Forums

Which oscilloscope to go for?

Started by faiyaz March 7, 2009
Dear all,

I am working on a hardware which has AT91SAM7S64, which works on 18.432
MHz main clock.  This is the maximum frequency which can be seen on
hardware.  Now we are planning to buy an analog oscilloscope, we have seen
several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ scope has
the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS,  30 MHz has around
11 nS.  Can you suggest what bandwidth we should go for?

Also if there is anybody who is from India and if used Scientific or Aplab
Oscilloscopes?  If yes than can you evaluate and compare them?
"faiyaz" <faiyaz.pw@gmail.com> writes:

> Dear all, > > I am working on a hardware which has AT91SAM7S64, which works on 18.432 > MHz main clock. This is the maximum frequency which can be seen on > hardware. Now we are planning to buy an analog oscilloscope, we have seen > several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ scope has > the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS, 30 MHz has around > 11 nS. Can you suggest what bandwidth we should go for?
Analog scopes are great, but for debugging microcontroller hardware a digital scope is generally more useful (if you have to choose). For example, a digital scope is superior for debugging: - power on/off & reset sequencing - code timing, latencies etc (by toggling an I/O) - I2C, SPI, UART, ADC interfaces - RS485, tranceiver direction switching etc Having said that, there are some cheap logic analyser pods that may perhaps be used for some of these instead, I am not that familiar with them. To address your actual question, it is indeed the risetime that you should be looking at rather than the absolute frequency. If possible you should get a scope that can measure the fastest rise/fall time in your system. So that you can look at things like ground bounce, switching glitches, reflection and crosstalk issues. Switching can be very fast on modern processors, even if the external clock is not. Sorry I do not have specific information on the SAM7S64. I do know that for an analog scope I would rather buy a 400MHz unit off ebay than a new 100MHz one for the same price. For digital there are some new decent-looking low cost ones around now. Although I think to compete with all features of a good analog scope too, a digitial one would still need to be a high-end model.
> Also if there is anybody who is from India and if used Scientific or Aplab > Oscilloscopes? If yes than can you evaluate and compare them?
-- John Devereux
"faiyaz" <faiyaz.pw@gmail.com> wrote in message 
news:c_udnSvW8O-7_i_UnZ2dnUVZ_gKWnZ2d@giganews.com...
> Dear all, > > I am working on a hardware which has AT91SAM7S64, which works on 18.432 > MHz main clock. This is the maximum frequency which can be seen on > hardware. Now we are planning to buy an analog oscilloscope, we have seen > several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ scope has > the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS, 30 MHz has around > 11 nS. Can you suggest what bandwidth we should go for? > > Also if there is anybody who is from India and if used Scientific or Aplab > Oscilloscopes? If yes than can you evaluate and compare them?
I would not get anything less than 100mhz scope.
faiyaz wrote:

> Dear all, > > I am working on a hardware which has AT91SAM7S64, which works on 18.432 > MHz main clock. This is the maximum frequency which can be seen on > hardware. Now we are planning to buy an analog oscilloscope, we have > seen several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ > scope has > the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS, 30 MHz has > around > 11 nS. Can you suggest what bandwidth we should go for? > > Also if there is anybody who is from India and if used Scientific or > Aplab > Oscilloscopes? If yes than can you evaluate and compare them?
These days go for a digital isolated scope. With the sort of money you invest in such instruments you get much more out of them than just the plain analogue (which are a similar ball-park price as the digital ones). See <http://www.tti.co.uk/products-resale/tektronix/tek-pdfs/tps2000-tti.pdf> for a datasheet on the Tek TPS2024 I bought recently (just under the &#2013266083;2k mark). 4 channels fully isolated to 200MHz and very light and portable too. -- ******************************************************************** Paul E. Bennett...............<email://Paul_E.Bennett@topmail.co.uk> Forth based HIDECS Consultancy Mob: +44 (0)7811-639972 Tel: +44 (0)1235-811095 Going Forth Safely ..... EBA. www.electric-boat-association.org.uk.. ********************************************************************
On Sat, 07 Mar 2009 06:05:58 -0600, "faiyaz" <faiyaz.pw@gmail.com> wrote:

