Neat throttle

[sirenwerks]

This is interesting http://model-railroad-hobbyist.com/node/25866

[mmagliaro]

I think it's awesome.  Now they just need to also make a steam version.
Finally, somebody is thinking about the major Achilles Heel of DCC control: the hideous user interfaces.

Phone?  No, sorry.  Tapping buttons and touch screen thingies on a phone is not a natural user interface.  That is another example of the human being trained to conform to the phone's interface.

Humans could pick up this throttle and intuitively use it, and remember how to use it even after they haven't seen it for a while.

[C855B]

Nifty idea, looks nice. I had seen pix of an earlier mock-up Yes, it's a prototype, but in the current packaging it's too thick and awkward to hold. Needs to be about half as thick. Definitely intended for operations, vs. train-running.

Max, say what you want about the other DCC user interfaces, but behind the (frequently poorly-implemented) complexity is a reason - there is more to a DCC throttle than controlling a single locomotive. How do you build a consist with this thing, and, especially, how do you throw powered points? Apparently you don't.

EDIT: I think I stand corrected. See below.

[jdcolombo]

 It IS interesting, but I've got a lot of questions about how it works.

For example, the speed lever says it has 8 positions plus idle.  OK, but does that mean that it has just 8 distinct speed "notches" - if so, that's something of a step down from 128-step speed control.  Maybe it is continuously variable, but with 8 detents, people using the throttle will naturally gravitate toward one of the detents.  It might work OK with massive amounts of momentum, but . . . I think I like the idea of a continuously-variable knob (either a pot or a digital encoder) to control speed, especially for switching duty, would be better (again, the 8-notch approach might be fine for mainline running with a LOT of momentum built-in).

What I'd really like to see is a variation on this throttle specifically designed to work with manual notching on a sound decoder.  This variation would have a knob for speed control, but would keep the 8-position lever for notching up and down.  Teamed with a sound decoder that allows manual notching (and proper programming commands), you would use the lever to set the notch sound, and the knob to control speed.   Now THAT would be cool.  The problem with most manual notching on current sound decoders (like ESU's Loksound) is that the notching is assigned to two function buttons, one for notch up and the other for notch down.  The function buttons on most DCC throttles are latching (except for F2).  That means that if you press, say, the F9 key for notch-up, the sound will keep notching up until you press F9 again to stop the notch up process.  Same for notch down.  It's certainly possible to do this while controlling speed with a knob, but it's not very intuitive.

Now think about a variation of this throttle where the notch lever works with the notch-up and notch-down commands of a sound decoder.  If you move the lever to Notch 4, the decoder sound notches up to Run 4.  Move to idle, and it notches back down.  Now THIS is intuitive, and good interface design.  Meanwhile, you control actual speed with a knob.  So, for example, say you are in the yard with a heavy train.  On leaving the  yard, you first notch up to, say, Run 4, then move the throttle knob to actually start the train moving.   After the train starts moving,  you move the lever to Run 8, and then use the knob to set an appropriate speed.   Taking a heavy train up Sherman Hill?  No problem - use the lever to set Run 8, and the knob to set the speed at a crawl.  Coming down hill?  Again, no problem - set the lever at idle, hit the button for Dynamics, and use the knob to accelerate the train to an appropriate downhill track speed.  This is sort of like ESU's Full Throttle Drive Hold feature, but not quite; with Drive Hold, the throttle knob controls the notching sound, but you can't independently control speed while Drive Hold is on.  And while you can certainly mimic this same behavior with manual notching on a current throttle, it isn't very intuitive or easy (the two buttons problem, again).

So what I'd REALLY like to see is a version of this throttle designed with the idea of using the lever to independently control notching, and a knob somewhere to independently control speed.  With proper programming, such a throttle could be used with ANY sound decoder that implemented manual notching.  And that would be very, very cool, IMHO.

John C.

[learmoia]

It IS interesting, but I've got a lot of questions about how it works.

