Function generator for scope calibrations.

[Zenwizard]

I have had a lot of requests to the channel to attempt a scope calibration of a Tektronix analog scope 4xx 7xxx or 2xxx but do it without the correct test equipment. My current function generators are Tektronix FG’s and are woefully under speced at 2Mhz.

To be honest I HAVE all the correct test equipment that is needed to calibrate up to a 1Ghz scope with out to much trouble however I have been asked to do the best I can with a function generator and not much else.

To that end I am looking to add a function generator to the lab and run the test to see how close some one can get without the requisite test equipment.
Frequency stability will be addressed with the 10Mhz ref in the lab. But I am looking to add a function generator to the lab for this project as well as for future use in the lab.
My requirements are as follows

New preferably given the experiment I would not like to have to validate the output of the FG after repair as this would defeat the purpose of the video.

Reasonably priced – I am looking in the sub $2000.00 range. I would like to make it more economical than acquiring all the gear to test and validate the specified test equipment. But I also need to get the cleanest output that I can because I will be using it for a calibration/alignment

Frequency range – as fast as possible with in reason.

Rise time – As fast as I can get but I need the edge speed here not frequency, so I am looking for fast edge 50% ish duty cycle and about a 1Khz rep rate.

I have been looking at Siglent or Rigol units but I am open to Keysight, Tek, or R&S as well unless there is a manufacture that I have not though about.

Thanks for the input in advance

Zen

[David Hess]

Forget looking for a single instrument for horizontal sweep calibration and transient response calibration.  Common pulse and function generators do not produce a clean enough signal for transient response calibration.

[Zenwizard]

I totally agree with you and I know exactly what you mean. I apologize if there is some confusion on my posts. The FG is NOT for my personal needs.
My lab is fully equipped with:
PG506
TG501 with a Rb time base
TG501 with High accuracy 5Mhz time base
SG503
SG504 (recently repaired)
067-0681-01 Tunnel Diode Pulser
2X type 284’s both repaired and rebuilt both with working TD’s
Type 106 High amplitude square wave generator
Type 184 Time mark generator
Type 191 Leveled sine wave generator
2x leo bodnar pulsers

I have a YouTube channel that currently has showcased rebuilding and aligning this listed equipment as well as scopes and I have gotten the function generator question SO much it is prompting a video as to WHY this is not the best solution and the appropriate gear should be used.
So the current idea is to do the alignment with a function generator taking care to make it as good as possible with available equipment. And then test the scope with the correct gear to see where it ends up. This might be good enough for some home techs. I can not speak to anyone else’s needs but my own. 
It also does my fellow technicians no good if I acquire a scope calibrator from Fluke as this will not help with the exercise or what I want to show. The question being asked of me is if this is all I have how good can I get?
I also do not want to get a 20 Mhz FG and try to do a 100 Mhz scope This would more than a bit ambitious. But I also don’t want to shoot the budget and acquire something that a hobbyist might not have as well.

A channel link is here for reference - https://www.youtube.com/c/ZenwizardStudios as well as in my signature

[tautech]

If I was back fixing scopes for fun and trying to do checks with just a few bits of gear I'd have given a left nut to have had an enhanced 500 MHz SDG6022X.

It's well within your budget and can perform most of the tasks your other gear can.

[Zenwizard]

Tautech, I looked at the SDG6022X looks promising and even has a good not great, but good rise time. Can you point me to some information on the enhancement? really curious as to what this entails.

Zen

[tautech]

Tautech, I looked at the SDG6022X looks promising and even has a good not great, but good rise time. Can you point me to some information on the enhancement? really curious as to what this entails.

Zen

AFAIK it's the same process as SDG2000X models of which there's lots of guidance here and even a dedicated hack thread.
I did one years ago with code supplied by another member....just to prove to us it could be done.
Today it seems the python script can also do SDG models.....it's almost getting too easy.

[David Hess]

Excluding transient response calibration, I have done horizontal and vertical calibration of an oscilloscope with a function generator but it requires care and some math.

Horizontal calibration up to the limits of the function generator is easy since it only requires a frequency counter, but some function generators are quite fast these days.

Vertical calibration requires external attenuation to get accurate low levels signals for the most sensitive oscilloscope ranges.  The output of the function generator can be accurately calibrated using an AC voltmeter to measure the amplitude of the square wave.  For a true RMS responding meter, this is easy because the peak-to-peak value will be exactly twice the RMS value.  But this calibration can still be done with an average responding meter, where a square wave will read 11.1% high.

To get small signal levels, the high level output from the function generator is calibrated, and then the output after the attenuator is measured in the same way to calibrate the attenuator.  Then the high level output can be lowered to a level which can still be measured, and the output from the attenuator will be known without measurement.  In this way an accurate 10 or 20 millivolt peak-to-peak signal can be generated without direct measurement.

The above is done at a relatively low square wave frequency to maximum AC voltmeter accuracy.

[med6753]

If you can find one the Heathkit IG-4244 combined with a frequency counter is an excellent tool to add to your scope calibration arsenal. Mine get's used often and I've successfully checked analog scopes up to 500MHz B/W.

[rf-loop]

Tektronix CG 5010 or CG5011 (with level head) is good tool for make example Tek analog oscilloscopes full service calibration procedure (example for 2445 - 2465 series etc). Least I have used in old times.