fastest but pratical way to measure time

Suppose I want to measure a time interval. How precise can I get? In AOE it says TAC's can go down to the ps's which I doubt I can get but I have to ask.

Basically I'm measuring how long it takes light to travel to use in triangulation. The issues is that I would like to get mm or atleast cm precision. i.e. 1/1000/c ~= 3*10^12= 3 ps.

Anyways, what I'm interested in is some way of doing this that is practical and doesn't take a research grant to do. Since the signal used is wireless I would also need a very accurate way to have two clocks at precisely the same frequency which I don't know if it is possible.

I was thinking, if say, I had a perfect capacitor I could use it to measure the time difference(since it is continuous and if ideal then the exact formula is known) but I don't know how error prone this is in the practical world?

Basicaly when the incomming signal is recieved it triggers the charging of a capacitor and then when the next clock pulse happens it causes the a sample and hold to happen on the capacitor's voltage. (so the initial voltage is Vo and the final is Vf which are known and if hte capacitance is known then the time it took to charge is known) I think this method is to error prone since I would imagine it would be impossible to get precise enough information on these factors and to get a capacitor that was close to ideal? (although maybe one could use a bank of capacitors and use some averaging effect?)

Anyways, is it possible to get what I want in a reasonable way(Which means that a hobbiest can do it)?

I'm still looking into Kalman filtering to see how much it will help but I'm trying to see how practical my idea is before I start investing to much time into it. For now I don't mind a somewhat inprecise implementation(say even down to the meter) if I know it can be improved to what I ultimately want(mm's).

Any ideas?

Thanks, Jon

Reply to
Jon Slaughter
Loading thread data ...

Jon Slaughter wrote: > Suppose I want to measure a time interval. How precise can I get? In AOE it > says TAC's can go down to the ps's which I doubt I can get but I have to > ask. >

So after calling everybody stupid back December, you want to know how to measure time? I'm pretty sure it won't happen with your non-existant perfect capacitor. BTW measuring charges on capacitors is the electronic analog of an hourglass. You'll notice nobody uses them for anything. Of course I'm no expert.

GG

Reply to
stratus46

huh, I called everyone stupid? No, I called those that seem to think that I'm suppose to take there word as it came from the bible as assholes. Its funny how those people who I ask only to explain there "It won't work" farther get pissed instead of actually explaining it. If your like that then there is no farther reason for you to respond to my posts.

I only ask those who want to respond and actually are interested in helping to respond and not those who are trying to look smart or have some ego problem. If you don't like it when I asked you explain your reasoning then go f*ck off. If you take it as an insult when I question your reasoning then go f*ck off(Its your problem for getting insulted and assuming I'm questing your intelligence. Since you assume that your probably not as smart as you think).

Only the true experts here have no issues with me. They are not offended even if I directly insult them, which I haven't. I haven't insulted any real experts here. Those people who I have "insulted"(which is usually only after the fact that they have insulted me) are not experts by any means. Experts do not get pissed at people asking them to explain themselfs and they do not have any need to prove how smart they are.

I'm pretty sure it won't happen with your non-existant

So I guess if I ask you why then I'm insulting you and calling you stupid. I guess if I also ask you to clarify yourself then I'm calling you stupid. Well, knowing your first sentence I already see that you think you are an expert yet you seem to have a slight attitude so I think I already know your response if I ask you to "Explain what you mean by hourglass"(microseconds, milliseconds, etc..). You know. That part of the response may or may not be valid and if you were a true expert you would have done better to ask "How long can you wait to measure a capacitor" or something to that effect. As far as you know I don't need to measure but every 3 hours. Maybe by hourglass you mean measured in lightyears? how bout milliseconds?

Anyways, I can see we won't get anywhere with this. I ask for those who actually do not mind sharing there knowledge AND explaining there reasoning and are not here to bitch and play little ego games. If no one responds to my questions then thats perfectly fine. I'd rather no one respond than someone with half a brain and a whole lotta arrogance turning what really should be "In my experience...." to something like "IT WON'T WORK!!! EVERY!! GO HOME AND BAKE A PIE!" kinda BS and then get pissed when I ask them why it won't work(which only tells me the really don't know why and hence it actually might work).

Point is, "experts" are a dime a dozen in these NG's. I'd said it once and I'll say it again that the only expert I have seen here is Winfield Hill. You can see this in is mannerism's when he responds, in is detailed knowledge about many things, and the respect he gets from others. I have yet seen him get pissed at anyone and respond in a childish way like I have from the so called other "experts". Ofcourse its all a matter of degree but people like Larkin who end up acting like a child are not experts(they might be intelligence on somethings but not as intelligent as they thing). A real expert will not even go into that real that many of you do because they have much better things to do and are confident in what they know and don't need validation on some NG like many of you do. Stop acting like kids and if you can't then don't even respond to me(and BTW, it is irrelevant how I act).

