Question about vertical offset and position and rant about Keysight scopes!

[analogRF]

after a couple of years of searching and waiting, I recently acquired a DSOX3054A which was dead due to a NAND corruption which I fixed here
https://www.eevblog.com/forum/testgear/agilent-dsox3054a-not-booting-possible-nand-corruption-help!/
This is only my second Agilent scope ever after my old 54845A which I sold quite a while ago.

Unfortunately Keysight scopes do not have separate vertical DC offset and vertical position. They (and many other non-Tek scopes) only have vertical position which also involves a DC offset but is very different in that it moves your ground level up/down. Still I cannot understand why Keysight/Agilent don't implement that feature in their high end expensive scopes. It is understandable to not have that in a cheap scope but 3000 or even 2000 X series are not cheap.

My Tektronix scopes (TDS3000, DPO7000, MSO3000) all have separate DC offset and vertical position and I know MDO3000 and 4000 also do
even my old analog Iwatsu 7840 A.K.A Lecroy LH314 does have this. But not on my digital Lecroys and the Keysight.

Alan Wolke (w2aew) has an excellent video about this on youtube.

Suppose I want to look at a small AC ripple/noise on top of a "'slowly varying'" DC and I do want to keep an eye on the DC level variations as well as the AC signal. For example imagine a not so good regulated supply under a heavy load which is turned on and off. I want to see the changes in the DC level due to the load being off/on as well as to see the ripple. Or you can imagine many other scenarios. e.g. keeping an eye on the temperature variations of a small AC signal riding on a DC voltage....
AC coupling does not cut it (cut off freq is usually at 3Hz or higher which is still too fast to catch slow variations) I tried to use MATH (like Ax+B) to emulate that but it does not work because you have to have the waveform in the screen in first place but for that the V/div is too high and the ripple is invisible so the MATH trick doesn't work.

Other trick is to use two channels, one to show the DC level and one to see the ripple in AC coupled, but again if the (slow) variations of the DC level are small you face the same problem 

Question: other than adding en external DC offset to the input signal, is there any other way to work around this?

On my Tektronix scopes I can do this very easily by setting the proper DC offset and then a high gain and still I can use the vertical position too, no need to AC couple.. so I can optimally use the whole ADC dynamic range.

perhaps a good topic for Dave to make a video about 

[Bud]

Is the Expand Around Center option may be what you need?

[nctnico]

I guess what you are looking for is a voltage offset which stays constant regardless the V/div setting. Otherwise you can turn the vertical position up & down but that won't be constant for changing V/div settings. Be aware though that internally both vertical offset and DC offset are likely controlling the same DAC (which feeds a DC offset into the front-end). At least the Yokogawa scope I owned worked that way; I found out while reparing it. In the end there is only some much DC offset the analog front-end can handle.

Isn't there some kind of setting somewhere to change the behaviour? If the vertical offset becomes an absolute voltage offset instead of a relative offset it should work for you as well.

Edit: it seems Bud has found the setting in the manual already!

[tautech]

Is the Expand Around Center option may be what you need?

In other brands it's called a Reference position and can be assigned to vertical or horizontal settings.
A fixed vertical reference position can indeed be handy however unless there's some flag to remind you you're in that mode also confusing when you don't need it.

[analogRF]

Is the Expand Around Center option may be what you need?

no it didnt work but thanks for showing that to me. I haven't noticed that. That's actually how I prefer it to be 

[analogRF]

I guess what you are looking for is a voltage offset which stays constant regardless the V/div setting. Otherwise you can turn the vertical position up & down but that won't be constant for changing V/div settings. Be aware though that internally both vertical offset and DC offset are likely controlling the same DAC (which feeds a DC offset into the front-end). At least the Yokogawa scope I owned worked that way; I found out while reparing it. In the end there is only some much DC offset the analog front-end can handle.

Isn't there some kind of setting somewhere to change the behaviour? If the vertical offset becomes an absolute voltage offset instead of a relative offset it should work for you as well.

Edit: it seems Bud has found the setting in the manual already!

yes, you are right, that's what I am looking for. That's how the Tek scopes work. My old Iwatsu is even better as it provides a wider DC offset range. I admit the amount of offset you can get depends on the V/div range and may not be enough for a particular situation (if the DC level is very large and the AC is very tiny) but I have compared a few non extreme examples on my Tek and the Keysight and it is possible to see it on the tek but not on the agilent.
Bud's suggestion didn't help. If both the vertical offset and position are applied to the same DAC as you pointed, then I think my problem is that the Keysight scope simply has a very limited offset range compared to the Tek. for example on 100mV/div on the Tek I can get +/-10V offset plus the (i dont know how many divisions) vertical positioning but on the Keysightat 100mV/div or even 200mV/div I only get 2V offset (I call it vertical position)
I just saw the datasheet for DSO7000B and that thing has a huge range of offset (position)!

maybe it seems so natural to me to have that DC offset thing because I have mostly used Tek scopes up to now...

[Bud]

Then you may need Keysight's Power Rail probe, but it may cost more than the scope 

[analogRF]

Then you may need Keysight's Power Rail probe, but it may cost more than the scope 

yeah, my research led me to that conclusion and then it scared me so I closed that page down quickly 

[analogRF]

in my old analog Iwatsu, I am pretty sure the DC offset is actually applied to the input signal, it is a totally different thing that vertical position of the trace (beam).

