rrinker(remember I use Digitrax, which does NOT shut off the main track while running the program track - I still don't understand why NCE does that
I think it allows a siding or section of layout that is gapped to serve as a program track, and that section is also connected to the main layout. You can run a locomotive onto the section, switch to program mode, which affects only that section of track, and program that loco without removing anything from the tracks.
York1 John
OP said all the locos have the same address, so to program just one, you have to pick it up and put it on the program track to change just that one loco. Or else they will all get the same change to each CV, negating any change.
They really should have individual addresses, then you can truly adjust them all on the fly, if you have enough continuous running space to let all of them run with only the loco being adjusted actually being under throttle control.
If they all have the same address, and you pick one up and adjust it on the program track (remember I use Digitrax, which does NOT shut off the main track while running the program track - I still don't understand why NCE does that), and then set it back on the main - it will immediately start running at the speed step that the others are going, as the command station is continuously sending speed and direction packets.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
hon30critterThat is the simplest and most easy to understand explanation of speed matching that I have ever read!
Except maybe for the programming of CV's 2, 5, and 6 on the programming track. I believe that normally any CV adjustment of speed related CVs is done on the main. After all, after adjusting one of these three CVs you need to run the loco to see what effect on speed the change you made was. Far as I know you can't do that on a program track.
railandsail I read this just recently from an experienced gentleman,... Put bluntly, you can not mix sound decoders from multiple vendors on your layout without becoming totally frustrated. Function key mappings vary, braking behavior and momentum effect settings vary. I'm sorry , but trying to consist locos with sound decoders from different vendors just does not work. That means you need to standardize on one vendor's sound decoders, and stick with them. Most layouts I have visited have sound decoders from multiple vendors, and those layout builders are quick to express their headaches to me. Even getting the function keys to work the same across multiple decoder brands leads to early baldness from all the hair pulling.
I read this just recently from an experienced gentleman,...
Put bluntly, you can not mix sound decoders from multiple vendors on your layout without becoming totally frustrated. Function key mappings vary, braking behavior and momentum effect settings vary. I'm sorry , but trying to consist locos with sound decoders from different vendors just does not work. That means you need to standardize on one vendor's sound decoders, and stick with them. Most layouts I have visited have sound decoders from multiple vendors, and those layout builders are quick to express their headaches to me. Even getting the function keys to work the same across multiple decoder brands leads to early baldness from all the hair pulling.
Put bluntly, you can not mix sound decoders from multiple vendors on your layout without becoming totally frustrated. Function key mappings vary, braking behavior and momentum effect settings vary. I'm sorry , but trying to consist locos with sound decoders from different vendors just does not work.
That means you need to standardize on one vendor's sound decoders, and stick with them. Most layouts I have visited have sound decoders from multiple vendors, and those layout builders are quick to express their headaches to me. Even getting the function keys to work the same across multiple decoder brands leads to early baldness from all the hair pulling.
This a fallacy. I use tsunami, tsunami 2, Locksound 3.5, Loksound 4, NCE and Digitrax decoders and even a few Lenz from. My go to for sound are Tsunami and Loksound depending on price.
The key to making them run together is making sure they ALL run together. I don't speed match to another locomotive. I speed match to this chart.
I do this on my work bench by having my locomotives on a set of rollers. I have a rolling speedometer (bachrus) which unfortunately is not longer in production taht works as part of JMRI. Bachmann makes a set of roller stand which i use to support the other wheels that are not on the speedometer. I start with the high speeds and work my way down. I program speed steps 1-7, 10, 14, 18, 22, 26, and 28. I let the sofware do the steps in between. This can be done with any decoder that supports speed curving.
It generally takes me about 30 minutes to do this per locomotive but in the end it means every locomotive in my fleet will run with any other. The only locomotives I don't do this to are switchers as they always independently on my layout. MY A-B-A sets of F units have mixed decoders and I have no issues matching them since I don't put sound in every unit.
