开云体育

Check power supply's for voltage and ripple first.

Then follow Z axis schematic and troubleshooting

Jon

Check out power supply, CRT, HV cal procedure first.

Likely failing caps or HV resistors

Jon

Hi All,

I have been working on a 2782 unit with questionable freq. and amplitude correction data making the unit not usable at the moment.

I have managed to get a working extension and serial IO card that gives me full keyboard service command control and feedback via trace mode, this allows me to look and store via screen dumps all correction data, good or bad.

I am reluctant to take the next step using time consuming manual commands and measurement setup to try to align FM narrow/freq. and amplitude correction data.

Does anyone have any prior knowledge or has worked on a 2782/4 in a past life?

My next step is to work through the manual Cal process and document all steps and what I have achieved along with all data/sw.

Now that I have a serial card and extender card working 100%, I may produce a bunch of extender cards with integrated serial interface to be made available if there is interest.

Any feedback is welcome.

Kind regards

Gerald
VK3GM

Here’s my story, Tek 485, a ham fest purchase, it was dirty and nonfunctional. It's a pretty common issue with ham fair test equipment. After cleanup of the 485, I found it had a shorted Tant in the -15v rail, a quick fix, and it was all operational. Then a few hours later, another Tant shorted, this time it was C1748 through a small choke. The resistance was the hint, it wasn’t 0v to ground but about 0.26 ohms to ground, so I knew it wasn’t a direct to ground, this helped locate C1748.

Even though now all rails are up where they should be (I won’t list them all here but they all came up fine from +-5v to +180v). However, now there’s an issue with the Z axis. I’ll try to describe it, the trace is Z axis modulated, as in a dashed line ~!~ Strangely, the frequency of dashes appears to change a bit in accord with the intensity control. If I turn the trace intensity down the trace disappears completely. Here’s the kicker: if I touch the case (Collector) of Q1772 or Q1752 as they're common, the modulation disappears and the trace looks fine and intensity has full control. There might be other issues but the trace modulation is the first I’ve ever seen and needs to be put to bed first. How do you suppose changing the brightness changes the apparent frequency of the spacing of dashes on the trace?

I’ll post a short clip, let me know if it’s best here or in the files section.

The following items have been added to the Files area of the [email protected] group.

* /Tektronix TDS420_460 Schematics.pdf ( /g/TekScopes/files/Tektronix%20TDS420_460%20Schematics.pdf )

*By:* Tim Wright <al7ds@...>

*Description:*
Factory schematics for the Tektronix TDS420/460 Digital Oscilloscope. Said to be 100% accurate with the exception of the Power Supply (Lecroy schematic of the same third party supply)

Files put into a single PDF and uploaded to the files section.

My 2465B suffers from a miss-tracking of focus with intensity.
When the intensity is turned up (75 to 100%) I have to re-adjust focus. When I do this the readout is out of focus (because it is a lower intensity).

Is this a problem that can be corrected by the "High Drive Focus" adjustment or is it likely to be a component failure in the z-axis amplifier?

I have been working on getting the schematics for the TDS 420/460 converted
from production engineering schematics to a form which can be used without
trying to create a manual. I have no background in manual writing, just a
technical background working at Tektronix for about 30 years. I was able
to obtain the drawings from one of the engineers who designed the TDS
series, now passed on. I tried to post on TekWiki via Kurt but I am not
savvy enough to figure out the posting process. So, I have placed the
schematics on my personal ISP site for the time being.



I had sent a couple of the trigger related schematics to Tim in hopes it
would help him find his problem. Kurt at TekWiki has also been informed of
the existence of the schematics.

With the exception of the power supply, I believe the schematics to be
accurate. The power supply information is from Lecroy and has been used
with success but may not be 100% accurate.

Best regards,

Ferrous Steinka

My caps were backwards as well. Looking at W140.com I see that they were added at SN 6040.

Hello Chris,
I am afraid there is no reasonable way how to rewind the transformer. It is very small (just 3x3x3cm), has 6 independent windings and broken one is closest to the core. It is also completely potted by some glue so probably the only way how to disassemble is by acetone and it will likely dissolve everything except ferrite core and wires.

Here is the photo of that transformer: /g/TekScopes/photo/300091/3876209?p=Created%2C%2C%2C20%2C2%2C0%2C0 , see the size comparison with RJ45 plug.

If I am not able to find the replacement, I am considering alternative way to build separate small power supply to be put inside of scope. It needs to provide 12V and galvanically separated +/-15V.

Vladimir

It's quite hard to believe that every 130LC was drawn and assembled incorrectly, and not picked up during what appears to be one ofTek's longest runs of production.

I do recall a piece of Marconi test gear where a similar arrangement was near the (single-ended and not centre-zero) meter and the note to the schematic said "polarity of capacitor to be chosen during calibration".

To my mind 10-12V DC pulses are too high to be normally tolerable when they could be avoided simply.

