix Mikes don't come with traction tires. In theory, how much less surface area would two curves make with one another vs a line and a curve?
wjstix It could just be that your programming track needs to be cleaned! It's happened to me that Decoder Pro will give me the 'no locomotive detected' error message, and I've slid the engine a 1/2" ahead or back on the programming track, and then - boom - it reads it fine. The programming track is low power compared to regular DCC, so is more likely to be finicky.
I think that's true to some extent, Stix. I did clean it prior to programming but it still had a bit of grime when I cleaned it again this morning.
As I mentioned in an earlier post, when Decoder Pro is reading the individual CVs of a TCS Wow decoder, the locomotive can either move a little (1/4") or lot (3-4") per CV. So, after 6 or so indexed CVs, I have to push the locomotive back to the other end of the 27"-long programming track. Given the number of indexed CVs per decoder, this becomes tiresome pretty quickly. That's why the roller bearing stands are more preferable in this situation.
Tom, a perfect circle would intersect another surface of any description at only one point. Theoretically. But also theoretically, for the materials we use, with their imperfections at the micro level, a curved tired should make more contact, have a larger footprint, with a level surface than it would with a curved surface where the two surfaces curve away from each other (the rollers on your loco stand).
Not taking into account the discussion about resistance making the difference, the loco 'should' have better contact on the rails, and have less dropout or misses in continuity in trying to programme the decoder than it would on your loco stand. At least, that's how it looks to me. Perhaps an engineer could point out some errors in my thinking.