For Sound locomotives I use what I can get, Tsunami, Loksound, Tsunami 2. For non-sound I have pretty much settled on NCE decoders.
My main requirement is that they have the ability to be speed curved.
Additionally any modern sound decoder supports function remapping, especially the Loksound. Its very easy to standardize functions across a fleet.
The key to this is taking the time to do it. I do it as part of decoder installation/ locomotive check in process. Every locomotive goes through this before it is allowed to run on the layout (other than testing).
I also program the locomotives on the fly. Once I get the into the system in JMRI, I use OPS mode program to tweak the speed curve as I go.
The last thing I do is apply momentum. This takes a bit of guesswork but I have narrowed down my settings pretty well depending on decoder type.
Colorado Front Range Railroad: http://www.coloradofrontrangerr.com/
I just recently speed matched a set of A-B-A F-units that all use the same address and run as a unit together. What I found worked best was to temporarily assign each a different ID number while doing OPS programming. I have a fairly long stretch of double track so I put one engine on each track and then adjust one so it matches the other in performance. Then when that one is done I take it off and add the other and adjust that to match the first one. Once they're all working together, I change them to all have the same ID number.
I tend to be a little fussier than most, I adjust speed and momentum CVs until the engines are as close as I can get them to starting, running, and stopping together as one unit.
My club has two programming tracks side by each: one hooked to JMRI; one set up for an NCE throttle. The test track for the NCE station is about 10' long. What I do for speed matching is this:
- Place the two locomotives on the NCE test track (as noted about 1 foot apart), consist them and see how far off they are. It is advisable to turn off all momentum on all the locos first.
- The unit that needs to be adjusted is removed from that track, placed on the JMRI and is set up in DecoderPro.
- I adjust the settings on the low value on the Motor Tab, save changes to that sheet only, and place it back up on the NCE track and test it.
- Adjust as needed. - Adjust the other values as needed.
I call it done when coupled they aren't straining the couplers. Don't worry if it's not 100%. It won't be. You don't need to take them out of consist until you are done.
The longest portion is having DecoderPro read all the values. While I only really care about 1-2 sheets, I usually just let it read all the sheets in case I want to adjust the sound levels. It doesn't matter if the units have the same DCC address since you are only writing the changes to the hardware on that specific decoder.
The only issue I have run into is the Bachmann DCC-On-Board decoder in my SD40 doesn't allow me to set much on the speed values. I am going to replace that decoder.
I make the same point over and over as well. I find it seems to be one of those "well, since you CAN..." but that doesn't mean you HAVE to. Even with an outlier, so far I have had no need to resort to the 20 step speed table, which really ends up being a total of 30 CVs to program. Nuts, i say. Just the 3 step settings, a total of 3 CVs to set, has thus far taken care of anything I needed to adjust.
But since so many of my locos are all the same - lots of P2K's - they all pretty much run the same with no adjustment. And the others that will tend to need to be run together - they are almost all Atlas/Kato. I'm not really deliberately selecting them to be the same, it's a matter of who made what, and instead of one of everything, I have groups of certain classes.
Randy, agreed, DCC does not really change any of that. I make the point about DC simply because so many people think speed matching is some real important deal.
In fact, the deeper we get into the modern age of model train production, the more these locos are geared similarly and the more they run the same, DC or DCC.
Sheldon
ANd DCC changes none of that - if they could work together in a train with DC, the same locos will work in a train together after installing decoders, provided you didn;t install decoders with vastly different operating characteristics. Even BEMF and non-BEMF decoders can work together, but if the two locos have motors that needed a decent bit of throttle to start in DC, then a BEMF decoder is going to get it moving well before a non-BEMF decoder on the same loco.
I have some context now - I still haven't received my notification that the new issue was available but I checked my download page and read it last night and I see where Brian's quote came from. A little context, Joe was talking about using something like the Proto Throttle to run trains, and yes, different brands of sound decoders do the braking differently, so even making the functions keys match, you might have issues using the Proto Throttle's brake with mixed brands of decoders - that's also one of the reasons I am considering changing the non-sound decoders I use, because the non-sound Lokpilot decoders support the running features (but obviously not sound) of the Loksound, so the brakes work, the drive hold for manual notching works, etc.
