Hi Guys,First post for me in this forum...
I'v been in the hobby since 1985 apox. when i joined a Sn3 narrow gauge guild, but now i'm modeling diesels in the H0 scale and looking forward to an 2x9 foot switching yard.
I thought the DCC seemed to be a step in the right way so i bought an inexpensive Roco mulitMAUS 10810 controller. And then ordered an NCE D13J decoder for my new WaltherProto EMD SD9 (SP).
It was just a plug and play 9pin connection, no problem there...And the mulitMAUS was used before so it was just a "start up and go".
The problem i trying to overcome is the jerky running, the loco starts very nice and slow, but when i increase the trottle... there is nothing for a while, then the loco jumps ahead to a faster speed, then there is nothing again, and the jumps ahead.
Its like there is: 1mph 5mph 10mph and nothing between.
I tryed to set different values on the CV's, but i'm not close to a smooth ride.(when i save a new value the loco jump forward half an inch, is that ok)
I have scanned through a lot of post in the forum but not found a workable solution. Hope some one here can shine a little light over the matter.
Its quite a high learning curve from DC to DCC if yu want to grasp all of it
Thank you very muchGreetings from: Pete
Pete,
It could be two things - one or both:
It's possible that the multiMAUS is fighting the "dithering" (BEMF) of the NCE decoder? You could try resetting the decoder to factory settings then confirm how it runs on address "3".
Tom
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
I'm thinking that the wires are not connected tightly on the decoder or else the gears need to be cleaned in the trucks. Since the loco is new, it isn't likely that the gear lube is the problem, but you never know.
Rich
Alton Junction
Thanks for your answer,
I set the CV3 to "0", no changeI set the speedstep to "128" no change
Then i set the CV30 to "2" (reset to factory) and there was a change, to loco can run much slower - prefect!, but it still "jumping ahead forward" when increase the trottle, but with smaller intervals, like - 1mph, 3mph, 5mph, 7mph
Anyway, progress :)
Where to go from hear, i have no clue...
RegardsPete
Rich,
I actually rotated the flywheels just to see i there was and binding of some sort, but i could not feel any hickup's in the gears.
I will change decoder, i tryed with an TCS T6X, will go back to the D13J
I'm voting for speed step mismatch.
It takes an iron man to play with a toy iron horse.
Since you have been messing with CV values, reset the decoder to factory defaults and see if the loco performs better. For NCE decoders, the reset is CV30=2.
Installed the NCE D13J again...
I did a reset with CV30=2, speedstep=128, Acceleration CV3=0
I can say... better than before, but the first "jump" of the loco is quite big, then there are just smaller ones.
I found a NMRA setting dokument: https://www.nmra.org/sites/default/files/standards/sandrp/pdf/s-9.2.2_decoder_cvs_2012.07.pdf
Variable 3 and 23 have something to do with speedsteps, i think.
And, the loco does jump ahead when saving the data, now it just flashes the lights two times.
Thanks,Pete
One way to tell if there is a speed step mismatch - if the headlight is on, with each speed increase does the headlight go on, then off for the next step, then back on for the next one, etc? The multiMAUS is capable of all 3 options, 14, 28, and 128 speed steps. Totally unfamiliar with that unit so I have no idea how you set it, but you might want to check that it is configured for 128 steps for smoothest operation. CV29 in the decoder needs to be set for 28/128 speed steps among the other options such as short address/long address and if you still want the decoder to operate on a DC layout.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Randy,
The multiMAUS can not read CV's, so i have no clue what value to enter in CV29.
The NCE handbook say's like this fo CV29:-bit0 1=direction reverced 0=normal-bit1 1=28 speed mode (always enabled)-bit2 1=analog enabled 0=disable-bit4 1=alternate speed table active 0=usw table defined by CV2,5,6-bit5 1=use long address in CV17/18 0=use short in CV1-bits 3,6,7 are ignored by the decoder
What number will be the correct one for this?
could it be dirty track, enough that commands aren't getting thru reliably. commands don't get thru until there's a section that cleaner and then there's a jump in speed.
greg - Philadelphia & Reading / Reading
Welcome to the forum--
I have the Lokmaus system, not the Multimaus, but the Lokmaus can be set for 14, 28, or 128 speed steps. Is the Multimaus set for the correct speed steps? Also, how does the locomotive run on DC? Is it possible to change back and see how it runs on DC?
