Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Differences in speed from different manufacturers

4013 views
14 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    December 2018
  • 6 posts
Differences in speed from different manufacturers
Posted by NS Fan on Thursday, December 27, 2018 4:29 PM

I was running a 2 loco consist using WiThrottle and with a MTH CW40-9 and BLI SD40-2.  I saw that the  MTH was struggling becuase the SD40-2 was holding her back. It seems my SD40-2 (and BLI SD9) ran a lot slower than all of my others so I started comparing them, side by side.  From slowest to fastest right out of the box  (of the ones I've dealt with so far) are BLI (SD40-2 & SD9), Atlas (GP38-2), MTH (CW40-9) and Athearn Genesis SD60E (9-1-1, First Responder). The Athearn is a fireball unless you toggle the 'half speed' button and its still the fastest even so.  I've played around with CV3,4,5&6 in the BLI's and the Athearn with little noticible change.

Can anyone suggest something to get some sort of speed matching?

Thanks,

Dave

  • Member since
    January 2011
  • 892 posts
Posted by PennCentral99 on Friday, December 28, 2018 9:39 AM

Just like Ford, Dodge, Chevy, Toyota, Honda, etc., they all run different. You need to get out your manuals and work on CV 66 (Forward Trim) and CV 95 (Reverse Trim)

Terry

Inspired by Addiction

See more on my YouTube Channel

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Friday, December 28, 2018 11:20 AM

 Those CVs ONLY apply if you use a 28 step speed table.

Not all the decoders support all 3 CVs from CV2-6-5. Which is hugely unfortunate, to the point where I no longer use any decoders that don't.

But there are some things that apply, even when these CVs are supported. First, with CV2 set as low as it will go (usually 0, a few only allow down to 1), that is as slow as the loco can go. So to match then, you have to speed up the slower ones - the one that runs the fastest on step 1 is your benchmark, if CV2 is already 0, it can;t start any slower. So the slower ones need increasing values in CV2 so they match at the start. At the top end, with CV5 at the default, at full throttle, the SLOWEST unit is the benchmark. By adjusting CV5 on the others, you can reduce their full throttle speed to match that of the other. This is probably all you really need to do, but for those that support CV6, you can do the same with half throttle, only in this case you can make ANY of the locos the benchmark and match to it since you can either speed up or slow down the half throttle speed by adjusting CV6 to some value between CV2 and CV5.

 Close, so one doesn't drag another around, is close enough. There is no need to make them so closely matched that you can run 5 laps of the layout and the distance between two doesn't change by so much as a fraction of an inch. 

                                             --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    May 2010
  • From: SE. WI.
  • 8,253 posts
Posted by mbinsewi on Friday, December 28, 2018 12:11 PM

Question, with the MTH in the mix, would it require some type of program from MTH to change CV's.  I'm not even sure if MTH calls them CV's.  Confused  I think they call it DCS ?

Mike.

Moderator
  • Member since
    June 2003
  • From: Northeast OH
  • 17,205 posts
Posted by tstage on Friday, December 28, 2018 3:48 PM

Mike,

DCS is the system but I believe MTH still refers to the programming variables as CVs.


Dave,

Even locomotives from the same manufacture can vary in speed.  And this can be contingent on a number of things: gear ration, linkage (steam), motor, lubrication, etc.  The key is to speed-match to the lowest common denominator.

The easiest way to accomplish that is to adjust CV 2 (VStart) & CV 5 (VMax).  Some decoders also allow for CV 6 (VMid).  As long as you can match the top and bottom speeds, as Randy suggested, you should be just fine.

Tom

https://tstage9.wixsite.com/nyc-modeling

Time...It marches on...without ever turning around to see if anyone is even keeping in step.

  • Member since
    December 2018
  • 6 posts
Posted by NS Fan on Friday, December 28, 2018 7:59 PM

I've tried playing with CV values in 5&6 on the Athearn with no change.  I slightlly modified the BLI's which sped them up a LITTLE.  I'm waiting for my interface cable from my Mac to my MRC system and will be giving this a shot using thre JMRI software.

  • Member since
    February 2002
  • From: Mpls/St.Paul
  • 13,793 posts
Posted by wjstix on Thursday, January 3, 2019 4:29 PM

Depends on what type of decoder is in each engine. Each manufacturer does things a little differently. On some, if you've enabled speed curves, the 'normal' speed CVs (2,5,6) don't work. Some support CV 5 but not CV 6. Also, not all CV values work the same. On some decoders, CV 6 can be 1 to 256, on some it only goes up to 64. Any number over 64 will just result in the same as 64 - full speed.