But to just grab a pair of locos and pull a train, sticking with the basics like horn and bell works fine with mixed brands. So long as it is just sounds being triggered, and not something that actually alters the running of the loco, mixing brands in consist is not a problem.
What you CAN do with DCC is fairly easily take two locos that don't run anything alike and make them run close enough that they can be used in consists.
To go to an extreme - I usually pull my 31 car (plus caboose) train on the club layout with a consist that is P2K GP7 - Atlas Trainmaster - P2K GP7. This is an order an old railroader told me the crews would try to get set up if they had a mixed consist of Geeps and Trainmasters, because the Geeps rode better. In my case, the two Geeps have TCS non-sound decoders in them, and the Trainmaster is an Atlas Gold with QSI sound decoder. I'm done NO adjustments to CV2-6-5 or speed tables. They are actually consisted using basic consisting - they all have the same address. The only thing I did was adjust the light functions so that none of the lights come on in the TM, and in forward, the lead Geep's headlight comes on, and in reverse, the trailing unit's headlight comes on, so there aren't any lights in the wrong place. They run fine, at all speeds - we are fully signaled so you need to reduce speed or stop depending on the signal indication, not jsut freely barrel around the layout.
ANd if you've only ever pushed or pulled 4-5 cars on 18" radius track - even on the 28" radius or so that the club has, when you move 31 cars by hand, you know you are moving something.
railandsail I thought you really needed to speed match very close if you were going to run some locos midstream in the consist, or as pusher service??
I thought you really needed to speed match very close if you were going to run some locos midstream in the consist, or as pusher service??
Well, maybe yes and maybe no.
You need to watch Dr Wayne's video of mid train helpers and pushers on his DC layout...........
Those of us in DC have been doing pushers and double/triple headers for decades without speed matching.
The key is this, all locos need a similar starting voltage and similar top speed. The starting voltage being more important than top speed.
If the train really requires the extra power, and the starting speeds are close enough, and the cars are properly weighted and good tracking, generally it works fine.
But if you are talking about twelve cars with two locos of different speeds on each end, around 18" radius curves, expect trouble........
I'm not so much into the pusher or mid train thing, the latter was pretty rare in my 1950's era.
But nearly all my trains are pulled with multiple powered locos, many are steam of mixed brand and wheel arrangement. And they slug my 35 to 60 car trains around the layout just fine - without DCC or speed matching. 35 cars at 3.5 to 5 oz each evens out the speed/load just fine.
And nearly all my freight cars have metal sprung/equalized Kadee trucks refitted with Intermountain metal wheels for superior tracking and low rolling resistance.
Brian
My Layout Plan
Interesting new Plan Consideration
You want to have the locos somewhat close in speed before consisting them.
Other than that, it makes no difference if the decoders are all the same brand, or different ones. Even with sound - in the real thing, when locos are in a consist, the horn and bell only work on the lead unit. Most DCC systems handle this automatically when you consist locos, unless you consist them by just setting each loco to the same address.
I thought 'speed matching' was an element of 'consisting' ? Perhaps I just don't fully understand the terminology yet.
Most sound locos these days end up with way too many functions that a) no one can remember all of them and b) many of them are just useless gimmicks to make it look better than another brand - station announcements, farmyard sounds, radio chatter (in a pre-1950's loco - riiiiight). And most allow decent function remapping, so you cna put the things you actually need all on the same function key.
There are good reasons to standardize, to make life easier when it comes to programming them, for example. And when you work with a decoder that allows you to change out the sounds, it means you cna always use the same decoder, and just load whatever sound set you need. I can't be the only one who never finishes projects in the order I start them in - so if I decide I want to finish another RS3 instead of work on the boxcab, I don;t have to go order another decoder, I can just use the one I was going to put in the boxcab and swap the right sounds in.
