开云体育

Considering Morris's deflection estimates, driving these CRTs would have been a pretty tall order, even in the golden age of vacuum tubes. Finding some schematics of the original deflection amplifiers would be interesting indeed, but maybe it wasn't as hard as one might think, in this particular application. Flashing back to every old movie and TV show I've ever seen, depicting radar sets and displays, I think it was always a polar display.. A rotating line (vector) and a bunch of dots, showing the badness getting ever closer.

It could be that in a simple polar display, the deflection signals could be provided by step-up transformers driven by more or less "conventional" amplifiers of any technology. The amplitude and phase info may be enough, in conjunction with the Z-axis(video), so all AC-coupling is OK, and the transformer's winding isolation could allow the common-mode deflection bias to be set wherever needed.

It's just a thought - I dunno for sure. If true though, it could be mean that it's relatively easy to draw a numberless analog clock face, but a PITA to do digital characters.

Ed

For television or comperable scanning display, DC coupling is not needed, so driving an electrostatic CRT for those applications is easier. The largest electrostatic jug I have seen and worked with was an 8JP1 used in old Allen automotive test scopes. No DC coupling was needed, if I recall correctly the deflection amplifiers were 12BH7 twin triodes. These were clever oscilloscopes that automatically varied the sweep frequency as the engine under test changed speed and kept all of the? firing impulse waveforms within the width of the screen. In addition, the impulses were accurately spaced across the screen regardless of the number of cylinders. I don't know just how effective these were in repairing engines, but they put on quite a show for the customer. I have seen some TV and other consumer electronics shops put Tektronix banners on the wall above the benches to impress the few customers who knew what it was all about.

??? Bruce Gentry, KA2IVY

Thanks to you both for pointing me in the right direction. I probed the inputs to the U825 with a differential amp and found that the signal from A horizontal wasn't there. The output of the U825 multiplexer was fine when I switched to the B horizontal - nice, sharp linear sawtooth waveform. So, luckily, the U825 is good.
I traced back and found that the two white plastic brackets that are on either side of the plugin's socket that support the pins on the main board were bowed to a convex shape and loose. So, I printed up two 75mm long wedge-shaped shims and stuck them along the sides of the two plastic brackets - pushing them down into the board slot to straighten the plastic brackets and give them some rigidity.
That solved the problem. I'm going to avoid changing plugins - the sockets aren't built for a 50+ year longevity.
Thanks again, I never would have found the problem without your direction. Oh right, my 7904 said thank you as well! :)
John

Hi Albert,
thanks for your patience.

Just now I ran the test I set out to do
I disconnected the base of Q1195 isolating it from the rest of the circuit.
I placed a 100 ohm resistor between the FG502 generator and the base.
I connected the ground of the generator to the 15V unreg GND (emitter of Q1195).

After that I carefully monitored two signals: the output of FG502 and the collector of Q1195.
I started powering the +15V and carefully regulating a DC offset to the base of Q1195, starting from 0V
Q1195 started to conduct gradually and I stopped with 0.7V approx and 150mA collector current.
At the same time I increased the sinusoidal signal superimposed on the dc offset.
A weak sine wave has appeared on the collector of Q1195. Gradually increased the sine wave applied in base but the signal was still weak.
After playing a bit with the offset and amplitude of the sine wave, the best I've been able to get is a collector signal of about 3Vpp superimposed on the 15V DC component.
Increasing the AC base signal beyond a certain limit, the waveform on the collector starts to clip in the upper part (as expected) but the amplitude of the oscillation always remains around 3 Vpp
I ran the test at frequencies between 41 to 45 KHz with no differences on the collector signal. No strange things.

Practically the same behavior he had at the beginning of this whole saga.
I am completely amazed and speechless.

Max

P.S.: the test was performed with the transformer still soldered to the pcb. All other windings left open (disconnected HV rectifier etc.)

