Large jumps in speed on one step speed changes

[Tomasz]

I noticed this on several of my locos, there are positions on the speed slider where going one step up or down causes too large of difference in speed whereas the steps before and after the change is much smaller. Most noticeable to me is going between step 44 and 45, the loco speed change is quite pronounced. I tried this with several profiles and both with and without speed table selected (CV29 value 16), but I can't figure out a way to smooth this out. I might have to pick up the Piko measuring car that has a build-in LCD that displays actual scale speed so I can prove it to myself that I'm not going crazy. Anyone else noticed this? 

[dBerriff]

Sorry but no, I have not noticed this. Not on OO or TT:120.

Edited Tuesday at 15:43 by dBerriff

[Tomasz]

I'm wondering if this is a DC only issue. Are you running DC current or DCC?

[96RAF]

Look at the complex speed table as see what speed step value difference there is at the jump point.

[Fishmanoz]

But Tomas says it occurs without the complex table selected too.

[LTSR_NSE]

Is the basic a straight line?

I would have imagined not & therefore the observed behaviour even more likely under the basic. 🤷‍♂️

Also if Basic selected surely Voltage min, med, high will need to be equidistant otherwise behaviour will be observed.

Edited Tuesday at 23:09 by LTSR_NSE

[Tomasz]

The basic was with default settings of 0,72,255. I changed it to 0,127,255, which would be a straight line and initially I thought the problem was fixed, however all that did was move the point at which the speed jump happens from step 45 to step 25. So the jump happens at the same actual speed regardless of the speed map (45 * 72 / 127 = 25.5). Either basic or advanced map does not explain why the speed increase in that one step is at least 2-3 times larger than the steps before and after. 

[What About The Bee]

0, 255,255?

Thus linear below 255.  If the mid cannot equal high, then 0,254,255

 

[LTSR_NSE]

Also (considering the speed jump is happening between min & med)

0, 1, 255

[Rallymatt]

The basic speed curve should start at 2 or more usually 3 not 0. That is the motor start up value. 
Asides that, it appears the issue is the motor response at a particular voltage. It could be worth doing an auto calibration to see if that smooths it out. 
If you want to go full on 🤓 there are other tools in the system to ultra refine motor performance. The Famous Manual will guide you in all these things. 

Edited Thursday at 09:33 by Rallymatt
Extra info

[Tomasz]

Hmm, I did the auto calibration at the very beginning and I didn't notice this issue until after. I can't say if it was there before or not. Unfortunately there seems to be no way to go back to non-calibrated values once the calibration has been run. I'll keep playing with it. 

[Rallymatt]

Do all locos displaying this quirk have the same motor in? 
When you did the auto calibration did you remember to do a full decoder reset (CV8) ? 
You can always re-run a calibration again too. 
 

[LTSR_NSE]

Alternatively to get back to default (non-calibrated) values simply reinstall the sound profile.

[Tomasz]

Thanks @Rallymatt, I forgot about CV8 reset ability. 

Not sure if they are the same motor, the issue is most prominent in some of my diesel locos, they are mostly Piko models. Some are earlier with an 8 pin connection, some are later with a plux22 connection, but the motor might be the same part. 

[Rallymatt]

That wasn’t quite what I was thinking with CV8, before running auto calibration all CVs need to be reset to factory settings for it to be effective. CV8 should put motor control settings back to for comparison 👍

[Tomasz]

No, I got it, I usually do the calibration as first thing before I modify anything else. I'll keep messing with settings trying to figure this out, but mostly I was just curious if this was a me thing or if others saw similar behavior. Looks like it's a me thing. 😄