Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Analog scope in X-Y mode

Status
Not open for further replies.
Hi,

The rise time is a measure of the scopes speed and can be used to calculate the time it takes for the trace to move from one position to the next, unless of course you intend to use a drive that does not take into account what the rise time actually is. Knowing the rise time allows us to predict and force the trace to go from one place to the next. The equation is:
v2=Vs*(1-e^(-t/RC))

and solving for RC as we measure the time from 10 percent to 90 percent we get a number very close to:
RC=t/2.2

Knowing this means we can calculate the time it takes for the trace to get say vertically with a drive of Vs volts. If that's not fast enough, then we can increase Vs. Let me show one example.

First solve for t:
t=-ln(1-v2/Vs)*RC

If Vs=1v and and v2=0.9 and RC=1 (just for an example) we have:
t=2.3 seconds.

Now make Vs=2v and v2=0.9 and RC=1, and we have:
t=0.598

This should illustrate how knowing the time constant helps us figure out how to get from one place to another in a given time, or how to calculate that time given the other parameters.

Settling time only comes into play when you are forced to work with a voltage that can not be changed on the fly. If that is the case however then you're stuck with that, but you dont have to accept that if you are willing to improve the design :)
 
I have messed with vector graphics, fairly recent, but it was with a tv crt repurposed.
Bright spot was a problem for me too, you need a good blanking amplifier that will switch off the beam quickly, a 'scope ought to be able to do this as you can easily display video on most 'scopes so the z input must have a b/w of around 4mc's.
I got around the issue by using another o/p on the micro for blanking the beam as well as the dac's for x/y/z, works well.
To get some ideas on blanking or even xy vectoring google amplifone monitor, they were a xy monitor popular for video games back in the 80's.

Needless to say, the popular vector graphics game was Asteroids! :D

However, looking at his picture he doesn't seem to be doing vector graphics, but a kind of 'bitmapped' system?.
 
Ok.

Yes asteroids total classic, what made the amplifone was star wars, surprisingly good for vector.
 
Hi,

To test the Z amplifier, simply apply a square wave or a pulse and 0v on the vertical and let the horizontal do its thing. Try to create small dots across the screen, both bright and dark dots. If you'd like to test at other beam angles simply input some voltage on the vertical axis.

One scope i had a long time ago was very slow on the Z axis so that limited my applications for the scope. It was single channel and i had hoped to add a second ghost channel by modulating the Z axis. It would work better with a faster Z axis control. I ended up building a fast multiplexor using analog switches :)
Used it for years but eventually gave it away.
 
Try to create small dots across the screen, both bright and dark dots
This video is full of small dots.
The last one on a character is too bright by 2X and
Any dot that if followed by a long black space is VERY BRIGHT.
I have designed many video amplifiers and can not think of a analog problem that will do this.
I am certain this is a digital problem. See above
muchas-ls-jpg.91887
 
Hi,

I was not concerned about his character generation technique. I was only addressing the subject of testing the Z axis amplifier. Testing the amplifier is much different than trying to generate characters.

As to the bright dot, maybe overshoot, maybe programming error, depending on which direction he is scanning the horizontal and vertical. I wont address this issue in detail however unless i could see the program at least.
 
maybe overshoot
not overshoot.
I have built "video cards" from before "PC" was a word. I made "CRT terminals" also known as dumb terminals. Designed CRT monitors and TVs for many companies. Done vector displays for radar. Designed video boards for computers.
maybe programming error
This is a timing error. At the end of a character the software goes off and leaves the last dot on, with out moving the dot to the next location. (When I say leaves the last the last dot on I mean; If the dot is on it leaves it on. If the last dot is off if leaves it off)
depending on which direction he is scanning the horizontal and vertical.
I have thought about this. It does not matter if the scan is Left to right or right to left or top to bottom or bottom to top. There is good evidence the scan is L to R.

Note the vertical part of the L does not have "over shoot". The last dot of the L is too bright.
Line 1 shows what happens when the dot is left on for 2 dots of time.
Line 2 shows a bright spot because the dot is on for 4X6=24 clocks.
Line 3 is very bright because the dot is on for the length of the orange line.
Line 4 is also a long line.
Line 6 and the L below it are examples if a long amount of time.
Now compare to Line 7 which is shorter.
upload_2015-4-11_9-8-31.png

If this is a simple video amp over shoot then all single dots should be too bright. If you look at the horizontal line of the L it is made up of 5 dots not a single line. The designer chose to make the video 50% on 50% off for a single dot. This way the frequency response of the video amp (Z amp) is not an issue. One dot, two dots......5 dots all have the same frequencies.
 
not overshoot.
I have built "video cards" from before "PC" was a word. I made "CRT terminals" also known as dumb terminals. Designed CRT monitors and TVs for many companies. Done vector displays for radar. Designed video boards for computers.

