The popular oscilloscope's 20 MHz BW limit ...

[BravoV]

As this number is already with so many scopes for decades, is there any history on how this limit was actually created from the 1st place ?

Also does this number still relevant on these modern days ? If not, what will be the "better" limit for today ?

[rstofer]

I think 20 MHz evolved from 10 MHz which evolved from 1 MHz which...  Nothing magic about evolution.

There is only one important specification for scopes: Bandwidth.  And channels...

The entry level Rigol DS1054Z will have a 100 MHz bandwidth with 4 channels and the newer Siglent SDS1202X-E will have 200 MHz of bandwidth (much better) but only 2 channels (hm...)

For exponentially more money you can get linearly more bandwidth once you move into professional grade scopes.

The problem is that over the years we have shifted to digital circuits.  The problem with that is the amount of bandwidth required to display even a modest square wave frequency.

Square waves consist of the sum of all the odd harmonics from DC to daylight.  A 10 MHz square wave will look kind of square with only up to the 9th harmonic (90 MHz) and 10 MHz isn't all that high.  Suppose you want to look at a 50 MHz square wave.  Same math and the 9th harmonic is at 450 MHz.

When we use a scope for sine waves, 100 MHz is a long way up there.  Sure, it's nowhere near the GHz range but still, for a hobby, it's pretty high.

Again, the most important spec is bandwidth.  Except for channels...

There are times when you want to decode a serial stream and having 4 channels works perfect for SPI.  Sure, you can use a logic analyzer but it really won't accurately show the setup time between the clock and data.  It will approximate it, and that may be good enough for low frequency but it the SPI channel is clocking up in the multiple MHz range, timing is everything.  Sometimes this can be done with just 2 channels.  In fact, I did it for many years...

I have the DS1054Z (100 MHz, 4 channels) but I sure like the looks of the SDS1202X-E (200 MHz, 2channels).  The DS1054Z is pretty much the darling of the folks around here but the Siglent should come on strong.  It is brand new so it still has a few bugs but they should have that cleaned up shortly.

[schmitt trigger]

Hmmmm.
I think the OP's question is related to how come 20 Mhz became the de facto BW limit for oscilloscopes. Am I correct BravoV?

In other words, where did this value came from? Why 20 Mhz and not 10 Mhz?

That is a good question....... I would also like to know the answer.

[vk6zgo]

I think the OP is talking about the 20MHz limits which can be switched in with 'scopes of much wider bandwidth.
AFAIK this is just a convenient compromise between a still reasonable response & rejection of the more common forms of higher frequency noise.

If the reference is to the total bandwidth of quite basic oscilloscopes, I can only point out that it isn't a "standard" at all, as  such instruments commonly have various bandwidths.
Certainly, with the very cheapest analog instruments, it is a matter of how good a response you can get without special design techniques.
My little "Digitech" analog  from the early 2000s is 3dB down at 10MHz, as is my 1970s BWD , although the latter is in most respects a better instrument.
The Telequipment "Serviscope" from the 1950s makes do with 5 MHz.

100MHz looks "easy" to many these days, but it is not.
The  1960s Tek 545B rolled off at 35 MHz, but was a sophisticated oscilloscope at the time.

[tautech]

As this number is already with so many scopes for decades, is there any history on how this limit was actually created from the 1st place ?

Also does this number still relevant on these modern days ? If not, what will be the "better" limit for today ?

I believe it is.
Most scope use is at low BW's and most often there is just no need to examine higher frequency harmonics.

High BW scopes can have multiple BW limits like the 1 GHz LeCroy gumpydoc's just got, 200 and 20 MHz.

With the EMI and RFI is the environment today there might be a case for some shifting of BW limit levels but as most newer DSO's do the BW limiting in a IC in the input stages the whole industry would need to rethink and design other IC's that might better serve the purpose.

Good question.

[EEVblog]

As this number is already with so many scopes for decades, is there any history on how this limit was actually created from the 1st place ?

Just sheer weight of models that had that bandwidth over a long period of time, and a few higher bandwidth models that started having a bandwidth limit button. Other copied that and bingo you have a defacto standard.

Also does this number still relevant on these modern days ? If not, what will be the "better" limit for today ?

No, because 20MHz is a defacto standard used in things like noise bandwidth measurement.

[BravoV]

Hmmmm.
I think the OP's question is related to how come 20 Mhz became the de facto BW limit for oscilloscopes. Am I correct BravoV?

In other words, where did this value came from? Why 20 Mhz and not 10 Mhz?

