I like to work with small HO diesel locomotives. Recently I have been working on two Grandt Line switchers. One is a 25 tonner and the other is a box cab. My goal was to compare the performance of one locomotive equipped with a traditional flywheel with another equipped with a Loksound Power Pack keep alive circuit.
I posted the video of the flywheel equipped 25 tonner here:
Tonight I was finally able to run the box cab with the Loksound Power Pack keep alive circiut. I’ll show a video soon, but all the video will confirm is that the keep alive circuit allows the locomotive to run incredibly smoothly with no hesitation. So far, I have only tested them on straight track.
However I have encountered a problem that needs to be resolved before I can make further tests. The Power Pack keep alive circuit seems to be causing a bit of strange performance in the locomotive.
When I select speed step 1 the locomotive jerks forward and runs at a speed much higher than it should. Same thing on steps 2 and 3. However, when I select speed step 4 the locomotive slows down to a crawl.
If I start the locomotive from a complete stop by selecting the ‘Incr Fast’ button (which starts at speed step 10) the locomotive moves off very slowly and accelerates to a speed which is less than what it was running when I selected speed step 1. I can then slow the locomotive down to a crawl by pushing the ‘Dec’ button. However, when I get down to speed step 3 the locomotive suddenly accelerates again.
I’m guessing that I have missed something re the CV settings required when installing a keep alive. Does anyone have any suggestions? What have I missed?
Ya, my guess would be you’re using a speed table and accidently have the first step set to a really high number. That’s a nice thing about Decoder Pro, you can see the speed table graph and easily spot a problem. When I first got Decoder Pro, I found I had done something like that, only the other way around…one step about midway thru the speed table was set to almost zero. Correcting that made things run better!
Well,yes and no. I’m pretty sure that with Loksound decoders speed table speed steps 1 and 28 are set by the values that are determined by CV 2 and CV 5, and then the intermediate points are manually input in the speed table. So if CV 2 is set too high, the starting speed would appear to be fast, and then the speed set by the CV for speed step 2 would act to slow down the unit before speeding up again.
But then all this pre-supposes that he has the speed table enabled.
I didn’t know that there was any BEMF calibration with a Loksound Select. At least I don’t remember reading that in the Select manual. I guess I’ll have to read through again.
I am finally able to run the two switchers side by side. The box cab isn’t finished yet but that has no effect on the test results.
Both switchers weigh the same, and both have identical BullAnt drives from Hollywood Foundry, except the CP switcher has a flywheel as supplied by Hollywood Foundry whereas the box cab has a Loksound Power Pack keep alive instead of the flywheel. The flywheel and the Power Pack both take up about the same amount of space. The BullAnts have a 60:1 drive ratio.
The locomotives are being tested on Atlas Code 100 NS track with a Peco curved turnout. The turnout has not been tuned up. Its straight out of the box. The track and wheels were cleaned with alcohol before the test. I am using an NCE Power Cab. I am using 128 speed steps.
The results are stunning to say the least.
The CP 25 tonner runs more or less steadily at speed step 40 but there are interruptions in the sound when slowed down to speed step 20 and it stalls a couple of times. Below speed step 20 it was hopeless. It stalled constantly. I stopped recording because it was so bad. If you want to see it, please ask and I will do another video.
The box cab on the other hand suffered no interruptions in sound even down to speed step 1. It crawled over the frog even though one of the wheels dropped noticeably into the gap between the rails. It runs so slow its like watching paint dry!
Here is the CP 25 tonner:
Here is the box cab:
It is interesting that Geoff Baxter from Hollywood Foundry was concerned that the BullAnt would not perform very well without a flywheel. I’ll send him a note.
The next step will be to see if I can get the CP 25 tonner apart to replace the flywheel with a Power Pack.
I’m not sure what the issue was. All I know is that when I did the BEMF thing with CV 54 the problem was solved.
I’ll have to look up CV 116 to see what I should do with it. Any suggestions?
EDIT: The Loksound Select manual does not refer to CV 116 directly in the index of CVs. Since the locomotive is running very nicely I don’t see the need to play with CV 116, unless you suggest otherwise. I haven’t got the patience to re-read the whole manual right now.
I also checked the CV listings for V4.0 and I couldn’t find any direct reference to CV 116 there either.
116 is not a user-defined CV in Loksound decoders … it is in Tsunami decoders though.
I had the same wheel drop problem with some of my turnouts. I added a plastic shim to the bottom of the frog, just thick enought for the flange to contact it. So when the wheel tread lost contact in the gap, the flange took over and kept things level.
Or was it cv113? Sorry guys…it’s the cv that you set to tell the decoder it have a keep alive attached…if you don’t set it…the keep alive has no effect.
CV113 sets the amount of time the Keep Alive module last. With a range of 0 to 255, the factory default is set to 50. Set it to 255 if you want the longest run time from your stay alive module.
It’s the TCS decoders that need to have a CV set to enable / disable the use of a stay alive module, not Loksound.
I don’t like the way it behaves over the frog but I deliberately chose a turnout with a large frog to see how it affected the locomotive. Despite the uneven rolling I think it proves absolutely that the Power Pack works wonders. When the box cab drops into the gap two of the four wheels are without power (the opposite rear wheel is lifted off of the track). It climbs back out of the hole every time. The other critter with the flywheel only stalled every time across the frog at anything below speed step 20 of 128. I was very pleasantly suprised with the slow speed performance with the Power Pack. In fact, it is almost too slow to be useful.
As I said, I haven’t done any tuning of the turnout so one aspect of that will be to shim the bottom of the gap at the frog as Mark suggested.
I’m going to leave the decoder at the factory settings as far as the Power Pack is concerned. They seem to work well.
Check the wheel gauge on the loco, too. Every loco that I had that bounced of Atlas #6 turnouts was proven to have wheels out of gauge when checked with an NMRA gauge. Adjusted the ones that needed it, they then were smooth even at low speeds.
I checked the wheel gauge on both critters and they are spot on.
However, I discovered a wee discrepancy between the two locomotives. The CP unit does have code 110 - 33" wheels, but the box cab has code 88 - 33" wheels. The mistake was mine when I ordered the second BullAnt. I had intended to order code 110 but obviously I checked the wrong box. Duh!
Rather than messing with 40+ turnouts I will give Geoff at Hollywood Foundry a call to order a set of code 110 wheels.