new killer scope in town - a true game changer from R&S - RTB2002 & RTB2004

[Joel_l]

For those who have had a chance to play with these now, is there a full screen mode where all the menus go away and the full screen is used for waveform display?

Joel

[maginnovision]

For those who have had a chance to play with these now, is there a full screen mode where all the menus go away and the full screen is used for waveform display?

Joel

No, the top and bottom menus are always visible.

[rf-loop]

So here is ci11's test and... it works!    Pretty darn amazing.

The first photo below shows the signal level. 492Vp-p = 246uV(peak) = 174uV(rms).  This is coming from a Krohn-Hite KH4400B audio oscillator (95% similar to their current KH4402B), which specs at 0.0005% THD at 20Hz.  The scope BW limit is set to 20MHz, although it really didn't make that much difference.  Keep in mind that the oscillator's attenuation network is a series of resistors.  Even though the THD is very low the oscillator is adding a pile of Johnson/thermal noise to the signal.  So that is not all scope noise floor visible in the signal, there is some incoming noise at this tiny level.   Just one 49K series resistor at 25C with the 100KHz bandwidth of the generator would add 9uV of noise.

The second photo shows the capture and trigger with the scope on high definition mode.

I've posted a bunch of additional photos (high res photos, I can only get two attached per post) on Google Drive here including unboxing, test setup, and averaging acquire mode.  Averaging acquire mode was almost as good as high res, I'll post that next (2 photo limit).

https://drive.google.com/drive/folders/0B67cJELZW-i8X294RF9XWFVkSDQ?usp=sharing

Where from this enormous noise?
Bit more wide freq BW do not explain this noise level. (what is this version BW?)

I do not try explain that S is better or what ever competition. This can not even do. This is not apples to apples.
But this is question about this R&S noise what looks bad in these images and question is why?

In R&S image there is 5 times average and 20M reject.
In 2mV image there is really terrible  noise. Nearly like noisy Rigol.
Do you have there some noise generator connected?


Here is cheap Economy Siglent.


2mVrms, continuous running, no average or other tricks for reduce visible noise, just raw  real time and full BW.


Attacments:
1. 174uVrms  realltime single shot without any kind of averaging or other tricks. Only analog BW rejected for 20M
2. 2mVrms realtime continuously running without any kind of averaging and now also BW full. (~200M)


Signal generator just Siglent 5082 funktion gen and between scope used Agilent 8495B step attenuator (60dB)

[agdr]

Here is ci11's tests with a better (fixed frequency, lower THD and lower noise) oscillator, one of Victor's famous ultra low THD 1KHz units.  Victor's oscillator produces less than -140dB THD, as noted here by a fellow who owns the current model Audio Precision distortion analyzer:

http://www.diyaudio.com/forums/headphone-systems/300491-neurochrome-hp-1-ultra-high-end-headphone-amp.html#post4918641

It beats the built-in variable frequency oscillator in the AP so he uses it for the lowest possible THD for a fixed frequency.

Victor's oscillator uses relatively few parts (relative to the variable frequency KH4400B) and is direct output, no resistive dividers to generate Johnson noise.  The oscillator uses two TL431's for a power supply and in this case I have them embedded in a power supply creation of mine.  Kind of went nutz for low noise.   This setup is using 5 9V batteries inside a fully shielded (copper spray inside) Hammond case.  The oscillator boards are individually fed with a constant current sink.

The lowest Victor's will go is about 366mV(rms), so I have added my own resistive divider on the output, as shown in the photo.  A 10K in series with a 5.1R, both 1% 50PPM/C metal film, for a 1960:1 voltage reduction, or 366mV(rms)/1960 = 187uV(rms) = 264uV (peak) = 528uVp-p, as shown in the attached photo (529uVp-p in the photo).  I had to keep the total resistance around 10K to keep from loading the oscillator, ideally it would be lower to reduce Johnson noise.  But to that end you might notice in the photo the resistors look a little frosty.    I've sprayed the resistors and their leads down with technical freeze spray to temporarily reduce the Johnson noise as low as I can.

The net result is the 2nd attachment!  This is done using the USB stick and the scope "camera" button trick.  Notice significantly less noise in the signal than with the variable-frequency KH4400B before.  The bulk of that noise previously was the output resistors and (old - LM318!) chips in the KH oscillator. The scope is in high definition acquisition mode again.  I couldn't get it to trigger this time using edge triggering.  I tried a few options and oddly enough "video" triggering did the job, which is what is used below.

For more photos of the test that were too many to post, see this Google Drive link:

https://drive.google.com/drive/folders/0B67cJELZW-i8VUowRlh1Q3ZtNXM?usp=sharing

[agdr]

Do you have there some noise generator connected?

Yep, see the post I just made above using Victor's oscillator (vastly lower noise than the KH4400B).  The noise isn't coming from the KH RTB2004 scope, it is coming from the old audio oscillator.  It is very low THD but not particularly low noise.  Victor's is though, see above.

Also keep in mind the scope is on normal acquisition there, not high definition yet.