Max,

It would not be my choice to put transistors at risk this way. You have seen now that no strange things occur with 6 V pp across the 15 V collector winding. That's only about 20% of the 30 V pp during normal operation but a lot more than found in circuit earlier.
I would reconnect everything except the multiplier and the heater winding, if possible. And then monitor the voltage across the collector winding (differentially!) all the time during startup. If the voltage does never exceed 6 V pp then there was no transformer problem (and then for now it's not needed to test the transformer externally with more input power).

Albert

I have a 321 I have put in new condensers and 1% resistors. It did work after putting in the new parts then started to fail. I did find the series pass transistor leakage to keep the 10V low. That was replaced with a Si type. The B+ was then able to be correct. The two power oscillators, Q700 and Q710, get hot with lug 12 and lug 16 disconnected to have only the heater load on the secondary. I do not know if changing out the two oscillators with Si types will cure the problem. I do not know if the transformer is bad. If it helps, the S/N is 2896. I would like to get this working. Would changing the Ge transistors out with Si be best? If so, what numbers would be ideal for replacements? Would the V and H outputs be best to be a TO-5/39 type to handle the higher power?

I did notice R602 was replaced with a switch in series with a resistor in mine when I got it. The original resistance for the grat. illumination is 100 ohms at a few watts.

Thanks.

Mark

Unfortunately I don't have a suitable power amplifier to make the test Miguel suggested.

The only idea that comes to my mind is to disconnect the base of Q1195 and try to excite Q1195 using the FG502, with a suitable base resistance and exploiting its possibility to provide a DC offset capable of polarizing the bjt in the active area with the superimposed sine wave of 41/45 KHz of suitable amplitude.

Another alternative is to disconnect pins 3 and 6 of the transformer from the pcb, to connect directly the base of Q1195 to the 5.6k resistor R1192 and coupling the sinusoidal signal coming from the FG502 with a 0.1 uF capacitor to the base itself.

Max

Hi Max,

Now your ellipse results in about 100 ohm resonance resistance, in line with Miguel's result.
Time to search elsewhere for the cause of insufficient oscillation in normal use?

Albert

You don't always need to DC couple the deflection plates. You could bias them at the right voltage with a divider in the usual way and capacitively couple a suitable deflection amplifier to them. That would ease the requirement for the active devices and I'm sure there are suitable MOSFTS that could do it. However the shift controls would have to be elevated to the A2 potential. This discussion needs to be taken off the Tekscopes list though, the Google Neonixie list is more suitable for it.

Morris

THIS is the point why I am looking for old circuits... To get at least an idea how to approach it dimension-wise.

I was, at first, considering 2kV Mosfets, then pulse pentodes like 3D21 but before I go into calculating I wonder how it was done in it's time...



Am 14.12.22 um 04:10 schrieb Morris Odell:

--
Mit freundlichen Gruessen---best regards----cordialement

Dipl.-Ing. Jan Philipp Wuesten (Telefon direkt: 04882-6054551)

Frag'Jan zuerst - Ask Jan First GmbH & Co. KG
Preiler Ring 10 ; D- 25774 Lehe, Germany
Phone: +49-(0)4882-6054551 Fax : +49-(0)4882-6054552
<> (Deutsch, English, Francais)
Email <FJZ@...>
HR A4788 PI, Vertretung: Wüsten Verw.GmbH, HR B 6295 PI
Registergericht Pinneberg. GF: Jan Wüsten; UStID: DE814610403

Wichtige Informationen:
Unsere Datenschutzerkl?rung: <>
Widerrufsformular als PDF: <>
Kontakt und Bestellabwicklung: <>
Unsere vollst?ndigen AGB: <>

Not in general but we are talking quite a few volts...



