Topics

Tektronix 2467B Auto Measure Faulty?

Leandro Lindemann
 

Hello everyone.

I have a 2467B that I bought a few years ago, and I never used the automatic measurement functions.

Today I decided to test this and I think the scope is defective in this sector.

When using my signal generator with a 1VPP sine wave in Ch1 with the generator providing a frequency of 100Khz with the CH1 selected for AC coupling (using the original P6137 probes) and pressing the Step / Auto key the displayed waveform in CRT it occupies 4 divisions (ok) but the sinusoid visualized is not stable on the screen (the time base has not been adjusted, I need to adjust it manually).

When using the Measure option and choosing the Volts option, the voltage voltage is almost correct (1.953V), but the waveform displayed on the screen is very large, at the bottom of the screen it informed 100mV, and using the Volts / div knob of channel 1, I need to adjust it to 500mV so that the sinusoid is displayed on the screen, located at the top of the CRT. In this situation the sinusoid also has the wrong time base, I need to manually adjust the SEC / DIV switch to view the sinusoid.

When I press the Measure key and select the FREQ option, I always receive the message. SMALL OR LOW REP RATE SIGNAL.

The measurements using the time and frequency cursor manually are correct.

I did the self test and no error was indicated.

Is there something I am doing wrong or am I really defective in scope?

Sorry for the bad English, I'm using Google Translate.

Leandro Lindemann
 

I believe that my scope really has some problem:

After performing the Auto DC Balance Routine procedure, a problem occurred:

I adjusted the trace to the center of the screen, with the volts / div switch set to 1V and the coupling in GND. When selecting the voltage of 500mV or 200mV, the trace left the center of the screen, moved 2 divisions! This has never happened before. To "resolve" this, I made a backup of Nvram Dalas 1225, which I had done when I bought the scopre and did a preventive maintenance, which includes the replacement of all the electrolytic capacitors of the source and also, of the capacitors that usually leak on the board A5. With the Backup performed, the trace is again in the center of the screen, regardless of the voltage selected in the Volts / Div switch.

I am very grateful if someone can guide me, because alone I do not have the knowledge to check what is happening.

Thanks.

Jean-Paul
 

Leandro, did you do a complete calibration? What was the cal status before backup of NVRAM?

Did you do a DC bal?

Your description points to lack of CAL.

Do you have the required CAL generators?

Ciao

Jean-Paul

Leandro Lindemann
 

Thank you for your help.

I'll try to explain it better:

When I bought the scope, when connecting it to 220V (voltage from the mains network) the Rifa capacitors exploded. That was 2 years ago. At that time, I ordered all the electrolytic capacitors + SMD + Nvram from Digikey. I changed both, and before changing the Nvram I did the data backup.

After that, the scope worked perfectly, with no calibration error indicated in the service mode and with all the automatic measurement functions operating perfectly.

Yesterday, when doing a medication for a waveform, I decided to press the Auto key and realized the problem. At this time, I accessed the service mode, no errors reported.

I got the instruction manual and after reading the question of DC Auto Balance, I decided to try it. It was after doing this procedure that the vertical displacement of the trace became absurd when using the volts-div switch.

I don't remember if I entered the service mode after that, but I certainly turned the scope on and off and no error was indicated in the CRT.

After that, I did the data backup on Nvram and the displacement of the trace was solved, but the automatic measurements do not work. Enter service mode, no error.

There must be a relationship between DC auto balance defects and automatic measurements.

Regarding calibration, I have an HP3324A function generator, another Hitachi V1100 scope, HP 8656B RF generator, frequency meter and HP power supplies.

Could the scope lose calibration alone? Without indicating error?

Leandro Lindemann
 

I searched a lot here on the group and found this topic that I believe has the same defect as my scope:

https://groups.io/g/TekScopes/topic/2465b_dc_balance_parametric/7649054?p=Created,,,20,1,0,0&jump=1

To my knowledge, if my problem was loss of calibration data, when I restored Nvram data the problem. Shouldn't it be over? I imagine that when I run DC Auto Balance, the circuit that controls this part does not work correctly and the error is recorded in Nvram. Therefore, I cannot run the test, because that way the scope is out of use. Until now it is usable, without the functionality of automatic measurements.

Is my thinking right?

Jelcke de Boer
 

Hello,
I have the same problem with my 2465B.
Auto setup and measurements not working
"Noisy or Aperiodic signal"
The problem was there when I got the scope rest seems to work fine.
So far I tried full recap, dc balance, loaded known good calibration into the nvram.
Measured voltages on the main header. (all seem fine)
Replaced hybrid U500 Ab trigger

No luck yet.
Someone suggested Cal 09 have not tried that yet.

Please keep us informed if you find a solution.

Thanks

Leandro Lindemann
 

Hello.

I hope to be able to resolve this defect with the help of the experienced technicians of this group. I hope that someone will have the patience and willingness to help me. In my ignorance, I think it is not a calibration problem, because if so, when restoring Nvram data the problem would have been solved.

I will hope and wait for instructions for the checks I need to carry out.

