A real head-scratcher

Want to know how something works? Harness the power of the Brainiacs!
Post Reply
kcbooboo
Posts: 452
Joined: Sat Dec 10, 2011 4:53 pm
Location: East of the Mississippi

A real head-scratcher

Post by kcbooboo » Fri May 12, 2017 2:09 pm

I'm trying to repair an old Motorola R2001B communications service monitor. It's got a CRT in it that can display text, swept traces, or a mix of both. The text is barely visible while the swept traces are nice and bright. The CRT has deflection plates and works essentially like one does on an ordinary oscilloscope. They position the beam and feed intensity modulation to the grid to make the text dots appear or not, sort of like an old dot-matrix printer.

The intensity modulation is a 0-35V signal that's coupled to the grid of the CRT through a 0.01uF cap. The amplitude increases as the front panel Intensity pot is rotated CW, which also increases the brightness of the display.

Here's the rub. I measured the voltages on the CRT with a Fluke 80K-6 high-voltage probe and Fluke 189 DMM. The cathode is -2,000V. The control grid is -2,018V at minimum intensity and -2,060V at maximum intensity. The focus grid is -1,860V. The 2nd anode is +4,115V.

My vacuum tube theory, at least for a triode, tells me that the more negative the grid, the less the tube should conduct, and the more positive it is, relative to the cathode, the more it should conduct. Does this also apply to the electron beam coming out of a CRT? This unit gets dimmer and actually has NO display when the grid is closer to the cathode voltage, and has a brighter display when the grid is 50-60V more negative than the cathode. This doesn't seem to make sense to me.

From what I can tell, as the intensity modulation signal goes positive, it causes the CRT to get brighter. I'm basing this on the fact that this signal has short narrow pulses on it for the text line on a mixed-mode display, and sits at +35V for the swept trace part of the display. But the intensity control on the front panel seems to contradict this action. NOTE: I may be in error about the intensity modulation polarity; it may be that the positive level dims the display, not brightens it. This has yet to be officially determined.

I've been scratching my head on this for weeks and figure I should share the aggravation with others. Is my theory at fault?

Bob M.

COMMENG
Posts: 95
Joined: Wed Dec 18, 2013 4:30 pm
Location: Midwest

Re: A real head-scratcher

Post by COMMENG » Wed May 17, 2017 2:09 pm

It sounds as if the Character Generator Z-axis output isn't properly modulating the cathode.

Somewhere between the Processor Board's Character Generator output and the Scope Amplifier board there should be an ORing function circuit that ORs the character generator output with other signals to modulate the Z-axis of the scope tube, the SCOPE Z-AXIS signal.


COMMENG

kcbooboo
Posts: 452
Joined: Sat Dec 10, 2011 4:53 pm
Location: East of the Mississippi

Re: A real head-scratcher

Post by kcbooboo » Thu May 18, 2017 4:43 am

There is a circuit with several "AND" and "OR" gates that does mix a blanking signal with the character generator dot signal. All of that is done at TTL levels and both signals are solid 0-5V levels, nothing inbetween. This modulation signal gets amplified through one stage that controls the maximum positive level as a function of the front panel Intensity pot. When the pot is set fully CCW, no Z-Axis modulation is present and the CRT is dark. At about 40% rotation, the signal goes up to about 35VDC for the positive level and to ground for the negative level. There's a separate trim-pot to synch the intensity control voltage and the modulation amplitude. The modulation goes through a 0.01uF 3kv cap to the grid of the CRT. The Intensity control operates an opto-isolator that runs a PNP transistor that lets the grid voltage go more toward the -2100V supply as the intensity is increased. The cathode is fixed at -2000V. So everything appears to be doing its thing.

That 0.01uF cap measures 0.011uF. That's from the board that feeds it all the way through to the pin on the CRT tube socket.

I'm still confused in that the grid going further negative than the cathode increases the intensity, and as it gets closer to the cathode voltage the intensity decreases. Perhaps it's just the way CRTs operate, in that the control grid is a cylinder around the cathode with a hole at the front that lets the electron beam through, unlike an ordinary tube where the grid is a fine wire wound around the cathode.

I've got a feeling that the tube is just very low on emission and is on the way out, but I'd still like to verify how the intensity system operates.

Bob M.

COMMENG
Posts: 95
Joined: Wed Dec 18, 2013 4:30 pm
Location: Midwest

Re: A real head-scratcher

Post by COMMENG » Thu May 18, 2017 3:44 pm

What is the CRT's industry tube number?


