Post by thread:Model RR and DCC

I am confining all my replies into one long post to reduce irritation to non-RR types on the PIC list... When I wanted to get back into microcontrollers after using them 8 years ago, I finally narrowed it down to the PIC 16C84 largely because I thought it would be an IDEAL chip for DCC. Building a DCC decoder has been my ambition for a while. The PIC's great for this application as 1) It is available in an 18 pin SOIC. Space is at a premium and it is a lot smaller than the 0.4"x0.4"x0.8" suggested by John Payson. I would say it is about 1" x 1" x 0.2". Remember the loco is already full of other stuff like the chassis, motor, etc and you are left with whatever space that is available. And John, as for your suggestion to switch to O guage, I really envy the guy using the PIC to open and close train doors! Bet he doesn't have any space problems with the 1:1 scale he is operating in :-) The 18 pin DIP is also great for breadboarding as some of the other chips come in a PLCC which is a pain. 2) It's got EEPROM. This is essential for storing the last settings without which no decoder is useful. Having to add an external one makes things a lot more difficult. 3) The ISP feature which is a boon for any circuit using SMT. There is no other way to upgrade the software short of a messy desoldering job which can be done only so many times before trashing the PCB. 4) It is fast and requires hardly any external components. To answer Roger Books question, the speed is necessary because the info is digital packets with an approx bit time of 100us. In fact, I would use a 10MHz part if I could find a single quantity source. And here are some replies to assorted points raised. There has been some discussion on protocols and a valid point about using your own instead of slavishly following standards and resulting in code bloat. I feel standards have their own place as a guide. They are usually quite thorough and handle situations you may not anticipate. They are also a good source of ideas. For example, I haven't seen a good multi-master protocol yet. The CAN standard, though not practical for a simple PIC implementation, does have a cool arbitration protocol which may be ideal for multiple PICs. Some day I shall look at implementing that... The DCC standard is however different. You have to be largely compatible with the standard. I do find the decoder a little daunting to build and may buy it off the shelf, for example. You do have the choice of not implmenting some part of the standards if you don't think you'll use it. Mike Keitz suggest a smart lamp post. This is being done to some extent. Once you have a single DCC signal, it makes sense to use it to drive every other accessory on the layout - switches, lamps, sound decoders... So a smart lamp post does exists. There is, however, room for a smart dog :-) I don't see myself using any other approach as DCC seems to be fairly comprehensive. The only problem with it is lack of feedback. The standard, as it stands now, just supports a single pulse response from the decoder though it does have scope for a more complex response. If you want to see the standards you will find them at http://www.tttrains.com/nmradcc/s91.html and http://www.tttrains.com/nmradcc/s92.html. If you are doing something in this area and would like to share experiences/circuits/software get in touch. May be we can have our own mailing list or web site and stop bothering the PIC list :-) Shel Michaels mentions writing some DCC software for the 16C74. Can it be used on the 16C84 or is the capture/compare essential? In any case would you like to share the code or perhaps plant it in your garden :-) Although there are a few manufacturers' DCC pages I have yet to find a 'roll-your-own' site. Has anybody seen anything on the WWW on this? John Payson wrote: > Normally my inclination for controlling a model railroad would be to use a > single power-switching transistor and a direction-switching relay. When > space is not a problem, this will save components and may save cost (esp. if > you'd want to opto-isolate things). In your case, however, there isn't really > room for a relay. :-( I did use that approach while building a block control circuit. I soon switched to an L293. The cost is roughly the same and the size is a lot smaller. You can also sense the current for occupancy detection or short circuit. One chip gives you 4 block which is ideal for 1 PIC to drive. Regards Prashant +----------------+ ------------------------------------------------- | | Prashant Bhandary | +---+ | Spatial Information Systems Section | | | | Roads and Traffic Authority | | | | Rosebery NSW 2018, AUSTRALIA | | | | Tel: +61-2-662 5299 | | +----+ | Fax: +61-2-662 5348 | | | | Email: spam_OUTprashbTakeThisOuTrta.oz.au | +--------+ | | Still a newbie | "2b|!2b" - William Shakespeare +----------------+ -------------------------------------------------

