• Please review our updated Terms and Rules here

20 MHz, 2 channel O-scope

ziloo

Veteran Member
Joined
Feb 7, 2006
Messages
990
Location
in the basement
Hello Folks,

I have been offered a 20 MHz analog O-scope by a friend;
What would be the use of such a device for vintage electronics,
or micro-controller circuits? 20 MHz sounds way too slow 🐌🐢.

ziloo :)
 
If you have a 2 MHz (or slower) machine it would be fine...

Of course, some machines still have higher clock frequencies for things like VDU circuits.

Dave
 
It's only recently that I've retired my 80's era 20 mhz Beckman Industrial scope in favor of a 90's 100 mhz Tekscope. 20 Mhz is fine for 8 bit systems, anything up to 10 Mhz. Main reasons I upgraded were size, power consumption, and not wanting to clean the switches.
 
Why only 10 MHz? Would it be adequate for a 16 MHz Arduino board 🐰
which is still lower than 20 MHz?

I think that recommendation has something to do with Nyquist sampling, IE, to really be sure of capturing an event of X duration "accurately" your sampling rate should be at least twice as fast as X. It's probably a decent rule of thumb, but strictly speaking it applies to *digital* sampling, and isn't really accurate when applied to an old analog scope.

Most of what you do with an Arduino is going to involve pins changing state a lot slower than 16mhz, so an old 20mhz scope should be perfectly fine. Even with assembly language I don't think you can trigger a pin on a 16mhz Atmega8 Arduino at faster than 8mhz, either with bitbanging or using the UART hardware.
 
Why only 10 MHz? Would it be adequate for a 16 MHz Arduino board 🐰
which is still lower than 20 MHz?

ziloo :)
You can mostly see what's going on over 10 mhz, but you start losing detail. I've used the 20 mhz scope on a 25 mhz signal and I could see it was there, but any kind of glitches would have been lost.
 
Last edited:
The bandwidth of an analogue oscilloscope is defined to be the point at which the voltage measurement has fallen by 3 dB (or approximately 70% of what it should be). That is one heck of an error!

Dave
 
I think the point remains that for novice/hobby electronics and (most hobby applications of) microcontrollers a 20mhz scope is going to be plenty useful. As I noted there's basically nothing an Arduino can do faster than 8mhz, and that's going flat out with assembly language code. If you're using the platform-independent "pin" constructs in Arduino C the maximum pin switching speed is somewhere between one and two orders of magnitude slower so resolution falloff after 10mhz is a complete non-issue.

If we're talking about troubleshooting a vintage computer, well, it should be fine for just about any "home computer" that used a television-frequency (NTSC/PAL) monitor, IE, anything from the 1970's through middle 80's. The fastest clocked thing you'll find in those will be the pixel clock and that's going to top out around 16mhz or so. (Also it's pretty rare that you even really need the scope to "see" that fine of detail if you're troubleshooting the video section; as long as you can see that there's "activity" during the active portion you're usually only really needing enough resolution to be able to troubleshoot the h/vsync signals. Even a 1mhz scope could give you a "usable" picture of those. A professional TV repairman back in the 1950's might have a scope with a resolution in the ballpark of half that.) The other parts of a machine like that which might run at "very high" frequencies would be parts like DRAM refresh signals, and honestly if you're really deep-diving into that you'll probably be itching to go for a logic analyzer instead.

I finally gave in this year and asked for a Siglent 200mhz digital scope for Xmas because working on incorporating interlace support into my perpetually stalled homebrew video card project made me wish for a sample/capture buffer to do really fine-grain timing analysis I can't quite figure out how to manage with the old analog scope, but unless you're doing that kind of stuff, again, a free 20mhz scope is way, way better than no scope at all.
 
The bigger problem with an analog scope is that it isn't a storage scope. It is good at showing repeating pattern signals, but not so good at capturing one off events. That is where a DSO will be much handier.

All that said, I've got an old BK 2120 on my bench because it is just plain fun to mess around with from time to time.
 
The bigger problem with an analog scope is that it isn't a storage scope. It is good at showing repeating pattern signals, but not so good at capturing one off events. That is where a DSO will be much handier.

All that said, I've got an old BK 2120 on my bench because it is just plain fun to mess around with from time to time.
Well that is not entirely correct.

It might depend on how much you know about analog storage scope technology.

