I received my Loksound Select Micro and speakers from TTE and temporarily installed them in my brass H20-44 switcher. The TDS SuperSonic "sugar cube" speaker and small enclosure combination sounds terrific! I bought two speaker/enclosures with the idea of wiring them in parallel. However, the single speaker is plenty loud enough.
Motor-control is good with this unit but isn't yielding anywhere the same results I was getting from the TCS M1P decoder I installed in it previously. Speed step 004 and > is nice & smooth; speed step 003 and < is somewhat jerky, pulsating, and actually runs faster than speed step 004.
Is this due to the BEMF of the Loksound Select over-compensating for the lowest speeds? Does anyone have any recommendations of what I can do to smooth it out? I did try adjusting CV54 & CV55 to the default values for a "can motor without flywheel", as suggested on pg. 22 of the Loksound Select Manual. However, it made no difference in speed step 003 and <.
I only have 27" of programming track so I can't really use the automatic motor tuning feature. Other suggestions in the manaul are to adjust CV52 (slow speed amplifier) by trial and error and turning off load compensation (CV49). I'm all ears if anyone has any other ideas to try.
Thanks,
Tom
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
Tom,
you should further experiment with CV 52, and try to smooth out slow speed performance, because this CV is meant for slow speed motor control optimization.
I recommend also going away also from "default" values for CV 54 and 55 because it is impossible that default values are appropriate for all types of motors. Sometimes even within the same motor type the same CV values do not yield the same results.
If there is no flywheel, than in general CV 55 should be lower than in case there is a big flywheel, so motor has high inertia.
Sometimes it can help to reduce a little CV 56, for degree of motor influence (for example from value of 255 to value of 80-120).
If you switch motor control completely with CV 49, then you loose all advantages of excellent ESU motor control.
Unfortunately Select does not have CV 51, which is responsible for motor control at speed step 1, as in 4.0 decoder version.
I have achieved good results with ESU Lokpilot 4.0 with motors where no other decoder could help, especially with 3-pole motors, or motors with huge cogging effect.
If everything is Ok with locomotive mechanically, I am sure it is possible to achieve satisfactory results with your decoder, but sometimes a more patience is required than for other locomotive.
Regards,
Hrvoje
tstageI only have 27" of programming track so I can't really use the automatic motor tuning feature.
Hi Tom:
I have used the ESU automatic adjustment feature on multiple locomotives and they all took less than two feet of track to go through the process, with one notable exception. I will qualify that by saying that none of the locomotives were brass. I don't know what effect that might have on the process. I would be ready to catch the locomotive just in case it does charge off the end of your test track.
FWIW, the notable exception was powered with a very old motor similar to what is used in slot cars. That critter (it was a scratch built railtruck) took off like a scalded cat and didn't seem to slow down at all before it ran off the end of the 6' test track. Eventually I did get it to run very nicely but I had to repeat the process several times.
This is the motor:
Dave
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
I used to have a toy race car set, each car ran on a pair of N size batteries (yes, there was such a thing) and the cars had a guide pin like slot cars, but the track was all plastic and the cars were free running. They had those motors in them. I wonder if the thing is even rated for 12 volts.
As for Tom - do try the autoconfig. The motor in that loco is a lot different than your typical Athearn/clonse style found in most locos, which is what the Loksound decoders seem to be set up to work best with out of the box. I don't think you will have the same problem as Dave, but make sure you have plenty of clear track.
Only if the autoconfig doesn't work, or still doesn't give great results, then start messing with the BEMF CVs. It does seem to be overcompensating - though the CVs are different, a good example is in one of the Digitrax decoder manuals, which explains how one CV is the force, another is the spring, and the third is the damper. Too much force and not enough damper and it oscillates back and forth trying to fix on a speed, too much damping and not enough force or spring and it won't be able to adjust the speed when necessary. And the happy medium, where it all works great.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
rrinkerI wonder if the thing is even rated for 12 volts.
Randy:
Based on the speed at which it travelled you may be right. I ended up setting the start voltage CV somewhere in the single digits. I had assumed that the motor was 12 volts or there abouts because it came in an ancient HO scale railtruck kit.
