CRT Deflection coils

[BrunoARG]

Hello everyone. I am saving to buy an oscilloscope but a good one as I need is pretty expensive where I live so I am planning to build my own one till I can buy it.

My idea is to use an old compaq CRT screen which works perfectly and modify it to work as an oscilloscope. I will generate the horizontal sawtooth signal with an integrator or some transistor based current sources charging a capacitor.

The horizontal deflection will be controlled by the variable frequency sawtooth signal.
The vertical deflection will be controlled by the input measured signal.

My question is: Are deflection coils (or are they plates?) fast enough to achieve a bandwith of about 500KHz? What current should I drive them with? I don't want a full screen oscilloscope (the screen size is about 17" but I am OK with a 5" covered area.

And how can I control the electron gun? Should I keep it turned on all the time? I don't care what color the image is, I just want to see the waveform.

 

[dick_freebird]

It's not the coils you have to worry about, it's the drivers.
Your horizontal rate is 15.x kHz and you want to go 30X
as fast, and you want to do it after a trigger event when
the CRT is free-running with a sync picked off some source.
Lock range I doubt runs 30:1. To sweep the horizontal 30X
as fast and keeping those coils, you will need 30X the drive
voltage if inductance sets the current ramp rate (and this
is not a bad bet - CRTs are optimized for cheapness in
manufacturing).

You also need a live vertical channel where the CRT control
is rasterized.

Beware CRT damage when the beam is allowed to sit in one
place (like when you're working on the drive scheme). Try
and back off the intensity a lot, the intensity you need for
a vector display is much lower than a full screen rasterized
display.

What about the new generation of USB digital 'scopes? I
have seen them with bandwidths well exceeding what you
say you want, and at a few hundred dollars. Many such,
and lower-end but serviceable no-name / little name regular
'scopes on eBay. Might try sniping for a while and see if you
get lucky before you get good at converting CRTs.

Oh, and one hand in your pocket if the other is in the guts.

 

[chuckey]

The line deflection coils as mentioned can push the spot across the screen at a rate of 15KHZ, so with some perseverance you could push this up. The problem That I see is the vertical scan, this is designed for 50/60 HZ, so when you apply your 10 KHZ audio input to them, the vertical deflection will be zero. A compromise would be to turn the set on its side. You need to take great pains in controlling the brightness, as you will only have one line on the screen instead of 525/625 and as a result the EHT current would be very low and the line output stage might go unstable. As said, get your self a spare time job and buy a digital adapter for your PC. In the end it will be cheaper and better then modifying a TV set.
Frank

 

[BrunoARG]

As far as I know deflection coils are 3 and 6 ohm. Is the driving current so high?

How about externally adding deflection plates? Should they be very close to the electron gun? Their response will be wide faster and better because they don't need current. I could use an low noise opamp bridge amplifier to control the plates, now my only worry is the opamp speed but it can be easily modified.

 

[chuckey]

They have much greater inductive impedance at their working frequency. Adding external deflection plates is possible BUT you will need thousands of volts to drive them. Also the EHT will have to be reduced as far as possible, while maintaining a viewable spot at maximum brilliance. This will help your deflection sensitivity. It is unlikely that the display deflection will be linear.
Frank

 

[Borber]

Deflection coils for 7" CRT have 150uH and 1mH for horizontal and vertical. Horizontal freq. is 15625Hz and vertical 50Hz.

 

[BrunoARG]

How a VGA computer screen works? If it does not detect any input signal it does not display anything and the electron gun is off. Could I control the flyback transformer mannualy to get it working again, or should I generate a VGA input signal?

Could I decrease the plates voltage increasing their lenght? I think so but anyway I will try to drive them with 100V and see what happens.

 

[D.A.(Tony)Stewart]

you can compute pixel frequency based maximum pixels and frame rate. Every CRT has a different BW with better ones like Illiad and, HItachi being > 300Mhz and lower end > 150MHz.

This only applies to the X axis and not the Y or Z axis.

YOu can also research the resolution BW on all monitors.

 

[FvM]

How a VGA computer screen works? If it does not detect any input signal it does not display anything and the electron gun is off. Could I control the flyback transformer mannualy to get it working again, or should I generate a VGA input signal?

Could I decrease the plates voltage increasing their lenght? I think so but anyway I will try to drive them with 100V and see what happens.

It has been said before, but your latest question prompts me to repeat a view points in a brief:

- It' effectively impossible to make a magnetically deflected CRT work as analog oscilloscope. Main reason is the low operation frequency and limited bandwidth of the deflection system. The vertical amplifier needs to convert input voltage into coil current, hard to achieve at higher frequencies without designing new low inductance deflection coils and high current drivers.

- A VGA monitor can be turned into a digital oscilloscope if the input signal is sampled by a sufficient fast ADC and the oscilloscope image generated by a processor or dedicated digital logic. DIY digital oscilloscope projects have been published at the internet, but it's a rather complex thing. Attaching a cheap USB oscillcope to your computer is probably the easier way.