I do not like touch-screens at all. I've got a cell phone and a tablet, and neither of them responds the way I would want. To get enough information on a screen, they keep making the characters/buttons smaller. For a guy like me, it is hard to reliably hit what I want on the screen. The latest Android "upgrades" seem to have made my screens more sensitive, so I frequently end up with the wrong button pushed. That's not so bad if I'm just browsing the web, but my train controls need better accuracy than any of my touch screens provide, and I am NOT going to walk around my layout with a 19-inch monitor.
It takes an iron man to play with a toy iron horse.
I think I have found my new lunchtime project.
Don - Specializing in layout DC->DCC conversions
Modeling C&O transition era and steel industries There's Nothing Like Big Steam!
DigitalGriffinHave you seen the home brew low cost DCC command stations using pi and arduino? We're a small market. It represents a couple thousand unit sales a year I would imagine.
they are low-cost if you're a DIYer and can build and debug the system yourself. A market of a thousand doesn't sound very profitable.
i am a DIYer and i've built an NCE compatible cab. probably $10 is cost and $1000+ in man-hours.
mine doesn't have a knob. I find it more comfortable to use than the powerCab. I'm beginning to like just using buttons to adjust the speed step. I'm satisfied with the LED coming on when programming the loco number and it going off when accepted by the command station.
i'm realizing i didn't need a 16 button keypad, 12 would be enough. Besides select and enter, i need the buttons to be digits when entering a loco number. those same ten buttons can be something else when operating, i currently use 5: forward, reverse, increase and decrease speed, 0 for headlight. Also placement of the buttons (human factors)
i've looked extensively at the DCC++ code. Thinking about building my own, but would need it to support a cab over rs-485. The NCE protocol is probably as good as any. would learn something about the pro/cons of that.
greg - Philadelphia & Reading / Reading
gregc khier Continue searching for execuses and convincing yourselves of the feasiblity to pay 250$ and more until a Chinese manufacturer flood the market with 30$ DCC devices that will work like LENZ, Digitraxx, ESU and even better. Only then you will get the point. in the 20+ years of DCC has a manufacturer taken advantage of these marketing blunders and come out with an inexpensive unit that provides better capability? Are there that many unsatisfied modelers? what do you think of the TCS system mentioned above?
khier Continue searching for execuses and convincing yourselves of the feasiblity to pay 250$ and more until a Chinese manufacturer flood the market with 30$ DCC devices that will work like LENZ, Digitraxx, ESU and even better. Only then you will get the point.
in the 20+ years of DCC has a manufacturer taken advantage of these marketing blunders and come out with an inexpensive unit that provides better capability? Are there that many unsatisfied modelers?
what do you think of the TCS system mentioned above?
Have you seen the home brew low cost DCC command stations using pi and arduino? We're a small market. It represents a couple thousand unit sales a year I would imagine. Given six months fulltime, I'm sure I could create a better throttle that is loconet compatible. I could even design the case using tinkercad and print it on shapeways.
I've done ground up before. I've written firmware and debugged satellite hardware. It's not that hard.Understand I'm not here to slam digitrax, NCE, or others. But I would hate to see them lose market share on what could be an easy win.I'm a principal engineer at a very large company and our main competitor is losing market share because the way they make things has made it too expensive to compete. They had to move to Mexico.
gregc in the meantime Walid, what do you think about the new TCS system? it uses today's technology.
in the meantime Walid, what do you think about the new TCS system? it uses today's technology.
Walid
i guess we'll have to wait and see.
gregc in the 20+ years of DCC has a manufacturer taken advantage of these marketing blunders and come out with an inexpensive unit that provides better capability? Are there that many unsatisfied modelers? what do you think of the TCS system mentioned above?
Just wait until they discover the MRR market as they discovered the RC world before. We will see then who will pay 800$ for an ESU Ecos or 700$ for Uhlenbrock Intellibox command stations.
Regards
khierContinue searching for execuses and convincing yourselves of the feasiblity to pay 250$ and more until a Chinese manufacturer flood the market with 30$ DCC devices that will work like LENZ, Digitraxx, ESU and even better. Only then you will get the point.