So far, from what I can see, this KH RTB2004 has amazingly low noise, which is exactly what I was hoping for.  For us audiophiles there is no such thing as "low enough noise". 

[mtdoc]

Where from this enormous noise?

As he says in his post, it is coming from the signal.

If you watch Mike's review, he shows the noise floor and it is very low.

[rf-loop]

Do you have there some noise generator connected?

Yep, see the post I just made above using Victor's oscillator (vastly lower noise than the KH4400B).  The noise isn't coming from the KH RTB2004 scope, it is coming from the old audio oscillator.  It is very low THD but not particularly low noise.  Victor's is though, see above.

Now set 2mV/div, full BW, just real time raw running without high res, without averaging, without any kind of tricks what reject displayed noise. Say example 20kHz.
Image.

[maginnovision]

By the way using USB MTP mode or ethernet you can more easily grab images from the scope. I always have a laptop with me so I have been using the USB. You can open it up and just grab images and data. If you like the usb drive way that works too, especially if you're unable to use the usb or ethernet ways due to distance, I just thought it was slightly cumbersome.

[mtdoc]

Now set 2mV/div, full BW, just real time raw running without high res, without averaging, without any kind of tricks what reject displayed noise. Say example 20kHz.
Image.

Watch Mike's video starting at about the 36 minute mark. He shows it running free with no input at 1mV/div no BW limit - so full 300MHz - noise floor is less than 1 division. Very impressive.  ?

[maginnovision]

Do you have there some noise generator connected?

Yep, see the post I just made above using Victor's oscillator (vastly lower noise than the KH4400B).  The noise isn't coming from the KH RTB2004 scope, it is coming from the old audio oscillator.  It is very low THD but not particularly low noise.  Victor's is though, see above.

Now set 2mV/div, full BW, just real time raw running without high res, without averaging, without any kind of tricks what reject displayed noise. Say example 20kHz.
Image.

Using two totally different setups I'm not sure what's supposed to be achieved here.

[agdr]

Here is another slick thing on the RTB2004, a button called "Quick Measure".  Just applies a typical set of measurements to the signal.  Hitting the button again turns them off.  Screen shots below are before and after.

I haven't looked, but I wouldn't be surprised if that quick-test set is configurable to other measurements.

[agdr]

Regarding noise, here are some tests with a BNC shorting cap on channel 1.  Vertical all the way up at 1mV/div, 20us time base.  The scope is set to measure the peak to peak voltage in the lower left corner.

The first two are normal acquisition mode, first on 20MHz BW limit and the second at full bandwidth.  270uVp-p and 900uVp-p respectively.

The next two are high definition acquisition, first on 20MHz BW limit and the final at full bandwidth.  138uVp-p and 152uVp-p respectively.

[rf-loop]

Do you have there some noise generator connected?

Yep, see the post I just made above using Victor's oscillator (vastly lower noise than the KH4400B).  The noise isn't coming from the KH RTB2004 scope, it is coming from the old audio oscillator.  It is very low THD but not particularly low noise.  Victor's is though, see above.

Now set 2mV/div, full BW, just real time raw running without high res, without averaging, without any kind of tricks what reject displayed noise. Say example 20kHz.
Image.

Using two totally different setups I'm not sure what's supposed to be achieved here.

Perhaps you did not find what reads in R&S screen.
(extremely low samplerate and HI res on what of course reduce displyed noise. This is far far away from real time full bw capture.
Also in mikes review it show quite high noise. 1mV/div  around 1mVpp and around 150uVrms noise. Not bad but but this is R&S ! what normally manufacture around state of art class equipments  - it is lot.

[rf-loop]

Regarding noise, here are some tests with a BNC shorting cap on channel 1.  Vertical all the way up at 1mV/div, 20us time base.  The scope is set to measure the peak to peak voltage in the lower left corner.

The first two are normal acquisition mode, first on 20MHz BW limit and the second at full bandwidth.  270uVp-p and 900uVp-p respectively.

The next two are high definition acquisition, first on 20MHz BW limit and the final at full bandwidth.  138uVp-p and 152uVp-p respectively.

Now it looks better, nearly like Sigl.  (meaningful is full BW 1mV/div raw realtime mode without any "cleaning" functions on.)

Your screen trace looks thin but still your measurements show 0.9mVpp. Why these peaks are not visible?
Look mikes review video, around same but trace in video looks much more thick.

Edit now I know why it looks thin.
Here it is, attached, sample from your "noise hide image"

Now it looks more like it really is.

[ebastler]

Now it looks better, nearly like Sigl.  (meaningful is full BW 1mV/div raw realtime mode without any "cleaning" functions on.)

Which Siglent model(s) are you comparing to?
Would you have any data? (Maybe they are already posted on a Siglent thread somewhere, but I can't recall.)
Thanks!

[tautech]

Now it looks better, nearly like Sigl.  (meaningful is full BW 1mV/div raw realtime mode without any "cleaning" functions on.)

Which Siglent model(s) are you comparing to?
Would you have any data? (Maybe they are already posted on a Siglent thread somewhere, but I can't recall.)
Thanks!