For example, the speed lever says it has 8 positions plus idle.  OK, but does that mean that it has just 8 distinct speed "notches" - if so, that's something of a step down from 128-step speed control.  Maybe it is continuously variable, but with 8 detents, people using the throttle will naturally gravitate toward one of the detents.  It might work OK with massive amounts of momentum, but . . . I think I like the idea of a continuously-variable knob (either a pot or a digital encoder) to control speed, especially for switching duty, would be better (again, the 8-notch approach might be fine for mainline running with a LOT of momentum built-in).

What I'd really like to see is a variation on this throttle specifically designed to work with manual notching on a sound decoder.  This variation would have a knob for speed control, but would keep the 8-position lever for notching up and down.  Teamed with a sound decoder that allows manual notching (and proper programming commands), you would use the lever to set the notch sound, and the knob to control speed.   Now THAT would be cool.  The problem with most manual notching on current sound decoders (like ESU's Loksound) is that the notching is assigned to two function buttons, one for notch up and the other for notch down.  The function buttons on most DCC throttles are latching (except for F2).  That means that if you press, say, the F9 key for notch-up, the sound will keep notching up until you press F9 again to stop the notch up process.  Same for notch down.  It's certainly possible to do this while controlling speed with a knob, but it's not very intuitive.

Now think about a variation of this throttle where the notch lever works with the notch-up and notch-down commands of a sound decoder.  If you move the lever to Notch 4, the decoder sound notches up to Run 4.  Move to idle, and it notches back down.  Now THIS is intuitive, and good interface design.  Meanwhile, you control actual speed with a knob.  So, for example, say you are in the yard with a heavy train.  On leaving the  yard, you first notch up to, say, Run 4, then move the throttle knob to actually start the train moving.   After the train starts moving,  you move the lever to Run 8, and then use the knob to set an appropriate speed.   Taking a heavy train up Sherman Hill?  No problem - use the lever to set Run 8, and the knob to set the speed at a crawl.  Coming down hill?  Again, no problem - set the lever at idle, hit the button for Dynamics, and use the knob to accelerate the train to an appropriate downhill track speed.  This is sort of like ESU's Full Throttle Drive Hold feature, but not quite; with Drive Hold, the throttle knob controls the notching sound, but you can't independently control speed while Drive Hold is on.  And while you can certainly mimic this same behavior with manual notching on a current throttle, it isn't very intuitive or easy (the two buttons problem, again).

So what I'd REALLY like to see is a version of this throttle designed with the idea of using the lever to independently control notching, and a knob somewhere to independently control speed.  With proper programming, such a throttle could be used with ANY sound decoder that implemented manual notching.  And that would be very, very cool, IMHO.

John C.

Watching the demo.. I'm 'guessing' that the Throttle is is translating the movements of the switches into throttle and function commands.. combining speed step increases and decreases with manual notching function.

Move throttle to Position = Increase throttle notch x 4 Delay + Increase speed step @ ##Steps/second..
Decrease throttle = Decrease throttle notch x # + Maintain current speed step.
Move throttle to Idle = Decrease throttle notch to 0 + decrease speed step @ ##Steps/Second (coast)
Brake Lever = Decrease Speed step @ ##Steps/Second. + Brake Sound

Needs to be scaled down in depth but looks really cool!

~Ian

[C855B]

Looking again, it appears that the DCC-related functions are probably incorporated into the display surrounded by the four small buttons. A little more clarity might address this. Maybe I should watch the demo. 

Interesting - the countersunk holes on the corners - it appears to be possible for it to be removed from the box as a panel and mounted to a fascia.

[u18b]

I like this.

but since N is so small, I wonder how it would work as far as realism and functionality.  For example, I never liked doing momentum in N scale.  I think it is just too had to control with the smaller physical distances.   I remember Atlas' first decoders were made by Lenz and had this momentum.   I didn't like it.