DigitalGriffin Now manufacturers have ZERO excuse for not setting up better interfaces these days. Devices like arduino and raspberry pi make development a breeze and the cost low. A LCD 2x40 display is $5.00 An arduino is $10.00. A 16 button keypad is $3.00 (4 arrow keys, an enter/Yes, a back/cancel/No + 10 digits)
Now manufacturers have ZERO excuse for not setting up better interfaces these days. Devices like arduino and raspberry pi make development a breeze and the cost low. A LCD 2x40 display is $5.00 An arduino is $10.00. A 16 button keypad is $3.00 (4 arrow keys, an enter/Yes, a back/cancel/No + 10 digits)
For those who philosophize the cost issue and argue with development cost (of a 30 years-old system),..etc...etc... Continue searching for execuses and convincing yourselves of the feasiblity to pay 250$ and more until a Chinese manufacturer flood the market with 30$ DCC devices that will work like LENZ, Digitraxx, ESU and even better. Only then you will get the point.Regards
The blasphemous OP
gregc DCC is not well suited for accessory (e.g. turnout) control. See LCC, the modern technology for non-locomotive control.
DCC is not well suited for accessory (e.g. turnout) control. See LCC, the modern technology for non-locomotive control.
I beg to differ...
It all comes down to user interface.
On my Z21 (especially when using a tablet), I can have several pages of track schematics with turnouts, routes, occupancy and signals.
And, the turnouts can be switched with regular buttons as well, as the switch decoders I use has connections for them.
Swedish Custom painter and model maker. My Website:
My Railroad
My Youtube:
Graff´s channel
DigitalGriffin Well I would consider the following nice to have dedicated buttons for Light (F0) Bell (F1) Horn/Whistle (F2) Coupler Clank (F3)Brakes (that work)Dynamic Brakes (that work)/ Cylinder blowdownThrottle/Effort UpThrottle/Effort DownDitch lightsThese are ALL common operations by REAL engineers. Yet NMRA never desginated common function buttons which are mapped all over the place between the big 4 (TCS/Loksound/Soundtraxx/QSI) It makes their operation cumbersome. And setting up consist is still a pain. Specifying which is lead, which is middle, which is rear, which is facing forward, or reverse.
Well I would consider the following nice to have dedicated buttons for
Light (F0)
Bell (F1)
Horn/Whistle (F2)
Coupler Clank (F3)Brakes (that work)Dynamic Brakes (that work)/ Cylinder blowdownThrottle/Effort UpThrottle/Effort DownDitch lightsThese are ALL common operations by REAL engineers. Yet NMRA never desginated common function buttons which are mapped all over the place between the big 4 (TCS/Loksound/Soundtraxx/QSI) It makes their operation cumbersome. And setting up consist is still a pain. Specifying which is lead, which is middle, which is rear, which is facing forward, or reverse.
And that's why I prefer a touch screen throttle.
Where you can map all functions to be where YOU want them to be!
The best hand control so far is the ESU Android throttle with physical throttle wheel and programmable physical buttons as well as the color display.
I hope that more manufacturers will follow suit.
After a four-decade hiatus from model trains (much of it consumed running the big trains) I decided to begin spending a little time with them again.
Of course, the entire paradigm of "control" had changed during my absence, so I began investigating the dcc systems available. To me, it was like I had come from outer space -- I had no prior experience or even exposure to the various systems out there. They were all new to me.
I ended up choosing the Roco z21 (small "z", ordered for a good price from Germany).
To steal a quote from Mr. Spock in the old Star Trek episode "The City on the Edge of Forever", compared to the wifi, graphical touchscreen interface of the z21 the others seemed like "stone knives and bearskins". As great a jump forward as was the Mac OS in 1984 when compared to MS-DOS.
That's my opinion and I'm stickin' to it!
DigitalGriffin Stevert DigitalGriffin Digitrax has the LNWI which works with JMRI or Engine Driver which works on android. That's not correct. The LNWI works with WiThrottle for IOS, Engine Driver for Android, and other throttle apps that use the same communication protocol. It does NOT require JMRI or the computer/WiFi router JMRI needs to communicate with those apps. Instead, the LNWI takes the place of JMRI and that computer/WiFi router. Not to argue the point, but I never claimed otherwise. :)
Stevert
DigitalGriffin Digitrax has the LNWI which works with JMRI or Engine Driver which works on android. That's not correct. The LNWI works with WiThrottle for IOS, Engine Driver for Android, and other throttle apps that use the same communication protocol. It does NOT require JMRI or the computer/WiFi router JMRI needs to communicate with those apps. Instead, the LNWI takes the place of JMRI and that computer/WiFi router.
