CV & Decoders

[Bernard]

One of my trains, the Kamome, is running slower than the other trains in my fleet. I've taken it apart and clean and lubed it and it still runs slow. I was wondering if it could have anything to do with the lenz decoder I've installed and/or the CV settings. Any thoughts?

[alpineaustralia]

CV05 operates the maximum voltage from values between 0 (0%) to 255 (100%). Have you tried changing this value to 255 (ie 100% voltage)?

[Martijn Meerts]

Not all motors will have the same characteristics. Even 2 motors of the same type and made in the same batch can have different performance.

 

A nice thing about DCC, is that you can set up a piece of track to to speed measurements using a computer. It'll measure the speed of a locomotive for each speedstep (or for a certain set of speedsteps, and calculate the rest of them). You can then adjust the speed curve a bit, and re-do the test. That way you can set max speed of trains by km/h rather than speed step.

 

I've read many reports on various forums about people who have the exact same loco, digital decoder factory installed, running on the same layout with default settings, and they can still show different running characteristics.

[CaptOblivious]

I do what Martijn suggests: I set up an oval of track (8 * "c280-45" + 2 * "s280" = 2319.24mm), take a loco, set midpoint voltage to just above start voltage (because it messes things up if it is higher than max voltage), and systematically vary the max voltage. Each change to max voltage, I do 10 laps around the track (fewer when I get to the really slow speeds: I can get my KIHA 110 down slow enough that one lap takes about 15 minutes) and carefully time it. I then use a spreadsheet to calculate the scale speed of the loco, and to calculate values for a 28-step speed table, carefully noting the max speed of the prototype. This is accurate enough that I have nearly perfectly synchronized several locos with very different motors each.

 

Incidentally, I've found that changing the values for BEMF functions, at least with Digitrax decoders, can really drastically alter the actual top speed without changing the max voltage CV. This is really annoying, I think. But, if you have a Digitrax decoder, you can boost the top speed slightly by increasing the amount of BEMF applied. This does bad things to low-speed operation, though. Do Lenz decoders do this too?

 

If you have a Mac, and use Apple's Numbers, I am happy to share this spreadsheet with others. I don't know if it will export correctly to Excel (I don't have a copy to try it on, either), but I can give that a try. It also serves as a handy reference for the CVs you've programmed for each loco.

[Martijn Meerts]

Cap, thanks for the offer, however due to my collective nature and (too) fast growing amount of trains, I think I'll invest in a computer program that'll do the measuring and calculating for me =)

 

For programming I usually use JMRI DecoderPro. Makes it a lot easier to program a decoder like the Lenz Gold with it's silly amount of CV's. On top of that, it keeps all the settings in xml roster files.

[alpineaustralia]

Cap

Can I take you up on the offer of the spreadsheet?

On another note I might also need you to help me out on headlights and tailights beacuse no matter how many times I read your posts (and I am sure that I am following them) my headlights/ tailights dont work. If you wouldnt mind, I'd like to send pictures of what I have done and perhaps you could tell me where I am going wrong. Could you do me that favour?

[CaptOblivious]

Cap

Can I take you up on the offer of the spreadsheet?

On another note I might also need you to help me out on headlights and tailights beacuse no matter how many times I read your posts (and I am sure that I am following them) my headlights/ tailights dont work. If you wouldnt mind, I'd like to send pictures of what I have done and perhaps you could tell me where I am going wrong. Could you do me that favour?

 

Yes, I would be happy to!

[Bernard]

Me too, can I have a copy of the spreadsheet! And I work on a Mac.

[CaptOblivious]

The spreadsheet is attached. Here's how it works.

 

On the scale speed page, you'll find a table with 5 rows, labeled 0, 32,

64, etc.

This is where you enter the time-trials data. Set the max voltage to some

value, and put that value on the first column. I start my trials with the

values 0, 32, 64, 128 and 255, hence the first 5 rows, but you are not

limited to those values. Just add another row to the table to add a

time-trial for a different max voltage.

IMPORTANT: Make sure the table is at all times sorted by the first column,

that is, that the values of the "CV 05 Value" column are always increasing

from the first row to the last. Otherwise that neat little chart below it

(and everything else) gets messed up.

 

Anyway, then enter in the total length, in mm, of the track you'll be

running on. If you're doing an oval, there's a calculator at the bottom of

the page to help you out with that.

 

Then, run your train at max voltage, and use a stopwatch to time it. Enter

the number of laps the train ran ("1" if you just ran it down a straight

length of track), and the time on the stopwatch, and the sheet does the

rest, calculating the speed in scale km/h and mph.

 

Once you've got enough data points (how much is enough? when your tired, I

guess), go to the "CVs" page, and enter the CVs in. I've got it configured

right now for Digitrax CVs, so some of yours might differ, just change the

labels to reflect your decoder.

 

Anyway, as you enter the values for CV02, CV04 and CV05 (voltage start,

max, mid), and again for CVs 65--95 (28-speed step table), the chart in

the upper right will change to show the speed of the loco in scale km/h at

each throttle setting, so you can tune the speed tables to give just the

response curve that you want, given the performance characteristics of

your particular loco.

 

Anyway, it's much easier than it sounds. Just play around with it, and

you'll see how it works. Do ask questions if you have any! And I'm always

open to suggestions on how to improve the sheet.

 

CAVEAT! BEMF settings mess with the throttle position-speed step-voltage relationship! If you change any of the CVs that control BEMF, you will have to re-run all your time trials! I don't think I can do anything about this, because the effects seem unpredictable (but I could be wrong!)

Speed.numbers.zip

[Bernard]

CaptO - Thanks you for the spreadsheet.

[alpineaustralia]

Capt - many thanks for the spreadsheet.

Unfortunately, you were right - I am having trouble opening it on XP.

[CaptOblivious]

Capt - many thanks for the spreadsheet.

Unfortunately, you were right - I am having trouble opening it on XP.

 

That's why I said "If you have a Mac, and use Apple's Numbers, I am happy to share this spreadsheet with others."  ;) I can see about trying to get something similar working in Microsoft Excel, but seeing as how I don't have a copy, it could take a little while...