Turn your Rigol DS1052E Oscilloscope into a 100MHz DS1102E
in [Design]
|
From: John Larkin on 9 Apr 2010 11:27 On Thu, 8 Apr 2010 22:26:29 -0500, "Tim Williams" <tmoranwms(a)charter.net> wrote: >"John Larkin" <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in message >news:ks6tr5h4k5jv51n0lp1is4hrs537v1rmse(a)4ax.com... >> Delay lines can't store milliseconds of pre-trigger data. The best >> they do is let you see a few ns of pre-trigger waveform. > >John John John, > >That's what entire spools of foamed teflon coax are for! ;-) > >Man, I can just imagine how many hours of cocaine you could buy for the >price of a few microseconds of that sort of stuff. Seems kind of >disappointing. > >Tim It's hard to store much information in an electromagnetic delay line. The losses kill you. It would be an interesting calculation to see how many bits you could store in any preferred length of, say, RG58. Fiber is a different story. You can stuff gigabits per second into a hundred kilometers of single-mode fiber and recover it perfectly. But RAM is a more sensible way to store information. We have a couple of spools, or rigid coils actually, of half-inch hardline coax in our lab, 50 ns ballpark. They make handy zero-jitter pretrigger delays for sampling scope situations. But the pulse that comes out the end is clearly degraded from what goes in. We also have a couple of Tek 7M11 delay-line plugins, which can be used standalone. It's dual-channel, 75 ns per, 2 GHz bw. They give up half the signal so that they can equalize for the losses. John
From: John Larkin on 9 Apr 2010 11:33 On 9 Apr 2010 03:46:01 -0500, The Phantom <phantom(a)aol.com> wrote: >On Thu, 08 Apr 2010 17:36:30 -0500, "krw(a)att.bizzzzzzzzzzzz" ><krw(a)att.bizzzzzzzzzzzz> wrote: > >>On 8 Apr 2010 14:14:01 -0500, The Phantom <phantom(a)aol.com> wrote: >> >>>On Thu, 08 Apr 2010 17:06:06 GMT, Jan Panteltje <pNaonStpealmtje(a)yahoo.com> >>>wrote: >>> >>>>On a sunny day (Thu, 8 Apr 2010 08:57:53 -0700 (PDT)) it happened brent >>>><bulegoge(a)columbus.rr.com> wrote in >>>><fb2980a4-6e02-48db-b20d-57eefa8df2a6(a)v8g2000vbh.googlegroups.com>: >>>> >>>>>On Apr 8, 10:01�am, John Larkin >>>>beep BAD SYNTAX >>>> >>>>>> I can't think of much. Maybe clean X-Y plots; the digitals are sloppy >>>>>> in X-Y mode. >>>>>> >>>>>> ftp://jjlarkin.lmi.net/Hills.JPG >>>>>> >>>>>> John >>>>> >>>>>When I am looking at video signals with higher power RF signals on the >>>>>board I will take an analog scope any day. >>>> >>>>I agree, for video an analog scope is great. >>>>In fact the *ONLY* reason for digital is storage, >>>>and even then good analog storage scope once existed. >>> >>>I can think of a particular aspect of digital storage that, AFAIK, analog >>>storage can't do. With digital storage, one can examine the signal BEFORE the >>>trigger point. Has there ever been an analog storage scope that could do that? >> >>Sure. Have you ever heard of delay lines? Also, delaying timebases were >>commonly used to look "before" the trigger event. > >Please explain how a delaying timebase could allow one to see events before (and >I mean much more than a delay line's worth of "before") the trigger event if the >trigger event is a one-shot event; no periodic waveforms involved. He probably refers to zooming in on the next recurrence of a periodic waveform. But signal period jitter and scope delaying timebase jitter can be nasty. Classic delayed sweeps and sampling scope timebases run in the ballpark of one part jitter in 20,000 to 50,000 of the delay time. John
From: Nico Coesel on 9 Apr 2010 12:38 "JosephKK"<quiettechblue(a)yahoo.com> wrote: >On Thu, 08 Apr 2010 16:49:13 GMT, nico(a)puntnl.niks (Nico Coesel) wrote: > >>"JosephKK"<quiettechblue(a)yahoo.com> wrote: >> >>>On Wed, 07 Apr 2010 17:17:53 GMT, nico(a)puntnl.niks (Nico Coesel) wrote: >>> >>>>"David L. Jones" <altzone(a)gmail.com> wrote: >>>> >>>>>John Larkin wrote: >>>>>> On Tue, 06 Apr 2010 19:34:32 -0700, >>>>>> "JosephKK"<quiettechblue(a)yahoo.com> wrote: >>>>>> >>>>>>> On Mon, 05 Apr 2010 21:45:57 -0700, John Larkin >>>>> >>>>>Yup, economy of scale and other factors. >>>> >>>>>and the 470MHz at $12K+: >>>>>http://www.tequipment.net/IwatsuSS-7847A.html >>>>>Ouch! >>>> >>>>They are a lot cheaper second hand because no-one wants an analog >>>>scope anymore. $500 should be enough. >>> >>>There is still a few us who want one. I am still on the lookout = >for=3D20 >>>a Tek 2465 or very similar. There are cases where even my TDS = >544A=3D20 >>>won't get me the same results. >> >>The Iwatsu SS-7847A is a very good oscilloscope and less old than the >>2465. > >The Tek 2465 has not been made for many years, the only ones available=20 >are used. And guess what, they are a damn sight less than $12000. My=20 >current use is more like (kind of well off) expert hobbyist. Guess=20 >which one fits my needs best. A second hand SS-7847A (Lecroy LA314H) is likely to be newer, better and cheaper than a Tek 2465. Just check Ebay. -- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico(a)nctdevpuntnl (punt=.) --------------------------------------------------------------
