EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: biank88 on January 18, 2017, 11:40:57 am
-
Hi there folks, its a pleasure to join this community!
I've just received my new Rigol 1054Z oscilloscope and I've noticed something that I've never considered before.
In particular, If I hook up CH1 (1X, 1V/div) to the test signal (3Vpp, 1Khz) I get the following trace:
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=285906)
Now, if I change the vertical resolution to 200mv/div I get heavly distorted signals like:
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=285904)
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=285908)
Can this be considered "normal" behaviour for the frontend amplifiers?
Thanks for your attention!
-
This is normal and every oscilloscope has this 'problem'. The phenomenon is called amplifier overdrive recovery. Once a transistor is saturated (overdriven) it takes a while to exit this state.
-
Yes, it's normal. When using a digital oscilloscope it's tempting to think that one can use the vertical controls to "zoom in" into a signal. Alas, it is possible only to a certain limited extent. You are not "zooming" but actually amplifying and adding some DC offset to the input signal fed to the A/D converter, which can indeed add distortion.
Anyone who hasn't made this mistake at least once, raise your hand! (I keep mine lowered) :popcorn:
-
Thanks for the quick reply.
I've studied the overdrive recovery for opamps but I've never seen any scope review giving any details about this effect.
-
Thanks for the quick reply.
I've studied the overdrive recovery for opamps but I've never seen any scope review giving any details about this effect.
A link to this document was posted on another similar thread some time ago.
http://cp.literature.agilent.com/litweb/pdf/5989-0068EN.pdf (http://cp.literature.agilent.com/litweb/pdf/5989-0068EN.pdf)
-
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286111;image)
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286113;image)
-
2 threads that discuss this apparent distortion in more depth:
https://www.eevblog.com/forum/testgear/ds1054z-distortion-issue/ (https://www.eevblog.com/forum/testgear/ds1054z-distortion-issue/)
https://www.eevblog.com/forum/testgear/rigol-ds1054z-offset-range-limits/ (https://www.eevblog.com/forum/testgear/rigol-ds1054z-offset-range-limits/)
Read ^^^ and understand. ;)
-
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286028;image)
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286030;image)
-
Hmmm.... I could swear that I've already seen those screenshots somewhere... :horse:
-
Rigol DS 1000Z series:
Offset Range (Probe ratio is 1X)
1 mV/div to 499 mV/div: ±2 V
500 mV/div to 10 V/div: ±100 V
-
is that guy spamming those screenshots or what?
-
Looks like everybody's posting in two threads.
-
is that guy spamming those screenshots or what?
He probably is but it does make me wonder whether Micsig has really solved overdrive issues or that the poster cherry picked a range where the analog frontend has the largest dynamic range. Still it is pretty impressive and usefull.
-
But why use an x1 probe setting? Not that it has anything to do with the original traces but still...
-
The signal is probably fed into the oscilloscope with a coax cable from a function generator but that doesn't really matter for what is shown on screen.
-
Don't doubt my tests! I just want everyone to know the truth, am I wrong? Mikeselectricstuff can prove that he can do the same test. Sorry, my English is not good.
Using the ground spring (the native 2V self-tuning signal)
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286230;image)
-
Don't doubt my tests! I just want everyone to know the truth, am I wrong?
What happens if you set the probe input to x10 mode (the usual setting)?
-
Don't doubt my tests! I just want everyone to know the truth, am I wrong?
What happens if you set the probe input to x10 mode (the usual setting)?
I set the probe input and scope to x10 mode, Test the native 2V self-tuning signal
Using ground clamp
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286297;image)
Using the ground spring
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286299;image)
-
This is of practical value,
When calibration probe compensation,
This is just a simple example, you should be good at discovering
(https://www.eevblog.com/forum/testgear/rigol-ds1054z-vertical-distortion-problem/?action=dlattach;attach=286447;image)
-
This is normal and every oscilloscope has this 'problem'. The phenomenon is called amplifier overdrive recovery. Once a transistor is saturated (overdriven) it takes a while to exit this state.
Not every oscilloscope suffers from overdrive induced distortion; sampling oscilloscopes are immune to non-destructive overdrive making them useful even at low bandwidths where a normal oscilloscope would usually be preferred. (1) Some old analog and digital storage oscilloscopes include cascode circuits for wider operating range and/or clamp circuits which limit the effects of overdrive.
The cause it not always transistor saturation. Low leakage diodes used as clamps usually have slow reverse recovery time. Amplifier stages which rely on balanced operation can suffer from thermal imbalance.
(1) This might explain why some Tektronix sampling instruments support real time operation with both sampling and real time timebases.
Yes, it's normal. When using a digital oscilloscope it's tempting to think that one can use the vertical controls to "zoom in" into a signal. Alas, it is possible only to a certain limited extent. You are not "zooming" but actually amplifying and adding some DC offset to the input signal fed to the A/D converter, which can indeed add distortion.
Analog oscilloscopes almost always suffer from the same problem.
Jim Williams discussed the problem in detail and shows examples of how subtle the problem can be and how to detect it in appendix B of Linear Technology application note 120 (http://cds.linear.com/docs/en/application-note/an120f.pdf) starting on page 14.
Thanks for the quick reply.
I've studied the overdrive recovery for opamps but I've never seen any scope review giving any details about this effect.
It is one of the things reviewers and marketing tend to ignore. In the later case because it would make their products look bad.