The real problem, if you notice is that one person starts the gravy train rolling and all you other fags jump on. One person thinks I should suck some others dick for some nonsensical reason and all you have to just jump in and say the same thing(reminds me a lot of HS) because, I suppose, you have nothing better to do... strange that your expertise is not need. If you actually would take the time out to follow the conversations in time you will see exactly what I mean. I do not insult anyone until I sense arrogance/ego or I'm insulted. I might insult different people at different times and others think I'm insulting them.

For example, goto the sci.electronics.design newsgroup with my last post about heatsinks. Two people responded with answers "It won't work". I ask them to explain farther because I have other people saying something different. Another poster responds that I should believe everything these people say because they are experts. I say that I will not believe everythign they say and they need to explain themselfs farther. I will ultimately decide what I believe. All of them take it as an insult. I get insulted by Larkin after that and then I insult him. Is Larking a real expert? No. If he was he wouldn't have insulted me(either before or after or ever) (ok, everyonce in a while an expert might act a little childish but I don't think this is the case with Larkin). So anyways, the other "experts" then chime in and insult me for insulting Larkin. Is this how real experts behave? If so then I don't want to be around them. I might as well go back in HS if I want to be around that kinda mentality.

Again, though, I want to point out that its not a crime for one to question athority. A true expert knows that he's not always right and past history does not ALWAYS explain present events. This kinda mentality is usually one from ego thinking that hte world revolves around them.

If someone wants to respond and wants to be treated with respect then when they choose to respond with an answer that requires farther explianation then they have a responsibility to explain when it is asked for or atleast explain why they can't or not explain anything at all. In any case a insulting respond is uncalled for and will only be answered by an insult by another person who is not an expert(like me, I am far from an expert and maybe Larkin is closer but he has not made it yet(probably because of the ego part)).

Reply to
Jon Slaughter

formatting link

24 picosecond single-shot resolution.

But the real issue is, what would you use for the pulsed light source and for the detector?

John

Reply to
John Larkin

Thanks John. I hope we might be able to get along in the future. I hope you do realize that I am not trying to insult you when I ask you to explain stuff. Please don't get upset. I do think your highly intelligent from what I have seen(yet I do not know if your an expert or not but thats not for me to decide). If you do get upset at something just remember that maybe the other person was having a bad day or maybe you took it the wrong way. Anyways, hopefully we can put that crap behind us ;) (sounds like a married couple but...) If you think it won't work or have issues with things then say them... don't get offended as usually that is not my intention(you always seem fairly diplomatic at the start but then get irritated for some reason).

Ok, on to what I want to do:

Well, my goal is to send a signal of some frequency over an antenna(doesn't matter the frequency as far as the speed of light is concerned except for enviromental effects) and then recieve it by another antenna. If, suppose, on both sides I have perfectly precide clocks running at the exact same frequency(I know, its impossible but I have no idea about the degree and how much I actually need for my app) then I can measure the distance(to be used in triangulation) using d = c*t where t is the time it takes. Ofcourse t is not precise because my measurements are not precise. If, say, I used a clock of M hz then t could be off by about 1/M.

Anyways basically I thinking

transimitter->reciever->TAC->app.

I assumed that its no issue transmitting a frequency over an antenna since it happens all the time. The main issue here is size, cost, and frequency(which frequency to use that is best for my app). I also probably don't have to worry about "reverberation" of the signal since I'm looking for the first instance and I believe that the reverberations will die out before I send the next pulse.

So the main issue(IMO, I could be wrong) is of measuring the time accurately so I get enough precision in my app to make it useful.

The way I was thinking about it is to have a clock that measures on the edge at a frequency of M. If I do this then my precision is 1/M. For mm then M ~= 300Ghz(~3ps). I might be able to get away with cm precision and use some type of averaging(kalman filtering well suited for this I think).

Ofcourse is there was a "continuous" way to keep track of this time then it would be much better assuming noise and errors are not an issue(Which I think they would be. Even with the clock I would be limiting my precision to jitter?)

The whole point is try to compute d = c*t by measuring t. Theoretically this is quite simple but as usually I get stuck in turning it into something practical.

There also might be other effects that I can use such as dopler but I'm not quite sure. To me the above is the easiest at this point as it is direct but it might be impractical(which is why I'm asking. Ofcourse I'm not asking someone to tell me it is impractical but why it is and how practical it can be).

So there are 3 main issues where I think error crepts in. One is clock synchronization/jitter, transmission, and time measurement.

