|
Post by Commodore on Oct 27, 2018 10:35:03 GMT -5
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 12, 2018 17:27:11 GMT -5
Now the "control console" is ready, it will be installed next monday (I need help to hold it), screwed with its rear panel in front of the layout's frame.
The PCB's (4 items) will only be installed inside (on the bottom) when the console is in place, for a better access to the screws. The panel can be lifted (for this reason, it is a little bit "longer" in front) even when installation will be finished, to get access to the preset trimmers for speed and momentum. There are rather few control elements, because the layout will be operated by a sequencer program (on Atmel 8515) : on the left hand side, the tumbler selects the "auto" mode, the train movements being selected by random numbers. When the layout is in "manual" mode, the yellow, white and grey "Digitasts" select one of 4 possible train movements (freight, railcar in both directions and gravel), the red one selects if the train will stop in the other station or run through and come back to its initial position (toggle function), the green one is for start. The tumbler on the right hand side is the emergency stop (it switches the power to the track off via a relay : this would have been necessary with the initial "vintage" version, I had only on-off (not SPDT) tumblers, and a red Led is on when power is off ... )
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 27, 2018 13:19:24 GMT -5
It works !
Under the lid : the wooden plate in the middle holds the PCB for the DigiTast's, the PCB at the bottom on the left-hand side controls the Led's in the DigiTast's. On the bottom : at the back, the five position detectors (right) and the 4 power transistors for the relays ; in front, on the right hand side, the power driver with OPA547 and the relay for emergency stop ; in the middle, the controls for speed and momentum ; on the left hand side, the board with the Atmel 8515 running at 6 MHz ; the program was written in Great Cow Basic (Open Source). In fact, my layout is analog, but digitally controlled.
|
|
|
Post by Commodore on Dec 27, 2018 21:46:18 GMT -5
That's Different....
Happy New Years 🎉
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 28, 2018 0:53:27 GMT -5
At least, I don't have "No user serviceable parts inside" , and vital SMD components ... But in a near future I will modify the processor's board and add into the free place a (readymade with SMD's, but non vital ! ) Emic-2 speech synthesis board, for a countdown before start when the layout is in automatic mode (it will be possible to deactivate this feature via software with a switch). I believe this will please the grandchildren of my friends when they come together with them to see the layout, because they grew up with an overdose of japanese SciFi animes and video games !
Happy New Year !
|
|
|
Post by markm on Dec 28, 2018 8:49:42 GMT -5
Alberich,
Nice clean design. I like the idea of adding sound. Going beyond what you propose a bit, there are sound chips available preprogrammed with sound. One that comes to mind is circuitron (se pgs 21, 39 of their catalog). Or you could consider a programmable board like: www.adafruit.com/product/2133
Of course you could also try out my first "sound system" , number 755A:
Mark
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 28, 2018 11:49:37 GMT -5
The main advantage of Emic-2 is that it has a built-in "text-to-speech engine", so no preprogrammed sentences are needed ; You send the text as an ASCII-string (via a serial connection), up to 1023 characters long (less with Great Cow Basic whose standard maximum length for strings is 40 characters ... but there is trick to use longer strings, thanks to the GC Basic forum ! ), and it speaks it ; as far as I understood, it was derived from the system used by late Prof. Stephen Hawking to talk, system which was used in the movie " Back to the Future II ", too.
If the 8K program memory of the 8515 are not enough, in case of extensive use of speech, it can be replaced with a Mega162 (same pinout, but 16K of prg mem) or bigger with more RAM. So, if I want, I can add more "unusual", hightech revivals of "vintage" features to my layout ... at any time ! It's simply a problem of modifying the software. But I think I will keep the minimum messages, countdown and alarm ; train announcements could only be interesting for kids, I will reflect about the problem, because I don't use sounds for engines ; too much sound can be boring. I must add that in 1994 I saw a device made by Busch, Germany, which provided sound effects for a station on digital basis. And it's easy for anybody to add sound effects, triggered by a switch, without beeing a tech geek, since the recording/playing devices used in talking greeting cards are available for a few bucks ...
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 20, 2019 18:51:40 GMT -5
Meanwhile I succeeded in programming the couple Atmel-Emic to say 4 sentences : one for departure of the Budd's, one when the railcars enter the station to stop, one when the railcars will run through a station without stopping, and one when a freight train enters the "main" tracks, coming from outside of the station or a "dead end" track ( "don't cross the tracks" ) ; the only problem with the railcars is that, when approaching the station, the speech is triggered simultaneously with deceleration, i.e. 44 cm before the first turnout, and about 75 cm before the front of the railcars reaches the beginning of the plattform. With a rate (programmable for the Emic ! ) of 350 words/minute (a higher rate produces a "Donald-Duck" effect), the whole sentence is about 3.5 s seconds in length, i.e. the railcar would run about 30 to 35 cm at full (scale) speed, but in fact there is the deceleration, even if there is no stop (the train runs slower accross the station) ; so there is a "satisfying" margin betwen the end of the sentence and the moment when the Budd's reach the critical, "dangerous" zone : on a model railroad the "real" distances cannot be followed. So I will modify my electronics with a relay which selects one among two speakers, according to the station concerned by the message, and simultaneously there will be a selection between two male voices (4 available) ; I am waiting for the new PCB's (one for the processor and one for the relay), the source programme has been already modified. I need an ATMega8515 or ATMega162 (in DIP40 package) because 34 I/O lines are used now ; but the most important device on the PCB will be the DIP switch to cut the voice off, when no kids (or geeks ? ) are present to see the layout ... Such voice announcements can be tiring for the modeller himself !
|
|
|
Post by markm on Jan 21, 2019 10:59:43 GMT -5
So when will we get to hear it?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 21, 2019 12:39:04 GMT -5
Here is a recording of the 4 available announcements (from my test assembly) ; it is a pure "synthetic" speech and was made with an old magnetic sensor (as used in the past for phone amplifiers) fixed to the speaker, then the overall volume was increased with Audacity.
|
|
|
Post by markm on Jan 24, 2019 11:08:07 GMT -5
It's unfortunate that you can't slow the messages down by about 10%, but you've already explained that and you really do have the messages cut to least number of words. Could you perhaps slow down the trains just a bit?
Mark
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 24, 2019 13:02:45 GMT -5
|
|
|
Post by markm on Jan 24, 2019 13:25:49 GMT -5
Sounds better. Listening again, you could also consider dropping "approaching" from message 3 and "freight" from message 4. Any possibility to get the announcement in the baritone range?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 24, 2019 13:36:17 GMT -5
Mark, do You mean "lower pitched" ? This is a "lower" male voice, there is a command on Emic2 to lower the pitch (one semitone ?), I applied it 3 times ; I think that "even more low" would sound too "robotic". I added a countdown at the end, by a female, "air hostess" like, voice : it will be used in "automatic mode" (but without train annoucements). I think these effects are first of all intersting for kids (or geeks ... ). They are a good demonstration of what can be done with electronics, today !
|
|
|
Post by markm on Jan 24, 2019 18:56:10 GMT -5
That sounds good.
|
|