Measuring bandwidth of OWON VDS3102 oscilloscope
using "Trigger Out" signal from "MULTI" connector.


Figure 1
Electrical connections and bandwidth calculating


Figure 2
Response on pulse slope oscillogram

Probe....Mode setting....Bandwidth, MHz
===============================
1:1........20M......................7.0
1:1........Full.......................7.9
1 :10.....20M....................22.6
1:10......Full...................100.6
W/O......20M....................21.7
W/O......Full...................134.6
===============================
Table 1
Bandwidth
--------------------------------------------------------------------------
http://inventrace.com/OWON-VDS3102-Bandwidth.html

(Note: Images pasted directly into the post by moderator)

 

Not sure what your question is (and I can't see any pictures)....

(Moderator's note: The images were made visible after this post was made.)

 

I have never seen this done viewing a trailing edge but using a rising edge. The pulses were generated by pulse generators affording a known very fast rise time. Tektronix was one manufacturer of oscilloscope calibration standards including such pulse generators. Another method is the use of a quality constant amplitude signal generator where frequency is increased to the -3 db points on a display.

Oscilloscopes and their Calibration reads in part:

Another instrument used for the calibration of frequency response is the fast-rise square wave generator. Sometimes referred to as ‘fast edge generator,’ and unlike the high amplitude square wave generator, its fast-rise counterpart is commonly limited in its output capability to a few volts, and specified for rise-time and flatness, the latter generally for a specified time epoch following the transition. The characteristics of the fast-rise generator may be used directly for oscilloscope rise-time calibration (with or without adjustment), as well as for vertical amplifier frequency response, which some manufacturers prefer to use in lieu of the swept frequency method. Perhaps the mostly commonly employed method of calibrating oscilloscope vertical frequency response is the swept frequency method using a constant amplitude signal generator (where an applicable device exists). By design, the constant amplitude generator sine wave output, typically ranging from several millivolts to a few volts, is internally leveled to maintain a peak-detected, flat output, in Vpp. This latter point is important to note in contrast to employing an rms levelling approach, which, unlike the peak responding device (oscilloscope) the generator’s output is designed to calibrate, instead also responds to distortion and noise products, inconsistent with the measurement.

So what is the question?

Ron

 

So what is the question?

The only clear question so far.

 

I have never seen this done viewing a trailing edge but using a rising edge. The pulses were generated by pulse generators affording a known very fast rise time.

= Congratulations! Now we can see it.
= What is the magic difference between leading and trailing edges?
According to Orwiler (1969, p. 22), the term "rise time" applies to either positive or negative step response, even if a displayed negative excursion is popularly termed fall time. davmar.org/TE/TekConcepts/TekVertAmpCircuits.pdf
= Does not matter what is real value of slope time of "Trigger Out" signal. It small enough to make such conclusions:
1. "Native" bandwidth of VDS3102 (without probe) is not smaller then 134.6 MHz.
2. Bandwidth of VDS3102 with 1:10 probe is not smaller then 100.6 MHz.
3. Bandwidth of VDS3102 with 1:1 probe is 12.7 times smaller then with 1:10 probe.
4. Owners of VDS3102 can evaluate bandwidth of their oscilloscopes using "Trigger Out" signal, without special pulse generator.
==================
To moderator:
What is right way to place pictures here in a forum?

 

1. "Native" bandwidth of VDS3102 (without probe) is not smaller then 134.6 MHz.
2. Bandwidth of VDS3102 with 1:10 probe is not smaller then 100.6 MHz.
3. Bandwidth of VDS3102 with 1:1 probe is 12.7 times smaller then with 1:10 probe.

Of course you need read "thAn" instead of "thEn" in my message above. I'm sorry...

 

I still have no idea what you measured, nor how. Could you please add to your pictures some textual explanation of what is going on?

If what I think is the most plausible explanation of what that mess could mean, then you possibly just measured the rise time of the trigger output.

 

According to Orwiler (1969, p. 22), the term "rise time" applies to either positive or negative step response, even if a displayed negative excursion is popularly termed fall time. davmar.org/TE/TekConcepts/TekVertAmpCircuits.pdf

I just refer to those characteristics as leading transitional edge and trailing transitional edge. A negative going pulse has a rise and fall time same as a positive going pulse.
I am still wondering what the first post was about? Was it just to be informative?

Ron

 

If what I think is the most plausible explanation of what that mess could mean, then you possibly just measured the rise time of the trigger output.

Yes, you are right. We just measure fall time (t_slope) of trigger output pulse and calculate bandwidth:
Bandwidth (MHz) = 0.35 / t_slope (us).

 

I just refer to those characteristics as leading transitional edge and trailing transitional edge. A negative going pulse has a rise and fall time same as a positive going pulse.
I am still wondering what the first post was about? Was it just to be informative?

It is not about "negative going pulse" and "positive going pulse".
It is about "negative going step" and "positive going step".
In attachment, page 20, we can see explanation:
"Time-domain studies are concerned with the transit time from one voltage level to another. When discussing time-domain methods one encounters the term step function. This waveform results when a voltage "steps" from one level to another."
Leading edge changes its level to more positive --> "positive going step".
Trailing edge changes its level to more negative --> "negative going step".
See attachment, Fig. 1-12., Fig. 1-13., Fig. 1-14., Fig. 1-15. on pages 20 and 21.
On pages 22 and 23 we have exactly our case - Fig. 1-16. and Fig. 1-17.
Leading and trailing edges are mirror curves and have identical time values.
=====================
Yes, it was to be informative, but seems it is more for discussion.

 

It is not about "negative going pulse" and "positive going pulse".
It is about "negative going step" and "positive going step".
In attachment, page 20, we can see explanation:
"Time-domain studies are concerned with the transit time from one voltage level to another. When discussing time-domain methods one encounters the term step function. This waveform results when a voltage "steps" from one level to another."
Leading edge changes its level to more positive --> "positive going step".
Trailing edge changes its level to more negative --> "negative going step".
See attachment, Fig. 1-12., Fig. 1-13., Fig. 1-14., Fig. 1-15. on pages 20 and 21.
On pages 22 and 23 we have exactly our case - Fig. 1-16. and Fig. 1-17.
Leading and trailing edges are mirror curves and have identical time values.
=====================
Yes, it was to be informative, but seems it is more for discussion.

I agree, I just called it differently. When doing scope calibration I simply looked at what I called a leading edge between the 10% and 90% points.

Ron