Tektronix 2445B Timing Check problem

[Eb2aoc]

Hi.
I am a new user of the Tektronix 2445B oscilloscope, bought on Ebay last month.
When I do the checks and adjustments following Chapter 4 of the user manual, when doing the Timing Check, something is wrong because the square wave does not cover the 10 divisions with the five cycles. The signal is bigger. I attach a photo.
The probes that I use are two Tektronix P6136 bought on Ebay.
What does this mean??
Is the oscilloscope wrong (uncalibrated)??
It is possible that what is not correct is the calibration signal of the oscilloscope?
Can the probes that I use be in poor condition and cause that defective signal frequency? The weird thing is that with the probes I get the same result

The funny thing is that if I measure the frequency of a signal with the oscilloscope, the measurement seems to be quite correct. The RF signal that I use comes from the RF Generator of a MiniVna Tiny. I have checked the RF signal of the MiniVna with the frequencimeter that I have and the signal seems reliable.

How is it possible that when performing the Timing Check of the oscilloscope, this is wrong, but then when making frequency measurements with the same oscilloscope the measurements seem correct ??

I'm sure I'm missing something so obvious that I can not see it. I apologize if this is so ....
I would appreciate any opinion that you can put me in the right way.
Thanks in advance

[tautech]

Probe compensation outputs on CRO's are not always exactly 1 KHz and it doesn't much matter when they're not.
What matters more is the fast edge and flat top.
Some have frequency adjustments and some do not.

Check the probe cal output frequency with another device.

[MarkL]

The calibrator output on the 2445B should be fairly accurate.  The specs say 0.4Vpp 1% vertical into 1Mohm, and 0.1% horizontal between 100ns to 100ms secs/div.  The output frequency changes depending on the secs/div setting, but should give you a clean 1 cycle every 2 horizontal divisions from 100ns to 100ms.

If you're not getting near that, something is wrong.  It's possibly the calibration, but could be plenty of other things too.

If you measure a waveform from the calibrator or from your RF generator with the 2445B cursors (not relying on the graticule lines), do you get an accurate result?  How about with the automated (parametric) frequency measurement?

Do you have the option 06 counter/timer option?  If so, how do those readings match up?


EDIT: I should add that the point of these questions is to determine if this is simply a problem with the CRT beam not aligning properly with the graticule lines, or if it's something more involved.  If it's just the alignment (gain & position), that can be adjusted separately from a full calibration, provided the power supply values are in spec (which you should also check).

A quick determination is to move the horizontal cursors 8 graticule divisions apart to see if you get the right readout.  Use 6 divisions for the vertical cursors, but the vertical looks ok from your photo.

[Eb2aoc]

Hi,
Sorry for the delay of my answer, but my free time to do tests is only in weekends.

Check the probe cal output frequency with another device.

Tautech, many thanks for your suggestion. I agree with you about the matter of probe compensation outputs. But it is that made me think that the oscilloscope time measurements were not well.
How can I prob cal output with another device?? Conecting the output to another oscilloscope throught a probe?? My other oscilloscope is a Kikusui 5041 but the frequency measurements are not goods and needs recalibration. For this I bought the Tek 2445B.
At the end I conected the prob cal output to my frequencymeter and I had obtained the attached measures. They are like expected. From 200 microseconds to 10 miliseconds my frequencymeter don´t measure stable frequencies

The calibrator output on the 2445B should be fairly accurate.  The specs say 0.4Vpp 1% vertical into 1Mohm, and 0.1% horizontal between 100ns to 100ms secs/div.  The output frequency changes depending on the secs/div setting, but should give you a clean 1 cycle every 2 horizontal divisions from 100ns to 100ms.

If you're not getting near that, something is wrong.  It's possibly the calibration, but could be plenty of other things too.

Hi MarkL, many thaks for your contribution. The signal is the same from 100ms to 500ns but out of range. From 200ns to up the signal is not a square signal, is a mixture between square and senoidal, but with the same error of frequency.

If you measure a waveform from the calibrator or from your RF generator with the 2445B cursors (not relying on the graticule lines), do you get an accurate result?  How about with the automated (parametric) frequency measurement?

I measured a RF signal from my generator (miniVNA Tiny) and the results are confused for me. At low frequencies the oscilloscope measures are OK (Auto Measures and Cursors Measures).
But from 5000khz to up the measures are smaller than spected.
The measures from the Auto and from the Cursors are very near like you can see in the attached image.
But the measures in the graticule are wrong.

For all this I can suspect that my scope have two problems. One problem with the graticule and another problem with the measurements 

Do you have the option 06 counter/timer option?  If so, how do those readings match up?

I think No. The seller didn´t told me anything about this. If I push the Measure button I don´t see any COUNTER selection. For this I suppose that haven´t this option.

EDIT: I should add that the point of these questions is to determine if this is simply a problem with the CRT beam not aligning properly with the graticule lines, or if it's something more involved.  If it's just the alignment (gain & position), that can be adjusted separately from a full calibration, provided the power supply values are in spec (which you should also check).

First thing to do. Open the covers of the scope and check the power supply values

Best regards

P.D: Sorry for my bad english. I hope that you understand all....