Tektronix 422 troubleshooting the vertical section (ch 2 deflected off screen)

[nixxon]

Hi!

I just bought an old TEK 422 (1965 - 1975) from a 82 year old retired engineer that used to work at "Elektrisk Bureau". http://en.wikipedia.org/wiki/Elektrisk_Bureau

Channel 2 is not working. I can only see a faint lit-up ghost passing the screen when I turn the INTENSITY knob all the way right. This ghost looks just like channel 1 when I purposely turn CH1 vertical POSITION knob all the way, either far left or far right (off screen). Because the similar looks between CH1 when deflected off-screen and CH-2, I believe that channel 2 is deflected off-screen as well. (the scope has no beam finder)

This situation is well described in the Tektronix publication: "Troubleshooting Your Oscilloscope" (Getting down to basics) on page 47. The publication can be found here: http://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf . My problem is that I just can't recognise which section of the scope, the low-res image is regarding.

I want to "common mode" "the stages", but I dont know which section of the scope circuit the illustration "FIGURE 1" on page 47 or "FIGURE 2" on page 48 is taken from. I would be very happy if someone could please tell me which circuit this figure 1 or 2 is taken from. I need to figure out where to apply a "clip lead to short the signal input pins of the amplifier stage" (quote from troubleshooting manual page 48 regarding "IC amplifiers" and not specifically "Push-Pull" amplifiers).

Again: which "stages" in what section of the scope are to be "common moded"? What points in the circuit and on which board do I connect to see if the beam is pulled onto the screen again?
Maybe this block diagram is the one: Instruction Manual for the 422 ( http://w140.com/tek_422.pdf ):  Block diagram page 169/170 "<4> VERTICAL SWITCHING AND OUTPUT AMPLIFIER" with symmetric looking transistor pair Q224 and Q234? Maybe I should common mode the Bases of these transistors?

nixxon

[nixxon]

I common moded the Bases of transistors Q224 and Q234, and sure enough, the channel 2 trace appeard nice and clear on the screen. The trace seems to be pulled up from below when I shorted the bases of the transistors. The same happened when I shorted the output signal from "channel 2 input amplifier <2>". This output signal is the input signal to the "vertical switching & output amplifier <4>" where Q224 and Q234 is located.   

I guess the output signal from the "channel 2 input amplifier" should be around 0 Volts (relative to each other) when there is nothing connected to the CH 2 input BNC-connector. However, I read 1.2 Volts across the 2 leads (labeled "C" and "D") on the CH2 output signal.

Relative to ground the schematics says that both "C" and "D" chould read +0.97 Volts. I measured C to +0.62V and D to -0.56V relative to ground. My next step now is to figure out why "D" is negative.

The fault could be either on the "channel 2 input amplifier <2>" board or on the "vertical switching & output amplifier <4>" board. I guess I can disconnect the C and D leads from the "vertical switching & output amplifier <4>" board to figure out which board that is causing the negative voltage on D.

[nixxon]

I have measured and measured various voltages across the CH2 input amplifier board <2> and labeled them on the enclosed copy of the schematic drawing.

The major voltage deviations that require further attention are encircled in red.

On the CH2 BNC input connector, I even measured negative voltages at center pin depending on VOLTS/DIV setting (from - 0.01 Volts @ 2 VOLTS/DIV to - 0.33 Volts @ 0.05 VOLTS/DIV).

I would be very grateful if anyone could & would suggest what to check next.

Is it obvious to any skilled EE where the fault is located at??

[PA0PBZ]

Can you swap the nuvistor between the 2 channels?
Looks like the labels in the diagram are for the wafers in the wafer switch, they have a front and rear.

[nixxon]

Thank you for your suggestion. 

Can a nuvistor be pulled straight out of a socket?

I wonder from where the nuvistor "V113 8056" could get its negative voltage though. It should be connected to the following 3 voltages: +20V, 0V and +1.2V...

After looking again, maybe I should check resistor R144. Maybe it has less than the rated "576" ohm resistance to the "-12V (DEC)". The "-12V (DEC)" is also slightly high, measured to -11.83 Volts. The deviation from the rated voltages are also highest (more negative) around here (below Q144 and Q134).

I wonder what the difference is between "-12v (OEC)" and "-12V (DEC)". Anybody know this?

[nixxon]

Can you swap the nuvistor between the 2 channels?
Looks like the labels in the diagram are for the wafers in the wafer switch, they have a front and rear.

Your suggestion made me find the faulty component.

The nuvistor, "V113 8056", was indeed broken. When I swapped this with the corresponding one on the CH1 input amplifier <1> ("V13 8056"), the problem was transplanted from CH2 to CH1. Channel 2 was then OK.

According to http://w140.com/tekwiki/wiki/Nuvistor :

"As high-impedance buffers (cathode follower), Nuvistors were replaced by JFETs. As voltage amplifiers, they were mostly replaced by BJTs. Some products such as the 1A1 and 453 started production with Nuvistors and then later switched to transistors.
 