From: JosephKK on 9 Apr 2010 22:09 On Thu, 08 Apr 2010 20:16:59 -0700, John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: >On Thu, 08 Apr 2010 19:52:47 -0700, >"JosephKK"<quiettechblue(a)yahoo.com> wrote: > >>On Thu, 08 Apr 2010 17:36:30 -0500, "krw(a)att.bizzzzzzzzzzzz" <krw(a)att.bizzzzzzzzzzzz> wrote: >> >>>On 8 Apr 2010 14:14:01 -0500, The Phantom <phantom(a)aol.com> wrote: >>> >>>>On Thu, 08 Apr 2010 17:06:06 GMT, Jan Panteltje <pNaonStpealmtje(a)yahoo.com> >>>>wrote: >>>> >>>>>On a sunny day (Thu, 8 Apr 2010 08:57:53 -0700 (PDT)) it happened brent >>>>><bulegoge(a)columbus.rr.com> wrote in >>>>><fb2980a4-6e02-48db-b20d-57eefa8df2a6(a)v8g2000vbh.googlegroups.com>: >>>>> >>>>>>On Apr 8, 10:01 am, John Larkin >>>>>beep BAD SYNTAX >>>>> >>>>>>> I can't think of much. Maybe clean X-Y plots; the digitals are sloppy >>>>>>> in X-Y mode. >>>>>>> >>>>>>> ftp://jjlarkin.lmi.net/Hills.JPG >>>>>>> >>>>>>> John >>>>>> >>>>>>When I am looking at video signals with higher power RF signals on the >>>>>>board I will take an analog scope any day. >>>>> >>>>>I agree, for video an analog scope is great. >>>>>In fact the *ONLY* reason for digital is storage, >>>>>and even then good analog storage scope once existed. >>>> >>>>I can think of a particular aspect of digital storage that, AFAIK, analog >>>>storage can't do. With digital storage, one can examine the signal BEFORE the >>>>trigger point. Has there ever been an analog storage scope that could do that? >>> >>>Sure. Have you ever heard of delay lines? Also, delaying timebases were >>>commonly used to look "before" the trigger event. There are applications >>>where I'd still like a calibrated delayed timebase. > >Delay lines can't store milliseconds of pre-trigger data. The best >they do is let you see a few ns of pre-trigger waveform. > >John > You clearly have forgotten how delayed sweep works. The sweep is not even on the same trigger cycle. Say alternate trace, just like is typical for multichannel analog scopes (dual gun scopes excepted).
From: JosephKK on 9 Apr 2010 22:37
On Fri, 09 Apr 2010 08:27:49 -0700, John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: >On Thu, 8 Apr 2010 22:26:29 -0500, "Tim Williams" ><tmoranwms(a)charter.net> wrote: > >>"John Larkin" <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in message >>news:ks6tr5h4k5jv51n0lp1is4hrs537v1rmse(a)4ax.com... >>> Delay lines can't store milliseconds of pre-trigger data. The best >>> they do is let you see a few ns of pre-trigger waveform. >> >>John John John, >> >>That's what entire spools of foamed teflon coax are for! ;-) >> >>Man, I can just imagine how many hours of cocaine you could buy for the >>price of a few microseconds of that sort of stuff. Seems kind of >>disappointing. >> >>Tim > >It's hard to store much information in an electromagnetic delay line. >The losses kill you. > >It would be an interesting calculation to see how many bits you could >store in any preferred length of, say, RG58. > >Fiber is a different story. You can stuff gigabits per second into a >hundred kilometers of single-mode fiber and recover it perfectly. Really? How much does it cost to do that? If you could back off to say 20 kilometers i would not have much reason to find fault. Fiber still does have attenuation and EDFA can only do so much. > >But RAM is a more sensible way to store information. Sometimes at least, depends on how much and how fast it is coming in or going out. > >We have a couple of spools, or rigid coils actually, of half-inch >hardline coax in our lab, 50 ns ballpark. They make handy zero-jitter >pretrigger delays for sampling scope situations. But the pulse that >comes out the end is clearly degraded from what goes in. > >We also have a couple of Tek 7M11 delay-line plugins, which can be >used standalone. It's dual-channel, 75 ns per, 2 GHz bw. They give up >half the signal so that they can equalize for the losses. > >John |