Funky programming on LokSound Kato E units.

I have a strange problem. I have 2 new Kato/Kobo PRR E units and the speed range is compressed into 60% to 99%. They don't move, although the sound ramps up at lower power levels correctly. It sounds like my engines are at run 6 and they barely move. Is there something else I need to try before I adjust the speed table? It just seems odd to me that I need 60% power to get these engines to move. My Kato/Kobo UP E units don't have the same issue.

Sent from my SM-T837A using Tapatalk

 

Is back EMF turned off or on? I also found something called 'reference voltage' on ESU decoders that will raise the top speed if the model is too slow. Have you tried CV 54 to zero? Watch some videos on ESU autotune to see what this does. The autotune will give decent Back EMF values, but it will often drop the reference voltage. You need to raise it back up to get higher speeds, especially if your speed table is already topped off.

 

I turned the back EMF off; with my grades, I figured that KISS is good for this newbie. Watched my first attempts at MU have the push-pull syndrome, so I disable it for now on all my engines. I'm happy with the top speed of both engines, but I'd really like them to move with less than 60% power. For reference, the UP Es start moving at about 4% power. Thanks for letting me know about the autotune videos.

Sound does not affect the start voltage. I'll make sure I try your hints first.

Thanks both!

 

I would agree with Rick. You pretty much need Back EMF with ESU decoders. I was having the same issue with a LokPilot when I turned Back EMF off.

The idea that Back EMF needs to be off for consisting is a bit of a myth. You need Back EMF for low speed control, but the way to fix the push-pull bumper car syndrome is to actually lower the motor control sensitivity. On Tsunami decoders, the default is 255, but 100-120 works well in my consists. There are similar settings on ESU, and I think you can specify a cutoff value, when the Back EMF control drops to zero at a certain speed step.

 

Okay, so I turned the BEMF back on, and set CV54 to 0 for the autotune. Now they start at 25% with sound off, down from 40%, but they CREEP, less than a scale 1 mph! The UP E units start at 4% with sound on. Better, but still not what I'm wanting.

Since the UP E8s arrived set up, I didn't put them on the programming track. I had an idea to have JMRI read the decoder settings, and see if I can clone them for the PRR E units. Will let you know if that works. I figure it may take an hour or more to read all the settings on the UP E.

Sent from my SM-T837A using Tapatalk

 

When you run CV 54 Autotune and after it does the high speed thing, you 'read sheet' to find what values it came up with. I usually lower them a bit, especially if the model 'surges'. That is usually caused by the Back EMF overcorrecting, so lowering K and I a bit will help to prevent that, just don't go too far; then you get to the other end, which is just stuttering. There is a sweet spot in the middle that you want to find.

When you are reading the sheets for the UP E-units, are you reading all sheets or just the one? If you just read the speed table sheet, it should take a few seconds. The long one is the function map.

 

Posted to a separate thread. Getting a LokProgrammer to help with this. I expect these won't be the last LokSound units I get.

 

I'd like to get a LokProgrammer too. The features and capabilities of the ESU decoder are so cool, with function remapping and custom sounds. I would buy ESU just for the function button startup rather than startup with track power like Tsunami. I think getting them programmed can have a bit of a learning curve, but I think they are worth it.

 

I mostly used the All/Global CV pane, until I ran into issues on the TCS WOW121 Diesel. Sometimes I have more useful results when using the other panes. And when a CV read fails, that CV error is also reported on the Global CV pane, where it can be re-read individually.

I also noticed a paragraph at the bottom of the page Global CV Listing Pane (jmri.org)
...Those new to DCC can safely ignore this panel without losing any capability at all.

Issues I had (and which could be user error):

1. I was trying to see the Global Volume. This requires setting CV 201, 202, 203, and then you can read CV 204. In the Global CV pane, it omitted the write to CV 203. When I read it in the Sound CV pane, it usually worked. (I was able to see the individual commands in the JMRI DCC++ traffic window.)
2. There are 38 sound CV settings to read, and it seems the process to read full sheet begins with the complex CVs. So it was taking too long and too many errors.
3. Perhaps there needs to be an option to only read the single CVs, or to read them first (since you can cancel once it seems like there are mostly errors).
4. One issue may be the result of the GEN-MB1 motherboard, which includes KA2 Keep-Alive. It seems there needs to be a longer pause between CV reads.