|
|
| Re: Canarsie CBTC (96616) | |
 |
|
| Home > SubChat | |
[ Post a New Response | Return to the Index ]
|
Page 6 of 8 |
||
|
| (100111) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Fri Jun 17 19:56:18 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Fri Jun 17 13:55:50 2005. It means you didn't understand my question, and chose not to ask me about the source of your confusion.Jeff's wrong, by the way. If your techincal arrogance is indicative of your profession in general, then I can understand the necessity for double blind experiments. |
|
|
| (100130) | |
Re: Canarsie CBTC |
|
|
Posted by AlM on Fri Jun 17 20:45:38 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Fri Jun 17 17:57:55 2005. "But the blanket assertion that GPS cannot be used at to support CBTC is nonsense."Certainly asserting this without evidence is foolish. But do YOU have the slightest proff that GPS signals ARE detectable in the subway. I have trouble enough getting a GPS to work on the streets f Manhattan because of the external interference from other signals. |
|
|
| (100187) | |
Re: Canarsie CBTC |
|
|
Posted by tracksionmotor on Sat Jun 18 00:07:29 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Fri Jun 17 17:57:55 2005. Geosynchronus Position Satellite requires a 'view' to at least sixsatellites. It will work quite nicely...at the earths surface. It will work with CBTC but the 'capture time' is excessive for what a short range rail system NYCTA uses. GPS is good for the long haul mainline trakkers because they don't have short signal zoning. CBTC requires instant comm at every comm zone...the system recognises trainset has entered what was formerly a 'signal rail zone' with the trainsets ID. System cannot wait for trainset to obtain GPS coordinates...just see how long your GPS hand-held takes to obtain accurate coordinates. Trainset cannot obtain GPS coordinates underground. CI peter |
|
|
(Sponsored) |
iPhone 6 (4.7 Inch) Premium PU Leather Wallet Case - Red w/ Floral Interior - by Notch-It |
|
| (100189) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 00:14:24 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Fri Jun 17 19:52:12 2005. "CBTC does not handle switching."You missed my point, which was that GPS accuracy would be insufficient to keep adjacent trains as separate objects - it's a little like watching two airplanes merge on a radar scope (leaving aside the transponders for a moment). GPS can supply position information (and if you have updated position information, then you can also calculate velocity). Your points about track assignments are correct; GPS could reduce the investment you'd need in wayside equipment. "How much of your own money will you put up, for the test?" A test like this would be well worth an investment from the US Dept. of Transportation. By definition that would involve my tax dollars (and yours). If the experiment fails we still will have learned a lot (actually we learn more from failures than we do from successes). Why not take some chances? Are you that risk averse? The Mercury and Gemini programs took a lot of risks that, in the retrospectoscope, look insane today. But look at the dividends they paid. |
|
|
| (100191) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 00:15:19 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 00:14:24 2005. I mentioned Mercury and Gemini because they led to Apollo and because YOU played a role in all of that. |
|
|
| (100192) | |
Re: Canarsie CBTC |
|
|
Posted by tracksionmotor on Sat Jun 18 00:16:38 2005, in response to Re: Canarsie CBTC, posted by Jeff H. on Fri Jun 17 16:45:42 2005. Jeff: Please don't beat the dead horse...the teeth may be good but the horse 'remains' won't pull the wagon. CI peter |
|
|
| (100193) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 00:17:17 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Fri Jun 17 19:56:18 2005. I'm not arrogant about this - just a lot more open minded than you are. Jeff is even more closed minded than you.There was a time when you were not so arrogant and closed minded - if you exhibited 40 years ago the attitudes you have today, the Apollo program would never have come to fruition. |