Bog standard SDS1kX.
When I saw Mikes vid I thought R&S might have done better than Siglent's best selling model but no.
I'd put some shots up but I'm out of stock, anyway there's plenty of examples in the SDS1000X thread.

[rf-loop]

Just for compare, (previous detail image from R&S  @agdr
Same settings (afaik)
Around same ballpark (note BW diffeerence R&S 300MHz  / Sig measured BW around 200MHz, nameplate  BW 100MHz)

[mikeselectricstuff]

For those who have had a chance to play with these now, is there a full screen mode where all the menus go away and the full screen is used for waveform display?

Joel

No, the top and bottom menus are always visible.

If you have lots of channels , decodes etc. going, the bottom menu can grow by another line.

[mikeselectricstuff]

Also - just a usage reminder - remember that pressing the vertical encoder knob goes from "coarse" to "fine" divisions which helps mitigate a bit that annoying thing Mike noted of the encoders having detents.

Having detents or not does not matter at all. These are rotary encodes, which have discrete output by design. I would rather complain that R&S skimped on the encoders, but with coarse resolution dials having detents is actually helpful, because it gives some feedback on adjustments.

The thing about the detents is that they prevent decent velocity sensing. Not a big deal for vert/horiz, but a pain for the general purpose adjust knob, e.g. adjusting holdoff , pulse trig width etc.

[maginnovision]

Also - just a usage reminder - remember that pressing the vertical encoder knob goes from "coarse" to "fine" divisions which helps mitigate a bit that annoying thing Mike noted of the encoders having detents.

Having detents or not does not matter at all. These are rotary encodes, which have discrete output by design. I would rather complain that R&S skimped on the encoders, but with coarse resolution dials having detents is actually helpful, because it gives some feedback on adjustments.

The thing about the detents is that they prevent decent velocity sensing. Not a big deal for vert/horiz, but a pain for the general purpose adjust knob, e.g. adjusting holdoff , pulse trig width etc.

When you first mentioned the detented knobs I thought I wouldn't care. I do understand where you're coming from though after having used it for 20-30 hours.

[EEVblog]

Just for compare, (previous detail image from R&S  @agdr
Same settings (afaik)
Around same ballpark (note BW diffeerence R&S 300MHz  / Sig measured BW around 200MHz, nameplate  BW 100MHz)

Noise comparisons have to use the same analog bandwidth, same memory depth, same sample rate, and same update rate, otherwise they are not apples-apples.

[mikeselectricstuff]

Also - just a usage reminder - remember that pressing the vertical encoder knob goes from "coarse" to "fine" divisions which helps mitigate a bit that annoying thing Mike noted of the encoders having detents.

Having detents or not does not matter at all. These are rotary encodes, which have discrete output by design. I would rather complain that R&S skimped on the encoders, but with coarse resolution dials having detents is actually helpful, because it gives some feedback on adjustments.

The thing about the detents is that they prevent decent velocity sensing. Not a big deal for vert/horiz, but a pain for the general purpose adjust knob, e.g. adjusting holdoff , pulse trig width etc.

When you first mentioned the detented knobs I thought I wouldn't care. I do understand where you're coming from though after having used it for 20-30 hours.

The touchscreen is what stops this being a serious issue as you can type values in,but there are times when it gets annoying. There is some velocity sensing but I think it's basically 2 'gears' and only barely useful.

[madires]

The thing about the detents is that they prevent decent velocity sensing. Not a big deal for vert/horiz, but a pain for the general purpose adjust knob, e.g. adjusting holdoff , pulse trig width etc.

Let's say, detents make it harder to sense the velocity. My observation is that users are turning an encoder a few steps automatically when turning fast. When you take the average time per step for a few steps (2 or 3) you get good results also for encoders with detents. And for higher velocities add one or two steps dynamically to prevent glitches (sudden high velocity). I've just tested this with cheap and crappy encoders, and it works reasonably well.

[nctnico]

Just for compare, (previous detail image from R&S  @agdr
Same settings (afaik)
Around same ballpark (note BW diffeerence R&S 300MHz  / Sig measured BW around 200MHz, nameplate  BW 100MHz)

Noise comparisons have to use the same analog bandwidth, same memory depth, same sample rate, and same update rate, otherwise they are not apples-apples.

It is not just about comparisons at the higher sensitivity settings (which are not often used anyway) but even at lower sensitivity setting a scope may make a signal look way too woollie to be usefull. A high update rate is less desireable for a close look at a signal because each sweep adds more noise.

[vokars]

Now it looks better, nearly like Sigl.  (meaningful is full BW 1mV/div raw realtime mode without any "cleaning" functions on.)

Which Siglent model(s) are you comparing to?
Would you have any data? (Maybe they are already posted on a Siglent thread somewhere, but I can't recall.)
Thanks!

Bog standard SDS1kX.
When I saw Mikes vid I thought R&S might have done better than Siglent's best selling model but no.
I'd put some shots up but I'm out of stock, anyway there's plenty of examples in the SDS1000X thread.

I am a little bit unhappy with foggy sales comments like this. Do you really think this is good for your business? Or does Siglent think so?