This is a timing error. At the end of a character the software goes off and leaves the last dot on, with out moving the dot to the next location. (When I say leaves the last the last dot on I mean; If the dot is on it leaves it on. If the last dot is off if leaves it off)

I have thought about this. It does not matter if the scan is Left to right or right to left or top to bottom or bottom to top. There is good evidence the scan is L to R.

Note the vertical part of the L does not have "over shoot". The last dot of the L is too bright.
Line 1 shows what happens when the dot is left on for 2 dots of time.
Line 2 shows a bright spot because the dot is on for 4X6=24 clocks.
Line 3 is very bright because the dot is on for the length of the orange line.
Line 4 is also a long line.
Line 6 and the L below it are examples if a long amount of time.
Now compare to Line 7 which is shorter.
View attachment 91943
If this is a simple video amp over shoot then all single dots should be too bright. If you look at the horizontal line of the L it is made up of 5 dots not a single line. The designer chose to make the video 50% on 50% off for a single dot. This way the frequency response of the video amp (Z amp) is not an issue. One dot, two dots......5 dots all have the same frequencies.

Hi Ron,

I am happy to hear that you had good success with your video cards in the past.

"This is a timing error. At the end of a character the software goes off and leaves the last dot on, with out moving the dot to the next location. (When I say leaves the last the last dot on I mean; If the dot is on it leaves it on. If the last dot is off if leaves it off)"
And where does that timing come from? The program. Thus we can in fact state that it is the program that is mostly at fault.

"I have thought about this. It does not matter if the scan is Left to right or right to left or top to bottom or bottom to top. There is good evidence the scan is L to R. "
I believe that is correct too, but overshoot could cause a brighter area in some places but that would normally be at the start of the character because after that any exponentials usually die down. This could be caused by various things, one of them an impedance mismatch. Also, when something doesnt work right, inconsistencies can appear without apparent good reason.
 
Looks like your right ron, the dots making the chars proves the hardware can do it without bright dots.
 
Looks like your right ron, the dots making the chars proves the hardware can do it without bright dots.

Hi,

I will probably agree, once the circuit is up and running perfectly. Until a circuit of any kind is working the way it is intended i never make any bets one way or the other :)
If we assume something in particular doesnt work right we can always disprove it later after it is working properly. I've learned not to jump to conclusions in the past.

But anyway, my point was really for testing the system to its capability, not testing it just to find out if it can generate characters of an already given size.
 
As long as I am pointing out things that could be changed.
Note that the left most L in each line is distorted. At first I saw the top an bottom left corner had distortion. Now that I look close I see the first L is distorted even then the L is near the center. My thought is that the X and Y amplifiers are being run at max speed (for the first item on a line). Going from L to L the amplifiers seem to be plenty fast. But going from the right edge to the left edge in only 6 clocks of time is a huge jump for the poor little X and Y amplifiers. There are several easy fixes but that is for another time.
upload_2015-4-11_11-55-57.png
 
Hola Ron,

Your analysis is correct in all counts but let me tell that the picture was taken in my very first try in 2009. As a personal challenge I decided to rewrite the whole code, which was a rather complex one (albeit worked OK).

Just by moving to RAM, the [delta X / delta Y] pairs defining where each dot goes, wrt the leftmost/bottom dot of each char, I could write a quite simple code. The flowchart is definitely a simple/straight one.

By now, the sole thing I need to correct, is the leftmost char in each line coming distorted.

But going from the right edge to the left edge in only 6 clocks of time is a huge jump for the poor little X and Y amplifiers. There are several easy fixes but that is for another time.
View attachment 91947

Could you mention now the easy fix for the remaining problem? If it is a delay, up to now I could not minimize the effect.

Bringing the clock down, brought flicker so I am back with a 16 MHz clock.

Since the beam, bright or dull is not a problem, it will stay out of the way for now. Once I buil the chars properly I will add beam control. I can make it brighter (as per user manual), make it dull or even to go from bright to dull in one step. My problem is not there.

Thanks for helping.
 
Could you mention now the easy fix for the remaining problem? If it is a delay, up to now I could not minimize the effect.
If you were doing raster scan the horizontal output would look like this, in black. (probably you aren't)
The red line is what happens when the scope is a little slow. The first character is misplaces a little because the dot did not get to the right spot yet.
Fix 1) Add some delay after a big jump. See the flat bottom in black trace two.
Fix 2) Scan Left to Right then scan from Right to Left. (Trace 3. )This way there is no big jump. I have used bidirectional scanning and had good luck.

upload_2015-4-11_18-7-30.png
 
Hola Ron

My self-esteem made me to insist in proper timing instead of going the (not discarded yet) bidirectional way.


NRR NO RASTER RON.JPG
 
Not messing with the Z axis yet.
 
Moving ahead as time permits. Beam turned off between dots.

St George.JPG
 
Hi,

Yes that's cute, and the date came out REALLY good <chuckle>
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top