Date: Wed, 19 Oct 94 04:30:03 PDT From: Advanced Amateur Radio Networking Group Errors-To: TCP-Group-Errors@UCSD.Edu Reply-To: TCP-Group@UCSD.Edu Precedence: List Subject: TCP-Group Digest V94 #233 To: tcp-group-digest TCP-Group Digest Wed, 19 Oct 94 Volume 94 : Issue 233 Today's Topics: KPC-9612 at 19.2Kb (3 msgs) Tekk Mini/Micro TEKK Prices (7 msgs) Send Replies or notes for publication to: . Subscription requests to . Problems you can't solve otherwise to brian@ucsd.edu. Archives of past issues of the TCP-Group Digest are available (by FTP only) from UCSD.Edu in directory "mailarchives". We trust that readers are intelligent enough to realize that all text herein consists of personal comments and does not represent the official policies or positions of any party. Your mileage may vary. So there. ---------------------------------------------------------------------- Date: Tue, 18 Oct 94 13:19:17 -0500 From: k5yfw@sacdm10.kelly.af.mil (WALT DUBOSE - K5YFW) Subject: KPC-9612 at 19.2Kb In Bruce Perens' message of 18 Oct 1994 at 0843 PDT, he writes: [Much Deleted] > Radio Shack $16.95 UHF television corner reflector antenna. I haven't > yet bothered to build a coaxial balun for the antenna - right now I'm > just feeding it directly with the TEKK radio. This is most interesting...I've always wondered about the antenna. A corner reflector was the antenna of choice for UHF links back in the 60's in the commercial world. Does anyone have more information on the antenna? > The only gotcha is that > the serial port baud rate of the KPC-9612 only goes up to 19,200 Baud. > That's bad for KISS operation, as if you are doing ping-pong > packet-then-ack operation you have to subtract the serial transfer > delay from the channel rate. An Ontario PI II card and TAPR modem > would be a better choice than the KPC-9612 for TCP/IP. Karl Metcalf runs two laptops, two KPC-9612s at 19.2Kb and two D4-10s in his packet demonstration and they seem to work well...across the room. He maintains that he has used them running NOS and they work well. Most individuals are aware of the problems associated with high speed serial port operation and for that reason most have gone to the Ottawa PI/PI2 card. My home AmprNet station is a 386 with PI2 card feeding TTL directly into a D4-10 at 19.2Kb. This configuration runs very well...thruput is exceptional in out half-duplex operation. I now have a ARES/Civil Defense/MARS requirement to run TCP/IP at 19.2Kb on our local AmprNet/MARS LAN and using a laptop computer. I have hopes of running a KPC-9612s into D4-10 radios to accomplish this task. Even with high-speed chips in laptop's the serial port, will I run into the a serial problem? Walt/K5YFW ------------------------------ Date: Tue, 18 Oct 94 16:41 PDT From: bruce@pixar.com (Bruce Perens) Subject: KPC-9612 at 19.2Kb > This is most interesting...I've always wondered about the antenna. A > corner reflector was the antenna of choice for UHF links back in the > 60's in the commercial world. > Does anyone have more information on the antenna? It's a Radio Shack U-75 (I don't have the longer part number, but it's the UHF-TV-only "75 mile" antenna in their catalog). It is a 300 ohm horizontal polarized corner reflector with directors. It doesn't come with the radio or anything, I just bought it in the Radio Shack and tried it, and it worked out of the box without matching about as well as a small 440 colinear array that I have. I would suggest that someone with the tools and more experience than I at hacking antennas do some real measurement on the antenna, design a coaxial balun or balanced feed, and figure out if there's a need to trim the driven element or add length to the directors. A proven design for 440 based on an assembled $16.95 TV antenna would be a nice thing to have. Thanks Bruce Perens ------------------------------ Date: Tue, 18 Oct 94 17:03 PDT From: bruce@pixar.com (Bruce Perens) Subject: KPC-9612 at 19.2Kb > Even with high-speed chips in laptop's the serial port, will I run > into a serial problem? The serial overrun problem that is handled by the high-speed chips is another issue. The problem that I was referring to is that the whole packet has to be received by the TNC before the TNC _starts_ to send the packet over the serial line. The TNC won't send the computer a packet that's incomplete or has framing errors, so it has to wait for the whole packet to come in first so that it can make sure there are no problems with the packet. Thus, if you are using a 19,200 radio link and a 19,200 serial line, you have to wait for the packet to go through _two_ sequential links at 19,200 - your radio and your serial line. The effective throughput for ping-pong packet-then-ack operation will be only 9600. If you have the same problem on the other side, you now get 7200 Baud throughput. If you are always running stream-oriented TCP/IP connections, you can hide this overhead by using a window at least twice the mss. Then you won't wait for an ack before you send the next packet, and the acks and packets will be going in different directions through the serial line at the same time. I have no idea if this really works in practice. Bruce ------------------------------ Date: Tue, 18 Oct 1994 11:32:24 -0800 (PDT) From: Glenn Elmore Subject: Tekk Mini/Micro Paulus writes: > What are the differences between T-Net Micro (KS-900) and > T-Net Mini (KS-960) ? Maybe someone with more information will post but after looking at and measuring the Mini and comparing with the schematic for the Micro the major difference appears to be in the use of the MC3371 in the Mini and the MC3359 in the Micro. I have a parts list only for