DCC Wishlist

Fluesheet

Member
If I could have anything as far as features....

A Momentum toggle - A switch that could toggle between two pre-set momentum parameters. Momentum A would represent running light, MB would represent a train.

Auto Spin - Programming that would be active during heavy throttle while using heavy momentum. I.e. if you apply more throttle than the acceleration curve allows, the motor would spin up. How could this be made to work with jack rabbit acceleration on our relatively light trains? I don't know, but I want it!

Auto speed match - place a control locomotive and and one with a new decoder in it on parallel tracks, accelerate the control locomotive to speedstep A, match that physicla speed with the second locomotive, assign the same speedstep to the second locomotive and associate the phyiscal speed with the speed step. This doesn't take into account the acceleration curves matching or not, so I'll have to work on that wish list item...

A retrofittable sound trigger for steam locomotives - I'm a steam snob and darn it, if you're going to have sound - it's got to match the actual mechanical goings - on!

Exhaust sound for steam is getting better - but how about tying it power being used? Right now there are several decoders that change volume depending on whether or not you are accelerating, but they are fixed levels. So if you accelerating with a train or light, you get the same sound. If this was tied more dynamically to motor power applied you would get quick acceleration with low sound levels (running light) and slow acceleration with hight sound levels when you've got a train.

I'm sure there's more in here, but I'll wait for the muse to come calling again.

Fluesheet.
 

Len Marinaccio

New Member
DCC wish list

Fluesheet said:
If I could have anything as far as features....

A Momentum toggle - A switch that could toggle between two pre-set momentum parameters. Momentum A would represent running light, MB would represent a train.

If your decoder uses CVs 23 and 24, you can plug in additions to the acceleration and deceleration rates, simulating a heavier train. It's not quite "toggle" simplicity but close.

Fluesheet said:
Auto Spin - Programming that would be active during heavy throttle while using heavy momentum. I.e. if you apply more throttle than the acceleration curve allows, the motor would spin up. How could this be made to work with jack rabbit acceleration on our relatively light trains? I don't know, but I want it!

Brakes in a weighted tender? Tall order there!

Fluesheet said:
Auto speed match - place a control locomotive and and one with a new decoder in it on parallel tracks, accelerate the control locomotive to speedstep A, match that physicla speed with the second locomotive, assign the same speedstep to the second locomotive and associate the phyiscal speed with the speed step. This doesn't take into account the acceleration curves matching or not, so I'll have to work on that wish list item...

Sounds like a side-by-side dyno would be the thing. With some of the PC based DCC controllers, I'll bet that a program could be written to monitor the speed of the new loco with respect to the older one running by its side and send speed table parameters to the new loco - completely automatically! It would require speed monitoring of each dyno to be fed into the PC but you can build that interface out of an old serial mouse.

Fluesheet said:
A retrofittable sound trigger for steam locomotives - I'm a steam snob and darn it, if you're going to have sound - it's got to match the actual mechanical goings - on!

You're killing me! One small miracle at a time, please!

Fluesheet said:
Exhaust sound for steam is getting better - but how about tying it power being used? Right now there are several decoders that change volume depending on whether or not you are accelerating, but they are fixed levels. So if you accelerating with a train or light, you get the same sound. If this was tied more dynamically to motor power applied you would get quick acceleration with low sound levels (running light) and slow acceleration with hight sound levels when you've got a train.

Now you're talking! That would be a great feature! I don't know if any decoders support it, but with minor hardware and software considerations in the decoder it would be possible. By selecting an H-bridge for motor control that provides current use feedback, you could loop that back to an A/D converter on the decoder chip. Now the decoder can know how much current draw there is with respect to the throttle setting it is delivering. When the ratio of current draw to throttle becomes high, the volume is increased. Once properly adjusted, this will produce automatic volume compensation for acceleration, heavy trains, grades, etc. The same thing could work on a diseasal engine but instead of increasing volume, you would increase or decrease simulated engine RPM too. I think the steam effect would be more realistic though.


Len
 

Fluesheet

Member
Aha! A like thinker! :thumb:

Of all the wish items listed, the sound to load matching would be the most interesting to me (closely followed by a sychronized exhuast note) for what I do.

Some minor additons I'd add would be to add a variable for steam pressure so you could mimic modern vs. low pressure (~pre-superpower days) and secondly, even exhaust notes. The two sound systems I've heard (Trix (locsound?) and the previous generation of Soundtraxx) both imitate a locomotive with out of time valve action - i.e. CHUFF chuff chuff chuff CHUFF chuff.... etc.

Thanks for the reply, I was kind of surprised this post didn't get much attention - perhaps it's previously been well discussed.
 

Fluesheet

Member
An article on Tony's Train Exchange website got me thinking about this again.

Another wishlist item hinted at above is exhaust sync:

Why not allow functionality that maps the BEMF signal to an exhaust event? The gearing on a locomotive is a constant. It never changes. Therefore, the motor will always rotate x many times for each complete driver revolution. I.e. every 13 motor revolutions = 1 chuff. The user would have to map this association (assuming it's not a factory installed decoder), but once set, exhaust sync would be accurate regardless of speed!