On the other hand, this would be pretty cool for an O scale layout.
And probably HO like in the video.

Just my 2cents

[TLOC]

I think the idea is great and to truely understand what they are trying to do you have to read the entire blog. Many of the ideas discussed here have been discussed in that thread and while they won't accomplish all you want, it's open source so you can make enhancements. Big issue maybe the pricing at roughly $500 that will put alot of guys off. They have been working with NCE and Loksound. It is really designed for the person that likes high momentum and switching. There have been a lot of changes since the 1st posting and many real life train engineers have supplied input. Just the fact some one is trying to make changes to normal control operations IMO should be appreciated, the manufacturers sure aren't doing it.
TomO

[C855B]

...but since N is so small, I wonder how it would work as far as realism and functionality.  For example, I never liked doing momentum in N scale.  I think it is just too had to control with the smaller physical distances.   I remember Atlas' first decoders were made by Lenz and had this momentum.   I didn't like it.

On the other hand, this would be pretty cool for an O scale layout.
And probably HO like in the video. ...

Agreed. Heavy momentum in N = loss of control, at least in my experience.

The lack of an emergency stop, as noted in one of the blog posts, is a HUGE issue. There needs to be a Big Red Button. Not to harp on the scale vs. scale issue, but outside of switching operations, N runs longer trains, with more that can go wrong. Forced momentum in this environment means increased likelihood of rolling stock hitting the floor.

But like you say, this would be fun/cool for "the operator", especially in larger scales.

[learmoia]

Agreed. Heavy momentum in N = loss of control, at least in my experience.

The lack of an emergency stop, as noted in one of the blog posts, is a HUGE issue. There needs to be a Big Red Button. Not to harp on the scale vs. scale issue, but outside of switching operations, N runs longer trains, with more that can go wrong. Forced momentum in this environment means increased likelihood of rolling stock hitting the floor.

But like you say, this would be fun/cool for "the operator", especially in larger scales.

I'm guessing the 'momentum' is in the throttle logic.. not the model (which may be a good or bad thing.. Not sure).. Emergency may be hidden in in the brake lever.. 
Throw the bake lever all the way over = Speed step = 0?

~Ian

[btrain]

It looks really great, but I feel that it ought to be mounted to the facia of a layout to get the proper feel of notching up or feathering the brakes. What I'd really like to see is a chest pack that they use for radio controlled engines like this one. Maybe they can even include the safety vest as well! 

[peteski]

I remember Atlas' first decoders were made by Lenz and had this momentum.   I didn't like it.

Momentum is a part of standard DCC design (NMRA specs) and has been implemented in DCC decoders since the beginning.  But it is configurable and usually disabled (set to zero) by default. See definition of CV3 and 4.  I suspect the locos you are talking about simply had a non-zero value programmed in those CVs

For example NCE throttles can change the momentum setting on the fly with just a single button press (then typing in a value). That actually reprograms CV 3 and 4 on the fly (programming on main).

I don't know if the throttle discussed here uses a momentum within the throttle or programmed in the decoder.

[C855B]

... Throw the brake lever all the way over = Speed step = 0? ...

Maybe, but in the model railroading context there really needs to be an unambiguous panic button. Case in point in the demo vid, the operator was fumbling around to find the brake when he was about to run out of track.

[Greg Elmassian]

Funny, the throttle pictured is an attempt to replicate the prototype, for example the notches in the throttle.

But many of the comments are anti-prototype, momentum, the speed control, etc.

Seems that if you wanted a prototype throttle, you would like to run like the prototype

(emergency stop excepted)

Greg

[learmoia]

Maybe, but in the model railroading context there really needs to be an unambiguous panic button. Case in point in the demo vid, the operator was fumbling around to find the brake when he was about to run out of track.

True.. and I did notice that...  an emergency button / lever or alerter would be a good idea.. I like the remote control pack idea better anyways..

~Ian