DigitalGriffin Digitrax has the LNWI which works with JMRI or Engine Driver which works on android.
Digitrax has the LNWI which works with JMRI or Engine Driver which works on android.
Not to argue the point, but I never claimed otherwise. :)
DigitalGriffin the LNWI which works with JMRI
Stevert DigitalGriffin Digitrax has the LNWI which works with JMRI or Engine Driver which works on android. That's not correct. The LNWI works with WiThrottle for IOS, Engine Driver for Android, and other throttle apps that use the same communication protocol. It does NOT require JMRI or the computer/WiFi router JMRI needs to communicate with those apps. Instead, the LNWI takes the place of JMRI and that computer/WiFi router.
DigitalGriffin In some ways I'm going to agree and disagree. For the most basic of functions, standard hand throttles are superior to touch screens. HOWEVER, Standard hand throttles are often a confusing mess when it comes to setting up and breaking down consist. They cannot be updated to handle differences between different decoders. They cannot have buttons with dedicated names like "dynamic brake", "brake", "coupler", "ditch lights", "dim", "mute" because these all shift between decoder brands. Now manufacturers have ZERO excuse for not setting up better interfaces these days. Devices like arduino and raspberry pi make development a breeze and the cost low. A LCD 2x40 display is $5.00 An arduino is $10.00. A 16 button keypad is $3.00 (4 arrow keys, an enter/Yes, a back/cancel/No + 10 digits) If they won't do it, they will find a bigger and bigger percentage of cottage industries doing customized ones. That means lost handheld sales. I will grant a concession that Lenz and Digitrax have finally (somewhat) updated their interface. Digitrax has the LNWI which works with JMRI or Engine Driver which works on android.
In some ways I'm going to agree and disagree.
For the most basic of functions, standard hand throttles are superior to touch screens.
HOWEVER,
Standard hand throttles are often a confusing mess when it comes to setting up and breaking down consist. They cannot be updated to handle differences between different decoders. They cannot have buttons with dedicated names like "dynamic brake", "brake", "coupler", "ditch lights", "dim", "mute" because these all shift between decoder brands.
If they won't do it, they will find a bigger and bigger percentage of cottage industries doing customized ones. That means lost handheld sales.
I will grant a concession that Lenz and Digitrax have finally (somewhat) updated their interface. Digitrax has the LNWI which works with JMRI or Engine Driver which works on android.
This is also touching on the OP's assertion that these things ought to be cheap because the components are cheap. Yes, for about $10 in components I can have a 4x20 LCD (still not completely sufficient to avoid some abbreviation of the mode words), a plenty powerful microcontroller, and a cheap keypad to drive it all. But I will not that the Serpac case Digitrax uses sells for $15 on DigiKey, and it is not big enoug for a 4x20 LCD and the full keypad (note the NCE Hammerhead is MUCH larger, at least at the top where the display is). I think they use a Serpac case as well, but I didn;t see that one - I know I've seen it available on DigiKey. Then you have the cable and all the miscellaneous parts. Not to mention board layout and design and coding. Oh and since it uses a cable, or an RF module, it needs to be FCC certified to be legal to sell - that testing alone is in the 10's of thousand of dollars - if you can actually sell several thousand of these units (not likely), that's 10's of dollars per unit just for certification costs. So no, it is NOT realistic to think we can have a fancy throttle with many character displays that can display every nuance of DCC operation and it should only cost $100. The R&D costs alone PER UNIT are probably that high. Add BOM cost and assembly costs, even automated assembly as much as possible, and it goes up from there. Dave Jones of EEVBlog fame has a good video on his YouTube channel about the true costs to design and manufactuer a product for sale. It's NOT as cheap as you think, despite you personally being able to buy the LCD from some Chinese supplier on eBay for $1, and the microcontroller likewise for $1, and buttons in packs of 100 for $5. You can build one for your own person use for cheap, sure.
