Tektronix 2465 With Ch 1, Ch 2 and Readout Vertical Issue. Odd Calibrator Glitch

[kozard]

I have attached two images from my Tektronix 2465. The power supply is known good and probes well. It is actually the power supply from my 2445 which has all new caps including the film caps.

The first image is with channel 2 attached to the calibrator output. The readout is obviously compressed and pushed up the screen.
The channel 2 trace itself is offscreen (high I believe) and can be found only with the beam finder.
Channel 1 acts exactly like Channel 2. The problem is not modified with any attenuator, coupling or timebase settings.

The second image is with channel 3 attached to the calibrator output. The readout is still obviously compressed and pushed up the screen.
At first it looks like channel 3 is working (and it basically is) but there are odd rising and falling edge glitches. I wonder if this is a clue? The power supply is good (from the working 2445).
Channel 4 acts exactly like Channel 3. If I zoom in with a shorter timescale there are glitches on both the rising and falling edges. (Not seen on my 2445 so I wonder if it is a clue as to what is wrong on this 2465.)

Note that swapping the channel switch hybrid with the 100% working 2445 scope changes nothing. The same with swapping the display sequencer.

I hope that someone recognizes this failure mode.



Following the "Vertical Troubleshooting Hints" the problem appears to be either the channel 1 and 2 preamps or the channel switch or the associated circuitry. [However, again, switching the channel switch with one from a working scope (2445) makes no difference.]

Also channel 1 and 2 have exactly the same problem so I wonder what are the chances that both (preamps) failed in the same way?



So I think something must be wrong that affects not only both the channel 1 and channel 2 preamps but also the readouts (channel 5). I wonder if it is some common surrounding circuitry on A1/main board? (But what involves the readouts, channel 1 and channel 2 but not channels 3 & 4?)

From step 2 of the "Vertical Troubleshooting Hints" I see the correct -4.6 to -5.25 volts at pin 17 of U100 and U200. (That is the input positioning voltage from the DAC on A5.) As a result I am guessing that A5 (microprocessor and analog control) and A4 (readout) are ok. But I am not 100% sure because perhaps there is some voltage reference or control signal that still affects the readout and channels 1/2 but not channels 3/4?



From step 3 of the "Vertical Troubleshooting Hints" I don't see the right voltages across the delay line output for the problem channels (1 & 2) but I do see the correct voltages across the delay line output for the working channels (3 & 4).

For a working channel (#3) I see -272mV across the delay line output when the position control is fully counter clockwise. (Service manual says -350mV.) I see +322mV across the delay line output when the position control is fully counter clockwise. (Service manual says +350mV.) When the trace is perfectly centered I see +46mV.


For a bad channel (#2) I see +515mV across the delay line output when the position control is fully counter clockwise. (Service manual says -700mV.) I see +1600mV across the delay line output when the position control is fully counter clockwise. (Service manual says +700mV.) The trace is always off screen (above the screen apparently) and this seems to make sense with the approximately 1.2V offset I am seeing between the expected delay line output and what I measure. 1.2V might be two Vbe? I wonder if that is a clue?



What could cause this? (Considering that a channel switch from a good scope makes no difference?)

How should I continue troubleshooting?

I am using the known good 2445 power supply because I have determined that the inverter of the 2465 is not starting up. Across the line one of the Rifa caps burned which was in series with a 30 Ohm carbon resistor which overheated split in two. I don't know if the failed inverter or failed Rifa has anything to do with the vertical problem that I am seeing. The photos and problem description are from using a known good 2445 power supply (both A2 & A3) in the problematic 2465. The problem is also identical if I use the known good 2445 inverter and the original (recapped) regulator board from the problematic 2465. (So I know that the 2465 regulator board is ok.)

[kozard]

Here is the solution in case it helps someone else fix their 2445 or 2465:

The problem turns out to be caused by failed transistors in both U475 and U485. [Those are on the A1 main board.] Those are CA3046 transistor arrays. Those are also one of the only circuit elements that involves Channels 1 & 2 AND the readouts (channel 5).

I found this by a combination of online reading and probing the U475 and U485. Once removed both CA3046 were verified to have failed (open) junctions. I fortunately had a tube of CA3046 from years ago.

The CA3046 are believed to be used here to fix a problem with the readouts (channel 5) in the Tektronix custom channel switch hybrid. (Something to do with leakage from one channel to another involving channels 1, 2 and 5. 5 is the readout channel.)