Digi-Key sells 10 mhz 16C84's in single quantity (from 12.10 to 13.30 US Dlls). phone 1-800-digi-key fax 218-681-3380 http://www.digikey.com At 10:43 AM 29/04/96 +1000, you wrote: {Quote hidden}>I am confining all my replies into one long post to reduce irritation >to non-RR types on the PIC list... > >When I wanted to get back into microcontrollers after using them >8 years ago, I finally narrowed it down to the PIC 16C84 largely >because I thought it would be an IDEAL chip for DCC. Building >a DCC decoder has been my ambition for a while. > >The PIC's great for this application as >1) It is available in an 18 pin SOIC. Space is at a premium and > it is a lot smaller than the 0.4"x0.4"x0.8" suggested by > John Payson. I would say it is about 1" x 1" x 0.2". Remember > the loco is already full of other stuff like the chassis, motor, > etc and you are left with whatever space that is available. And > John, as for your suggestion to switch to O guage, I really envy > the guy using the PIC to open and close train doors! Bet he doesn't > have any space problems with the 1:1 scale he is operating in :-) > The 18 pin DIP is also great for breadboarding as some of the other > chips come in a PLCC which is a pain. > >2) It's got EEPROM. This is essential for storing the last settings > without which no decoder is useful. Having to add an external one > makes things a lot more difficult. > >3) The ISP feature which is a boon for any circuit using SMT. There > is no other way to upgrade the software short of a messy desoldering > job which can be done only so many times before trashing the PCB. > >4) It is fast and requires hardly any external components. To answer Roger > Books question, the speed is necessary because the info is digital > packets with an approx bit time of 100us. In fact, I would use a 10MHz > part if I could find a single quantity source. > >And here are some replies to assorted points raised. > >There has been some discussion on protocols and a valid point about using >your own instead of slavishly following standards and resulting in code >bloat. I feel standards have their own place as a guide. They are usually >quite thorough and handle situations you may not anticipate. They are >also a good source of ideas. For example, I haven't seen a good multi-master >protocol yet. The CAN standard, though not practical for a simple PIC >implementation, does have a cool arbitration protocol which may be ideal >for multiple PICs. Some day I shall look at implementing that... > >The DCC standard is however different. You have to be largely compatible >with the standard. I do find the decoder a little daunting to build and >may buy it off the shelf, for example. You do have the choice of not implmenting {Quote hidden}>some part of the standards if you don't think you'll use it. Mike Keitz >suggest a smart lamp post. This is being done to some extent. Once you have >a single DCC signal, it makes sense to use it to drive every other accessory >on the layout - switches, lamps, sound decoders... So a smart lamp post does >exists. There is, however, room for a smart dog :-) > >I don't see myself using any other approach as DCC seems to be fairly >comprehensive. The only problem with it is lack of feedback. The standard, >as it stands now, just supports a single pulse response from the decoder >though it does have scope for a more complex response. If you want to see >the standards you will find them at http://www.tttrains.com/nmradcc/s91.html >and http://www.tttrains.com/nmradcc/s92.html. If you are doing something in >this area and would like to share experiences/circuits/software get in touch. >May be we can have our own mailing list or web site and stop bothering the >PIC list :-) Shel Michaels mentions writing some DCC software for the 16C74. >Can it be used on the 16C84 or is the capture/compare essential? In any case >would you like to share the code or perhaps plant it in your garden :-) >Although there are a few manufacturers' DCC pages I have yet to find a >'roll-your-own' site. Has anybody seen anything on the WWW on this? > >John Payson wrote: >> Normally my inclination for controlling a model railroad would be to use a >> single power-switching transistor and a direction-switching relay. When >> space is not a problem, this will save components and may save cost (esp. if >> you'd want to opto-isolate things). In your case, however, there isn't really >> room for a relay. :-( >I did use that approach while building a block control circuit. I soon >switched to an L293. The cost is roughly the same and the size is a lot smaller. {Quote hidden}>You can also sense the current for occupancy detection or short circuit. One >chip gives you 4 block which is ideal for 1 PIC to drive. > >Regards > >Prashant >+----------------+ ------------------------------------------------- >| | Prashant Bhandary >| +---+ | Spatial Information Systems Section >| | | | Roads and Traffic Authority >| | | | Rosebery NSW 2018, AUSTRALIA >| | | | Tel: +61-2-662 5299 >| | +----+ | Fax: +61-2-662 5348 >| | | | Email: .....prashbKILLspam@spam@rta.oz.au >| +--------+ | >| Still a newbie | "2b|!2b" - William Shakespeare >+----------------+ ------------------------------------------------- > >