One thing I'm not getting either - how many digits do you need for a speed display? The throttle goes fom 0 to 100% (Digitrax way) so there is no concern about speed steps. If you display actual speed steps, that makes it MORE complicated, because then it might go 0-14, 0-29, or 0-126 (yes, 128 speed steps reserves a few, it's not a full 128 unique steps). I'll take 0 to 100^, thank you very much, which is how most systems display it. That's a mere 2 1/2 digit display. Until every loco has an encoder wheel to measure actual speed (and unless that is some mechanically calibrated device rubbing on the track, it will need to be calibrated in the decoder to convert RPM of the shaft to actual speed based on gear ratio and wheel diameter - you can;t shortcut this) and direct feedback (both the Lenz Railcom and Digitrax's Transponding have shortcomings, and even the direct radio systems that have nice 2 way communication with the loco receiver still need some mechanism to calibrate any sort of scale speed readout), the only way you have to make the throttle read scale speec is to calibrate each loco so the speed percentage of throttle shows the scale speed the loco is configured to run at.
And what sort of feedback is needed? You press the headlight button, the lights come on or off. Press the horn button and the horn sounds. What more do you need? Most systems also show the function status on the display. All was good originally with 4 functions, and then 8, and even 12 - most systems can directly display the status of F0-F12 on their displays. Then they had to go ahead and make it 29 functions. No one wants an additional 10-15 buttons so that F0-F28 each get their own button on the throttle, so they ALL have some sort of shift to access functions past F10 or F12. You can't get around that, without making the controller even bigger. Well, you can - most every sound decoder, for operating in a realistic manner, needs more than 12 functions. Many times the extra functions include station calls, cattle noises, wayside sounds, non-prototypical radio chatter (even on locos from before the common use of radio), even train wreck noises. Unnecessary. And then the 'lay' buttons - there's already a momentary horn/whistle, but then they add a dedicated button to blow a grade crossing. Not needed, just use the horn. So, do we really need 29 functions, with distinct indicators for each? No. Leaving that out already makes the throttle much simpler in terms of UI. Having it optionally selectable means there's some understanding needed to access that stuff, but if you don't need or use it, it also doesn't get in the way of basic operation and you can forget it's even there.
Everyone is different. My ex father in law is anything but a computer genius, yet he had no difficulties figuring out how to run his trains using my Digitrax Zephyr. That has about as small a display as you can get - 4 LED digits. It only displays the selected address - no speed. Speed is controlled by a knob exactly like found on DC power packs for decades - the more you turn the pointer towards FULL, the faster you go. A simple but very effective user interface. One that doesn't take a computer expert or an engineer to figure out. Or even a college degree. Conversely, the old Digitrax DT100 throttle, discontinued at least 20 years ago, is an example of how NOT to do it - but iot comes from the early days of DCC when short addresses ONLY were the norm. Even selecting a long address with the DT100 is not as simple as newer devices. The ultimate in difficulty though is trying to program CVs with one. You have an 8 character LCD, plus a few extra segments to indicate modes. There are only 8 buttons. Simple? Hardly. I can figure it out, if I read the manual. I've memorized how to select a loco with it, but forget programming, without the manual handy. It can't even program many of the CVs in modern decoders anyway. In the balance between compact and easy to use, this one went all the way to compact - it is nice and small, and even that small it can actually run 2 locos at once. But the sacrifice for that was any sort of usability whatsoever. ANd it was discontinued in favor of newer products with far better interfaces a long time ago.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
khierbut certainly an interface where I can read clearly "ADDRESS" not "Adr." or "Add" on a monochrome LCD screen.
assuming you mean address of loco on a throttle, an LCD screen seems like unnecessary hardware. I'm comfortable with a smaller throttle without a display when setting the loco address i want to control. I'm also less willing to spend $$ on one that does have a display nor do i want a larger throttle where "address" can be spelled out.
khierWhere I can read clearly speed information not a dual seven-segment display.
khierWhere I can see acknowledge of my commands not a LED flashing twice.
it's good to use a throttle with a display when programming and testing a locomotive, but during regular operation, something small and lightweight is more comfortable. I've become comforatble with a cab without a knob, just using buttons to set the speed step.
human factors is more than just displays. It seems comfort is also a consideration
BMMECNYCAt the risk of sounding like an infomercial for TCS ... Im buying this when its released to replace all of the above ... Because that throttle was the most comfortable model train throttle I have ever held. Also because of Layout Command Control.