Initially the way to do it was with a long persistence phosphor. The classic one was invented in the USA for radar in WW2. It goes by the suffix "P7". in this case they mixed a long yellow and a short persistence blue phosphor on the CRT face.

The typical CRT used in American radar systems of the WW2 era was the 5FP7. Later this tube was used in ECG monitors in Cardiac ICU's in hospitals. To filter out the short blue, they often put an acrylic orange filter over the CRT faceplate. So you will see these cardiac monitors, with orange filters, to capture slow events, like the electrocardiogram, in old medical TV dramas like Marcus Welby MD.

Also, in the UK in the 1960's, the Telequipment Company, made the D52 scope. A dual trace scope. It also came with an option for a tube with the equivalent slow P7 phosphor (although it had a European designation). I have one of these scopes, photo attached. I have used this scope in medical research, work involving slow events. It is a 10MHz scope.

Later, Tektronix saw the need for an analog scope that could store slow events, so they came up with a different idea. The storage tubes in scopes like the 464 or 466 model, 100MHz scopes from the late 1970's to 1980's era. I have these scopes too and they are great for storing long term events, and they are analog. They have moderately complex electronics to support the storage function. I have used this scope on many an occasion to record slow events. But it requires a screen photograph to "save it forever"

But, it is true that they don't have the "permanent" storage result of a digital file, of the digital scope, but, for most cases, they work for your investigations.

I am a little bit of a scope lunatic , I think I have at least 15 or 20 different scopes. So I am well aware of their uses and limitations. But like any instrument in a laboratory, the best one is the thing between the ears. You need to understand the test instruments, to get the best out of them.

(One interesting thing; the Telequipment scopes, if you look very carefully you will see, these scopes were used in the Thunderbird Super-Marionation TV shows from the UK, on the control panels of the various craft and if they were good enough for Virgil, Scott, Alan and Mr. Tracey, they were certainly good enough for me, oops I forgot Parker & Lady Penelope)

On another note, I recently have seen an interesting method of permanent waveform recording from a dual trace analog CRT. An unexposed film was run across the CRT faceplate and the beam exposed the film. This is how the designers of the World's first Satellite, Sputnik 1, recorded the waveforms, back in 1957, of the "manipulator" which switched the two transmitters in that spacecraft alternately on and off. It created a permanent record on film, possibly even longer lasting into our future than a file from a digital scope stored in a flash memory. It might pay to remember, that flash memory, in the long run, is forgettory and it would rarely last longer than books or film.

For your interest I have attached an image showing some of the manipulator recordings from Sputnik 1.
 

Attachments

  • D52.jpg
    D52.jpg
    107.1 KB · Views: 6
  • Tek464.jpg
    Tek464.jpg
    142.1 KB · Views: 6
  • errors.jpg
    errors.jpg
    219.7 KB · Views: 6
Last edited:
Also, I have attached a very clever photographic recording from a CRT face, created in the 1936 era by Manfred Von Ardenne ( an early televsion researcher in Germany). He created it to visually demonstrate how the TV scanning raster was formed on a CRT's face. It was an interesting method and his permanent recording remains, to this day, in print. I have posted this image on the net before, so "the Cat is already out of the bag" on how he did it. If you don't know it is a bit of a brain teaser.
 

Attachments

  • rastaform.jpg
    rastaform.jpg
    201.9 KB · Views: 9
Amazing discussion.........Thunderbirds are go Hugo!!!! ;)
I am so glad I asked and for further inquiries:


Thanks so much to everybody for your comments !!!!

ziloo 😊
 
Last edited:
Amazing discussion.........Thunderbirds are go Hugo!!!! ;)
I am so glad I asked and for further inquiries:


Thanks so much to everybody for your comments !!!!

ziloo 😊

While we are on the topic of Thunderbirds, the Hero of the show, in my opinion, was Brains. But he was somewhat socially awkward (not unlike myself). For example, despite the fact he was able to design and build all of the Thunderbird rockets and the entire Tracy Island, he struggled with simple things, on one episode he said: " I h..have f..f..finally d..d..done it Mr. Tracy, I have finally made a Martini".

