tek 2465a delayed sweep time problem

[steve]

The scope is a Tektronix 2465a, I don't have the S/N offhand, but I
can get it tomorrow if needed. It's a newer unit, with all
pushbuttons instead of toggle switches.

The problem is that when in delayed sweep, measuring delta t using
overlap gives wrong readings (i.e. overlapping two marks from a time
mark generator at 50us reads 52.5us), but using the cursors, still in
delayed sweep, it reads on (i.e. putting cursors on rising edges of
two pusles). The horizontal axis also is correct, and the cursors are
correct again. So, the only problem is when in B swp, overlapping two
pulses to get a reading gives wrong numbers. Using cursors in A or B
swp gives proper readings,

We tried going through cal 01, 02, 03, and 04 with no luck. The offset
error is uniform for all B swp times for each A swp time (i.e. for
A=200us, the error will be the same for all B times, appx 10.2us, for
A=50us, the error will be the same for all B times, appx 2.5us).
Sorry if this is confusing, I will gladly clear anything up if I can.

Any help would be greatly appreciated, we're just a small calibration
lab, family run, and right now our only other standard scope (another
2465), is waiting on pots from sphere.co.

On a different note, the other 2465 we have had a problem with the
display. After warming up for probably half an hour, some dots appear
on the screen. They are on vertical graticule lines, and turning the
trace and readout intenities down doesn't change them. I've freeze
sprayed all I.C.'s with no effect. We aren't as concerned with this
problem, because the all time readings on this scope are dead on, and
we just need a NIST traceable reference so we can keep calibrating.

Thanks for any help you can give, and if I can clear anything up just
let me know.

Steve Kamego,
Calibration Specialty Inc.

 

[Jim Yanik]

steve wrote in

The scope is a Tektronix 2465a, I don't have the S/N offhand, but I
can get it tomorrow if needed. It's a newer unit, with all
pushbuttons instead of toggle switches.

The problem is that when in delayed sweep, measuring delta t using
overlap gives wrong readings (i.e. overlapping two marks from a time
mark generator at 50us reads 52.5us), but using the cursors, still in
delayed sweep, it reads on (i.e. putting cursors on rising edges of
two pusles). The horizontal axis also is correct, and the cursors are
correct again. So, the only problem is when in B swp, overlapping two
pulses to get a reading gives wrong numbers. Using cursors in A or B
swp gives proper readings,

Do you magnify the sweep to get the best resolution for your overlap?
Or just overlapping a 50us/div A swp and a 50 us/div B swp? (for example)
Is this when you are following the cal procedure?

We tried going through cal 01, 02, 03, and 04 with no luck. The offset
error is uniform for all B swp times for each A swp time (i.e. for
A=200us, the error will be the same for all B times, appx 10.2us, for
A=50us, the error will be the same for all B times, appx 2.5us).
Sorry if this is confusing, I will gladly clear anything up if I can.

Any help would be greatly appreciated, we're just a small calibration
lab, family run, and right now our only other standard scope (another
2465), is waiting on pots from sphere.co.

You could try swapping the A and B sweep hybrids and recalibrating;maybe
there's a problem with one of the sweep hybrids.Or use the hybrids from the
2465(non-A)

On a different note, the other 2465 we have had a problem with the
display. After warming up for probably half an hour, some dots appear
on the screen. They are on vertical graticule lines, and turning the
trace and readout intenities down doesn't change them.

Possible Z-axis problem? (unblanking of the CRT,grid bias problem.)


If the dots appear on the readout areas,it indicates a calibration problem.

I've freeze
sprayed all I.C.'s with no effect. We aren't as concerned with this
problem, because the all time readings on this scope are dead on, and
we just need a NIST traceable reference so we can keep calibrating.

Thanks for any help you can give, and if I can clear anything up just
let me know.

Steve Kamego,
Calibration Specialty Inc.

One thing I always recommmend is that one read the Circuit description in
the service manual and follow it on the schematics,to learn fully how the
scope works.


Disclaimer;its been 6 years since I worked on any 2400 scopes,and I no
longer have any manuals to refer to,so I work from my sometimes faulty
memory.

 

[steve]

steve wrote in

Do you magnify the sweep to get the best resolution for your overlap?
Or just overlapping a 50us/div A swp and a 50 us/div B swp? (for example)
Is this when you are following the cal procedure?

The resolution isn't producing the error, it is present in B swp from
main time all the way down to 5ns. If you set the delay to where the
scope reads 50us, the two pulses will be apart by as much as an entire
division.

