Welcome to Australia's only Vintage Radio and Television discussion forums. You are not logged in. Please log in below, apply for an account or retrieve your password.
Australian Vintage Radio Forums
  Home  ·  About Us  ·  Discussion Forums  ·  Glossary  ·  Outside Links  ·  Policies  ·  Services Directory  ·  Safety Warnings  ·  Tutorials

Special Projects

Forum home - Go back to Special Projects

 A more accurate condition tester for vintage CRTs
« Back · 1 · Next »
 Return to top of page · Post #: 1 · Written at 8:17:51 PM on 14 May 2025.
Ian Robertson's Gravatar
 Location: Belrose, NSW
 Member since 31 December 2015
 Member #: 1844
 Postcount: 2592

Fred's regular contributions have prompted me to contribute something!

Before I go to the trouble of bringing up a vintage TV to raster stage, I like to know if it's going to be worth the trouble. Do I put in the effort for a TV with a bad CRT? Or does it become a parts chassis? What I've discovered over the last few years is that conventional CRT testers can read a CRT as poor, but when connected to a chassis that same CRT produces a great picture.

The issue seems to be the linearity of the emission reading, or at least how it relates to actual tube performance.

Some experiments with a range of CRTs using different voltages and currents passed through the G1 to Cathode diode, revealed that using a voltage of about 40 with a series resistance of 33k or 39k gave a pleasingly linear relationship between the current measured and the operational brightness of the tube. This means that a strong tube will pass just on 1mA under test.

I soon got sick of setting up two power supplies, clipleads, a resistor and a meter every time I needed to do this so I built a purpose – designed self contained tester.

CRT Tester


 
 Return to top of page · Post #: 2 · Written at 11:15:08 PM on 15 May 2025.
Brad's avatar
 Administrator
 Location: Naremburn, NSW
 Member since 15 November 2005
 Member #: 1
 Postcount: 7511

Document uploaded.


‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
A valve a day keeps the transistor away...

 
 Return to top of page · Post #: 3 · Written at 7:54:16 AM on 17 May 2025.
Fred Lever's Gravatar
 Location: Toongabbie, NSW
 Member since 19 November 2015
 Member #: 1828
 Postcount: 1366

Hi Ian.
Thats what I like, something that actually works by basic principals and not full of complications!
It works on the part that matters in the CRT, the hot bit right in the base of the tube!

Dumb questions:

I guess the phosphor screens last forever ?

About estimating the heater voltage in a running set by eye?
Can you go by the colour of the heater?
Assuming you can see it, of course.

Fred.


 
 Return to top of page · Post #: 4 · Written at 8:27:05 AM on 17 May 2025.
Ian Robertson's Gravatar
 Location: Belrose, NSW
 Member since 31 December 2015
 Member #: 1844
 Postcount: 2592

Yes Fred, it was originally going to be a 40V DC supply with an adjustable buck regulator and a 2nd meter for the heater voltage. But then I found these rotary switches while going through the boxes in the garage looking for something else (as you do!)

Since Jaycar had a transformer with some useful tap voltages (even if it was 2x as big as I needed) I took the simple approach.

The phosphor does dim a little over the life of a tube, but unless there is burn-in of a stationary image it was possible to re-gun a tube without replacing the screen, which is a very interesting process. Have a look at the videos on the Early Television Foundation's website of the very last tube rebuilder operating anywhere in the world, before they closed.

Heater voltages are (or should be) tightly controlled as they have a big effect on tube life. Most mono CRTs are 6.3 volts and pull either 600 or 300mA.

As far as judging the voltage by the colour of the heater, no, not really. Gun and cathode structures vary quite a lot. Of course, if you up the heater voltage it will glow brighter, becoming yellow instead of red or orange.

Although the colour comparison method was used for getting a reading of effective heater power with EHT rectifiers. It's a bit hard to measure a narrow pulse waveform sitting 20kV above ground! So the way it was done was to set up a 2nd identical valve powered from an adjustable DC supply and compare the colours.


 
 Return to top of page · Post #: 5 · Written at 8:13:48 PM on 27 May 2025.
Ian Robertson's Gravatar
 Location: Belrose, NSW
 Member since 31 December 2015
 Member #: 1844
 Postcount: 2592

Basically, some success has been found drawing current through the grid-cathode diode. Pretty similar to what my tester does.

The following is an edited version of some posts I made on the UK site, in a long-running thread about non-destructive CRT rejuvenation.

Interestingly, while testing a few tubes that I'd replace if I could, I tried increasing the heater voltage to see what the tube could do. In 3 cases so far, I noticed an interesting effect. Here is one example, a 25" Thomas CRT in my HMV Y2:

The emission came up, as expected, to about 80% and stabilised. But then, after about 30 seconds to a minute, it suddenly jumped to 95%! This 95% reading remained when the heater was switched back to 6 volts.

Even better, after a 5 minute cooldown, switching the tester on again set to 6 volts, the emission rose rapidly to 95%. This from a tube that only got to 25% before and took 10 minutes to get there.

Plugging it back into the chassis, it looks almost as good as a new tube.

I'm wondering - does this parallel the way the cathodes were activated in new tubes?

It just seems too good to be true, or maybe just too good to last.

I'll let the Y2 sit for a week, then power it up and see if the tube still looks good.

<<<<<<<<<<<<<< 2 weeks later >>>>>>>>>>>>>>>>

Well, it's been nearly 2 weeks and I just powered up the HMV Y2-D3

Look at that!

What a pleasant surprise!

It's come straight up looking like a new tube!


 
« Back · 1 · Next »
 You need to be a member to post comments on this forum.

Sign In

Username:
Password:
 Keep me logged in.
Do not tick box on a computer with public access.