Large Scale Central

Bluetooth and Battery powered DCC

In my quest to find silly things to spend my money and time on, I have wired up my new TCS Wow decoder to control it from a cheap android phone. It’s the same firmware and Android App I did for the Economi but the basics worked out of the box on the TCS. The TCS does sound really good but it looks to have some quirks in the control chain so I’ll be noodling with that as I go along. Anyhow, if you are interested, I have some prose, pics and a video on my site-

http://martinsant.net/?p=3568

Martin

Martin, you might want to be sure to run the motor calibration and turn on “traditional” motor control, that will help a lot.

I’ve not found a way to stop the sound dropouts, and there are also bugs in the custom speed table.

This might be helpful: https://elmassian.com/index.php?option=com_content&view=article&id=440&Itemid=504

Greg

Greg, thanks for that. Hmm. After reading your page I’m pretty sure I must have a newer rev of firmware. Or something. I didn’t configure the decoder at all, I just hooked it up to my widget and sent it 128 speed step messages. Reducing the throttle to zero worked fine, I didn’t have to send it any brake commands at all. It did seem to have some acceleration and deceleration configured already as it took a bit to speed up and then some to slow down if I closed the throttle suddenly. But it wasn’t very long- no runaway sort of thing. I have not tested it hardly at all so I can’t be sure but with the default engine selected, I didn’t hear any drop outs and I thought the sound was pretty good. Perhaps I just have not run it enough. Or maybe they did fix these issues? Not sure. I hope to have time to poke at it some more this weekend.

Then they changed the default… can you check which motor control mode is in force? (traditional vs prototype)… the online manual states prototype is default. (perhaps someone else changed it for you?)

Anyway, was just warning you if you got the weirdness…

Greg

Well, it was fresh out of the box. Or static envelope as it were. I power it up and send it idle packets. About 3ms or so apart and it just sits there and I hear the idle sound. Nice idle. We shall see how it goes.

Straight out of the packaging, “Prototype” throttle mode is default. This mode integrates the motor’s BEMF with the prime mover sound so that it reacts to changes in speed and load. At any given speed step, the motor could be at notch 8, idle, or anywhere in between depending on what the locomotive is doing. “Traditional” throttle mode simply ties the prime mover “notches” to speed steps. (i.e, the faster the throttle is set, the higher the RPM of the prime mover.) “Prototype” mode is the bread-and-butter of the TCS software, and once calibrated, it’s a thing of beauty. (Calibrating it–not so much. Lots of trial and error, and the manual isn’t as clear as it could be.)

By default, the deceleration CV (CV4) is programmed to 60 in what TCS calls “mainline” mode. At this setting (on my loco) it takes about 53 seconds to come to a complete stop from full speed. With CV4 set to 255 (maximum), my loco takes about 3.5 minutes to stop from full speed. I currently have it set to 180, so it takes about 2.5 minutes to come to a complete stop from full speed.

For those who think 2.5 minutes to come to a complete stop is far too long, that’s what the brake function is for. TCS uses a progressive 5-step brake structure, where each successive press of the brake key (F7) increases the braking amount by a programmable amount. By default, it’s 20%, 40%, 60%, 80% and 100%, though I’m not sure the percentages are quite accurate. On my loco, without brakes, it takes about 2.5 minutes to come to a complete stop from full speed. Pressing the brake key once decreases that dramatically to just under 30 seconds. Pressing it twice drops the stopping time to around 22 seconds. Three times 16 seconds, 4 presses stops me inside of 10 seconds, and a fifth press stops me inside of 5 seconds from full speed.

TCS also has what it calls “switching” momentum. This (by default) is set to 5 for both acceleration and deceleration, giving you near instant response time to changes in the throttle. You can toggle between “switching” and “mainline” momentum by pressing the F15 key. (This is handy for kids or adults who don’t quite grasp the notion of using a separate key for slowing down the locomotive.)

Note: The values for acceleration and deceleration in “mainline” and “switching” mode are set via the 4-CV programming protocol. Simply changing CVs 3 and 4 does not affect these values. For example, the decoder comes with the “mainline” acceleration and deceleration values of 20 and 60 respectively. I find that too quick, so I re-program CVs 3 and 4 to 60 and 180. The loco reacts based on those new settings. I change to “switching” momentum via the F15 key. Doing so writes a value of 5 to both CV3 and CV4. I switch back to “mainline” mode. The default values for “mainline” mode are still 20 and 60, so the decoder will write 20 and 60 to CV3 and CV4 respectively. In order for my 60/180 to become permanent for “mainline” mode, I must re-program the default using the 4-cv programming (or Decoder Pro). Then, I can toggle back and forth to my heart’s content. (Another place where more clarity in the manual might be helpful.)