Am 13.12.22 um 19:39 schrieb Martin via groups.io:

--
Mit freundlichen Gruessen---best regards----cordialement

Dipl.-Ing. Jan Philipp Wuesten (Telefon direkt: 04882-6054551)

Frag'Jan zuerst - Ask Jan First GmbH & Co. KG
Preiler Ring 10 ; D- 25774 Lehe, Germany
Phone: +49-(0)4882-6054551 Fax : +49-(0)4882-6054552
<> (Deutsch, English, Francais)
Email <FJZ@...>
HR A4788 PI, Vertretung: Wüsten Verw.GmbH, HR B 6295 PI
Registergericht Pinneberg. GF: Jan Wüsten; UStID: DE814610403

Wichtige Informationen:
Unsere Datenschutzerkl?rung: <>
Widerrufsformular als PDF: <>
Kontakt und Bestellabwicklung: <>
Unsere vollst?ndigen AGB: <>

Tubes that size are usually magnetically deflected. When you do the arithmetic for electrostatic deflection it's not hard to see why - the datasheet quotes the X deflection sensitivity as (1345/Va3) mm/volt. With a typical operating voltage of 6 kV on A3 that's 0.224 mm/volt. So to deflect the beam from the centre to the edge of the 235 mm diameter screen would take 235/(2*0.224) = 524 volts. The X deflection amplifiers would need to have a range of +/- 524 volts around something near the A2 voltage of 1.6 kV. The Y plates would need even more voltage. That's not impossible but certainly non-trivial! The power supply and amplifier design would be fun, to put it mildly.

Morris

I deleted last photo and inserted a new one : /g/TekScopes/photo/281947/3540141

There was a terrible error due to a false contact in the AC-GND-DC plugin.
I redid the measure.
Using the FG502, it has almost 10Vpp unloaded.
Loaded with the 10Rin series withe the 15 V winding the voltage drops to 6.4 V approx
The voltage at the winding now is exactly 6

The last XY photo was obtained in resonance @ 40.8 KHz, to obtain a perfect line as suggested by Miguel
Noting also that touching with a finger the 3500V winding (as I see in the video Miguel did), the line becomes practically a vertical one.

I apologize for the mistake.

Max

Hi Max,
Now you have almost 6 V pp across the winding, in contrast with the maximal obtainable 1.8 V pp earlier. Apparently something was wrong earlier or didn't you use the FG502 now? With the present result I also don't understand how you could have that fast decay in the ring test.
Albert

It seems a good transformer, at leats at these voltages. 0,05v/10ohms= 5mA at 2V = 400ohms, mine shows less 150ohms, but my secondary 1-7 has less resistance than yours too. Do you have a power amplifier, audio for example, capable to output 30v pp to excite 15v transfomer windingin? It's the only thing that comes to my mind right for now

Following Miguel's suggestions, just obtained this photo: /g/TekScopes/photo/281947/3540121

Vertical is the voltage across the 10 ohm resistor in series with the winding,
Horizontal is the voltage across the 15 V winding (pins 4-5)
Resonant frequency was 41.5 KHz.

What do i have to infer?

Max

On 13. Dec 2022, at 12:55, Jan Wuesten <fjz@...> wrote:
...I just do not want to design deflection for those myself...

Deflection seems electrostatic, that should not be very difficult, should it?

cheers
Martin

Is a VI GRAHP XY, X is V, I is Y, horizontal is voltage, vertical current, just like a curve tracer. Try to sweep frequency and measure impedance at resonance.

CV1085 = 10E/11 = VCR85 CRT pinout:
appears simple enough...

Some refs bellow !
rick

Your particular CRT seems to be pictured here { upper right corner } :




data sheet & tube dimensions here:




UK radar power tubes:




Early UK Radar:


UK AMES type 1 radar:

On Tue, Dec 13, 2022 at 03:12 PM, Albert Otten wrote:

Remember that Vpp vertical is measured across the pin 1-7 of winding, NOT

the

series 10R+winding.

True but I converted that to the 15 V winding and ended op with a very low
resistance value. IIRC the disadvantage there was that the calculations also
relied on the open generator voltage and its output impedance. Now only the
ratio of vertical and horizontal amplitude is needed as described.
You can also create an ALT display of the voltages and tune such that both
waveforms are in phase.
Albert

Oh no, it was really the 15 V winding you tested in message 196988:

Using the FG502 ....
Now I reached a maximum of 1.8 Vpp on the 15 V winding (leaving HV windings open). No strange things observed.

That results for the winding in resistance 1.8/8.2*50 ohm = 11 ohm.
Albert