You could try swapping the A and B sweep hybrids and recalibrating;maybe
there's a problem with one of the sweep hybrids.Or use the hybrids from the
2465(non-A)

Hadn't even thought of that, that's why I posted here. I've been so
busy trying to calibrate avionics that I forget simple things like
swapping out parts with working ones. I'll definantely try that that
tomorrow once I get back in the lab.

Possible Z-axis problem? (unblanking of the CRT,grid bias problem.)


If the dots appear on the readout areas,it indicates a calibration problem.

The dots aren't in the readout area, but some of them vary in
intensity with readout intensity. Others are on all of the time. I
may bring the camera in tomorrow and see if I can put some pictures to
my words

One thing I always recommmend is that one read the Circuit description in
the service manual and follow it on the schematics,to learn fully how the
scope works.


Disclaimer;its been 6 years since I worked on any 2400 scopes,and I no
longer have any manuals to refer to,so I work from my sometimes faulty
memory.

Thanks so much for the reply, I do appreciate it. Now back to the
broken HP 8551b I bought off of ebay. Anyone have any experience with
those?
Steve

 

[steve]

The scope is a Tektronix 2465a, I don't have the S/N offhand, but I
can get it tomorrow if needed. It's a newer unit, with all
pushbuttons instead of toggle switches.

The problem is that when in delayed sweep, measuring delta t using
overlap gives wrong readings (i.e. overlapping two marks from a time
mark generator at 50us reads 52.5us), but using the cursors, still in
delayed sweep, it reads on (i.e. putting cursors on rising edges of
two pusles). The horizontal axis also is correct, and the cursors are
correct again. So, the only problem is when in B swp, overlapping two
pulses to get a reading gives wrong numbers. Using cursors in A or B
swp gives proper readings,

We tried going through cal 01, 02, 03, and 04 with no luck. The offset
error is uniform for all B swp times for each A swp time (i.e. for
A=200us, the error will be the same for all B times, appx 10.2us, for
A=50us, the error will be the same for all B times, appx 2.5us).
Sorry if this is confusing, I will gladly clear anything up if I can.

Any help would be greatly appreciated, we're just a small calibration
lab, family run, and right now our only other standard scope (another
2465), is waiting on pots from sphere.co.

On a different note, the other 2465 we have had a problem with the
display. After warming up for probably half an hour, some dots appear
on the screen. They are on vertical graticule lines, and turning the
trace and readout intenities down doesn't change them. I've freeze
sprayed all I.C.'s with no effect. We aren't as concerned with this
problem, because the all time readings on this scope are dead on, and
we just need a NIST traceable reference so we can keep calibrating.

Thanks for any help you can give, and if I can clear anything up just
let me know.

Steve Kamego,
Calibration Specialty Inc.

Today I swapped out all of the hybrids with the good ones, ran through
cal again, and had the same results. Any other ideas?

Thanks,
Steve

 

[Jim Yanik]

steve wrote in

Today I swapped out all of the hybrids with the good ones, ran through
cal again, and had the same results. Any other ideas?

Thanks,
Steve

Perhaps a leaky cap (0.1uf IIRC)on one of the sample&hold IC inputs that
'buffer' the delta time pots before AD conversion.Maybe nonlinear.
(here's where the manual's circuit description comes in handy)

When at TEK,I would have used my test microprocessor board,or one from a
similar unit,but you don't have that option.

 

[steve]

Perhaps a leaky cap (0.1uf IIRC)on one of the sample&hold IC inputs that
'buffer' the delta time pots before AD conversion.Maybe nonlinear.
(here's where the manual's circuit description comes in handy)

When at TEK,I would have used my test microprocessor board,or one from a
similar unit,but you don't have that option.

Thanks for your replies, they've been most helpful. As soon as I get
some time, I'm going to dig into it, but for now we'll just use main
time and use the other one for delayed sweep.

Thanks again for the help,
Steve

 

[tekman]

steve wrote:
...snippety..

On a different note, the other 2465 we have had a problem with the
display. After warming up for probably half an hour, some dots appear
on the screen. They are on vertical graticule lines, and turning the
trace and readout intenities down doesn't change them. I've freeze
sprayed all I.C.'s with no effect. We aren't as concerned with this
problem, because the all time readings on this scope are dead on, and
we just need a NIST traceable reference so we can keep calibrating.

Thanks for any help you can give, and if I can clear anything up just
let me know.

Steve Kamego,
Calibration Specialty Inc.

Hi Steve, yes, this dot problem I have had too. I had to readjust the
grid bias and all other settings in the HV section.

After that, the problem disappeared: no more dots, etc disturbance.

I have never traced the problem totally down. Funny: All other voltages
in the LV supply, ripple, etc. show normal values.

So, maybe you give a try on your 2465 and let us know in the newsgroup
if it helped you too.

hth,
Andreas