I’ve got probably around 40 hours or so on my latest TCS WowDiesel decoder (the same one Greg and Martin have). I’ve not had any of the issues Greg has experienced with respect to sound dropping out or runaway locos. I don’t use custom speed tables, so I can’t speak to the issues Greg seems to have with them. Outside of an occasional stutter at slow speeds, mine has been rock solid. TCS decoders do occasionally go berserk and need to be reset in order to behave properly once again. I’ve seen it with other TCS decoders I have in use, and have read of others’ experiences as well. Perhaps that’s why TCS has a “no questions asked” return policy–just in case.

Later,

K

I have videos of the sound dropping out, and I have the latest firmware. It drops out more frequently when you stop and reverse direction quickly. It also often has the prime mover running at a high speed when the loco is stationary.

I have videos of all of this, but have not posted them on Youtube, and don’t intend to. I have no ill feelings towards TCS, just won’t use their stuff any more.

I get what they want to do with prototype mode, but they should have done something better with the brake function, and I hope they do.

Greg

I certainly don’t doubt you Greg. I’ll keep an open mind for now and see what happens. Thanks to both you and Kevin for the comments and research. Much appreciated.

Greg, can you e-mail me a link to the sound dropping out video? I’d like to hear what you’re referring to, so I can keep my ears open for it. When I stop and reverse direction quickly, I get the motor sound spinning down to idle (the result of stopping), then not immediately spinning back up as the loco begins to move the opposite direction. It sounds (to my ears) like it’s playing out the “wind down” sound clip before it starts playing the “wind up” clip in response to it moving, even though the loco is 2’ down the track already. There’s a bit of a lag (5 - 10 seconds, maybe). It never completely drops out (i.e., goes silent), rather it just drops to low idle, which is quiet. If you wait a few seconds between moving again once you’ve stopped, the sound keeps up.

I also get the motor spinning up to high RPM when just starting to move (never stationary). You can adjust how much “wandering” there is in the motor sounds in response to motor BEMF. (4-CV programming mode = Momentum Effect) These parameters are what I refer to when above I write that calibrating the Prototype Mode takes some trial and error, and could be better-explained.

I’ll have to disagree with you on the brake function. I find it rather intuitive. Tsunami and Zimo just have one brake rate, which I think can be somewhat inconvenient when switching. (I have Tsunami’s “momentum override” mapped to F9 as a “stop NOW” key. It’s combined with the coupler clank sound.) With the TCS, you press the brake key to begin to slow down. If you need more brake, you press it again for a bit more pressure on the line. Still not slowing down fast enough? Make another application. It’s actually similar to how QSI does theirs, though QSI adjusts the brake rate in connection with how long you hold down the key, not the number of times you press it. A quick tap of the brake key makes a light application, whereas holding it down makes a heavier application. Also, the sensitivity of QSI’s braking seems to be tied to the value set in CV4. If you have a lot of momentum programmed in, you have more control over braking rates than if you do not. The downside is that it still has a maximum brake rate, which still takes 15 or so seconds to stop from full speed. (And none of that is documented anywhere, mind you. I learned that through observation and experimentation.) The Airwire throttles do not support emergency stop, so there’s no way to stop locomotives immediately in an emergency. (Rather prototypical, mind you, but it does lead to more time in the shop.)

BTW, did you ever get your prime mover selection issue sorted out?

Later,

K

I prefer the single press for brakes, and increasing stopping “pressure” the longer you hold it down.

5 different levels of braking with no feedback on which you have is not intuitive.

I’m not uploading the video for the reasons given and I am not getting into a pissing match with you Kevin about what this decoder does or does not do. I spent a number of hours with this decoder.

Greg

Everyone relax.

Martin, I do run my DCC just under 24 volts, and perhaps this decoder is sensitive, like the ill-fated “quasinami” that went weird at 21 volts. I doubt you are running that high a voltage.

I went to the link you provided, but did not see any TCS-specific comments, is there another page?

Was interested in the “quirks” you found to see if they parallel mine.

Greg

14.8v via 4 cell lipo. I modulate the DCC packets on top of that. I don’t particularly like the DCC interface but it is what it is. I NEED A KNOB. I do not like the slider on a phone. I am working on that.

I hear you, what about perhaps buying empty cases from NCE that have just the rotary encoder and a keyboard? I wonder if they would sell those …

Greg

Dave Bodner built his own DCC controllers. He knows where to get the parts like the keypad and knob assembly.

Well, I want to keep the phone as the UI because not only does it have a nice colorful touch screen but also the integrated Bluetooth and the Wifi fills the needs of where I want to go with all my network ideas. Plus I can program it in Python which is one of my all time favorite languages.

