Post by thread:[PIC]:[EE]: A KISS TNC based on a pic

See comments below <> ----- Original Message ----- From: "John C. Frenzel" <spam_OUTjohnfrenzelTakeThisOuTHOTPOP.COM> To: <.....PICLISTKILLspam@spam@MITVMA.MIT.EDU> Sent: Thursday, July 06, 2000 6:44 PM Subject: Re: [PIC]:[EE]: A KISS TNC based on a pic {Quote hidden}> I would like to know if it would be easy to design a KISS TNC(similar to = > serial modem for a PC but operates at 1200 bd half duplex) with a PIC =nd > little to no other circuitry to keep the cost and complexity down. > > TNX + 73--- > Andy N1YEW > > Andy, > Take a look on the tapr website http://www.tapr.org they have one already built > John W2SP > <>Well I couldn't find it. I still wish to build my own *unless* theirs is all set(complete in that i dont have to modify anything (plug+play :-) TNX Andy N1YEW -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

To my knowledge, TAPR does not have a KISS based PIC TNC out at this point. I've recently completed the design of one, however, and it is tentatively scheduled to appear as a construction article in November's QST (give or take publication schedule slippage). It is based on the 16F877 and uses a 32K static ram for storage. It has two modes, one is simply a straight KISS TNC (though multidrop KISS is not currently supported) that has a 28K transmit buffer and a 4K receive buffer. The other is an APRS Tracker mode that takes a GPS NMEA input and transmits a position report at regular intervals using the compressed MIC-E format. In this mode it will also display received APRS UI frames to a simple terminal. Complete kits of parts will be available in the early fall. John, W2FS At 08:40 PM 7/6/00 -0400, you wrote: >See comments below <> > >{Original Message removed}

> To my knowledge, TAPR does not have a KISS based PIC TNC out at this > point. I've recently completed the design of one, however, and it is > tentatively scheduled to appear as a construction article in November's QST > (give or take publication schedule slippage). It is based on the 16F877 > and uses a 32K static ram for storage. It has two modes, one is simply a > straight KISS TNC (though multidrop KISS is not currently supported) that > has a 28K transmit buffer and a 4K receive buffer. The other is an APRS > Tracker mode that takes a GPS NMEA input and transmits a position report at > regular intervals using the compressed MIC-E format. In this mode it will > also display received APRS UI frames to a simple terminal. Complete kits > of parts will be available in the early fall. > > John, W2FS That is so cool! Congrats on the QST publication. I was thinking of the PIC encoder. I am looking forward to seeing your work. John -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Well I sort of half designed one, It's based on an AVR part however. You can take a look at the project on my AVR page, you might find some aspects of the design useful. http://www.users.interport.net/~carlott/projects.html At 06:38 PM 7/6/2000, you wrote: {Quote hidden}>I would like to know if it would be easy to design a KISS TNC(similar to a >serial modem for a PC but operates at 1200 bd half duplex) with a PIC and >little to no other circuitry to keep the cost and complexity down. > >TNX + 73--- >Andy N1YEW > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

At 11:20 AM 7/7/00 -0400, you wrote: >You shouldn't need a recieve buffer using KISS protocol... you might need a >transmit buffer though. unless you are communicating to the PC at the same >rate you are transmitting(1200 bd) The reason that you have to have a receive buffer is that you have to hold an entire frame in memory before you can do the CRC check to determine whether it should be passed out the serial port or not. If it fails the CRC check, you must discard it. The KISS protocol is responsible for bit-stuffing and the CRC check. Secondly, if you are planning to receive continuously you have to be accumulating data for the second frame while the first frame is being sent out the serial port. Thus, for continuous reception, you need storage for at least two receive frames. It is my understanding that those who are using KISS TNC's for IP work routinely use packet lengths of at least 1K. As a result you need at least 2K worth of storage in order to ensure continuous reception. I chose 4K (in a ring buffer configuration) simply because it was easier to code than 2K and I wasn't sure that the IP guys were limiting themselves to packet lengths of only 1K. You need a bigger transmit buffer because the unit should hold off transmitting until the channel is free. You have to check to ensure that the frequency is free before transmitting or you might step on someone. I've done this by using a software DCD, not by simply detecting activity of any kind on the frequency. Thus, you can run this unit with the squelch open. During the period when you are holding off transmitting, however, the TNC needs to continue to accumulate data over the serial line, since the KISS protocol has no flow control at all. I started this design by using two 2K FRAM chips for storage, one for transmit and one for receive. This involved a lot fewer I/O lines, but I was concerned that 2K (particularly with a potential packet length of 1K) would not be adequate. It turns out that static ram chips are incredibly cheap and switching to a 32K SRAM actually lowered the cost of project, though it does use quite a few I/O lines. >I was going to make a KISS TNC minus a buffer for simplicity and communicate >w/ the pc at the same rate that we are transmitting...although I hadn't >thought about trying to transmit AND recive data at the SAME time...(I think >thats called FULL DUPLEX :) The TNC that I have done does not do full duplex either. This is because the modem chip I'm using (the MX-614) doesn't do full duplex. I could have used two of these chips to do full duplex, but it seemed to me that most users wouldn't be operating full duplex anyway. Really the hardest part of this project is simulating a kind of "multitasking." While the TNC is receiving data over the radio link it must also be receiving data over the serial port and routing received data out the serial port. Interleaving these processes is an interesting challenge, but do-able. It would be a bit more difficult to add transmitting data over the radio link as well, since the radio link is bit-banged, but the serial link to the computer can use the PIC's USART. John Hansen W2FS -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