Leandro Lindemann
 

Vídeo with the fail:

https://youtu.be/9RO12A-TQbs

Leandro Lindemann
 

I'm trying to understand where to start looking for the defect and I really appreciate any help.

Checking the service manual, there is a defect flowchart. In the Parametric measuremet trobleshooting section I have the 1st doubt:

The first first question is:

Is volts Acurate? (Must be timing measurement)

If the answer is no, check triger system
If so, perform cal 01 and cal 09.

My doubt is, is the measure of time that needs to be checked measured by the Auto function or manually?

From the video, you can see that using the Auto option the displayed waveform is outside the time base, but without using the Auto function everything is normal.

Thanks.

Leandro Lindemann
 

The service manual states that a Fluke 5820A calibrator is required to calibrate the 2467. This instrument is inaccessible here in Brazil.

I believe that there must be other instruments capable of being used. Can someone tell me what they would be?
 
I intend to rent the instruments I need here in Brazil. I really want to resolve this defect.

Thanks.

Chuck Harris
 

First thing you need to know is that the cursor measurements are
calibrated to be in the center of the noise on the trace, not on
the peak of the noise on the trace, so all of the 1.06V, or so
measurements you have made with the cursors are not being done
correctly.

There are two principle measurements done with the Parameteric
section: voltage measurements and time measurements.

Voltage measurements use the trigger set point on the hybrid to
measure the most positive voltage of the waveform, and the most
negative voltage of the waveform that they can trigger upon.
They take the difference to get the peak to peak voltage.

Time measurements use the trigger hybrid to trigger the A and B
sweeps to measure the center crossing of the first and second cycle
of the waveform, and then read the time between those two points.

Automatic setup uses all aspects of the measurement system to give
you a 1 cycle, or so, waveform that fits on the screen in amplitude.

The trigger hybrid is the center of both timing and voltage parametric
measurements. It must work for voltage measurements, or it will not
work for either voltage or timing measurements.

But, it can work for voltage measurements, and still not work for
timing.

And, it can work manually for all measurements, and not work at all
for parametric measurements... they use different control paths.

I find that contact problems on the contact pins of the trigger
hybrid socket are the number one problem. You can put in a dozen
new hybrids, but if the socket pin isn't making contact, in just
the right way, none of them will work reliably.

My suspicion is that the pad on one of the hybrid terminals is
done badly, and relies on the socket pin to touch it in just the
right place for it to work. More investigation is necessary.

DeOxit, and some wiping on the socket pins is often helpful.

The Parametric Measurement's PAL's socket is another trouble
area. It should be cleaned with DeOxit.

Now, on to the flowchart:

The first step asks if volts is accurate (must be timing measurement)

This wants you to set the measurement mode to measure volts on an
AC waveform.

If the volts measured is wrong, or simply goofy, then you need to
troubleshoot the trigger system.

If the volts is correctly being measured, then the problem could be
that the parametric measurements section needs to be calibrated.
(CAL01 and CAL09).

In my experience, that rarely, if ever, fixes the
voltage measurement.

-Chuck Harris


leandro.lindemann via Groups.Io wrote:

I'm trying to understand where to start looking for the defect and I really appreciate any help.

Checking the service manual, there is a defect flowchart. In the Parametric measuremet trobleshooting section I have the 1st doubt:

The first first question is:

Is volts Acurate? (Must be timing measurement)

If the answer is no, check triger system
If so, perform cal 01 and cal 09.

My doubt is, is the measure of time that needs to be checked measured by the Auto function or manually?

From the video, you can see that using the Auto option the displayed waveform is outside the time base, but without using the Auto function everything is normal.
Thanks.

Chuck Harris
 

You are looking at the military service manual... it is not the best
manual to have. The US military standardized on the Fluke calibrators.

You can calibrate your scope with the usual SG503/504, PG506, TG501
and TD pulser arrangement.

Go to Didier's website, and get this manual:

<http://www.ko4bb.com/getsimple/index.php?id=download&file=Tektronix/Tektronix_-_2465B_Oscilloscope/Tek_2465B_2467B_Service_1993.pdf>

-Chuck Harris

leandro.lindemann via Groups.Io wrote:

The service manual states that a Fluke 5820A calibrator is required to calibrate the 2467. This instrument is inaccessible here in Brazil.

I believe that there must be other instruments capable of being used. Can someone tell me what they would be?
 
I intend to rent the instruments I need here in Brazil. I really want to resolve this defect.

Thanks.

Leandro Lindemann
 

Hello Chuck.

First of all, thank you very much for helping me.

Thank you also for informing that you were using the cursors in the wrong way. I learned that now.

Regarding hybrid ICs, I never worked with them. To remove them, and do a cleaning with DeoxitD5, is it necessary any extra care? Honestly they are intimidating, the fear of damaging them and failing to replace them is enormous.

I will wait for your answer on this before I do the cleaning.

I will hope that the problem of my dear 2467B is badly contacted.

Big hug and thank you very much.

Leandro Lindemann
 

Thanks for the link.