Gary
Scandinavian Its quite a high learning curve from DC to DCC if yu want to grasp all of it Thank you very muchGreetings from: Pete
That is true.
First, I am puzzled why the problem improved after a re-set. The decoder was new; so, unless you were mucking about, your re-set wouldn't change anything. Now, "mucking about" can be very useful and informative. And perhaps cause problems.
Second, a bit of clarification, please. When you say there are sharp increases in speed, does this happen exactly when you change the throttle speed step? Or does it happen in a different way? It's important to exactly describe the event/problem.
I have some locomotives I am programming right now. Some of them are doing what appears to be what you describe: when I go from speed step 0 to step 1, there is a quick start. When I go from 1 to 2, there is another sharp increase in speed. And etc. I have fixed this by modifying the speed tables in the decoder. On completion, there is a very smooth speed increase.
On the other hand, if the sharp changes in speed do not coincide with changes in speed steps; that's a whole different problem.
So, please describe again. Very accurately.
Ed
Scandinavian's initial posts are being moderated so there is a time delay in his responses to our suggestions and questions.
Scandinavian Randy, The multiMAUS can not read CV's, so i have no clue what value to enter in CV29. The NCE handbook say's like this fo CV29:-bit0 1=direction reverced 0=normal-bit1 1=28 speed mode (always enabled)-bit2 1=analog enabled 0=disable-bit4 1=alternate speed table active 0=usw table defined by CV2,5,6-bit5 1=use long address in CV17/18 0=use short in CV1-bits 3,6,7 are ignored by the decoder What number will be the correct one for this? Thanks,Pete
There is a CV29 calculator here:
http://www.2mm.org.uk/articles/cv29%20calculator.htm
Assuming you set the loco address to be 1234, you would use 34 or 38 for CV29. Not sure what Roco uses for a short address cutoff but assuming it's like Lenz, if you address is say 50 you would use 2 or 6.
Sounds like your decoder is set to 28/128 speed steps and your controller is set to 14 steps. There should be something in your controller's documentation for setting the speed steps it puts out.
Mark.
¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ
This is the digitrax decoder manual. Scroll down to page 34 and you will find a chart that list all the settings for CV29. These settings should work on any decoder.
Check the speed step setting in your controller before you play with any further setting on your decoder. Your symptoms really sound like your controller is set for 14 speed steps.
richhotrain Scandinavian's initial posts are being moderated so there is a time delay in his responses to our suggestions and questions. Rich
Thank you, very true...I was a litte pussled first, but thats the way it is right now :)
Well, i will look through all your posts guys, but first i will just say this - i changed the 18v transformer to a 12v... and the problems disappeared
I will do some more testing and be back with some sort of result.
Thank you so much!Pete
Okay, i done some testing.
-gregccleaned the track and no change
-garyaThanks :) Yes, it runs on 128. The loco runs jerky on DC also, skipping speedsteps
-7j43kThe loco starts slow, and then I turn the knob like "4 speedsteps" and nothing happends, the on the 5:th it jumps away. Then the same, 4 steps nothing happends... and then jumps to a much faster speed.
-rrinkerActally found the calculator, and entered 38. Loco address, it says someting like L0003 in the window.
-Mark R.The multiMAUS is set to 128, and 128 steps entered in CV29.
-SouthPennThanks, got it.
The 12v trafo did make a big difference, but the loco does not start as smooth compare to 18v.
And as I found out, the loco have the same jerky behavior on DC with a simple Fleischmann trafo, how booring...
And if not that is enough, the loco have a crooked frame... so i have to dismantle the whole thing and try to straighten it out.
Any good suggestions what to do next?