Nuvistors, like all tubes, eventually need to be replaced. They are no longer manufactured but they are still (as of 2012) available online. Another repair option is to convert the circuit to use a transistor. (...)"

And here comes the best part:

" (...) In at least one case (the 422), JFETs can be plugged directly into the Nuvistor sockets with no further circuit modification, and proper function of the scope is restored."

 

And finally, according to http://w140.com/tekwiki/wiki/422 :

"The 8056 Nuvistor used in the input amplifier of the 422 can be replaced by a MPF102 JFET"

Again - PA0PBZ - thank you very much for you suggestion!

(I still havent figured out where all that negative voltage came from though)

Happy camper,
nixxon 

Now I just have to find some MPF102's..

[PA0PBZ]

Good to see that you managed to locate the fault. After staring at the diagram and the voltages you measured the nuvistor was in my opinion the best candidate, all the other voltages made kind of sense. I also remembered that the negative voltage sometimes "just happens" when a tube goes bad, it's been a long time since I did anything with tubes so my memory is a bit faint on this but I guess it has something to do with the filament still emitting. Anyway, the MPF102 should not be very hard to get, there's lots of them on EBay. Looks like you have to buy at least 10 but for a total of $ 2.50 that is hardly a problem. Even the nuvistors are still available there, but you pay a lot more.

[amyk]

A Nuvistor teardown is also welcome.

[nixxon]

A Nuvistor teardown is also welcome.

8056 datasheet: http://tubedata.tubes.se/sheets/049/8/8056.pdf

I will dremel off the top half and take some pictures.

I expect to see something like this: http://www.thevalvepage.com/valvetek/Nuvistor/nuvistor.htm

Edit: Also good info and cutaway views here: http://tubedata.milbert.com/other/RCA/RCA_Nuvistor_Tubes.pdf

illustration, "nuvistor guts 2", explanation:
Fig. 265 - Cross section of a Nuvistor valve where:
 1) Cathode
 2) Grid
 3) Heater
 4) Plate
 5) Metal shell
 6) Ceramic base wafer
 7) Indexing lugs

[nixxon]

The following information should be of interest to anyone who wants to transistorize their Tek 422 scope.

The MPF102 JFET has 3 pins (1. Drain, 2. Source and 3. Gate)

The nuvistor 8056 has 5 protruding pins (#2 Plate(P) , #4 Grid(G), #8 Cathode(K), #10 & #12 Heater(H))

1. Drain = #2 Plate(P)
2. Source = #8 Cathode(K)
3. Gate = #4 Grid(K)

The JFET pins go into the corresponding nuvistor socket as shown on the enclosed drawing (bottom right).

[Martin.M]

there is two versions of 422, an earlier and a later.
The earlier have only one field an the front for the serial number, its the one what use Nuvis at the inputs. (and there is a eth Nuvi on the vertical board also, not only the 2 what is wired for the inputs.
The later version have a second field beside the serial No, for numbers of options. This is allready using FET.

It may be possible to convert a 422 to Fet input, but its no good idea! Look in the bay to get that nuvi and the old Tek is originally and allready working. Nuvistors are made for a very long life, that scopes also.

greetings
Martin

[nixxon]

there is two versions of 422, an earlier and a later.
The earlier have only one field an the front for the serial number, its the one what use Nuvis at the inputs. (and there is a eth Nuvi on the vertical board also, not only the 2 what is wired for the inputs.
The later version have a second field beside the serial No, for numbers of options. This is allready using FET.

It may be possible to convert a 422 to Fet input, but its no good idea! Look in the bay to get that nuvi and the old Tek is originally and allready working. Nuvistors are made for a very long life, that scopes also.

greetings
Martin

I probably have the 23rd unit of the Tektronix model 422 that was manufactured at Tektronix Guernsey (serial 100023).

Each channel has a 8056 nuvistor that is used as cathode follower input stage. I have replaced both nuvistors with MPF102 JFETs. The JFETs should preserve the high imput impedance to the input circuit and hopefully isolate the input circuit from the remaining amplifier cicuitry, just as the original nuvistors. Some say that this doesn't require any modifications at all. Yet others say that the scope will be less robust against abuse. I don't know who is right. Maybe both?

The traces are both working again on the display, though they are both a little high, even with the POSISTION knobs turned all the way down. I am trying to figure out where to make the correct vertical adjustment to bring the traces down again with POSITION knob centered. Does anybody know how to calibrate the vertical positions of the traces? Somehow I need to adjust the single ended output of the Feedback Amplifier Stages and/or balance the Paraphase Amplifier's conversion to "push-pull".

After that, I have to figure out why either trace disappears when the corresponding VOLTS/DIV knob is turned down to it's lowest values, 0.2 or 0.1.  volts/div.

[SeanB]

Offset voltages in the first stage are out, thus when you turn up the sensitivity the traces run off screen. R135 and R121 are prime points to adjust to get the offset back to zero.

[lordvader88]

I'm about to buy 1 of these thats working.

2-3 weeks time I suppose.....