Chuck Harris
 

It doesn't take much care to work with the
"finned" hybrids. They each have 4, #4-40 nuts,
with captive star lock washers, that hold them
to the socket. Some of the hybrids use a hex
standoff as one of the nuts...

To remove the nuts, either use a finger to
press the middle of the hybrid, and just spin
the nuts off, or if that is too many hands,
remove two diagonal nuts, and then loosen the
remaining two diagonal nuts 3 or 4 turns at a
time.

To put the hybrids back, you will need to use
both hands. One to arrange, and press the
hybrid into the socket, so its aluminum fins
contact the board on all sides.. and at the
same time, put on a diagonal pair of nuts to
finger tight. I use a plastic nut starter to
make this easier. The nut starter stretches to
fit the nut snugly, when you press it onto a nut
that is laying on your bench...

Then the other diagonal pair, and then tighten
all 4 nuts to just 1/4 turn past contact.

The only danger to the hybrids is if you tighten
the nuts when one side is hung up... that will break
the ceramic substrate. That is why you hold the
hybrid down against the board before you put on
a diagonal pair of nuts.

-Chuck Harris

leandro.lindemann via Groups.Io wrote:

Hello Chuck.

First of all, thank you very much for helping me.

Thank you also for informing that you were using the cursors in the wrong way. I learned that now.

Regarding hybrid ICs, I never worked with them. To remove them, and do a cleaning with DeoxitD5, is it necessary any extra care? Honestly they are intimidating, the fear of damaging them and failing to replace them is enormous.

I will wait for your answer on this before I do the cleaning.

I will hope that the problem of my dear 2467B is badly contacted.

Big hug and thank you very much.



Leandro Lindemann
 

I understood your explanations Chuck.

I will do this today, and I will return with the result.

With much hope of being something simple, like bad contact.

In this case, the hybrid is the U500, correct?

Thank you.

Leandro Lindemann.

-----Mensagem Original-----
From: Chuck Harris
Sent: Wednesday, February 26, 2020 2:54 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] Tektronix 2467B Auto Measure Faulty?

It doesn't take much care to work with the
"finned" hybrids. They each have 4, #4-40 nuts,
with captive star lock washers, that hold them
to the socket. Some of the hybrids use a hex
standoff as one of the nuts...

To remove the nuts, either use a finger to
press the middle of the hybrid, and just spin
the nuts off, or if that is too many hands,
remove two diagonal nuts, and then loosen the
remaining two diagonal nuts 3 or 4 turns at a
time.

To put the hybrids back, you will need to use
both hands. One to arrange, and press the
hybrid into the socket, so its aluminum fins
contact the board on all sides.. and at the
same time, put on a diagonal pair of nuts to
finger tight. I use a plastic nut starter to
make this easier. The nut starter stretches to
fit the nut snugly, when you press it onto a nut
that is laying on your bench...

Then the other diagonal pair, and then tighten
all 4 nuts to just 1/4 turn past contact.

The only danger to the hybrids is if you tighten
the nuts when one side is hung up... that will break
the ceramic substrate. That is why you hold the
hybrid down against the board before you put on
a diagonal pair of nuts.

-Chuck Harris

leandro.lindemann via Groups.Io wrote:
Hello Chuck.

First of all, thank you very much for helping me.

Thank you also for informing that you were using the cursors in the wrong way. I learned that now.

Regarding hybrid ICs, I never worked with them. To remove them, and do a cleaning with DeoxitD5, is it necessary any extra care? Honestly they are intimidating, the fear of damaging them and failing to replace them is enormous.

I will wait for your answer on this before I do the cleaning.

I will hope that the problem of my dear 2467B is badly contacted.

Big hug and thank you very much.



Leandro Lindemann
 

I understood your explanations Chuck.

I will do this today, and I will return with the result.

With much hope of being something simple, like bad contact.

In this case, the hybrid is the U500, correct?

Thank you.

Leandro Lindemann.

Chuck Harris
 

Yes.

Leandro Lindemann via Groups.Io wrote:

I understood your explanations Chuck.

I will do this today, and I will return with the result.

With much hope of being something simple, like bad contact.

In this case, the hybrid is the U500, correct?

Thank you.

Leandro Lindemann.



Leandro Lindemann
 

Hello Chuck.

I removed the IC U500. I cleaned with Deoxit D5 its terminals and its socket.

Unfortunately it didn't. Problem persist.

Then I removed the IC PAL, cleaned his contacts and the problem persists.

I will leave the scope open, hoping that I have time to continue helping me.

Thank you.

Chuck Harris
 

My next step after cleaning U500 is to replace it with
a known working U500 from one of my test scopes.

They are a very common failure part.

-Chuck Harris

Leandro Lindemann via Groups.Io wrote:

Hello Chuck.

I removed the IC U500. I cleaned with Deoxit D5 its terminals and its socket.

Unfortunately it didn't. Problem persist.

Then I removed the IC PAL, cleaned his contacts and the problem persists.

I will leave the scope open, hoping that I have time to continue helping me.

Thank you.