Cudos to Loksound for designing a decoder which could tame the wild motor
Did you have to drop the top speed to keep it reasonable? I sort of remembe rin my reading of every MR from the first one (when I got the complete collection DVD, I did indeed start with the very first one and read them all in order, up until where I first subscibed and had them all) on what I think was a similar or maybe the same rail truck where they mentioned in the review that the motor must be intended for less than the typical 12V of model railroads.
hon30critter tstage I only have 27" of programming track so I can't really use the automatic motor tuning feature. Hi Tom: I have used the ESU automatic adjustment feature on multiple locomotives and they all took less than two feet of track to go through the process, with one notable exception. I will qualify that by saying that none of the locomotives were brass. I don't know what effect that might have on the process. I would be ready to catch the locomotive just in case it does charge off the end of your test track.
tstage I only have 27" of programming track so I can't really use the automatic motor tuning feature.
Dave & Randy,
I ended up trying the automatic motor tuning feature and now everything is copacetic! Thanks for encouraging me to try that.
All I had to do was change CV54 to a value of 0, hit F1, and the locomotive took off like a shot. I thought it would start out fast then slow down gradually so I kept my hand at the other end of the programming track. In reality, it took only ~12" to perform the tuning and the locomotive came to an abrupt stop, which surprised me. Now my H20-44 crawls along as well as the TCS M1P decoder that I had installed in it before.
Just out of curosity I checked CVs 52-55 to see what the values were: [Default]...New
Quite an adjustment. I think I would have been playing around with values for a fair bit before coming up with those specs.
BTW, I abandoned the whole shroud idea and end up laying a 3/4 x 2-3/4 x 1/8" piece of styrene on top of the motor housing. It hangs off both ends by ~3/4" and covers the set screws protruding out of the drive shaft coupling nearest the motor. Because the motor is round and the shelf overhangs the sides slightly, it allows for a place to tuck the motor/decoder wires out of the way.
Here's a couple of pics of the install:
The only space for the speaker was up in the front nose of the shell. I actually tried using the open portion of the platform on the left side. However, the speaker enclosure is too tall and the shell won't seat all the way on that end.
Overall, I think it's a simpler design and won't require as much disassembly should I need to do maintenance - e.g. lubrication. Next on the agenda is to add front and rear LED headlights. I'll keep you posted...
tstageBTW, I abandoned the whole shroud idea and end up laying a 3/4 x 2-3/4 x 1/8" piece of styrene on top of the motor housing.
Tom:
Nice simple solution! Nice clean wiring job too.
I'm glad to hear that the automatic adjustment worked for you.
rrinker Did you have to drop the top speed to keep it reasonable?
Yes, I did. I dug out my programming record. The actual numbers I ended up with are:
CV 2 = 1
CV 5 = 12
CV 6 = 6
I had forgotten how low the numbers were. They could be a bit higher but I wanted the railtruck to be fairly slow.
Yeah, if you have to use 12 as the TOP speed to keep it reasonable - that's not a 12V motor. Pretty sure those were meant to work on like 3V If it has any part number markings on it it might be possible to find a data sheet.
Tom - nice job, keep it simple, the platform is a much less complex thing than that whole shield idea. And not bad on the speaker, it looks like it's right by the stacks so it's a good place for the sound to be coming from.
Thanks, fellas. Again, I appreciate your input and now understand a little better what the Loksound Select decoders can do.
Sorry, I kind of hijacked your thread a bit talking about my own stuff. My bad!
No offense taken, Dave. Sometimes those make for important and interesting side-notes to the main discussion. And we can always blame Randy...
tom
Sure, since I replied to it, thus taking the heat off Dave.
Anyway, now we know how to hande a less than 12V motor, too.
A question for all concerning the CV54 auto calibrate feature. Instead of having it on the layout with plenty of track, has anyone done this by having the loco on rollers on a programming track? I'm assuming when doing this the decoder is recording load parameters from many different speed steps, then configuring itself to work at the premium?
The reason I ask is I like to have the decoder completely tweaked using JMRI on the rollers. I guess I could try it out, just wondering about anyone elses experiences? It seemed to work fine for Tom by just holding the loco on the programming track.
"The true sign of intelligence is not knowledge but imagination."-Albert Einstein
http://gearedsteam.blogspot.com/
Like I mentioned, Dean, it only required ~12" of my 27" programming track to perform the calibration test. I could be wrong but I don't see why it would make any difference whether the track were linear or "circular".
Only thing I can see is that there may be slightly different rolling characteristis with the loco movign it's own weight vs just spinning rollers.
But since you have the settings saved in JMRI - why not run an autoconfig on the layout and then read back the BEMF CVs and see how different they are from what it calculated using the rollers? Since the roller values are saved, you can just put them back, or keep the new on-rails numbers.