I also carry 10 Mhz 16C84's for what I believe is the cheapest price available on the Net in this quantity: $8.50 USD for less then 10 $9.00 USD for 10 or more. See my WEB page for more details. Ben, Ben Wirz Check out My Home Page for Great Deals on Bulk Buy's Nitinol Wire, LMD 18200 H Bridge and 10 Mhz PIC 16C84 EraseMEblw2spam_OUTTakeThisOuTcec.wustl.edu http://cec.wustl.edu/~blw2/index.html On Mon, 29 Apr 1996, Rogerio Odriozola Belden wrote: > Digi-Key sells 10 mhz 16C84's in single quantity (from 12.10 to 13.30 US Dlls). {Quote hidden}> phone 1-800-digi-key > fax 218-681-3380 > http://www.digikey.com > > > > At 10:43 AM 29/04/96 +1000, you wrote: > >I am confining all my replies into one long post to reduce irritation > >to non-RR types on the PIC list... > > > >When I wanted to get back into microcontrollers after using them > >8 years ago, I finally narrowed it down to the PIC 16C84 largely > >because I thought it would be an IDEAL chip for DCC. Building > >a DCC decoder has been my ambition for a while. > > > >The PIC's great for this application as > >1) It is available in an 18 pin SOIC. Space is at a premium and > > it is a lot smaller than the 0.4"x0.4"x0.8" suggested by > > John Payson. I would say it is about 1" x 1" x 0.2". Remember > > the loco is already full of other stuff like the chassis, motor, > > etc and you are left with whatever space that is available. And > > John, as for your suggestion to switch to O guage, I really envy > > the guy using the PIC to open and close train doors! Bet he doesn't > > have any space problems with the 1:1 scale he is operating in :-) > > The 18 pin DIP is also great for breadboarding as some of the other > > chips come in a PLCC which is a pain. > > > >2) It's got EEPROM. This is essential for storing the last settings > > without which no decoder is useful. Having to add an external one > > makes things a lot more difficult. > > > >3) The ISP feature which is a boon for any circuit using SMT. There > > is no other way to upgrade the software short of a messy desoldering > > job which can be done only so many times before trashing the PCB. > > > >4) It is fast and requires hardly any external components. To answer Roger > > Books question, the speed is necessary because the info is digital > > packets with an approx bit time of 100us. In fact, I would use a 10MHz > > part if I could find a single quantity source. > > > >And here are some replies to assorted points raised. > > > >There has been some discussion on protocols and a valid point about using > >your own instead of slavishly following standards and resulting in code > >bloat. I feel standards have their own place as a guide. They are usually > >quite thorough and handle situations you may not anticipate. They are > >also a good source of ideas. For example, I haven't seen a good multi-master > >protocol yet. The CAN standard, though not practical for a simple PIC > >implementation, does have a cool arbitration protocol which may be ideal > >for multiple PICs. Some day I shall lo

Ditto from the factory. We try to do as much research into applications as we can to better understand how to configure our products. These application discussions are a wealth of info for us. We are doing visits to companies doing industrial machine/process control now. Rgds, Brian. ______________________________ Forward Header __________________________________ Subject: Re: Model RR and DCC Author: "Mark A. Corio" <Mcoriospam_OUTAOL.COM> at Internet_Exchange Date: 4/29/96 1:06 PM Hi all, Prashant suggested moving to a new list to keep from boring the PIC people. I for one, hope you do not. A list that discussed only the technical nature of the PIC parts, without discussions of the applications in which they are used, would be boring. Some may disagree with me but I like the discussions that go into the application areas. This thread is a good example. I could never have predicted some of the application areas that the PIC has gone. Keep the application discussions going.....my 2 cents. Mark A. Corio Rochester MicroSystems, Inc. 200 Buell Road, Suite 9 Rochester, NY 14624 Tel: (716) 328-5850 Fax: (716) 328-1144 e-mail: @spam@McorioKILLspamaol.com ***** Designing Electronics For Research & Industry *****