Yes, exactly.

Let me rephrase the 1st question, why and where this magic number 20 MHz pop out in the 1st place ? 

[vk6zgo]

Hmmmm.
I think the OP's question is related to how come 20 Mhz became the de facto BW limit for oscilloscopes. Am I correct BravoV?

In other words, where did this value came from? Why 20 Mhz and not 10 Mhz?

Yes, exactly.

Let me rephrase the 1st question, why and where this magic number 20 MHz pop out in the 1st place ?

As I said above,there are many lower performance oscilloscopes with different band limits, but the switchable 20MHz function probably comes from a statistical analysis of the frequency distribution of noise many years ago.
There was a time when Engineers did stuff like that instead of farting around on the Internet.
 Certainly, there are a number of loading factors used in the measurement of noise.
 
I once worked at a TV site where the vision carrier was around 64 MHz.
You could still display this on a 545B if you wanted to, but it didn't cause much noise on other stuff you looked at.
We received a nice Tek 7000 series CRO which had one vacant plugin position without a cover.
(And No!,the supplied plugins didn't,have a 20 MHz limit switch)

Compared to the older instrument, the trace was a lot thicker,& we grizzled about it.
A cover was obtained, & the noise level was reduced to an acceptable level.

[RJFreeman]

I do recall BWD Service manuals saying of their 25Mhz scopes, such as the 539D, that the aim was for them to be usable for trouble shooting 27Mhz CB radio.

[oldway]

There was a good reason for this limitation to 20 Mhz BW : the crt manufacturing technology.

A crt of simple manufacture and low cost, without post-acceleration, had a practical limit of frequency 20 to 30 Mhz.

It was possible to reach higher frequencies but the brightness was becoming too low.

If the acceleration voltage (between -1200 and -2500V) was raised to increase the brightness, the sensitivity decreased, causing another problem, that of the voltage required for the deflection of the electron beam.

The oscilloscopes of 50 Mhz and more therefore use another technology, that of the post acceleration (5 to 25Kv's) which allows to have both good luminosity and good sensitivity.

But this technology is more expensive, that' is why low-end oscilloscopes have a bandwidth limited to about 20 Mhz

[danadak]

The 20 Mhz limit predates processors, just an observation.

Still important, the limit. I think the addition of digital filter in the DSOs
much more useful, where one can set BW, and capability, like LPF, HPF,
Notch, BPF.


Regards, Dana.

[Cerebus]

Dave and Oldway, between them,  have hit this one on the head.

The physics of early CRT oscilloscopes limited the bandwidth of a 'good' scope to around 20 MHz. People started using calibrated 'scopes to measure things like broadband noise levels. The natural bandwidth limit of a 'good' AC coupled scope was about 20 Hz - 20 MHz. So when someone wanted to characterize, say, a DC power supply they'd use their scope and quote the results as something like "noise and ripple 30 mV peak to peak from 20 Hz to 20 MHz" (but they probably said 20 c/s to 20 Mc/s).

Better scopes come along and someone makes a noise measurement that comes out as "noise and ripple 70 mV peak to peak from 10 Hz to 100 MHz". A pointy headed boss asks why our noise figures are worse than the competition. The reply is "They're not worse, they're better, but ours are measured in a way that makes them look worse if you don't understand the measurement specification". PHB: "Measure them the same way as the competition so that they look better". Engineer goes off and builds a 20 Hz to 20 MHz filter.

Enough engineers do this that a smart 'scope manufacturer "brand X" realizes there's a market for a 'scope with a built in filter that makes fast modern equipment behave like slow old equipment and they pick 20 MHz as the magic number. The other scope manufacturers lose a few sales because of this and their PHBs say "copy them" and so 20 MHz gets entrenched as the standard.

[danadak]

Not exactly on scope CRT bandwidth.


Tek 519 1 Ghz in 1961 - http://w140.com/tekwiki/wiki/519


Regards, Dana.

[oldway]

Not exactly on scope CRT bandwidth.


Tek 519 1 Ghz in 1961 - http://w140.com/tekwiki/wiki/519


Regards, Dana.

20Mhz was the max bandwith for oscilloscopes WITHOUT POST ACCELERATION.....
CRT of Tektonix 519 has a post acceleration voltage of 24KV's....

[MK14]

Not exactly on scope CRT bandwidth.


Tek 519 1 Ghz in 1961 - http://w140.com/tekwiki/wiki/519


Regards, Dana.