Because that throttle was the most comfortable model train throttle I have ever held. Also because of Layout Command Control.
There are some that see LCC as the solution to all, but I don;t care what sort of hand control you have, or how facny the UI is, switching back and forth between sending commands to the train you are running and operating accessories like turnouts is going to be a pain. ANd on complex layouts - you'll need some sort of labeling to tell you what turnout it it. Maybe a fancier UI can use a location name, like a prototype, which is good for the seasoned modeler. Labels on the layout are certainly prototypical, by for scale reasons the lettering has to be fairly small. Something other than just numbers, corresponding to the stationary decoder address, make it easier to figure out WHAT to operate, but still doesn't remove the cumbersome change from one mode to another. This is why I still prefer the fascia controls with a track schematic. And individual buttons over toggles. Press the button on the track you want the train to go on. If the light comes on, you're lined up. Doesn't get much easier for a UI than that. If behind the scenes this is all hooked up to some sort of control bus, to allow centralized control or whatever - that's fine. It doesn't have to intereact with the DCC system at all, really. Or it can - one of the reasons I went with Digitrax is that the Loconet control bus, unlike many, is quite capable of handling the traffic from detectors, stationary decoders, and signal controllers right along with the DCC train control traffic. That's probably why I am not as excited about LCC as others. And why I am not just using Loconet comes down to a complexity issue. If I was using commercial interface devices - I'd just use Loconet to interconnect it all. But for DIY - first I'd have to make sure I am using correct Loconet packet types to communicate with my devices. Then I'd still need to build a protocol into that to send the actual data back and forth. And make sure that each of my devices properly behaves to share the Loconet bus correctly per the specifications. Same with other protocols like Ethernet. I briefly toyed with the idea of using Ethernet for the command and control bus. The physical protocol is handled by the interface chips, but then I am effectively connecting a more powerful micro to the 8 bit Atmel one just to get a physical interface that runs far faster than I'd ever need, and faster than the small micro can even keep up with. And I'd still need to take a protocol on top of the physical layer (like TCP or ICMP).
gregc khier I am curious to know what do you think of the available command stations, which one you prefer, and which fuctionalities you wish to see. it's more than just a question about less than modern technology, it's also economics and the available market. Any new, more modern system today needs to compete with existing products -- it would need to be significantly better but at about the same price. you said it yourself khier but if you look at the ones with TFT display and easier use like ESU Ecos or Viesmann 5300, you are abocve the 600$ level. many existing modelers rewired and replaced their DC systems with DCC because DCC allowed them to operate trains in a way that you can't with DC and much simpler, justifying the time , effort and replacement cost. There's not a big market for modeler to replace there existing DCC system with one that is more modern but does about the same thing. DCC is not well suited for accessory (e.g. turnout) control. See LCC, the modern technology for non-locomotive control. Many modelers, like Randy, build there own for block occupancy, turnout position monitoring, and turnout and signal control
khier I am curious to know what do you think of the available command stations, which one you prefer, and which fuctionalities you wish to see.
it's more than just a question about less than modern technology, it's also economics and the available market. Any new, more modern system today needs to compete with existing products -- it would need to be significantly better but at about the same price. you said it yourself
khier but if you look at the ones with TFT display and easier use like ESU Ecos or Viesmann 5300, you are abocve the 600$ level.
many existing modelers rewired and replaced their DC systems with DCC because DCC allowed them to operate trains in a way that you can't with DC and much simpler, justifying the time , effort and replacement cost. There's not a big market for modeler to replace there existing DCC system with one that is more modern but does about the same thing.
DCC is not well suited for accessory (e.g. turnout) control. See LCC, the modern technology for non-locomotive control. Many modelers, like Randy, build there own for block occupancy, turnout position monitoring, and turnout and signal control
Have to agree about turnout control thats why I have used DC and/or ground throws on all the layouts I have built.
Joe Staten Island West
How on earth did so many come to the idea I specifically mean touch screens when I talked about better interface? It may have touch screen, it may not be, but certainly an interface where I can read clearly "ADDRESS" not "Adr." or "Add" on a monochrome LCD screen. Where I can read clearly speed information not a dual seven-segment display. Where I can see acknowledge of my commands not a LED flashing twice. Walid
Forward and reverse, speed control, horn, bell, coupling and uncoupling sounds, air down, getting a verbal speed report and some accessory control I can do by feel alone. I really like the fact I don't need to look at a screen or the controller to do these functions. Is it possible to do these on a screen without looking at it?