As Randy says, you can't use a CV to make an engine go faster than it's top speed. If engine A is faster than engine B, you have to slow down engine A. You can't speed up B.

Stix
  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, January 3, 2019 4:40 PM

 Make sure you don;t have CV29 set to use a 28 step speed curve, or CV2-6-5 will be ignored. Otherwise, you should see some very different operation if you reset CV5 to the default vs setting CV5 to some smaller value, like 127 if the decoder goes to 255, or 32 if the decoder only goes to 64 (not sure which ones do that - ESU has a limit like that on the MOMENTUM CVs). 127 in CV5 as opposed to 0 or 255 should run at half the speed at full theottle (step 1 will not be affected). If the decoder supports all 3 values, then CV5 > CV6 > CV2, any other order and the decoder may do some really strange things - all that means is the top speed setting in CV5 has to be bigger than the mid speed setting in CV6 which has to be bigger than the start speed setting in CV2. Anything else doesn;t make sense anyway.

 The Athearn with a Tsunami or a custom Tsunami won't likely have CV6, a newer one with Tsunami 2 will.

                                        --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    June 2004
  • From: From Golden, CO living in Puyallup (Seattle), WA
  • 750 posts
Posted by Renegade1c on Thursday, January 3, 2019 5:03 PM

Speed matching is a bit of a art in itself and everybody has their own method of doing it. 

Personally, I do not speed match locomotives to other locomotives directly. I set the speed of a locomotive based on speed step. See chart below. Green highlighted speed steps are the ones i directly program. The lighter ones are interpolated (more on that below). 

 

Speed Chart

I have a set of rollers and Bachrus speedometer  on my test track on my work bench. Each locomotive is placed on the roller ands I use JMRI to adjust the speed steps until they match the corresponding speed of the locomotive. This is done with Programming on the Main, which means the changes take place immediately, requiring to stopping the locomotive to change the CV. 

I start at the highest speed steps and work my way down to the lowest. This gets the motors up to operating temp and gets the lubrication to what it needs to be during normal operation. 

I would highly suggest you get all your locomotive maintanence in before doing this since it will really make the process much easier.

I don't match all 28 steps (decoders run at 128 but everything is interpolated between the 28 steps. I use the check boxes in JMRI to indivdually set speed steps 1-7, 10, 14, 18, 22, 26 and 28. All other speed steps are linearly interpolated by JMRI and are close enough that I haven't noticed. 

I have done this with approximately 25+ locomotives now and they can all run in sync with each other. Manufacturers include athearn (BB,RTR and Genesis), Proto 2000 (Life like and Walthers), Atlas, Kato, and even a few brass models. 

I have found that certain decoders (namely early digitrax) do not support speed curves which means they got replaced. My standard fleet decoder (NCE DSRJ13) does support speed curving. Almost all modern (Tsunami, LokSound, TCS wow) sound decoders support speed curving. One thing that has also helped is putting similar type decoders in sets that will stay together. For example, I have two A-B-A sets of F-units. The B-unit has the Soundtraxx Tsunami sound decoder and the A-Units have soundtraxx non-sound decoders. This makes the speed matching much easier since it has the same electrical drive circuitry but no sound. 

It takes about 30-45 minutes per locomotive but its very much worth it since all my locomotives run together no matter the manufactuer or decoder. 

 


Colorado Front Range Railroad: 
http://www.coloradofrontrangerr.com/

flag

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, January 3, 2019 7:09 PM

 I had someone from the NCE list send me some test results from his club, showing how the NCE decoders don't properly interpolate when running in 128 step mode - which I think is one factor why a lot of NCE users say there is no difference between using 28 step (which NCE systems default to) and 128 step mode. 

 Speed curves - the biggest thing DC users laugh at us DCC users over. I've just never seen the point. I toss any decoder that doesn;t support CV2-6-5 (CV6 is the one usually left out, sorry Tsunami) and using just those 3 I cna get even performance for ewach class of loco. Poor drive/poor implementation of BEMF (sorry, Digitrax) are also out, so my locos all start on step 1. CV5 is set so the top speed is something reasonable for the prototype - I may turn that down for my layout so that wide open the locos run not at their maximum prototype speed, but at the maximum speed that suits the compressed nature of my layout (or anyone elses, unless they build in an airplane hanger). CV6 comes in to tweak response over the new speed range, more powerful locos I can push it above the exact midpoint between CV2 and CV5, for faster acceleration off the line, for locos more suited to drag freights, I can set CV6 slightly below the mid point, so it takes longer for them to get moving.