|
|
| (100194) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 00:18:40 2005, in response to Re: Canarsie CBTC, posted by tracksionmotor on Sat Jun 18 00:07:29 2005. You point out some limitations that should be examined. of course, GPS won't work in the tunnel. |
|
|
| (100202) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 00:27:57 2005, in response to Re: Canarsie CBTC, posted by Jeff H. on Fri Jun 17 16:45:42 2005. "I went through the trouble of explaining why "eliminating relays"could not be legitimately argued to be an exclusive benefit of CBTC," It's not an exclusive benefit of CBTC; it's that CBTC is the technology currently most capable of improving on it. I never said that CBTC has an exclusive lock on being the only technology which can replace relays. as usual, you ignore what you don't want to hear. I put to you a simple challenge: Tell us about Siemens' scheme specifically and then criticize it to your heart's content. Need to find out more about it - you have the Internet, a telephone and email. "And, BTW, if you would like to tell us all what Siemens is doing, please go ahead and do so." You're the one who had isssues with it and insisted on posting authoritatively about CBTC without knowing much about it. But since you don't want to learn anything, I just might do that, even though it means spoonfeeding you (and you have a background in EE; why should I be spoonfeeding you? OK. What the hell... If you want to be taken seriously about this stuff, you have to post like an adult. You don't want to do that, then I won't treat you like one. That's fine. |
|
|
| (100203) | |
Re: Canarsie CBTC |
|
|
Posted by tracksionmotor on Sat Jun 18 00:28:45 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 00:14:24 2005. I know ATS/ATO/CBTC and everything else. I worked LORAN. GPS aquisition is not fast enough for NYCTA CBTC even if the trainsetis above ground and can be located. What you need to do is go down into the pit and take apart some New Tech speed sensors. Obviously, you have not been reading my past posts...'seven braked wheels.' Doumm...doumm...doumm...doumm... CHU Canada Eastern Daylight Savings Time...Zero Hours, Twenty Eight Minutes. Yee-up...worked that too. CI peter |
|
|
| (100204) | |
Re: Canarsie CBTC |
|
|
Posted by tracksionmotor on Sat Jun 18 00:32:16 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 00:18:40 2005. GPS aquistion time is not compatible with CBTC. Works OK with long haul mainline block signalling. |
|
|
| (100229) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 01:29:09 2005, in response to Re: Canarsie CBTC, posted by tracksionmotor on Sat Jun 18 00:32:16 2005. OK. I'll take your word for that. Should we expect acquisition time to be the same as it is now, say, in five years? |
|
|
| (100233) | |
Re: Canarsie CBTC |
|
|
Posted by Jeff H. on Sat Jun 18 01:34:42 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 00:27:57 2005. It's not an exclusive benefit of CBTC; it's that CBTC is the technology currently most capable ofimproving on it. I never said that CBTC has an exclusive lock on being the only technology which can replace relays. as usual, you ignore what you don't want to hear. We were talking about (or at least I think we were talking about) CBTC vs other signal technologies, including wayside block and ASC. Now, will you agree that in that context, CBTC does not have any special advantage in terms of reducing relays? If not, please elborate. If you want to be taken seriously about this stuff, you have to post like an adult. If you like, we can have a subchat poll on whose posts are more childish. You claim to have some inside knowledge of what the vendor is doing and presumably that would shed some light on this debate. Show your cards, or just forget about it. Doesn't matter to me. This isn't a contest and I'm not going to win any prizes. |
|
|