I'm not sure about the first to being huge issues compared to the second for my problem. It might turn out there is no precise way to synchronize the clocks for the precision I'm after(even if the second two are ideal). The second case might be a problem too as the speed of light is c only in a vacuum and probably not constant(but I think its constant enough for my app, I just gotta choose the right frequency to send at(whcih might be an issue because of FAA)). The third, IMO, seems the real issue though. If I can't know the precise time then I'm limited. This limitation might be enough make the project useless. All the factors combined are the real issues but I think the 3rd is more important than the others and hence is the one I'm trying to investigate first.

I'm sure there are other issues involved such as delays created by the electronics and stuff. I'm not sure how much these factor into the final result and if they can be taken care of or not. I know that a transistor takes a finite amount of time to turn on and this delay propagates through the system an is absorbed into the t I'm trying to find by I might be able to remove it through calibration or some other means(such as averaging).

Basically I have a set of recievers that are trying to triangulate the position of an object(which w.r.t to c is practically fixed. Ofcourse if the object was truely fixed I probably could use some other means to measure its distance). I was thinking I could also use an accellerometer but I think this would be much more inaccurate? If my plan above is impractical then I'll look into some other methods and see how they work(or if anyone has any good ideas for me to look at).

Thanks, Jon

Reply to
Jon Slaughter

a.. 12-channel, 32-bit resolution timestamper b.. Measures time intervals up to 6.71 seconds c.. Each of 12 channels is independent timestamper d.. 256-sample FIFO memory for each channel e.. Jitter typically below 75 ps RMS f.. Includes option of up to 8M RAM for event buffering g.. 32-bit VME module

24ps?

It would seem the jitter limits my accuracy to about 1 cm if it truely is

24ps but would be about 1/2 a foot or so. I wonder the costs of these things though as it might be to much(over 1k would be to much for it being practical in my case)?

Also, what I was thinking is that I could have several fixed receivers for triangulation and then take all the combinations by 3 (n,3) and could then average the results. 3 receivers would give me an approximate 3d position and if I space the receivers properly then I would have n!/3!/(n-3)! which grows exponentially with n. So if I had n = 12 receivers I would have

12!/3!/9! = 6*11*10 = 660 points to average over which should give me a very good measurement as long as all the other factors I talked about are not to huge of an issue.

Thanks, Jon

Reply to
Jon Slaughter

(ok, maybe I need 4 for 3D)

Reply to
Jon Slaughter

formatting link

Sorta points to what I can use. (I call this triangulation in general). Actually I was going to use trilateration

formatting link
but I suppose either case will work.

Reply to
Jon Slaughter

and

formatting link

might be usefull too.

Reply to
Jon Slaughter

Jon, the people on these newsgroups don't owe you a damn thing. You have a piss poor attitude and make demands of others. It doesn't work that way. Learn to use a good search engine along with the USENET archives of the newsgroups on "Google Groups". This isn't a "Have it your way" burger joint. The people that you condemn have spent their lives STUDYING electronics, not asking stupid questions, then crying like a two year old when they aren't spoon fed the answer the want to hear. Keep that attitude, and the only replies you'll get are from the nut jobs.

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
 Click to see the full signature
Reply to
Michael A. Terrell

Did I ever once say they did? This is your first problem.

Yep, sure do. When people like you come in here with your ego then I'm not going to take it up the ass and smile. This is your MAIN problem.

It doesn't work

So your really have an issue with me just asking? You don't like when people ask for help because you think there lazy? Well to f****ng bad.

Don't f****ng respond then. When I post a question I don't except or demand a response. This is your third problem.

lol. your last sentence says it all. (I guess I have to say it: So your implying that you are a nut job?)

Look, I have no problem with people replying or not and saying whats on there mind. But when I ask for help and all I get is BS ego then I'm not going to shut my mouth(maybe I should and sometimes I do). Its people like you that think you own these NG's and have some right to tell others how to behave in it. You have no right. The behavior is etiquette and not required. If I act like baby then thats my problem. If you don't like me asking for help then thats your problem. If I want to be lazy and have everyone else do all the work for me(which is not the case) then thats my problem not yours. But yet you seem to want to make all my problems your problems. Don't you have your own problems?

I'LL SAY IT AGAIN! I WILL NOT TAKE EVERY WORD BY ANY MOTHER FUCKER(AN EXPERT ON HERE OR NOT) AS THE WORD OF GOD AND I WILL ALWAYS QUESTION WHAT THEY SAY TO SOME DEGREE OR ANOTHER. TAKE IT AS AN INSULT IF YOU WANT.

I can't believe some of you with attitudes like this even graduated college.