That is AMAZING.
3 GHz in the early 1960's (you saying 1 GHz, is the FIRST version, the 519C version circa 1963 went to 3 GHz, but they are NOT 100% it was ever sold/produced).

So 1 or even 3 GHz, with the ability to view the trigger signal as well (45 nano second delay line for 1 GHz version).

But it had a number of serious limitations as well (please see wiki for details).

[BravoV]

There was a good reason for this limitation to 20 Mhz BW : the crt manufacturing technology.

A crt of simple manufacture and low cost, without post-acceleration, had a practical limit of frequency 20 to 30 Mhz.

It was possible to reach higher frequencies but the brightness was becoming too low.

If the acceleration voltage (between -1200 and -2500V) was raised to increase the brightness, the sensitivity decreased, causing another problem, that of the voltage required for the deflection of the electron beam.

The oscilloscopes of 50 Mhz and more therefore use another technology, that of the post acceleration (5 to 25Kv's) which allows to have both good luminosity and good sensitivity.

But this technology is more expensive, that' is why low-end oscilloscopes have a bandwidth limited to about 20 Mhz

So its the CRT tubes for TV, includes the booming of repair/service industry for the CRT TVs ?  CMIIW

[oldway]

The cathode ray tubes for TV's operate on a completely different principle.

The cathode-ray tubes for oscilloscopes use the principle of deflection of the electron beam by an electrostatic field, whereas in a TV cathode-ray tube, the deviation is made by a magnetic field.

In the tv's, the deviation frequency is very low, 100 or 120 Hz in vertical frequency, and of the order of 25Khz in horizontal frequency ...
Also Tv's use fixed vertical and horizontal frequencies, only brightness of the beam(s) is modulated.

The 20Mhz are related to electrostatic deflection (tubes for oscilloscopes), not to TV's tubes.

[danadak]

@oldway, thanks for the correction to my post. I should have re-read yours
before pontificating.


Regards, Dana.

[jmelson]

Not exactly on scope CRT bandwidth.


Tek 519 1 Ghz in 1961 - http://w140.com/tekwiki/wiki/519


Regards, Dana.

I had a Tek 661 sampling scope.  It had a 1 GHz bandwidth.  It used tunnel diodes in the sampler bridge, and about a 2" long "snap off line" to generate the sample aperture.  It was almost all discrete Germanium transistors, which do have very good bandwidth.  The trigger circuitry was running right on the edge of working at all, a mix of Germanium transistors and tunnel diodes.  it also had some tubes in the low-bandwidth display section of the scope.  The probes had tiny planar triode tubes in them, and were connected to the sampler inputs with RG-8 cables with GR-874 connectors.

A real-time scope with 1 GHz bandwidth is REALLY hard to do.  You need a power amplifier that can deliver GHz signals at several hundred Volts amplitude.  You also need segmented deflection plates with delay lines between segments, so the signal follows the electron beam as it progresses toward the screen.  The Tek 7104 had that capability.

The 519 had no vertical amplifier, which severely limited the applications.  It could be used to check the drive signals for electro-optic modulators, though.

Jon

[danadak]

661 One of the scopes missing in my collection. I have the 4S1 sampler and time base,
manuals, but not the mainframe.

Regards, Dana.

[David Hess]

In other words, where did this value came from? Why 20 Mhz and not 10 Mhz?

Some oscilloscopes did use a 5 or 10 MHz bandwidth limit function instead of 20 MHz.

In some cases this was simply because their maximum bandwidth was not much higher than 20 MHz so a selectable 20 MHz bandwidth limit did not make much sense and a lower one was included.  The 60 MHz Tektronix 22xx models with their 10 MHz bandwidth limit are an example of this.

Another reason a bandwidth limit lower than 20 MHz would be used is to support a sensitivity higher than the typical 2 or 5 mV/div on an otherwise higher bandwidth oscilloscope.  The 50 MHz Tektronix 2225 series combine their 5 MHz bandwidth limit with their x10 vertical magnification to achieve 500uV/div.  The 7000 series 80 MHz 7A15A does the same thing.  The 7000 series 100 MHz 7A13 vertical amplifier has a 5 MHz bandwidth limit to make maximum use of its 1mV/div and high common mode 10mV/div sensitivity. (1)

Later oscilloscopes with much higher bandwidths seem to have settled on supporting both 20 and 100 MHz bandwidth limits which I think points to the reason Dave cited.  20 MHz become the lowest common standard and the informal standard for power supply noise measurements.  Later 100 MHz did as well although to a lessor extent simply because 100 MHz oscilloscopes are so common.