----- Original Message ----- From: "John Hansen" <johnKILLspamHANSEN.NET> To: <.....PICLISTKILLspam.....MITVMA.MIT.EDU> Sent: Friday, July 07, 2000 7:55 PM Subject: Re: [PIC]:[EE]: A KISS TNC based on a pic > Really the hardest part of this project is simulating a kind of > "multitasking." While the TNC is receiving data over the radio link it > must also be receiving data over the serial port and routing received data > out the serial port. Interleaving these processes is an interesting > challenge, but do-able. It would be a bit more difficult to add > transmitting data over the radio link as well, since the radio link is > bit-banged, but the serial link to the computer can use the PIC's USART. Right.... thats what i meant. i was going to bitbang both the input and output for the tnc. > > John Hansen W2FS Perhaps my question was supposed to be 'can i make a total tnc with pics only. (tone decode and generation, etc (but no CRC) 73- Andy N1YEW -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

part 0 44 bytes his is a multi-part message in MIME format. part 1 1109 bytes content-type:text/plain; charset=koi8-r (decoded 7bit) It's OK, but take a look this: Abt 60-80% hams cant find those FX,MX,CMX IC's.. How to avoid them using PIC's??? Any ideas?? See ya... Henry Carl Ott wrote: {Quote hidden}> Well I sort of half designed one, It's based on an AVR part however. You > can take a look at the project on my AVR page, you might find some aspects > of the design useful. > > http://www.users.interport.net/~carlott/projects.html > > At 06:38 PM 7/6/2000, you wrote: > >I would like to know if it would be easy to design a KISS TNC(similar to a > >serial modem for a PC but operates at 1200 bd half duplex) with a PIC and > >little to no other circuitry to keep the cost and complexity down. > > > >TNX + 73--- > >Andy N1YEW > > > >-- > >http://www.piclist.com hint: The list server can filter out subtopics > >(like ads or off topics) for you. See http://www.piclist.com/#topics > > -- > http://www.piclist.com hint: The PICList is archived three different > ways. See http://www.piclist.com/#archives for details. part 2 201 bytes content-type:text/x-vcard; charset=koi8-r; (decoded 7bit) begin:vcard n:Pavlica;Milan x-mozilla-html:FALSE org:SuperSonic Systems adr:;;;;;; version:2.1 email;internet:EraseMEmpavlicaspam_OUTTakeThisOuTptt.yu title:Chief fn:Milan Pavlica end:vcard part 3 103 bytes -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestspam_OUTmitvma.mit.edu>

At 11:54 AM 7/8/00 +0200, you wrote: >It's OK, but take a look this: >Abt 60-80% hams cant find those FX,MX,CMX IC's.. >How to avoid them using PIC's??? >Any ideas?? >See ya... Why do you say that? The parts are readily available via mail order, just as PICs are. TAPR (http://www.tapr.org) sells the MX-614 in single quantities, MAX232's are available from a number of sources, including JDR, as are static ram chips. Probably 90 % of hams can't buy a PIC 16F877 locally either. For almost all of us these things have to be acquired mail order these days. John Hansen W2FS -- http://www.piclist.com hint: To leave the PICList @spam@piclist-unsubscribe-requestKILLspammitvma.mit.edu>

That was my intention for building a cheap readily availabe tnc. 73- Andy N1YEW ----- Original Message ----- From: "Milan Pavlica (YU7AEC)" <KILLspammpavlicaKILLspamPTT.YU> To: <RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU> Sent: Saturday, July 08, 2000 5:54 AM Subject: Re: [PIC]:[EE]: A KISS TNC based on a pic > It's OK, but take a look this: > Abt 60-80% hams cant find those FX,MX,CMX IC's.. > How to avoid them using PIC's??? > Any ideas?? > See ya... > > Henry Carl Ott wrote: > > > Well I sort of half designed one, It's based on an AVR part however. You > > can take a look at the project on my AVR page, you might find some aspects > > of the design useful. > > > > www.users.interport.net/~carlott/projects.html > > > > At 06:38 PM 7/6/2000, you wrote: > > >I would like to know if it would be easy to design a KISS TNC(similar to a > > >serial modem for a PC but operates at 1200 bd half duplex) with a PIC and {Quote hidden}> > >little to no other circuitry to keep the cost and complexity down. > > > > > >TNX + 73--- > > >Andy N1YEW > > > > > >-- > > >http://www.piclist.com hint: The list server can filter out subtopics > > >(like ads or off topics) for you. See http://www.piclist.com/#topics > > > > -- > > http://www.piclist.com hint: The PICList is archived three different > > ways. See http://www.piclist.com/#archives for details. > -- http://www.piclist.com hint: To leave the PICList spamBeGonepiclist-unsubscribe-requestspamBeGonemitvma.mit.edu>