If you wanna keep your ego when replying to me then I'll do my bes to keep mine. If you want to share your knowledge and experiences with no strings attached then I'll do my best to make use of it. If you wanna bitch about me not believing 100% of everything you say on faith then just put me on your ignore filters. ITS THAT FUCKING SIMPLE!

Reply to
Jon Slaughter
[snip]

Your points would have been much more forcefully made without the potty mouth. Just sayin', that's all.

Reply to
Greg Neill

D- A half assed rant from an immature little boy. Needs work, or you will fail at being a troll. The only thing that is f****ng simple, is you.

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
 Click to see the full signature
Reply to
Michael A. Terrell

J, one response from someone outside the discussion, if I ever see you "asking" another question, and remember your name, I will not give you an answer even if you beg me to. Now go and proceed to justify your rudeness.

Reply to
Sjouke Burry

Thats fine. I have no issues with that. Its not your job to answer my questions. I never said it was. What pisses me off, maybe because I'm immature, is that you really think I have an issue with it and you think you can use it against me in some way like I care.

What really pisses me off, and maybe the reason I say the things I do, is simply that you people who respond like do it only to try and prove some point that you think is valid. The only thing your doing is wasting my time(well, I'm wasting it for responding), NG bandwidth, and causing other people to waste there time(by wading throught the crap responses or having htem get in on the conversation. People like you tend to clump together). These newsgroups are suppose to be able knowledge and not about ego yet thats all I see. (and which is what pisses me off because I believe it should be about learning and knowledge instead of other bullshit)

I really don't see the point in folks like you posting this nonsense though. If what you say is true then why respond in the first place? I know, the exact same reason why I'm responding. In my case I have a definate point though which is to try and make you understand that that these NG's are for knowledge and you should check your ego at the door.

I'm here to learn, what are you here for? [ Ofcousre the de facto response to that is that you will say I'm just here to call people stupid and insult them. You might ask yourself why you believe this first and then maybe try and find real proof of it. (if you see me insult someone first ask yourself why I responded that way and remember that I'm not going to take any BS from anyone when it deviates from the main goal of learning(which for some of you means understanding why not not just taking it on faith). ]

Any time I post to help someone that is my one and only goal. If they don't believe me then thats there choice. If they say I'm wrong then maybe there right. I don't get upset(ok, once in a blue moon I do but its usually because they get there ego in there and the conversation turns from knowledge to BS).

If I choose to click the reply button on a message where someone is being asked for help then its my duty try and help them and not anything else. If I do anything else then its my fault and not theres. If they call me a jack ass then maybe I have a right to get pissed but it depends why. My best option is just to just put them on an ignore filter if I really don't like it.

I think the issues involve din these newsgroups is that you have a few people that start shit and then everything goes down hill from there. One person reads a post and takes a few things out of context and tries to start some BS and then others jump in trying to act like they got some intelligence and then it just results in a big waste of time. If people would just try to stick to only spreading knowledge then 99% of these issues would be solved.

Reply to
Jon Slaughter

message

Can you bounce the signal off the target and rely on one clock?

R
Reply to
Roger Dewhurst

I don't think so ;/ I'm not sure how hard this would be but there are actually multiple targets and not always a clear LOS(although its mostly LOS the objects can get in each others ways).

Essentially I am implemented a GPS like system on a much smaller scale. The site I gave a link to says something about a maximum error of 100m or so is required by some standard for GPS and I figure that since I'm using a much smaller scale I can get much better accuracy. Ofcourse if I'm relying on time differences only I suppose it really just has to do with the "clock" and independent of the actual scale(well, except for the noise issues that come along with scaling). I guess though I might have look into the different possible solutions a lot more as its not going to be as simple as I though.

Thanks, Jon

Reply to
Jon Slaughter

--
Light-years is a measure of distance, not time.
Reply to
John Fields

But your GPS system doesn't have to include the vacuum of space. How about a system that sends out an RF pulse as a timing reference, and a sound wave signal to be used as a distance measure. RF travels just about 1 million times faster than sound, so its delay can be ignored. Just measure the sound delay. And since sound travels only about a foot per millisecond, your time measurement resolution doesn't need to be nearly so good, to measure distances to an inch resolution.

Reply to
John Popelish

Yes, I thought about using sound but surely that is not practical? It would require a great deal of power send over the distances unless the frequency is very low and it is very noise prone(cannot distinguish it from other sounds)? Also this project might be used with people around and it should be transparent to them. Maybe using a high frequency sound > 20khz but wouldn't this have a lot of issues with direction? (and would need a huge amount of amplification + would require a much larger size transmitter)

Thanks, Jon

Reply to
Jon Slaughter

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.