| (100238) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 01:46:07 2005, in response to Re: Canarsie CBTC, posted by Jeff H. on Sat Jun 18 01:34:42 2005. "You claim to have some inside knowledge of what the vendor is doing"No, I don't, and none of my posts even hint at that. I do know (and you do too) that Siemens' tehnology is being adopted by a number of transit systems around the world; they obviously believe that it does work. CBTC itself is already deployed and working in some places. You claim to understand the limitations of CBTC, regardless of vendor and method, without knowing enough about CBTC to have any kind of intelligent discussion. I asked you to at least research what you were criticizing. Think Siemens is screwing up? Fine, tell us what Siemens is doing and then blast away. But you can't, because you prefer being a child. I like dealing with adults and with children whose chronological ages are appropriate for their level of function. Yours isn't, and that's too bad, because now I have to treat you like a kid because you act like one. We all lose because of your attitude. That's a shame. |
|
|
| (100240) | |
Re: Canarsie CBTC |
|
|
Posted by Jeff H. on Sat Jun 18 01:50:37 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 01:46:07 2005. That's nonsense Ron. You've told us about the reams of materialwhich the vendor has gleefully supplied to you. What do they say? |
|
|
| (100266) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sat Jun 18 06:53:43 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 00:14:24 2005. GPS can supply position information (and if you have updated position information, then you can also calculate velocity).Just how do you propose to calculate velocity? |
|
|
| (100267) | |
Re: Canarsie CBTC |
|
|
Posted by AlM on Sat Jun 18 07:02:57 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 00:14:24 2005. "GPS can supply position information (and if you have updated position information, then you can also calculate velocity). Your points about track assignments are correct; GPS could reduce the investment you'd need in wayside equipment."Let's forget about the fact that satellite signals almost certainly aren't strong enough to be detected reliably underground. They're WAY weaker than cellphone signals, and have you every tried using a cellphone in the subway? There is some error in GPS measurements. I believe it is 20 meters, though it is unlikely that two consecutive measurements would be off in radically different ways. So if you take two position measurements a second apart, and they are off by even 5 meters, the resulting velocity estimate would have far too much inaccuracy to be useful. Only over longer term periods is GPS a useful measure of average (not instantaneous) velocity. "A test like this would be well worth an investment from the US Dept. of Transportation." How do you know it hasn't been done? It wouldn't be expensive to take a GPS receiver down into a subway tunnel and check if you can detect a sufficient signal. If the answer is not a chance, it might not have gotten much publicity. |
|
|
| (100268) | |
Re: Canarsie CBTC to Jeff |
|
|
Posted by Brother Pielet on Sat Jun 18 07:08:55 2005, in response to Re: Canarsie CBTC, posted by Jeff H. on Sat Jun 18 01:34:42 2005. Jeff its really a shame you have to continue explaining yourself to this moron. A posting to him is a dead frog in science class. I really dug into the relay posts and after playing with a GPS device the last 5 months myself I can see it will not be used for anything related to train signalling, safety and position. Shown bottom a typical handheld using Mapsource software shows the eastbound and westbound Kearny area at Hudson Yards from PA rails on three round trips. I'm not sure what type of GPS Terripan STation uses but I recall having a discussion over speed clockings with this as well being plus or minus 5 MPH. Had a friend in DOT try out a new radar gun to callibrate a speeodometer mod for me. Off over 7 MPH. Takes almost 40 seconds for the 17 satellites to even lock in and other GPS receivers don't fair much better today. To be 15 feet off and only work outdoors will help lost motorists and hikers. Not railroads.