So it’s not so much the parts but how to connect them to the phone. In theory, I can connect an OTG cable to the phone’s USB port and then talk serial from the phone app to a custom micro that reads the knob and buttons. I have a circuit cobbled together but have not really put any effort into it yet so I don’t know if that’s viable or not.

Greg Elmassian said:

Was interested in the “quirks” you found to see if they parallel mine.

Greg, this is exactly the dialogue which I’ve been trying to engage in with you. It’s obvious that both of us have had widely differing experiences with identical decoders. The question is, “why?” What am I doing that you’re not, and vice versa? When you first posted that you were having trouble assigning a prime mover sound, my response was not in any way intended to insult you or discredit you in the least. Everything I have posted with respect to my decoder, its settings, and its performance as a result of those settings is to give you a baseline against which you can compare your decoder. If you know I have done steps A, B, and C, and gotten good results, it stands to reason that if you follow steps A, B, and C and do not get good results, something is amiss and should be addressed by the manufacturer. You wouldn’t know which steps I followed if I didn’t post steps A, B, and C in the first place.

Not being able to assign a prime mover sound via Audio Assist = not normal behavior

Locomotive not stopping = not normal behavior

Sounds cutting out = not normal behavior

These are some of the issues you seem to have had which I have not. I’ve tried my best to describe the settings I use and the results I see, the intent being that you can compare them to your board and seek resolution. (I can’t help you with your speed table issue at the moment, but I’ve got a new board on order and would gladly set one up to your specifications once it arrives to see if I see the same things you’re seeing.)

It never has been my intention to start a “pissing match” (your words) with you over this. Believe me–knowing our history over the years, even agreeing with you is fraught with peril. I sincerely thought I could provide information to help you get better performance out of your decoder. I’ve got 4 TCS Wow decoders in operation right now, and they all work very well. It hasn’t been a smooth road on any of them to get them running that way. Each has had its idiosyncrasies. It’s taken phone calls and e-mails to TCS. It’s taken other modelers e-mailing me their decoder settings so I can compare them with mine to determine if there may be something wrong with the software. (It happens.) It’s required resets. It’s required me to send boards back to TCS for reprogramming. Through it all, I’ve learned a lot about these boards. I figured I’d pay it forward to someone else who was having issues. I don’t care who that “someone” is. If I have some information which I think may be beneficial to help a modeler resolve an issue, I’m going to offer it. You have your web pages you point folks to; I write individual e-mails. The spirit is the same–to provide information that may make things work in the end.

It’s up to you. In another thread on another forum recently, another modeler misconstrued your sincere intentions behind something you sent to him and publicly took you to task as a result. Your response was–in sum and substance–“I was just trying to help.” Same thing here. Just because there may be a “history of acrimony” (again, your words) between people does not mean every action between them is motivated by malice. Folks don’t have to like each other to be helpful to them. I’m not asking you to like me. I’m simply suggesting that–perhaps–you might look at my posts in a different light; one of simply trying to be helpful and resolve your decoder’s performance issues.

If you’re interested in sorting out weaknesses you see in your TCS decoder, to see if my “quirks” parallel yours (as you invited Martin to do), my phone and in-box remain available.

Later,

K

Kevin,

Would you please post Greg’s results if he contacts your phone or inbox as others would like to learn as well.

Thank you

Rooz

Hmmm, if I made private communications to Kevin, would he share them with the world?

I would hope the response would be “it’s up to Greg”

But don’t worry, this is not happening, my results are on my web site, as always, and people can make their decisions to use that information as they want, and as my site says, email me with errors or corrections. Facts though, not opinions.

Greg

Just one more data point- I got a response from TCS Tech Support yesterday and he indicated that there had been some hardware changes to the 5amp since it was first released. He didn’t go into what they were and was not clear if the firmware changed or not. Mine is pretty new, I got it from RP about a month ago. He said there was no way to read the s/w version so I don’t know exactly what rev of firmware is in there.

I will defer to Greg and Kevin as they have way more experience with DCC than I have, and I am doing kinda weird things to mine- I generate the DCC packets in my own firmware- but so far I like the TCS. There are some things that seem, uh, eccentric, but I really like the brake squeal and how that works. Since I’m using my phone, holding the ‘brake button’ down is the same as tapping it so I can blast out the messages really fast if I want to stop quickly. I think it’s kinda cool how it ‘coasts’ when you shut down the throttle.

And not directly related to TCS- but I like the phone too. $50 from Amazon, nice display, great range, I’m getting 100ft easy.

BLU Studio X8 HD

I really really miss a knob though (http://largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-cry.gif)