Tektronix SPC and Calibration for TDSXXX oscilloscopes

[rplst8]

Hello,

First post and new member here.  First I'd like to thank everyone for their contributions to this forum because it's helped me to better understand the oscilloscope I recently bought and provides a near endless source of information.  However I'm still a little confused about how SPC, calibration, and what Tek refers to as "adjustment" via their TDS700 Adjustment Software relate to one another.  First, WRT Tek TDS scopes, are "adjustment" and calibration the same thing?  Next, does the scope need to pass SPC *before* an adjustment can be done?

I recently acquired a TDS 744A scope that is in great cosmetic and (I think) working condition.  It passes the power-on self test, All the channels seem to work.  They all show a trace, properly display the probe compensation signal,  relay/attenuators (10x/100x/1000x, AC/DC/GND, 50/1M ohm) audibly click on and off, signals are measured correctly on all channels, and other features of the scope seem to work just fine.  It holds the correct date and other settings so I think the NVRAM is OK, although when I first powered it up the date was correct except for the year - though that may have been an oversight by the last operator.

The scope will not pass the SPC however.  It will run (about 5 minutes or so) but eventually stops and reports a failure.  The calibration page shows the voltage reference, frequency response and pulse trigger as "initialized".

I've read conflicting things however on how to remedy this.  At least one post has said that if the cal. status page shows those in the initialized state, it either lost it's calibration constants, or someone started a calibration with the ADJ700 software but didn't finish it.  Another post said that calibration of the scope depends on completing the SPC correctly first.  However in the service manual, it states that the all other adjustments are dependent on the Voltage Reference passing, though in table 5-1 SPC is listed last of the CATS tests that is run during that procedure.

I read about "leaky caps" being an issue with some of these early to mid 90s TDSXXX scopes so I opened the case and took a look.  The processor board and power boards were marked with a p-touch label that they were recapped on Jan of 2020, and the area below the relays/attenuators had written in sharpie "relays replaced 1-2020" on the metal frame.  The acquisition board in this model didn't appear to have any electrolytic capacitors on it and I read that later revisions of the boards removed those (this unit was manufactured in 1996).

So far my hunch is that this just needs to be taken through the full adjustment procedure to write the correct constants back to the NVRAM.  But I don't want to pay through the nose for a calibration or bother with setting up a home calibration rig if it should pass SPC first.

Any help would be greatly appreciated.  Thank you.

[nctnico]

If the SPC doesn't pass then you likely have leaking capacitors (but that should have been taken care of) or a faulty attenuator (but that also has been taken care of). Maybe replacing the capacitors or attenuators went wrong. What is the precise error message you get from the log?

SPC is an internal test for DC offsets. With the adjustment software you can run a test which also includes time alignment but you need a whole bunch of test equipment and a specific GPIB card to run it. It is not easy. I would not go this route.

[rplst8]

What is the precise error message you get from the log?

Code: [Select]

ERROR: Cal meas clipped at min dig ou
    ( 2 @  b201): -5.119 divs


[Jwalling]

What is the precise error message you get from the log?

Code: [Select]

ERROR: Cal meas clipped at min dig ou
    ( 2 @  b201): -5.119 divs


CH3 attenuator is the likely cause.

[Tantratron]

Hoping you have solved the issue... otherwise could post a picture showing your SPC complete screen result to see all sub-items (Signal, Voltage, Freq, Trigger...) ?

[stcoso]

Got a TDS784C with SPC compensation fail... Replaced all the relays on the hybrid attenuators and solved the problem.

[capt bullshot]

The calibration software will not complete if SPC fails.
As said before, the complete calibration procedure is quite lengthy (up to four hours afair) and requires special equipment (a signal generator capable of leveled amplitude up to or above 10dBm and up to 1GHz, some known voltages and a particular ISA Bus GPIB card in an ancient PC running MS-DOS), so not recommended if there isn't a particular reason for it.

Failed or just worn out relays on the attenuator hybrids apparently is a common reason for SPC failures.

[stcoso]

By the way... when I've replaced those relays I've noticed something strange (at least for me)... the solder was really "opaque" and seemed like it have becomed "fragile"...

I haven't tested the old relays to check for their contact resistance but maybe the problem lays in those solder joints.