|
|
|
| (100271) | |
Re: Canarsie CBTC |
|
|
Posted by H.S.Relay on Sat Jun 18 07:27:40 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Tue Jun 14 22:14:45 2005. I'd never given that question any thought because I've never believed that a"moving block" system provided any significant capacity increase over a properly designed "fixed block" system.Wow. I'm shaking my head . . . I completely agree with you. |
|
|
| (100275) | |
I'd like proff. |
|
|
Posted by H.S.Relay on Sat Jun 18 07:50:45 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Thu Jun 16 09:10:27 2005. "...because in the end, its still all about signaling and safely controlling train movements, which is a craft distinct from ATM machines, web browsers and (yes SB) industrial controls."False statement, and incredibly ignorant on your part. Pure bullshit with nothing -ZERO - to support it except your prejudice towards recent technologies. Okay, so you supplied ZERO to support YOUR argument. -do you see the irony? Lay it out, point-by-point. Prove to me that Signaling is NOT a distinct craft with a distinct design approach from ATM machines, web browsers and/or industrial controls. -obviously you will need working knowledge of the design philosophies of all four disciplines. Be objective and inclusive, and don't opinionate. Show your work, and cite sources. This is a thread about technology, keep you discourse in the technical realm. Go. Can you do it? |
|
|
| (100278) | |
Welcome to 2nd grade: No- YOU'RE a liar!!! |
|
|
Posted by H.S.Relay on Sat Jun 18 08:14:07 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Fri Jun 17 07:53:51 2005. CBTC is far cheaper to maintain than the type of equipment used now. It isn't just insulated joints; it's the relays that run to $60 million worth of equipment to control just 2,000 feet of route (Chambers St as example).The chief maintenance cost is labor, will there be fewer maintainers? The TA doesn't know how many man-hours will it take to replace a vandalized radio antennae, which must be lined up to within millimeters to operate properly. (In case you're wondering, that little tidbit came from a large German electrical concern) Chambers was an interlocking. No one disputes that CBTC will save some cash by eliminating block signals. Chambers St: 21 Interlocked signals, 6 automatics, 6 switches, 2 traffic levers. $60 million, you say. Third Avenue: 8 Interlocked Signals, 4 switches, 4 traffic levers Cost? I dunno. Both plants control about 2000 feet plus or minus. both are/will be "CBTC Ready," with CBTC already "plugged-and-played" at 3rd Ave. Can you tell the group what the price tag for 3rd Avenue interlocking was? [BTW, as mentioned before, its an all-relay interlocking.] |
|
|
| (100279) | |
_three_ manufacturers |
|
|
Posted by H.S.Relay on Sat Jun 18 08:17:28 2005, in response to Re: Canarsie CBTC, posted by Jeff H. on Fri Jun 17 16:45:42 2005. What's Alstom? Chopped Liver? And where's "Alstom R160A" to defend their namesake? |
|
|
| (100280) | |
Re: Canarsie CBTC |
|
|
Posted by H.S.Relay on Sat Jun 18 08:25:20 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 00:27:57 2005. If you want to be taken seriously about this stuff, you have to post like an adult. You don't want to do that, then I won't treat you like one. That's fine.You're joking, right? And speaking as someone FROM the inside, Jeff's CBTC synopsis a couple of posts back was a COMPLETEY ACCURATE description of Siemens' scheme. So, in the timeless words of Lt. Zachary Garber: "Shaddap, will-ye?" |
|
|
| (100281) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 08:27:55 2005, in response to Re: Canarsie CBTC, posted by Jeff H. on Sat Jun 18 01:50:37 2005. Go back and look at what I wrote. I didn't ask for material from Siemens, but I told you that manufacturers are likely to be very cooperative if you approach them with professional interest. I mentioned that the French were especially eager to cooperate.When I was developing the Ride and Seek subway game some years ago (before Bombardier had a website), I asked Bombardier for specific information about the R68 subway car. I didn't have a specific address, only "Paris-La Defense France" and I don't speak or write French. So I wrote to "Public Relations" at the general address (I didn't even have a street number). A week later I got a parcel in the mail with glossy brochures, technical specs on the R68 car (in English, thankfully), information about ANF Industrie (the carbody builder), diagrams (and not just the kind you get in brochures sometimes), you name it. and they sent me information packets in French on RATP and other transportation networks in France. And business cards for different managers in Canada and France. Any questions, they asked? Give us a call - we'll tell you anything you want to know. I think Siemens would do the same for you - no, even better, if you tell them your educational background. If nothing else they'll be showing off for you (try a little stroking, see what you get). Or you can skip this whole thing and prove that you have no respect for yourself, never mind anybody else on this board. |