ANd having similar running characteristics right out of the box makes setting them all up to run together easier as well. So much so that I will probably switch to a different motor decoder for non-sound locos, so they get the same running characteristics as the sound version. I'm not going to rip out all my old decoders, but going forward, they will all be even more identical.
I don't think there is any mention of sound here Brian, just speed matching locomotives that all have the same address.
I have the DB150, and does not support a program track, so I use the layout main, with only the loco I'm programing, on the track.
Mike.
My You Tube
On systems with a dedicated program track: anything on the program track gets programmed. Loco address does not matter.
On systems that support programming on a program track but don;t have a dedicated connection for one (PowerCab, Digitrax DB150), the whole layout becomes the program track and the above applies.
For programming on the main, or ops mode as some systems call it, program commands are directed, on the main track, to the decoder address selected. Of course if there are 3 locos on the track with the same address, they would all get programmed. But nothing says you can;t park two of them in an isolated siding, or take them off the track so only one gets programmed. Of if when speed matching, one of the three starts up faster than the other two, but the other two are pretty evenly matched. They both have to have CV2 increased so they start with the faster one, so take the faster one off, and program both slower ones at the same time.
Generally you only want to program one loco at a time. But I could use a DCC system like a DC power pack and leave a collection of 20 locos all set to address 3, and set unique configurations in each one. Programming information is all stored in the decoder, not the system. So if you program one loco #3 so that F3 is a short horn toot, the next one that is also #3 does not necessarily have F3 as a short toot unless you then program that one.
I agree! I thought to program locos, they have to have their own number.
Thanks Randy.
Hopefully, the OP has his answers.
Randy,
That is the simplest and most easy to understand explanation of speed matching that I have ever read!
Thank you!
Dave
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
Sure you can, doesn;t matter how they end up running together - same address, Digitrax's command station based consisting, CV19 advanced consisting.
If you ARE running them all on the same address, to adjust one in Ops Mode on the main, you have to take the others off the track so they don;t get programmed as well. Or pick up the one that needs to be adjusted and put it on the program track, and program using the program track.
It's really straightforward, at least if you have enough room on your layout and a continuous run. Just put them all on the track, NOT coupled to one another, but a foot or two apart, then see what happens when they start moving as slow as they will go. I quite frankly refuse to use decoders that do not support CV2-5-6 for 3 step speed tables, so I keep it simple and never mess with complex speed tables. It's not necessary. So at whatever speed they all start moving, you need to increase CV2 on the SLOWEST ones to match the fastest one. (I am assuming no one has messed with these locos before and the fast one isn;t fast because someone already put too high a value in CV2).
Then try full throttle. Here you will need to adjust CV5 on the FASTEST one to match the slowest one.
Finally run them at half throttle, 50 on a Digitrax throttle with a display. Here you can do whatever you want, adjusting CV6 to speed up the slower one, or slow down the faster one. Your choice.
They do not have to be absolutely perfect, if one gains on another a little bit each lap around the layout, it's fine - once the weight of a train is on them, they will even out and each pull their share. If one loops your layout in a minute at half throttle, adn the other takes 2 minutes - they need to be adjusted.
There's really no need for fancy speedometers or anything. Yes, there is a script for JMRI that automates things using one of the roller stands that has a speedometer in it, but it's no big deal to just do this yourself. And close is good enough. People ran multiple locos on a train long before DCC and the ability to adjust the running of each loco like we can now, and it worked fine. DCC doesn't change that.
I'm not sure if you can, if the locos have the same address.
I have many locos with the same address, and put together locos that run well together, just like I did in the DC days.
If there close, they tend to even out and run fine together.
I apologize for posting a question that has been addressed a bazillion times in the past... I read a reference to a MR issue that addressed the issue of "speed matching" multiple locos w/same address. I would appreciate any tips on how to speed-match multiple loco, using Digitrax. Thanks