COMMENG

kcbooboo
Posts: 452
Joined: Sat Dec 10, 2011 4:53 pm
Location: East of the Mississippi

Re: A real head-scratcher

Post by kcbooboo » Fri May 19, 2017 3:39 am

Nothing I can identify. The "P/N" numbers in white ink are Motorola part numbers. Nothing else seems like a number I can relate to.

Bob M.
Attachments
crt-labels.jpg
R2000-series CRT labels

COMMENG
Posts: 95
Joined: Wed Dec 18, 2013 4:30 pm
Location: Midwest

Re: A real head-scratcher

Post by COMMENG » Fri May 26, 2017 2:16 pm

Well, here is what we know so far:

The CRO is an AEG/Telefunken 140mm (diagonal measurement) scope tube. "D" is single trace, direct view, "GH" is a P31 screen phosphor; 521006 is the neck connection pin out configuration (not common), and "390" is an internal series designation which some believe is the gun assembly length in mm.

A total potential of 6.25kV exists between the cathode and screen coating.
On the HV Board, the top rail of the Intensity Modulation circuit is -2,013V, while the bottom rail is at -2,100V, as determined by VR1.

A CW movement of the front panel Intensity control results in a more negative voltage at opto-isolator U2's photodiode, increasing conduction in phototransistor in U2 which brings the top of R14 more positive. This forces the pnp Q2 to conduct less which allows the top of R15 to become more negative, which also forces the CRO's grid more negative.

The voltage at the top of R15 and collector of Q2 is a nominal DC voltage of -2.011kV. The CRT-Z Axis signal is capacitive coupled to R16, so the voltage at this junction is a DC voltage with an "informational" plus blanking pedestal riding on it.

I don’t know the exact P-P voltage of the informational plus blanking signal, but for a 35V P-P signal at +17.5 volt (half of that) signal would cause the grid to go to -1994V and which would bias the tube into conduction for spot display.

Also, there is a lot of interaction between the various voltages as seen in the voltage sampling voltage dividers which are used for comparative purposes.


COMMENG

kcbooboo
Posts: 452
Joined: Sat Dec 10, 2011 4:53 pm
Location: East of the Mississippi

Re: A real head-scratcher

Post by kcbooboo » Sat May 27, 2017 5:23 am

I concur with your analysis, but why does the grid voltage seem to go further negative (closer to -2100) when the intensity is turned CW, making the trace brighter? I thought noving the grid further negative would cut off the electron flow more, like it does in a triode or tetrode, reducing plate current or even cutting it off, yet this CRT seems to work in reverse. That's my big concern.

The overall problem, intensity pot rotation ignored for now, is that the scope traces look fine, but the data that's painted on the CRT is barely readable. Even on a mixed-mode display (spectrum analyzer has the center freq in text at the top, the normal SA trace for the rest of the screen), the trace is bright, the text can't be seen.

It almost acts as if there's a huge capacitor on the grid to either ground or the cathode, and while that lets the low-freq blanking signal go through just fine (and swept traces look great), it swamps out the high-freq data pulses (character text). The 35Vp-p Z-Axis Modulation signal was measured on the board that drives the 0.01uF coupling (to the grid) cap. I can't accurately view the signals on the CRT base with the Fluke HV probe as it's only rated to maybe 1 kHz. These pulses are in the microsecond-wide range. The manual claims the low-freq intensity goes through the opto, while the high-freq data goes through the coupling cap. The scope traces look great through all horizontal sweep ranges.

I can't say what's non-standard about the base pinout but the rest of the tube number seems to be quite useful. Where did you find this information? Thanks for providing it.

Happy Memorial Day.

Bob M.

kcbooboo
Posts: 452
Joined: Sat Dec 10, 2011 4:53 pm
Location: East of the Mississippi

Re: A real head-scratcher

Post by kcbooboo » Thu Jul 27, 2017 10:32 am

There is joy in Mudville today.

I purchased a Sencore CR70 CRT tester/rejuvenator for $65 shipped. It looks brand new. Came with all the adapters, including the "universal adapter" (a pod with EZ-clips at the end of five wires) and has the setup book. It passed the field tests. I subsequently bought a CD for $15 with the service, setup, and operating manuals on it.

I connected it to my CRT per the book, that actually had an entry for D16-390GH, and it showed zero emission using both pin 4 (Focus grid) and pin 7 (Accelerator grid) as the G2 connection point. I went through the rejuvenate and restore cycles and it still showed zero emission using pin 4, but had loads of emission when I used pin 7.

I put the CRT back into my service monitor and it looks beautiful. Plenty of brightness, sharp, focused, and no blooming. The intensity increases throughout its range as I rotate the pot clockwise. I would not have thought this tube was repairable. Now that I can read the screen, I can do a complete alignment.

Bob M.

Post Reply