|
|
| (100282) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 08:32:30 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Sat Jun 18 06:53:43 2005. What happens when you have a series of successive data points regarding position, each with a time-stamp?I'll give you a hint: It's not just GPS that can do that. My wife contributed software to something called "Track while scan" (which is actually a bit of a misnomer). |
|
|
| (100286) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 08:39:18 2005, in response to Re: Canarsie CBTC, posted by AlM on Sat Jun 18 07:02:57 2005. "Let's forget about the fact that satellite signals almost certainly aren't strong enough to be detected reliably underground. They're WAY weaker than cellphone signals, and have you every tried using a cellphone in the subway?"Let's not forget about it. If you had bothered to read through this thread, you would have seen my posts where I stipulated to the fact that GPS is not useful underground. We acknowledge that - perhaps technology in future will change that (???) But not today. "There is some error in GPS measurements. I believe it is 20 meters," Try 5 meters. The military is no longer demanding to degrade signals for civilian use. And we may see better wiithin 5 years. "though it is unlikely that two consecutive measurements would be off in radically different ways." Agreed - meaning we can quantitate the likely error and account for it. "So if you take two position measurements a second apart, and they are off by even 5 meters, the resulting velocity estimate would have far too much inaccuracy to be useful." Your caveat is important - but it would be useful to find out where the possibilities and boundaries really are. "It wouldn't be expensive to take a GPS receiver down into a subway tunnel and check if you can detect a sufficient signal. If the answer is not a chance, it might not have gotten much publicity." Refer to the top of this post. |
|
|
| (100287) | |
Re: Canarsie CBTC to Jeff |
|
|
Posted by RonInBayside on Sat Jun 18 08:45:54 2005, in response to Re: Canarsie CBTC to Jeff, posted by Brother Pielet on Sat Jun 18 07:08:55 2005. Well at least you actually ran an experiment and collected real data worthy of discussion (and even provided a map). That's a hell of a lot more than Jeff does (sleeping is a hell of a lot more than Jeff does) so I congratulate you.As to your attitude, well, I'd say that as long as you don't torture Brother Thienpont too much I don't care what you do. You remember the old Grimm fairy tale Great Claus and Little Claus? Great Claus was an idiot (you play that role) and Little Claus was the smart one (Elias can play that role). In the end of the story Little Claus takes Great Claus' fish catch and Great Claus dies (mostly from his stupidity). In real life Brother Thienpont can kindly explain to you the error of your ways (you're a monk, not a monkey, remember? I know it's hard) and send you to your room to meditate. |
|
|
| (100288) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 08:49:39 2005, in response to Re: Canarsie CBTC, posted by H.S.Relay on Sat Jun 18 08:25:20 2005. "You're joking, right?"I wish I were. "Jeff's CBTC synopsis a couple of posts back was a COMPLETEY ACCURATE description of Siemens' scheme." I don't think so. But then, it would be easy for you to prove me wrong. Just post some specifications and offer a reference to which dept. at Siemens this comes from. These are not state secrets. |
|
|
| (100289) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sat Jun 18 08:50:47 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 08:32:30 2005. What happens when you have a series of successive data points regarding position, each with a time-stamp?You need to update position and velocity data at least once per second. Does GPS give you position data that frequently? |
|
|
| (100290) | |
Re: Canarsie CBTC |
|
|