the Micro but it *appears* that they both use the same 15 KHz wide 21.4 first IF filter and that this filter dominates the second IF filter which appears to be a Murata 455C or 455D filter, either 20 or 25 KHz wide. I know that there is a current rumor that the Micro is "better" for 9600 than the Mini but I don't see anything yet which would support that. If the IF first filter dominates, I wouldn't expect the two to be very different at 9600 bps. The MC3359 doesn't have the RSSI output and the micro doesn't seem to offer the carrier detect line. From my measurements, this is no great loss since the response is heavily filtered and too slow anyway. When used with TAPR 9600 modems with data derived DCD, this function isn't necessary. Hopefully, the slow DCD of the Mini doesn't glitch the data on weak or noisey signals. If it does this could be a difference between the two but the slow DCD circuit can easily be disabled. I can believe that tuning, netting and modulation setting of individual radios could make one look considerably better/worse than the other but I don't yet see anything to support a fundamental difference between models. Can anyone who's carefully compared BER of the two radios under operating conditions comment? Glenn Elmore n6gn amateur IP: glenn@SantaRosa.ampr.org Internet: glenne@sr.hp.com ------------------------------ Date: Tue, 18 Oct 94 08:43 PDT From: bruce@pixar.com (Bruce Perens) Subject: TEKK prices The KS-960 is somewhat larger, as its cabinet has space for TEKK's 4800 Baud modem (which we don't want). The KS-960 also has a received signal strength indicator output, which can drive an S-meter. I've also heard that there is some difference in the crystal filters used on the two models, though I can't confirm that by looking at my KS-960 and KS-900. The KS-900 seems to work as well as the KS-960. We don't need the S-meter. My KS-900 runs cool, while the KS-960 is always a bit warm. I don't really understand why they bother having two models that are so similar. Both radios seem to work fine with the Kantronics KPC-9612 and with the Radio Shack $16.95 UHF television corner reflector antenna. I haven't yet bothered to build a coaxial balun for the antenna - right now I'm just feeding it directly with the TEKK radio. The only gotcha is that the serial port baud rate of the KPC-9612 only goes up to 19,200 Baud. That's bad for KISS operation, as if you are doing ping-pong packet-then-ack operation you have to subtract the serial transfer delay from the channel rate. An Ontario PI II card and TAPR modem would be a better choice than the KPC-9612 for TCP/IP. Bruce Perens ------------------------------ Date: Tue, 18 Oct 94 12:26:30 -0500 From: David Rush Subject: TEKK Prices >What are the differences between T-Net Micro (KS-900) and >T-Net Mini (KS-960) ? The Mini is tiny, the Micro is tinyer. The only differences that I know of, other than size, are that the Mini provides a few bells and whistles, such as a pin on its DB9 where you can attach a signal strength meter. I think the Mini also has power, xmit lites, too. The Micro is a black box with a BNC and a DB9 - that's all. The Mini has some chrome. David, n0oxh, rush@erg.sri.com ------------------------------ Date: Tue, 18 Oct 1994 13:39:36 -0500 (CDT) From: Gerald J Creager Subject: TEKK Prices The Mini is at least nominally servicable, too. The Micro is too small for me to deal with, even with a magnifier and usually adequate tools. 73, gerry N5JXS ------------------------------ Date: Tue, 18 Oct 94 16:42 PDT From: bruce@pixar.com (Bruce Perens) Subject: TEKK Prices > I think the Mini also has power, xmit lites, too. Nope, only holes for the lights on the TEKK 4800 Baud modem to show through. Bruce ------------------------------ Date: Tue, 18 Oct 94 16:51 PDT From: bruce@pixar.com (Bruce Perens) Subject: TEKK Prices > The Mini is at least nominally servicable, too. The mini has a lot of surface-mount components on both sides of the PC. The micro gets nearly everything on one side. I thought there were fewer surface-mount components in the micro, but I'd have to look again. It is tighter. The micro looks perfect for baloon work. It's so small and light. The PC can be removed from the cabinet in one piece and placed in a lighter enclosure. It would need testing through the radical environment that baloon hardware is subjected to. Bruce ------------------------------ Date: Tue, 18 Oct 1994 20:51:33 -0500 (CDT) From: Gerald J Creager Subject: TEKK Prices For servicability, I'd at least like to be able to get the tweezers onto the SMCs. That's the basis, from my view, of the comment... I agree that the Micro is good for balloon work, but I'd be inclined to leave it in the box it comes in. Actually, it's not too much mass, and should be pretty reasonable. Me? I'm using 25 of the Mini's for an APRS-like project with the University on -- cows -- with Motorola GPS engines. Biggest problem is the battery packs to run the Mini, the GPS, and a TNC for 5 days... 73, gerry ------------------------------ Date: Wed, 19 Oct 1994 14:14:06 +1000 (EST) From: Dave Horsfall Subject: TEKK Prices On Tue, 18 Oct 1994, Gerald J Creager wrote: > Me? I'm using 25 of the Mini's for an APRS-like project with the University > on -- cows -- with Motorola GPS engines. Biggest problem is the battery packs > to run the Mini, the GPS, and a TNC for 5 days... Ummm, make them methane-powered? -- Dave Horsfall (VK2KFU) | dave@esi.com.au | VK2KFU @ VK2AAB.NSW.AUS.OC | PGP 2.6 Opinions expressed are mine. | E7 FE 97 88 E5 02 3C AE 9C 8C 54 5B 9A D4 A0 CD ------------------------------ End of TCP-Group Digest V94 #233 ******************************