Author Topic: Speed matching - MRH newsletter  (Read 15304 times)

Alan

  • Conductor
  • ****
  • Posts: 1073
    • LK&O Railroad
Speed matching - MRH newsletter
« on: December 04, 2021, 07:34:28 PM »
Anyone get the Dec 4 MRH newsletter with a link to DCC loco speed matching thread?

https://model-railroad-hobbyist.com/node/41120#comment-452981

Man, am I glad I don't have to deal with that. What if they wanted to speed match every loco in their fleet so the could make any consist they wish? Even with my modest roster I can't imagine how many hours that would take to do.

With RailPro I just consist as I please without changing anything and 99% of the time everything runs fine. Thanks RailPro.
Alan

LK&O Railroad website

When I was a kid... no wait, I still do that. HO, 28x32, double deck, 1969, RailPro

AlexW

  • Fireman
  • **
  • Posts: 11
Re: Speed matching - MRH newsletter
« Reply #1 on: December 04, 2021, 09:19:06 PM »
Anyone get the Dec 4 MRH newsletter with a link to DCC loco speed matching thread?

Man, am I glad I don't have to deal with that. What if they wanted to speed match every loco in their fleet so the could make any consist they wish? Even with my modest roster I can't imagine how many hours that would take to do.

There's another speed matching thread going on over at MRH. I don't know why people think it's so difficult. I did two locomotives the other day. One took about 5 minutes, one took about 7 minutes because it was an ornery old QSI that was behaving oddly. And at 5-7 minutes per loco, and I can have a menagerie of different DCC decoder brands, and they all work together!

https://model-railroad-hobbyist.com/node/43979

KPack

  • Conductor
  • ****
  • Posts: 784
Re: Speed matching - MRH newsletter
« Reply #2 on: December 05, 2021, 10:25:15 AM »
Anyone get the Dec 4 MRH newsletter with a link to DCC loco speed matching thread?

Man, am I glad I don't have to deal with that. What if they wanted to speed match every loco in their fleet so the could make any consist they wish? Even with my modest roster I can't imagine how many hours that would take to do.

There's another speed matching thread going on over at MRH. I don't know why people think it's so difficult. I did two locomotives the other day. One took about 5 minutes, one took about 7 minutes because it was an ornery old QSI that was behaving oddly. And at 5-7 minutes per loco, and I can have a menagerie of different DCC decoder brands, and they all work together!

https://model-railroad-hobbyist.com/node/43979

Congrats.  I hated everything about speed matching back in my DCC days and I don't miss it one bit.  Could I do it?  Yes.  Doesn't mean that I enjoyed it.

I just ran an consist of 3x2 (three up front, 2 DPU) of Athearn Genesis, MTH, Scaletrains, Scaletrains, Athearn Blue Box with zero problems.  I did no speed matching, a couple taps on the screen, and I was off.  Yes DCC has some cool things, but speed matching is not one of them.

-Kevin

Alan

  • Conductor
  • ****
  • Posts: 1073
    • LK&O Railroad
Re: Speed matching - MRH newsletter
« Reply #3 on: December 05, 2021, 11:03:19 AM »
I can understand how matching two locomotives would not be hard. But doesn't the job become exponentially more difficult with a large number of locomotives? It seems the effort should increase by the square of number of locos.

The MRH thread states that speed matching is done on a single parallel track location presumably flat or nearly so. What about sharp curves or steep grades?

Earlier this year I built a helix - 32"R, 2%G, 5T. Regular readers know that I build everything on the workbench and then install on the layout once I know it is tight-n-right. Sidebar: Even my trackwork was done this way - image 2. My helix was not an exception. It too was built on the bench and I did extensive testing before declaring it layout-ready. During testing I observed something new and unexpected. Some of my locomotives (alone, not consisted) that otherwise have similar running characteristics on the layout react quite differently from each other while climbing the helix. Some require considerably more throttle than others to maintain speed up the helix. Had I not built a helix it is likely I would have never noticed this difference. Been running trains for a long time and hadn't noticed until now.