Posted by AlM on Sat Jun 18 08:54:29 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 08:39:18 2005. "Refer to the top of this post."OK, so we're only talking about above ground. By the way, when the military was insisting that GPS signals be randomized, the amount was much more than 20 meters. I think it was 200 feet. However, I would agree that for pure above-ground travel, it is not intrinsically impossible that you could measure velocity with sufficient accuracy using GPS. In particular, whatever errors occur are likely not to vary rapidly with time, so two successive measurements of position are likely to have similar errors in both magnitude and direction. Still, I SUSPECT it's not possible to measure velocity accurately enough, based on anecdotal evidence, but certainly can't prove it. |
|
|
| (100291) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 08:59:04 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Sat Jun 18 08:50:47 2005. "You need to update position and velocity data at least once per second."Agreed that this is a reasonable requirement. "Does GPS give you position data that frequently?" That's probably the upper-limit of what can be done currently - which is why I would also stipulate to GPS' being a subject of research for transit applications. Other posters have pointed to its usefulness when trains are further apart on larger geographic areas. (Freight trains) If we move from transit and go to commuter railroads with say 15 minute- to half hour intervals, we might see a point where GPS could be used. Defining the current boundary of its application is useful. Over time, that boundary will change. |
|
|
| (100292) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 09:01:16 2005, in response to Re: Canarsie CBTC, posted by AlM on Sat Jun 18 08:54:29 2005. "By the way, when the military was insisting that GPS signals be randomized, the amount was much more than 20 meters. I think it was 200 feet."Yes; you are correct. However, the current unadulterated signal is good down to 5, if ot better. But I must also concede that this is an average value, that is there is an error bar associated with this. But if we know about it, we can compensate for it. |
|
|
| (100293) | |
Re: I'd like proff. |
|
|
Posted by RonInBayside on Sat Jun 18 09:07:50 2005, in response to I'd like proff., posted by H.S.Relay on Sat Jun 18 07:50:45 2005. There is mno irony here. I merely pointed out that Jeff doesn't know what he;s talking about regarding CBTC. He'd deny it even knowing that it's already in use. That's still true."Signaling is NOT a distinct craft with a distinct design approach from ATM machines, web browsers and/or industrial controls" Do you read Nuclear News? Go to the American Nuclear Society's website and look trhrough some of their literature. Nuclear News is a good place to start. Signaling is an application with a mathematical and empirical base. CBTC can serve it just as electro-mechanical relay design serves it. Both are elegant theoetical designs; CBTC offers advantages and capabilities that the 1920s relay system cannot. |
|
|
| (100294) | |
Re: Welcome to 2nd grade |
|
|
Posted by RonInBayside on Sat Jun 18 09:15:05 2005, in response to Welcome to 2nd grade: No- YOU'RE a liar!!!, posted by H.S.Relay on Sat Jun 18 08:14:07 2005. "The TA doesn't know how many man-hours will it take to replace a vandalized radio antennae, which must be lined up to within millimeters to operate properly. (In case you're wondering, that little tidbit came from a large German electrical concern)"How do you know the TA doesn't know how do it? Siemens told you but didn't tell them? "Chambers St: 21 Interlocked signals, 6 automatics, 6 switches, 2 traffic levers. $60 million, you say." No, the TA told the NY Times $60 million. The TA was also quoted as saying the lead time for ordering relays is measured in months. Recall that the emergency repair program they did was enabled by using spares that were taken off the D line and fortunately were still available. Sounds to me like your posting title is correct. I thank you for welcoming me to your second grade class. When do you graduate to third grade? |
|
|
| (100295) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sat Jun 18 09:17:35 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 08:59:04 2005. That's probably the upper-limit of what can be done currently - which is why I would also stipulate to GPS' being a subject of research for transit applications.Assume that you get position measurements every second. You said the accuracy is +/- 15 feet. How accurate will the computed velocity be? How accurate would the position measurements have to be for the velocity accuracy to be within +/- 1.5 ft/sec (1 mph)? |
|
|