Noted re HT on the meter - that became clear looking at the schematic: we're measuring small deltas riding on a large baseline value. Bit like measuring DC on the mains...

I wonder if 15V back-to-back Zeners might provide enough protection? I've used this on an input 'protector' for an AF sig gen and/or curve tracer I didn't want to blow up poking around in valve gear.

Replying by memory, I believe it is correct.? In most cases, if the fuse that blows is the .250 mA fuse, check the tantalum caps nearby. The culprit mostly will be the HV block. There was another thread where I wrote some in detail.

I don't recall for sure, but on my two 130s I think those caps were also reversed from the "proper" polarity, but it didn't seem to make much difference to operation. The net DC voltage on the meter movement is pretty small, although the AC applied to the meter circuit is fairly large, low frequency narrow pulses when it's near the bottom of the scale. You can look with a scope and see what's going on, then decide appropriate polarity - maybe backwards is better - I don't remember which way I picked.

I decided to put in about five to ten times more C for filtering to get less bouncing around and easier reading at low values below about 5% of FS. The tradeoff is overall slower response, but no hurry with something like this. I used good hermetic Ta caps.

BTW be very careful in poking around in the meter circuit. The meter floats on the B+ and the signal operates around it. The series R to the B+ there limits the current to a low average value, but the B+ filter capacitor stores enough energy to likely easily blow the movement's coil if anything is inadvertently shorted to ground. I had planned to modify the circuit for meter protection but didn't do it yet. The apparent easy solution of adding clamp diodes does not work here because of the large (around 10-12V I think) pulse voltage applied then averaged by the range filter caps. I did that first and it totally messed up the readings, until I figured out what was going on.

Ed

Hi,
not what you want to hear, but have you considered re-winding it ? Either on the same or a similar former. If you're very lucky, the primary winding might be the outer one, or a separate former ? (can't find my photos ATM). If you don't feel like counting turns, and if you still have your friend's T3 you could determine the turns ratio. Not a magnetics pro here but I assume with turns ratio, inductance, operating frequency and guestimating the current loads (should be fairly low) you can probably get pretty close ?

Chris

Bad SMD? Not likely due to age. However, for the long term every board should be pulled (only one or two at a time) and inspected anyway, see. See /g/TekScopes/photo/285143/3590895? Chemical hour meter will be leaking if still installed. Then the power supply needs attention, x2 y2 caps etc. on mains. I refer you to John Miles' notes which has detailed support info. Manuals are easily found by Google. Units are very maintainable too with parts available.

Bummer. It was worth trying.

On Wed, Jan 15, 2025, 11:02 AM Tim Wright via groups.io <al7ds=
[email protected]> wrote:

Just checked, and it does not. Different ACQ board altogether. I have a
540, and it is similar to that. Makes sense, same series.

Just checked, and it does not. Different ACQ board altogether. I have a 540, and it is similar to that. Makes sense, same series.

Same here, my caps were backwards. I also replaced a couple of the film caps there because they were DiFilm with high dielectric absorption which warped the low-scale reading. I found that polyester was too good, and ended up adding RC’s to simulate a selected amount of DA. That might be taking it too far but it was so satisfying to run the problem to ground at last.

Dave Wise

There are many ways to approach it. You should be able to do this with a 7904 + 7B92A in delayed trigger mode.

You'll need to display a complete frame with the main sweep and display the scan lines with the delayed trigger sweep. Just a few lines should do nicely.
You can input the RGBHV lines to the vertical channels as needed for viewing and selecting pulses of interest.

IIRC, you can put VSYNC directly into the main trigger EXT input and HSYNC into the delayed trigger EXT input and use both at the same time.
That might give you better triggering depending of the operational status of your 7B92A.

Play with it and see. I've used similar setups to watch live video on my 7904A.

The TDS520 component level service manual is available, not sure if it
applies to the 420.
Radu.

On Wed, Jan 15, 2025, 9:36 AM Tim Wright via groups.io <al7ds=
[email protected]> wrote:

All SMD caps replaced, board cleaned and inspected, via's reflowed if
showing corrosion signs. Still getting trigComparatorTest FAIL, ERROR ID
151 diagnostic test failure trigComparatorTest, TRIGA status after trigger:
exp = 1, act =0. ** alpha trigger (XTRIG), on the RS232 Debug port.
This tells me that there is still an open via somewhere. Since schematics
appear to be un-obtanium, I am wondering if anybody with experience on
these scopes can at least identify the part of the ACQ board I should look
at closer under the microscope. What chips handle the triggering? should
be close to the EXT trigger BNC connection on the PC, Yes? Only reason I
am fooling around with repairs on this "Antique" (Not so antique as my 511A
though) is for the challenge. It has become a big one. It would be great
if someone, ex-Tek repair tech, had the schematics on these things, but I
am not holding my breath.