We received a nice Tek 7000 series CRO which had one vacant plugin position without a cover.
(And No!,the supplied plugins didn't,have a 20 MHz limit switch)

Some of the 7000 series vertical plug-ins had a switchable bandwidth limit.  I prefer the 200 MHz 7A26 (and single channel 7A16/7A16A) over the 75 MHz 7A18 even in a slower mainframe simply because of their 20 MHz bandwidth limit and not because of their higher bandwidth.

Compared to the older instrument, the trace was a lot thicker,& we grizzled about it.
A cover was obtained, & the noise level was reduced to an acceptable level.

Depending on what oscilloscope you replaced, this may not have been do to higher bandwidth.  Like most new oscilloscopes at that time, the 7000 series used scan expansion trading better deflection sensitivity for worse spot size.  I have heard a story about one TV studio that replaced their out of production 50 MHz 547s with 7000 series mainframes and then switched back because of this.

A real-time scope with 1 GHz bandwidth is REALLY hard to do.  You need a power amplifier that can deliver GHz signals at several hundred Volts amplitude.  You also need segmented deflection plates with delay lines between segments, so the signal follows the electron beam as it progresses toward the screen.  The Tek 7104 had that capability.

It is 10s of volts for the 1 GHz 7104.  The 7104 CRT itself has about 3 times the deflection sensitivity as the 500 MHz 7904 CRT so a lower voltage higher frequency process could be used for the vertical output amplifier.

(1) The 7A13 is about 5 times noisier than a typical 100 MHz input simply because of the design necessary to support differential inputs with a wide common mode range although it is still less noisy than some modern DSOs.  Despite this it supports a higher input sensitivity then its contemporary 100 MHz amplifiers making a bandwidth limit function unusually useful.

[w2aew]

I have the *real* answer regarding the reason that 20MHz was originally chosen...

...I had the great fortune of being on the Tek campus this week, and visited the guys at vintageTEK.  I asked around, and learned where the 20MHz BW limit came from... 

It turns out that it was chosen by the designer (John Addis) of the 7A11 vertical module, in 1968 or so. John told me that during the development of this plug-in, he was getting interference in the preamp from the broadcast transmitter for TV channel 2 (50ish MHz).  So, he decided to set the optional lowpass filter to be low enough to effectively reject this interference.  Being the established market leader on scopes, other manufacturers simply copied what Tektronix had done...

True story - and I heard it from John himself. 

[BravoV]

I have the *real* answer regarding the reason that 20MHz was originally chosen...

...I had the great fortune of being on the Tek campus this week, and visited the guys at vintageTEK.  I asked around, and learned where the 20MHz BW limit came from... 

It turns out that it was chosen by the designer (John Addis) of the 7A11 vertical module, in 1968 or so. John told me that during the development of this plug-in, he was getting interference in the preamp from the broadcast transmitter for TV channel 2 (50ish MHz).  So, he decided to set the optional lowpass filter to be low enough to effectively reject this interference.  Being the established market leader on scopes, other manufacturers simply copied what Tektronix had done...

True story - and I heard it from John himself.

Wow ... Alan , thank you Sir ! Its clear as crystal now ! 

[oldway]

This does not explain why almost all manufacturers of analog oscilloscopes had essentially two product lines:
Oscilloscopes for beginners, amateurs and repairers of TV and audio with a bandwidth of 20 Mhz and another line of industrial products with a bandwidth of 100Mhz.

Tektronix had certainly some influence on the market, but I think that it is not the choice of only one designer of Tektronix on only one vertical module (7A11) that has changed all the market....

Hard to believe ...

In my opinion , there was a technical limitation with low cost CRT 's (without post-acceleration)....

[macboy]

This does not explain why almost all manufacturers of analog oscilloscopes had essentially two product lines:
Oscilloscopes for beginners, amateurs and repairers of TV and audio with a bandwidth of 20 Mhz and another line of industrial products with a bandwidth of 100Mhz.

Tektronix had certainly some influence on the market, but I think that it is not the choice of only one designer of Tektronix on only one vertical module (7A11) that has changed all the market....

Hard to believe ...

In my opinion , there was a technical limitation with low cost CRT 's (without post-acceleration)....

Perhaps that question isn't answered, but that is not the OP's original question. The OP wants to know why the "BW Limit" switch on nearly every high bandwidth scope ever made, analog and digital alike, limits bandwidth to 20 MHz instead of some other number. I got one believe that a single decision by Tek could influence the entire industry back then.