| (100298) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 09:30:18 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Sat Jun 18 09:17:35 2005. The error bar here is 30 feet (+ or - 15 feet). So if GPS is used entirely unassisted, you could be off, in the worst worst case by the length of a commuter rail car within six seconds; it is likely your error will be less than that.But you can "zero" your error with the use of a double-check (such as a fixed beacon). Recall that pilots can correct the error o an inertial navigation system by double-checking using a magnetic compass (set aside GPS and other aids for a moment in that industry). If you use fixed beacons, using GPS could allow you to use fewer beacons because you would use beacons at intervals to correct the error of your GPS. You would then define how long an interval you could tolerate (how bad the error gets before you have to correct it). That interval could vary depending on where you are. Would your tolerance for position error on GPS not be greater in a marshaling yard, train station or busy interlocking (like Harold) than out on a stretch of track where trains pass once per hour? |
|
|
| (100301) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sat Jun 18 09:42:30 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 09:30:18 2005. The error bar here is 30 feet (+ or - 15 feet).I asked you for the accuracy of the computed velocity. So, if the error in the distance measurement for a 1 second interval is 30 feet, then what is the accuracy of the computed velocity? |
|
|
| (100312) | |
Re: Canarsie CBTC |
|
|
Posted by Brother Pielet on Sat Jun 18 10:37:09 2005, in response to Re: Canarsie CBTC, posted by Jeff H. on Sat Jun 18 01:50:37 2005. It says Ron has an interest in Siemen. Its appearance, texture and taste |
|
|
| (100314) | |
Re: Canarsie CBTC |
|
|
Posted by AlM on Sat Jun 18 10:38:24 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Sat Jun 18 09:17:35 2005. "You said the accuracy is +/- 15 feet. How accurate will the computed velocity be?"Depends how rapidly the error can change with time. If each error is totally independent of the preivous error, then of course 15' is too much. If each error of position is very closely correlated in magnitude and direction with the prior error, then it would be possible to get extremely accurate measurements of velocity. |
|
|
| (100317) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 10:44:55 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Sat Jun 18 09:42:30 2005. If you are traveling at 50 mph, that would be 50 x 5280 feet per mile = 264,000 feet per hour. Divide that by 3600 seconds per hour, and you get 73.33 feet per second.But we want to try to measure that by GPS. So at time point 0 you are at point zero on the track. At time 1 you are, by GPS, somewhere between 58 feet and 88 feet down the track. The maximum error at that first time point is + or - 15/73 or 20.5%. At time 2 the error could go to +/- 30 feet or 41%; after 3 seconds 62%. So let's say your true velocity is 73 feet per second. Between time point 0 and time point one GPS could estimate your velocity at between 58 and 88 feet per second; at time point two, your estimated velocity by GPS is between 58+43/2 = 50.5 feet per second and 88+ 103 feet per second/2 = 191/2 = 95.5 feet per second, which is 24-25% off one way or the other. This error then grows with succeeding time points. So it's t1(p1*error factor) + t2(p2*error factor) = t3(p3*error factor)...tN(pN)*error factor) / t --> velocity estimate for time interval tN. |
|
|
| (100318) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 10:46:59 2005, in response to Re: Canarsie CBTC, posted by Brother Pielet on Sat Jun 18 10:37:09 2005. Are you expressing a sexual fetish? Is celibacy right for you? Either that, or you like sucking on electrical components. Do you have an iron deficiency, perhaps, or plumbism (high serum lead levels - that might explain your ranting). |
|
|
| (100320) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sat Jun 18 10:51:16 2005, in response to Re: Canarsie CBTC, posted by AlM on Sat Jun 18 08:54:29 2005. Still, I SUSPECT it's not possible to measure velocity accurately enough, based on anecdotal evidence, but certainly can't prove it.While there are many numerical integration techniques, there are very few for numerical differentiation. Here's one of the problems. Consider Fourier transforms (moment generating functions for those used to statistics) and the operation for differentiation in the time domain with what happens in the frequency domain. Time domain differentiation is equivalent to multiplying the Fourier transform by the frequency. So, if there is a small additive high frequency noise component it's going to swamp the derivative. The reverse is true for integration; the transform is divided by the frequency. That means that high frequency noise will have very little effect. This indicates that whatever numerical differentian technique is chosen, it must incorporate some form of band limiting to be immune to noise. Band limiting introduces delays into the result. So, if one wants to derive velocity from position measurements, one has the choice of making the computed velocity either timely or accurate but not both. I don't propose that this is a proof, but it should point you in the proper direction. |