Matt
 
L

lester perry

I love it but keep it simple for nontechnos like myself.
Les.
 

Fluesheet

Member
Agreed! Simple is best BUT I'll take complex if that's what it takes. After all, how much more complex can sound decoders get? If you can get a Tsunami sounding decent without having a personal CPU crash, a couple more CV's *probably* won't melt you down! :D

By the way, a much more wordy (and more current) lamentation on this topic can be found here:
http://tonystrains.com/forum/viewtopic.php?t=458
 

pgandw

Active Member
An article on Tony's Train Exchange website got me thinking about this again.

Another wishlist item hinted at above is exhaust sync:

Why not allow functionality that maps the BEMF signal to an exhaust event? The gearing on a locomotive is a constant. It never changes. Therefore, the motor will always rotate x many times for each complete driver revolution. I.e. every 13 motor revolutions = 1 chuff. The user would have to map this association (assuming it's not a factory installed decoder), but once set, exhaust sync would be accurate regardless of speed!

Matt

The Tsunami decoders presently set the steam chuff based on voltage or a cam on a driver axle - your choice. Now, I don't own one, but those who do claim they can get pretty good synchronization at low speeds without a cam. One issue is that the sound can be at the correct interval but not properly timed to the position of the pistons, which could be detectable at very slow speeds. The second issue is getting the cutoff sound to be appropriate for the work the engine is doing. Tsunamis allow you to adjust the cutoff manually, and have an automatic profile which varies with voltage. Neither is entirely satisfactory.

Using BEMF and the gear ratio would probably be somewhat more accurate interval sych than just voltage.

But it is the number of poles in the motor times the motor RPM that determine the number of BEMF pulses. And the smoother and more efficient the motor, the less sharp the rise of the BEMF pulse. All in all, it could be a better solution than an axle cam depending on how precisely the BEMF pulse can be timed.

However, measuring BEMF still doesn't handle the cutoff all that well because our motor loading scenarios don't even come close to replicating the prototype. Again, it might well be better than the current situation. It sure would be a significant increase in processing, though.

my thoughts, your choices
 

woodone

Member
Lights

I would like to see some CV that would control voltage to the lights.
It takes more time to set up lights than to install a sound decoder and speaker. This is very true if you have to deal with 1.5 volt lamps, and there are more than one.
 

Fluesheet

Member
The Tsunami decoders presently set the steam chuff based on voltage or a cam on a driver axle - your choice. Now, I don't own one, but those who do claim they can get pretty good synchronization at low speeds without a cam. One issue is that the sound can be at the correct interval but not properly timed to the position of the pistons, which could be detectable at very slow speeds. The second issue is getting the cutoff sound to be appropriate for the work the engine is doing. Tsunamis allow you to adjust the cutoff manually, and have an automatic profile which varies with voltage. Neither is entirely satisfactory.

Using BEMF and the gear ratio would probably be somewhat more accurate interval sych than just voltage.

But it is the number of poles in the motor times the motor RPM that determine the number of BEMF pulses. And the smoother and more efficient the motor, the less sharp the rise of the BEMF pulse. All in all, it could be a better solution than an axle cam depending on how precisely the BEMF pulse can be timed.

However, measuring BEMF still doesn't handle the cutoff all that well because our motor loading scenarios don't even come close to replicating the prototype. Again, it might well be better than the current situation. It sure would be a significant increase in processing, though.

my thoughts, your choices

Thanks for your thoughts pgandw!

I currently own one Tsunami, and while I've been very happy with the sounds themselves, the auto chuff has been disappointing outside of a narrow speed range, in spite of a lot of experimenting. Where in the piston stroke the exhaust event triggers is important to me, but comes second to having four of them per revolution! :mrgreen:

I wouldn't mind using a physical cam, though those raise their own issues (retrofittability, wiper setup, etc.). Hence the wish for a motor-triggered cam. Your point about five poles is a good one that I'd also thought about; in combination with gear ratio, that would allow for a high degree of potential accuracy.

Sometime after starting this thread, I found that Zimo had created a sound module that could be triggered off their motor decoder's back emf and just today learned that Bluline decoders apparently have a magnetic reed switch triggered by the flywheel fo exhaust synchronization. I glad to hear that there is development in these areas.

Point taken about processing power; while I think that power is certainly available in the sizes required, companies servicing this niche market certainly don't have the flexibility / capital to constantly re-platform their products. As far as I know what I'd like to see would require a completely new decoder architecture, though on the surface it seems the necessary inputs are there for sync. Just a few more tables...

Cutoff I'll have to leave to yourself and other folks more experienced with steam than I. Here is where I picture a throttle handset with a Johnson bar on the side to set cutoff on the fly. Who knows - maybe there'll be a time and command station / decoder powerful and flexible enough to require the user to manipulate both appropriately to get a train started and not run yourself out of steam!

Thanks again for the input.
 
Top