Later, somebody came up with the idea of a much more confident and outgoing Brains and created what I think is a brilliant music video:

 
Last edited:
While a 20 MHz analog scope would be useful up to about 10 MHz, the same can't be said for a digital scope. It is true that digital scope have the value of persistence of single sweep signals, analog scopes seem to take less fiddling. Most inexpensive digital scope don't sample slow sweep rates at the full rate and can suffer aliasing when used out of range. The analog scope has the advantage there when looking at a cat's eye pattern for aligning the tracking of a floppy drive. One can clearly see the 1 MHz output of the drive when the full sweep is looking at a 60 Hz repetition. One does still have to be more creative when looking at short single pulses on an analog scope, as compared to a digital. I prefer the combination of a logic analyzer and analog scope but each to their own. Eight channels of a logic analyzer out gun a digital scope in most applications.
Dwight
 
While we are on the topic of Thunderbirds,
The husband & wife team of Andersons created some of the best sci-fi series of
the sixties and seventies. The elaborate sets and very detailed machines and props
they designed and built were phenomenal. When I read movie reviews about today's
highly computerized CGI and special effects, there are so many even young generation
audience that express their saturation with the mindless stories and over the top fast
paced action scenes.

I prefer the combination of a logic analyzer and analog scope but each to their own. Eight channels of a logic analyzer out gun a digital scope in most applications.
And I remember you had developed some other techniques of troubleshooting
using software that one of these days you should write a blog about. ;)

ziloo 😊
 
Last edited:
The elaborate sets and very detailed machines and props
they designed and built were phenomenal. When I read movie reviews about today's
highly computerized CGI and special effects, there are so many even young generation
audience that express their saturation with the mindless stories and over the top fast
paced action scenes.


ziloo 😊

The same thing applies to the Original Planet of the Ape's movie from 1968, which was beyond brilliant, all masks and makeup, no CGI. It is impossible to beat it with the new CGI versions, along with the original story too.

Get your stinking paws off me, you damned dirty ape!

One thing I liked about that movie, it forced people to look at the way we treat what we foolishly believe to be "dumb Animals".
 
Last edited:
The husband & wife team of Andersons created some of the best sci-fi series of
the sixties and seventies. The elaborate sets and very detailed machines and props
they designed and built were phenomenal. When I read movie reviews about today's
highly computerized CGI and special effects, there are so many even young generation
audience that express their saturation with the mindless stories and over the top fast
paced action scenes.


And I remember you had developed some other techniques of troubleshooting
using software that one of these days you should write a blog about. ;)

ziloo 😊
If the machine has anyway of running code, I always do that first. If it can take an EPROM I write a small piece of code to check that it can at least run a small loop. I build up from there. I made some small replacement modules for the KIM 1, 6530 chips. For that I made an external ROM board to put diagnostics on. It had a large EPROM with address range selects from some slide switches. each section of the EPROM had a specific test. The first section tested that it could run code and blink a light. The next section tested the RAM on the KIM. A 6502 is almost useless without page 0 and page 1 RAM. That was tricky to do because it needed to run in the 6502 without using any RAM. Again, it reported failures with the blinking light. For this, I used every register in the 6502, including the stack pointer.
Anyway, it is all based on testing one particular part at a time. One needs to keep the test tightly dedicated to one narrow objective. Trying to do too much in each test is a bad idea. Each test is a very pointer test to do just one thing. Being able to repeat the test in a loop is a good idea. There is a difference between a diagnostic test and a confidence test. A diagnostic test just does the same thing over and over. A confidence test does a sequence of multiple things.
I now back to working on my Nicolet 1080 ( actually a Lab80 ). It has a front panel so I can manually enter diagnostic loops.
 
While we are on the topic of Thunderbirds, the Hero of the show, in my opinion, was Brains. But he was somewhat socially awkward (not unlike myself). For example, despite the fact he was able to design and build all of the Thunderbird rockets and the entire Tracy Island, he struggled with simple things, on one episode he said: " I h..have f..f..finally d..d..done it Mr. Tracy, I have finally made a Martini".

Later, somebody came up with the idea of a much more confident and outgoing Brains and created what I think is a brilliant music video:

I have to say I never watched the Thunderbirds but I do remember Fireball XL5. One of my favorite shows as a kid. Created by those same Andersons.
 
If the machine has anyway of running code, I always do that first............ A 6502 is almost useless without page 0 and page 1 RAM. That was tricky to do because it needed to run in the 6502 without using any RAM. Again, it reported failures with the blinking light. For this, I used every register in the 6502, including the stack pointer........
I vaguely remember you mentioned something about using
"Forth" language (or I may be imagining things 😵‍💫 ) for some of
your troubleshooting. Would you please elaborate on that?

ziloo 😊
 
Back
Top