|
|
| (100322) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 10:54:20 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Sat Jun 18 10:51:16 2005. Would the magnitude of the delay depend on the degree of accuracy you want? |
|
|
| (100329) | |
Re: _three_ manufacturers |
|
|
Posted by ALSTOM R160A on Sat Jun 18 11:00:44 2005, in response to _three_ manufacturers, posted by H.S.Relay on Sat Jun 18 08:17:28 2005. ALSTOM est ici et prépare pour défendre ! |
|
|
| (100330) | |
Re: Canarsie CBTC |
|
|
Posted by AlM on Sat Jun 18 11:03:51 2005, in response to Re: Canarsie CBTC, posted by RonInBayside on Sat Jun 18 10:44:55 2005. This error then grows with succeeding time points.Um, no. The error in position as measured by GPS never exceeds 15 feet from your true psoition. Therefore, the more time between measurements, the less the error in velocity. If I measure my position at t=0 and t=15, the error of each positional estimate is still only 15 feet. Subtracting, my maximum possible mistake is 30 feet, or 2 feet per second. Now if I measure position at t=0 and t=30, my maximum possible positional error in the difference of the measurements is still only 30 feet, so my maximum error in the velocity calculation is 1 foot per second. And as I mentioned elsewhere I think the error in velocity is much less because successive position erros are strongly correlated. There is one other issue, however. What if you don't get a signal? The satellite signals are so weak that some unauthorized nearby interference could kill some of them temporarily. As I mentioned, this happens frequently in Manhattan with a hand held GPS. |
|
|
| (100332) | |
Re: _three_ manufacturers |
|
|
Posted by RonInBayside on Sat Jun 18 11:05:57 2005, in response to _three_ manufacturers, posted by H.S.Relay on Sat Jun 18 08:17:28 2005. Alstom is never chopped liver! But goosefat and fresh garlic would be ok! |
|
|
| (100333) | |
Re: Canarsie CBTC |
|
|
Posted by RonInBayside on Sat Jun 18 11:08:04 2005, in response to Re: Canarsie CBTC, posted by AlM on Sat Jun 18 11:03:51 2005. Thank you for correcting me."What if you don't get a signal? The satellite signals are so weak that some unauthorized nearby interference could kill some of them temporarily. As I mentioned, this happens frequently in Manhattan with a hand held GPS." A valid concern. Less of a problem out of the canyons of Manhattan or center city Philly. What about the Hudson Line? |
|
|
| (100336) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sat Jun 18 11:11:06 2005, in response to Re: Canarsie CBTC, posted by AlM on Sat Jun 18 10:38:24 2005. Depends how rapidly the error can change with time. If each error is totally independent of the preivous error, then of course 15' is too much. If each error of position is very closely correlated in magnitude and direction with the prior error, then it would be possible to get extremely accurate measurements of velocity.You must make the assumption that the errors are statistically independent. The fact that the position reading on a stationary GPS receiver does not vary means only that successive readings are repeatable. That's not necessarily a bad thing, if the application is to find the same spot over and over again. Repeatability and accuracy are two different animals. You need accuracy to compare successive readings from two different locations. |
|
|
| (100338) | |
Re: Canarsie CBTC |
|
|
Posted by AlM on Sat Jun 18 11:14:10 2005, in response to Re: Canarsie CBTC, posted by Stephen Bauman on Sat Jun 18 11:11:06 2005. "You must make the assumption that the errors are statistically independent."Why must I make that assumption? I don't believe that surveyors who use GPS in their job make that assumption. I may be wrong, but I'd actually be interested if you have some information proving me wrong. I have the strong impression from everything I've read that position errors are strongly correlated in both space (i.e., location of the measurement) and time. |
|
|
| (100340) | |
Re: Canarsie CBTC |
|
|
Posted by Stephen Bauman on Sat Jun 18 11:21:25 2005, in response to Re: Canarsie CBTC, posted by AlM on Sat Jun 18 11:14:10 2005. I don't believe that surveyors who use GPS in their job make that assumption.They use only one reference point. All their other points are still measured optically. They can find that reference point again, thanks to repeatability. |
|
|
|
Page 6 of 8 |
||
