Decoder Question

[jeep35]

I've got a pair of old Life Like Proto 2000 BL2s and I want to install decoders and LEDs in them. Both have had the original Athearn style motors removed and replaced with I believe are Alco Products Helix Humper (I know) motors. One reference source said to use a NCE D13W decoder. Would this still be appropriate for  this type of motor?

Thanks

Jim

[BigDaddy]

Got a meter?   An important thing to know is the the stall current.  On DC you measure the amps the engine draws when you either grab the flywheel or press the engine on the track to stop it's forward momentum.
You want that value to be less the the limits of the decoder, which in this case is .75 continuous and 1.2 amps max

[Overmod]

Just out of curiosity, is this the type with the large-diameter multiple-thread worm?  We will have members who likely know the exact provenance and 'care and feeding' of the can motors in these.

One point is that you will need to remap the voltage-speed map in the selected decoder as the reported response of this drive under test was slower than 'ideal' at low commanded voltage, and faster at high.  On the other hand it unquestionably produces more 'drawbar TE' when weighted than most other drives... see some of the test data near the bottom of this page, with curves.  It might be easier to program this in an environment like JMRI (see here if unfamiliar with JMRI), especially if wanting to take advantage of low-speed performance with 128 steps.  I suspect most decoders that support speed curves and adjustable BEMF values would work just fine.

You didn't say if you were interested in sound.  That might affect your range of choices for these.

 

 

[rrinker]

 I question the applicability of those charts - especially the speed one, as the start speed is abnormally high in all cases because the tests were all conducted using pure filtered DC. Just about any decoder will be able to start the loco FAR slower than pure DC. 

 ANd they run at different speeds - compared to what? The same loco with a differnet motor? Both the OP's locos have the same motor, so what does it matter? The amount of insnae metrics some people generate for little to no actual benfit sometimes astounds me. I guess I just have better things to do with my time than take 5 or 6 different motors and keep swappign them int he same loco and run a bunch of performance tests.

My metric is - does it start slow and run steadily at a reasonably slow speed? Does it smoothly increase speed up to some maximum, without a lot of motor or gear noise? OK, it's a good loco. 

 A 1.2 amp stall decoder shoudl be fine for the motor used int he Helix Humper kit. I forget where he used to get them from - the original creator of them was a friend of the club I used to belong to and would stop by regularly to show the latest kits, until he passed away and the business transferred. That was some 23-24 years ago.

                               --Randy

[jeep35]

Thanks for the information. As it turned out I had a couple of Train Control System T1 decoders I had forgotten about that did the trick. Thanks Again.

[Overmod]

rrinker

I question the applicability of those charts - especially the speed one, as the start speed is abnormally high in all cases because the tests were all conducted using pure filtered DC. Just about any decoder will be able to start the loco FAR slower than pure DC.

Starting was not the issue in question -- obtaining linear response across the speed steps was.  (I saw no reason to point out adjusting the lowest range of speed steps "differently" if desired for enhanced starting and low-speed performance -- something others here know more about than I do.)

... and they run at different speeds - compared to what?

The author says, if you read the description that accompanied the chart, his tests are compared to the 'ideal' speed/voltage curve for a DC voltage-controlled motor.  What was pointed out is, for a specific Alco Helix-Humper motor, the slower-speed response was smaller for voltage increment change than the "ideal", and the higher-speed response more speed for equivalent voltage incremental change -- presumably using either voltage at midrange, or a smooth division of rotational speed by voltage for increments between 0 and full voltage; I didn't read to see how he defined 'ideal linear response'.  The moral does remain that if he wants a more linear response -- and judging by some threads and posts on this subject in the past, there are people who do -- he will want to edit his speed map or table so the increments (whether 28 or 128-step) produce equal resultant speeds.

My metric is - does it start slow and run steadily at a reasonably slow speed? Does it smoothly increase speed up to some maximum, without a lot of motor or gear noise? OK, it's a good loco.

This is true, and in the pre-1990 world it would probably be as good as you'd get without tinkering with the drive electronics, in what might be arcane ways applicable only to particular locomotives.  And they are certainly prerequisites for any further 'tuning' when using decoder rather than DC drive.  But if it were mine -- and I were running in 'twist the knob to go faster' operating mode -- I'd want equal response across the range of my speed control, and I can get that to a reasonable surety by editing my speed table values.

[rrinker]

 That seems to be less than ideal, actually. A low geared road loco (protoype) I'd want tune for more throttle = less speed change at the low end, with maybe the top of the speed range reserved for "unrealistic hostling" when you want to move the loco to another spot ont he layout but operating at a prototypical speed is more or less just a time waster. Likewise for a passenger loco, quicker speed incremements for a given change int he throttle at the low end, with finer speed control at the top. ANd a switcher? Almost all throttle response in the low end.

 Just seems odd - the kind of people doing these studies are generally the type looking for realistic operation, not simple "twist the knob to go faster" types. 

 I accomplish the varying curve slopes with the simple expedient of using the start, mid, and top settings. I won't use a decoder that doesn't have a mid, as setting a 28 step table for everything, regardless of the use of JMRI or whatever, combined with the varying ways decoders interpolate the 28 entries for 128 step operation, makes it needlessly complicated. I'm not sure what people are looking for - the closest to realistic operation these days is none of that, rather lots of momentum and a decoder with a functioning brake, controlled by a throttle that looks and works like the prototype's control stand, rather than using any knobs. 

                                                --Randy