My helix experience tells me speed matching is more than simply getting two locos to move at the same speed in one particular scenario. Speed matching appears to be dynamic, not static and therefore is not a once and done thing. And that lies at the heart of RailPro's one undeniable advantage - a feedback loop and the necessary processing to dynamically act upon it. I'll admit I have a perfectionist streak in me. It drives my wife nuts. The differences I observed may not in actual operation manifest themselves as derailments, wheel slippage, or other adverse operation thus rendering the whole discussion moot. But for perfectionist buttwads like me, and more importantly, for sake of accuracy in the discussion I don't believe DCC speed matching with respect to consisting actually works. I think it is a case of close enough. In fact, the absence of a feedback loop in DCC makes it impossible to speed match under varying conditions. RP is effectively speed matching on-the-fly depending upon conditions. Bereft of feedback DCC can't possibly do that.   

Full disclosure: My railroad has been RP since day one. I have no DCC experience. I'm not anti-DCC, just sharing my observations and trying to fully understand.





Alan

LK&O Railroad website

When I was a kid... no wait, I still do that. HO, 28x32, double deck, 1969, RailPro

AlexW

  • Fireman
  • **
  • Posts: 11
Re: Speed matching - MRH newsletter
« Reply #4 on: December 05, 2021, 04:33:43 PM »
Quote
I can understand how matching two locomotives would not be hard. But doesn't the job become exponentially more difficult with a large number of locomotives? It seems the effort should increase by the square of number of locos.

No, it's linear. The "traditional" method is to start with one locomotive and match everything else to that, the newer and more precise method is to use a standard like the La Mesa 30/60 standard, and match everything to that. It takes 5-7 minutes per locomotive, and is fairly easy. If you want to really do it right, you would service, lubricate, and then warm up each locomotive first. People who do automated layouts with RR&Co have found that locomotive performance can change up to 10% as the locomotives warm up combined with fluctuations in room temperature.

Quote
Speed matching appears to be dynamic, not static and therefore is not a once and done thing. And that lies at the heart of RailPro's one undeniable advantage - a feedback loop and the necessary processing to dynamically act upon it.

It's an overly complicated method to avoid 5-7 minutes of speed matching per locomotive. It's one of those things that is kind of cool from a technology perspective, but is totally unnecessary in the real world. Doing the speed matching on straight, flat track gets it close enough for whatever scenario and allows you to calibrate it to a specific scale speed.

Quote
But for perfectionist buttwads like me, and more importantly, for sake of accuracy in the discussion I don't believe DCC speed matching with respect to consisting actually works. I think it is a case of close enough. In fact, the absence of a feedback loop in DCC makes it impossible to speed match under varying conditions. RP is effectively speed matching on-the-fly depending upon conditions. Bereft of feedback DCC can't possibly do that.

It works fine, the couplers hold together any slight differences in speed. Once you have a decent load behind the locomotives it ends up spread among the units anyway.

We tend to take speed matching for granted today, as it is quite simple to do, but it's one of the major features of DCC, and is generic enough in its implementation (other than differences between decoder brands) that it allows for different approaches to doing it, i.e. master locomotive, calibrated to a speedometer, and allows for having all locomotives on a layout speed matched, or different groups of locomotives matched to different speeds, i.e. switchers geared lower than road units so that any two switchers can be MU'ed, and any two road units MU'ed.

ON28

  • Conductor
  • ****
  • Posts: 260
Re: Speed matching - MRH newsletter
« Reply #5 on: December 05, 2021, 09:00:15 PM »
My friend, a retired railroad engineer, has built and operated for 25 years a large two-deck HO DCC layout with a large six-turn helix and long trains. I took my RP locos over there once for a demo and pointed out the great load-matching feature. I asked him how he did his speed-matching. He said he has never speed-matched, just confines his MU consists to locos of identical brands. I wonder how many DCC users don't bother to speed match. Works for him, and us, as his layout operates very well.