Brent
"All of the world's problems are the result of the difference between how we think and how the world works."
PC101Bottom line, the Lenz system runs trains.
For me, that's the bottom line. My Lenz System 100 has been running trains for over a dozen years now.
When I was looking to buy, I saw that Lenz was a 5-amp system right out of the box. The "stack" is a whopping 32 locomotives, large enough for my entire roster and then some, plus consists.
The thing that got me, though, is the great big buttons on the throttle. It is not like trying to make calls or do texts on a cell phone. I love that throttle just for the buttons. I can see the buttons and identify them with my old eyes, and I can press the one I want every time with my large fingers. That's why, when a thread comes on about "Which DCC system should I buy," my answer is to try them all out and pick the throttle you like best, because that's going to be your interface to your trains.
Ahh, yes. The DCC system debate...
My opinion is that the DCC manufacturers are now trying to make the user interface as user friendly as possible.
That's why you see a lot of touch screen devices. They make the whole thing much easier for the user (if made right).
Example:
I use the ROCO z21 DCC system. It is a self sustained system with its own app and router for the mobile devices.
It also has Xbus connections for Lenz and ROCO hand controls.
When I want to connect it to my computer, I just plug in a LAN cable, and it's ready to connect to JMRI so I can program my decoders.
With the new sound decoders with 28 functions, I don't need to remap them, I just assign each function to a button in the app.
No need for a computer engineering degree anymore to use a high end DCC system.
I understand that many users of 20 year old systems like them. But surely you must acknowledge that the user interface of them isn't top notch.
I currently have a NCE DCC PH-Pro 5amp system for the layout and the Power-CAB 2amp system for programming at the workbench. Why? Its what my club used. Its easy to use. Programming 4 digit addresses is only requires entering the menu, then inputing the 4 digit address of my choice (you dont need a CV calculator or ever need to touch CV 17/18)
Ive really only touched CV29 to shut off DC operation. Other CVs Ive touched only by following the exact instructions in the decoders included instruction sheet on some TCS decoders.
Speaking of,
Their sound menu programming is great. Obliviates in most cases the need to ever program a CV.
At the risk of sounding like an infomercial for TCS (no, I dont work for them):
Im buying this when its released to replace all of the above:
http://tcsdcc.com/commandstation
http://tcsdcc.com/throttle
http://tcsdcc.com/mini-throttle
RR_MelI run trains not the controller, simple knob for speed and a couple of buttons for direction, lights and horn.
To me, that's the whole purpose of DCC and it's why I love it.
When I got back into the hobby 14 years ago, I realized many of the things I'd been missing. I had become more of a servant to my computers, both the ones at home and the ones at work. I did Flight Simulator and Train Simulator. Like with my layout, I built an elaborate Flight Sim world, adding scenery and aircraft to my digital world. I loved it, but when I went back to trains, some of the first things I needed to do involved my subway systems and scratchbuilding with styrene and hydrocal. I got plaster and paint all over my hands. I reveled in the processes of building something physical. At the same time, I was running the trains on DCC, again something physical, not just images on a screen.
Between computers and trains we have a continuum of activities. I personally prefer to keep them separate, the digital stuff on one side and the physical stuff on the other. Yes, there is overlap, but I'd rather keep them as far apart as possible.
I have a Digitrax Zephyr DCS50 with two jump power supplies. I Bought them back in 19 whenever and they still work fine.
I'm still using my nearly twenty year-old Digitrax DCS-100. Over the years, I've updated my throttles, fast clocks and even a couple of boosters as improved ones, or ones that better meet my needs have become available, and the DCS-100 continues to support them just fine.
If I were to need to start completely over, the ESU Cab Control system has some very nice technical merits to its architecture, but the deal-killer for me is the cost of the throttles, given the number that I would need. I don't feel that a visual display would enhance my crews' experience enough to be worth the price.
-Fritz Milhaupt, Publications Editor, Pere Marquette Historical Society, Inc.http://www.pmhistsoc.org