>Dear all, > >I am working on a hardware which has AT91SAM7S64, which works on 18.432 >MHz main clock. This is the maximum frequency which can be seen on >hardware. Now we are planning to buy an analog oscilloscope, we have seen >several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ scope has >the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS, 30 MHz has around >11 nS. Can you suggest what bandwidth we should go for? > >Also if there is anybody who is from India and if used Scientific or Aplab >Oscilloscopes? If yes than can you evaluate and compare them?
A digital scope will be much more useful for any processor related work. Lower-end digital scopes do have limitations so it can sometimes be useful to have an analogue one available. If you really want a decent analogue scope, find a used Tek 2465 or similar, which will be way better than any of the few analogue scopes still available, but even a cheap digital will be a lot more useful than the best analogue for most of the time. the order of usefulness would be : 1) low-end Digital scope 2) low-end Digital scope plus good (i.e. used Tek) analogue scope 3) mid to high-end digital scope (deep memory, intensity display, preferably MSO) - no need for analogue once you have the fast update/intensity functionality.
On Sat, 07 Mar 2009 12:46:47 +0000, the renowned John Devereux
<john@devereux.me.uk> wrote:

>"faiyaz" <faiyaz.pw@gmail.com> writes: > >> Dear all, >> >> I am working on a hardware which has AT91SAM7S64, which works on 18.432 >> MHz main clock. This is the maximum frequency which can be seen on >> hardware. Now we are planning to buy an analog oscilloscope, we have seen >> several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ scope has >> the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS, 30 MHz has around >> 11 nS. Can you suggest what bandwidth we should go for? > >Analog scopes are great, but for debugging microcontroller hardware a >digital scope is generally more useful (if you have to choose). For >example, a digital scope is superior for debugging: > >- power on/off & reset sequencing >- code timing, latencies etc (by toggling an I/O) >- I2C, SPI, UART, ADC interfaces >- RS485, tranceiver direction switching etc > >Having said that, there are some cheap logic analyser pods that may >perhaps be used for some of these instead, I am not that familiar with >them. > >To address your actual question, it is indeed the risetime that you >should be looking at rather than the absolute frequency. If possible >you should get a scope that can measure the fastest rise/fall time in >your system. So that you can look at things like ground bounce, >switching glitches, reflection and crosstalk issues. Switching can be >very fast on modern processors, even if the external clock is not. > >Sorry I do not have specific information on the SAM7S64. I do know >that for an analog scope I would rather buy a 400MHz unit off ebay >than a new 100MHz one for the same price. For digital there are some >new decent-looking low cost ones around now. Although I think to >compete with all features of a good analog scope too, a digitial one >would still need to be a high-end model. > >> Also if there is anybody who is from India and if used Scientific or Aplab >> Oscilloscopes? If yes than can you evaluate and compare them?
There is an irritating trend to charge additional fees for digital 'scopes to enable firmare for the I2C, SPI, RS-232 etc. Best regards, Spehro Pefhany -- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com
Spehro Pefhany wrote:
<SNIP>
> > There is an irritating trend to charge additional fees for digital > 'scopes to enable firmare for the I2C, SPI, RS-232 etc. > > > Best regards, > Spehro Pefhany
So, you don't think that the extra SW effort required to convert a captured analog stream to a digital stream, then display it appropriately, should be paid for somehow? RB
On Sat, 07 Mar 2009 17:17:42 -0500, Rube Bumpkin
<Someone@somewhere.world> wrote:

>Spehro Pefhany wrote: ><SNIP> >> >> There is an irritating trend to charge additional fees for digital >> 'scopes to enable firmare for the I2C, SPI, RS-232 etc. >> >> Best regards, >> Spehro Pefhany > >So, you don't think that the extra SW effort required to convert a >captured analog stream to a digital stream, then display it >appropriately, should be paid for somehow?
It's not particularly complex, first off. In addition, the instrumentation is itself rather complex as a whole and should be sold for its features and capabilities. I've no problem with Spehro's irritation. I'm irritated that these days companies separate out a shipping charge and bundle it with a handling charge, as standard practice, and profit greatly from the handling charge part of it. In my earlier days of life, this practice was not only rare but it would have caused customers to shy away, if engaged. Somehow, now, the younger crowd has grown to know nothing else and accept it without question. 30 years back, the case where extra was charged was for heavier objects which required unusual shipping costs. Otherwise, the product price usually included the shipping and whatever handling was required. It made comparisons on price far easier. Which is why the new practice developed. Which makes it irritating. Jon
Spehro Pefhany <speffSNIP@interlogDOTyou.knowwhat> writes:

> On Sat, 07 Mar 2009 12:46:47 +0000, the renowned John Devereux > <john@devereux.me.uk> wrote: > >>"faiyaz" <faiyaz.pw@gmail.com> writes: >> >>> Dear all, >>> >>> I am working on a hardware which has AT91SAM7S64, which works on 18.432 >>> MHz main clock. This is the maximum frequency which can be seen on >>> hardware. Now we are planning to buy an analog oscilloscope, we have seen >>> several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ scope has >>> the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS, 30 MHz has around >>> 11 nS. Can you suggest what bandwidth we should go for? >> >>Analog scopes are great, but for debugging microcontroller hardware a >>digital scope is generally more useful (if you have to choose). For >>example, a digital scope is superior for debugging: >> >>- power on/off & reset sequencing >>- code timing, latencies etc (by toggling an I/O) >>- I2C, SPI, UART, ADC interfaces >>- RS485, tranceiver direction switching etc >> >>Having said that, there are some cheap logic analyser pods that may >>perhaps be used for some of these instead, I am not that familiar with >>them. >> >>To address your actual question, it is indeed the risetime that you >>should be looking at rather than the absolute frequency. If possible >>you should get a scope that can measure the fastest rise/fall time in >>your system. So that you can look at things like ground bounce, >>switching glitches, reflection and crosstalk issues. Switching can be >>very fast on modern processors, even if the external clock is not. >> >>Sorry I do not have specific information on the SAM7S64. I do know >>that for an analog scope I would rather buy a 400MHz unit off ebay >>than a new 100MHz one for the same price. For digital there are some >>new decent-looking low cost ones around now. Although I think to >>compete with all features of a good analog scope too, a digitial one >>would still need to be a high-end model. >> >>> Also if there is anybody who is from India and if used Scientific or Aplab >>> Oscilloscopes? If yes than can you evaluate and compare them? > > There is an irritating trend to charge additional fees for digital > 'scopes to enable firmare for the I2C, SPI, RS-232 etc.
Yes, I've noticed this too when looking at the newer Tek models. Yet the cheaper brands seem to include it automatically, and there are some very cheap logic analyser pods that do it too. The UK pound seems to have dropped again - Tek scopes now start at &pound;141,094,210,200 according to their web site (from here) :) And now I have momentarily clicked on the pricing section, as sure as night follows day, a Tek rep will call me on Monday... :) -- John Devereux
On Sat, 07 Mar 2009 17:14:24 -0500, Spehro Pefhany <speffSNIP@interlogDOTyou.knowwhat> wrote:

>On Sat, 07 Mar 2009 12:46:47 +0000, the renowned John Devereux ><john@devereux.me.uk> wrote: > >>"faiyaz" <faiyaz.pw@gmail.com> writes: >> >>> Dear all, >>> >>> I am working on a hardware which has AT91SAM7S64, which works on 18.432 >>> MHz main clock. This is the maximum frequency which can be seen on >>> hardware. Now we are planning to buy an analog oscilloscope, we have seen >>> several 30 MHz , 60 MHz and 100 MHz oscilloscopes. The 100 MHZ scope has >>> the least rise time - 3.6 nS, for 60 MHz it is 5-6 nS, 30 MHz has around >>> 11 nS. Can you suggest what bandwidth we should go for? >> >>Analog scopes are great, but for debugging microcontroller hardware a >>digital scope is generally more useful (if you have to choose). For >>example, a digital scope is superior for debugging: >> >>- power on/off & reset sequencing >>- code timing, latencies etc (by toggling an I/O) >>- I2C, SPI, UART, ADC interfaces >>- RS485, tranceiver direction switching etc >> >>Having said that, there are some cheap logic analyser pods that may >>perhaps be used for some of these instead, I am not that familiar with >>them. >> >>To address your actual question, it is indeed the risetime that you >>should be looking at rather than the absolute frequency. If possible >>you should get a scope that can measure the fastest rise/fall time in >>your system. So that you can look at things like ground bounce, >>switching glitches, reflection and crosstalk issues. Switching can be >>very fast on modern processors, even if the external clock is not. >> >>Sorry I do not have specific information on the SAM7S64. I do know >>that for an analog scope I would rather buy a 400MHz unit off ebay >>than a new 100MHz one for the same price. For digital there are some >>new decent-looking low cost ones around now. Although I think to >>compete with all features of a good analog scope too, a digitial one >>would still need to be a high-end model. >> >>> Also if there is anybody who is from India and if used Scientific or Aplab >>> Oscilloscopes? If yes than can you evaluate and compare them? > >There is an irritating trend to charge additional fees for digital >'scopes to enable firmare for the I2C, SPI, RS-232 etc. > > >Best regards, >Spehro Pefhany