I have to watch Alan's video about this again but I vaguely remember he said it is applied to the scope's input and he was using an MDO4000 to demonstrate that...it's been a while since I watched it....

[Bud]

I'd think in digital oscilloscopes there is no other way to achieve that other than to add bias to the input signal. This is because all has to be done prior to hitting the ADC . The ADC has a range and everything including the signal AND vertical position adjustments must be there and comply to the ADC voltage range.
In analog scopes beam position control would likely be possible to do at the display tube level. There is no such controls on a digital display.

[analogRF]

I'd think in digital oscilloscopes there is no other way to achieve that other than to add bias to the input signal. This is because all has to be done prior to hitting the ADC . The ADC has a range and everything including the signal AND vertical position adjustments must be there and comply to the ADC voltage range.
In analog scopes beam position control would likely be possible to do at the display tube level. There is no such controls on a digital display.

yes I know but why and how Tek scopes have both a DC offset (usually in menus) and vertical position (knob) separately if they are doing the same thing?
DC offset adjustment does not move the ground position (the reference level) it just adds voltage to it, position knob however works like other scopes meaning it moves the reference (ground) position on display.
actually in my DPO7000, the position knob does both functions when you push it, it adjusts the DC offset and when you push it again
it switches to adjusting position of the reference level so there is no need to go to menus to adjust the DC offset

but on the other hand I agree with you that all of this should be done prior to ADC

actually the usual offset (position) like in the keysight would do the job I am asking only if there was more range to it
I think some scopes have much larger offset range

[nctnico]

I'd think in digital oscilloscopes there is no other way to achieve that other than to add bias to the input signal. This is because all has to be done prior to hitting the ADC . The ADC has a range and everything including the signal AND vertical position adjustments must be there and comply to the ADC voltage range.

Absolutely not! The DC offset is applied in the analog domain before the ADC. How much offset can be applied depends entirely on the design of the analog front-end. This is also part of the specification of every oscilloscope.

[analogRF]

here is Alan's video on the subject





I wish he would join and comment on this. He knows how Tek scopes work better than anybody I know

[w2aew]

You've got the description of the problem correct, as well as the operation of Offset.  The Offset function "moves" the vertical reference point, and thus the "expansion" point away from ground, as I described in the video.  This offset is applied in the analog domain before the ADC, so that you can position the signal (DC level) within the digitizing range, and increase the sensitivity (lower v/div) around that.  The offset is indicated by the little channel reference icon along the left edge of the graticule.  The Position control moves this up/down on the display, the Offset control determines if this point is ground (default), or if it is some user defined DC value.  The amount of offset that you can add/subtract is a function of the vertical scale setting.  For situations where you need more offset voltage, then a Power Rail probe is in order.  Tek has power rail probes that offer +/-60V offset, with only 1.25x attenuation, making it easy to characterize very small disturbances on large DC rails.

[David Hess]

I admit the amount of offset you can get depends on the V/div range and may not be enough for a particular situation (if the DC level is very large and the AC is very tiny) but I have compared a few non extreme examples on my Tek and the Keysight and it is possible to see it on the tek but not on the agilent.

...

Bud's suggestion didn't help. If both the vertical offset and position are applied to the same DAC as you pointed, then I think my problem is that the Keysight scope simply has a very limited offset range compared to the Tek. for example on 100mV/div on the Tek I can get +/-10V offset plus the (i dont know how many divisions) vertical positioning but on the Keysightat 100mV/div or even 200mV/div I only get 2V offset (I call it vertical position)

The difference between position and offset is where the DC signal is applied in the attenuation chain.  Offset is applied earlier and position is applied after all switchable attenuation.  A modern DSO with only a single DAC might play games with this but the result is the same.

Commonly offset is applied after the high impedance input attenuators which is why the offset range changes with groups of attenuation settings rather than for each one.  So if you have only one input attenuator, there are only two offset ranges.  With two input attenuators, there might be three or four offset ranges depending on exactly how many input attenuation configurations there are.

maybe it seems so natural to me to have that DC offset thing because I have mostly used Tek scopes up to now...

Tektronix sometimes listed offset range in marketing literature in divisions so on their 7A13 differential comparator, the offset range was up to 20,000 divisions which sounds hilarious.

[wizard69]

Oh come on hit that buy it now button.

Then you may need Keysight's Power Rail probe, but it may cost more than the scope 

yeah, my research led me to that conclusion and then it scared me so I closed that page down quickly 

[analogRF]

Oh come on hit that buy it now button.

Then you may need Keysight's Power Rail probe, but it may cost more than the scope 

yeah, my research led me to that conclusion and then it scared me so I closed that page down quickly 

i can't...even my mouse got scared and stopped clicking so I panicked and Ctrl+W 
you can buy a good scope with the price of one of these probes...

[David Hess]

i can't...even my mouse got scared and stopped clicking so I panicked and Ctrl+W 
you can buy a good scope with the price of one of these probes...

But that probe can do things that no oscilloscope can do without it.