 In reality today - few of my locos have any settings other than CV2 to make sure they start on step 1, and most of the BEMF ones, at least ones with good BEMF, like TCS and ESU, don;t need CV2 raised for that to happen. An MU consist I use at club shows has a pair of P2K locos with TCS T1 decoders and an Atlas Trainmaster with QSI. I did nothing to them - they run fine together, yet if I put them on the track a few feet apart, eventually one will run into the other. But coupled to a train - they start up smoothly and pull the train for hours at a time. If I do something silly, like go from stop to full throttle instantly, then one drags the other two for a little while. But throttle up smoothly - they MU together just like DC locos, in other words, no special work to make them pull together. 

 One thing is, I standardized on all the same decoders early on. I never wuite understood buying all sorts of different brands of decoders, at least not in HO scale. N scale can be a bit different because it's not easy to just make one decoder work in all locos, and not all manufacturers have a decoder for every brand of N scale loco. But in HO - just save yourself a lot of grief and buy the same brand decoder for them all. I use TCS for motor, and ESU for sound (that Atlas Trainmaster is my only sound loco with a different decoder). That helps a lot to get them to run together. It also, sinc emy era and prototype has no fancy light effects, makes it super simple to program, JMRI not needed. Because rally all i do is set an address. Maybe tweak CV2. No multiple CV setups to remember, or need JMRI to do for me, since all my locos are the same it wasn;t hard to just remember the CVs I need, which is few indeed. The system handles the address - all current systems do that, so no calculating CV17,18 and 29. No ditch lights or flashing beacons, so no figuring out what value makes the ditch lights turn on witht he headlight and flash with the horn. And since I doon't use 28 step speed tables, that eliminates anoother good reasoon to use DecoderPro. 

 Yes, I am quite opinionated. Now get you kids off my lawn Laugh

                                        --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    December 2018
  • 6 posts
Posted by NS Fan on Friday, January 4, 2019 4:01 PM

I received †he MRC 1516 USB cable I canno† figure out how to conenct my Mac to the MRC Prodigy Advance2 command station.  Is there a good forum where I can get help on †his and other issues I will have with JMRI going forward?

Thanks,

Dave

  • Member since
    December 2018
  • 6 posts
Posted by NS Fan on Friday, January 4, 2019 8:30 PM

To be clear... yes I can plug in the USB connector and connect the other end to a 'Cab' jack.   PanelPro preferences, setting up the MRC connector only shows a 'bluetooth' port and no serial port.

  • Member since
    October 2005
  • 1,033 posts
Posted by betamax on Saturday, January 5, 2019 6:15 AM

NS Fan

I received †he MRC 1516 USB cable I canno† figure out how to conenct my Mac to the MRC Prodigy Advance2 command station.  Is there a good forum where I can get help on †his and other issues I will have with JMRI going forward?

Thanks,

Dave

 

 

You will need software, and I doubt MRC even considered having something written for the Mac OS.

Since JMRI supports MRC, go to their website (jmri.org) or just google "jmri mrc".

For support, join the JMRI email list at Groups.io.  Many of the developers hang out there, so when you run into problems, they can help.

Current version of JMRI is 4.14. Follow the instructions to install it, and be sure to install Java first (or check that it already installed and the correct version).

Moderator
  • Member since
    June 2003
  • From: Northeast OH
  • 17,205 posts
Posted by tstage on Saturday, January 5, 2019 8:47 AM

I totally agree with Randy on standardizing decoders.  While I'm an NCE guy, I'm not particularly a big fan of their decoders.  TCS has been my go-to decoder brand for a few years now and their motor-control is hard to beat.  (Only ESU comes close.)  Rarely do I have to tweak anything, as they generally roll along at speed step 001 at 1sMPH or < right out of the box.  Makes any speed matching fairly easy; since all I need to do is match CV5 (VMax) to the top speed of the other locomotive - or vice versa.

Tom

https://tstage9.wixsite.com/nyc-modeling

Time...It marches on...without ever turning around to see if anyone is even keeping in step.

  • Member since
    February 2010
  • From: Baden, PA
  • 96 posts
Posted by wm #734 on Monday, January 7, 2019 8:48 AM

The thing about MTH decoders that I haven't seen anybody mention is that they don't use your typical speed steps. With MTH locos each speed step equals one scale mph. Just keep that in mind when speed matching them. Don't try to change the MTH loco, change the other ones to match it.

       

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!