Guys, thanks for all you input, priceless! Pete
Well, if it does the same thing on DC, then it's definitely not a programming issure with your system OR the decoder.
That crooked frame may have something to do with it, putting a bind of some sort on the drive. That may be requiring a higher voltage to get things moving / increasing. Tear it down and see if there is a bind anywhere.
ScandinavianThe loco runs jerky on DC also, skipping speedsteps
There's your issue, Pete. If it operates poorly on DC, DCC will NOT be a panacea/cure-all for it. You might try operating the locomotive with the shell off to see if the motor is binding.
Mark R. Well, if it does the same thing on DC, then it's definitely not a programming issure with your system OR the decoder. That crooked frame may have something to do with it, putting a bind of some sort on the drive. That may be requiring a higher voltage to get things moving / increasing. Tear it down and see if there is a bind anywhere. Mark.
That's a bummer since he bought it new. May need to send it back. To me, it still seems like something is binding in the truck gears.
If you have a meter, connect it in series with one of your DC track leads and check the current load. If is indeed a bind, it will show up as excessive current draw. Be sure to set your meter to amps, not milli-amps.
And I guess that answers that, it works like NCE - if you cn run it oon address 3 with CV29 set to 38. Most systems, address 3 is a short address, and CV29 would be something like 6.
But if the loco is jerky on DC, then it will be just as jerky on DCC. Good running DC locos will usually be good runners on DCC. Poor runners on DC will always be poor runners on DCC, there's no magic in DCC that can fix that. A cracked frame, especially on a steam loco, is almost certainly going to jam things up because of the side rods.
rrinker A cracked frame, especially on a steam loco, is almost certainly going to jam things up because of the side rods.
A cracked frame, especially on a steam loco, is almost certainly going to jam things up because of the side rods.
Also sorta sounds like maybe the loco's running at 28 speed steps but the throttle is outputting 128 steps. Thus steps 1-4 on the throttle would all be step 1 on the loco.
NO. There IS NO DECODER SETTING for 28 vs 128.
If the system is putting out 28, then the decoder will act like there are 28 steps. If the system is putting out 128, then the decoder will act like it is getting 128.
It COULD be a speed table, but he has said he has reset the decoder, plus the value of 38 in CV29 does not enable a speed table. It's not momentum, because momentum increases the time it takes to move between speed steps, if anything it would smooth out drastic speed changes. NCE decoders also support CV2-6-5 3 step speed tables, but again the OP has indicated the decoder has been reset, which should wipe any of that if it got accidently set.
The only exception is the 14 and 28 step modes - that IS controlled by CV29 but again the value of 38 reported is for 28 steps (OR 128 if sent). Another symptom of 14/28 mismatch is the headlight going on and off with alternate speed steps, because of the way the two packets are defined in the NMRA standard.
Hi guys,
After a meticulous examination of everything i found that all the 9 cables that is coming up in the rear of the main circuit board, the plastic cover on the cables were peeled away a little to far, and the cables were pressed down hard and twisted right there, crossing each other on some places.
And after reading the manual a little more...Well, in the beginning I noticed a STOP sign and a FLASH in the trottle window, and now... i know that the FLASH means - short circuit!
Shall we all agree upon that the main circuit board is toasted ?
It can't be a "binding" caus if i left it running at the lowest starting speed it just continued without a hesitation. The problem came when i increased the speed, um... tryed to increase.
My english is not the best, you have to excuse me for that.
Your command of English is fine. Better than some peopel who have lived int he US all their lives. Perfectly understandable what you are trying to explain.
I don't think it's a short circuit though - if it were, the speed step you are trying to run on would have no effect. You'd get the same indication on the throttle if it were an overload - which is what I think is goign on here, the mechanism is binding and stalling the motor, causing it to draw more current than the MultiMaus can deliver. At slow speeds it may be able to inch itself past the bind, but when you try to go faster, it can't. That it does the same thing on DC is pretty telling. I would suggest putting a decoder in a different loco, one you know runs fine on DC, and sending this one off for repairs, as annoying as that is, but if it's new - that's what the warranty is for.