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

[Kean]

Thanks a lot! Can some more people with official access to the firmware also post the SHA2-256 sum of the firmware? (MD5 has been mentioned, but that's been obsolete for almost 30 years.)

Code: [Select]

% sha256sum RTB22_Firmware_v03.00.zip
4989a4f92f61b2e6f5cfc7cfd5653bc0326056d4edadbd962c12377511453dd9  RTB22_Firmware_v03.00.zip
% sha256sum RTB24_Firmware_v03.00.zip
56de9016b75385af57d7d2e550ac682933edf67af0ff2956f920df96830d3d9c  RTB24_Firmware_v03.00.zip

Edit: Good point noted by another member, so here are the values for the FWU files after extracting from the ZIP.

Code: [Select]

% sha256sum RTB*
ced7c8cca2053d9bbd78b4c99aec8c83c729ea8be71d0c05697bbae4b7769c1f  RTB2002.FWU
73d6a8a695ea9338f4cf9a6672721d82d5afc4a8bbe1ab6dfefa2607872bb3c2  RTB2004.FWU

But as mentioned by tv84 below, the R&S firmware upgrade process almost certainly checks firmware signature and it would be quite some task to fake that let alone reverse engineer the firmware to do whatever it is you think someone wants to do to your scope... implement Doom I presume.

[tv84]

Thanks a lot! Can some more people with official access to the firmware also post the SHA2-256 sum of the firmware? (MD5 has been mentioned, but that's been obsolete for almost 30 years.)

  Do you have any clue of the probability of someone creating a different R&S signed FW package with the same MD5 hash as the one you downloaded?

Only R&S would be able to try such thing (because of the signature) and if you don't trust them then no good using their products.

[peka]

I am too late...
Could somebody share the Firmware 3.0 again for my RTB2004?
Thanks

[Fenstergucker]

The firmware is still there, I have not removed it.

Peter

[peka]

It was my fault, I did not read your post exactly.
Now it works, my RTB2004 has got the new firmware.
Thank you very much!

[Jeroen3]

Just for checks, I can confirm the official R&S download at 2025-04-15:

Code: [Select]

Naam: RTB24_Firmware_v03.00.zip
Grootte: 23624610 bytes : 22 MiB
SHA256: 56de9016b75385af57d7d2e550ac682933edf67af0ff2956f920df96830d3d9c


Code: [Select]

Naam: RTB22_Firmware_v03.00.zip
Grootte: 23444557 bytes : 22 MiB
SHA256: 4989a4f92f61b2e6f5cfc7cfd5653bc0326056d4edadbd962c12377511453dd9


[bayjelly]

Do you have any clue of the probability of someone creating a different R&S signed FW package with the same MD5 hash as the one you downloaded?

MD5 is so well and truly broken, individuals were able to easily create collision *20 years ago*.

Does that make it probable that someone would want to create fake firmware? No. Is it easy to use SHA256 instead of MD5 to guard against the very unlikely chance that some troll is having some fun? Yes.

Only R&S would be able to try such thing (because of the signature)

As a professional who works with bootstrapping including firmware validation, I am well aware how many buggy implementations of validation there are out there (I know you are, too). I don't know about R&S's, and I also think my scope is a very unlikely attack vector, but...

... it's so easy to get a SHA256 hash instead of an MD5 one, so why not just do it? If not anything else, it fosters hygiene and the death of MD5.

[tv84]

... it's so easy to get a SHA256 hash instead of an MD5 one, so why not just do it? If not anything else, it fosters hygiene and the death of MD5.

Sure but, as always, horses for courses.

The MD5 is being used here as a binary integrity checksum, not as an authenticity tool. For that, R&S included the signature inside the FW package.

So, MD5 can/will continue to exist as long as you know what you are doing with it. We're not killing CRCs just because they are dummy-easy to hack...

[salomonander]

Hey there,
I have a total noob question:
So far i only used the supplied probes. Now i want to connect the scope directly to my audio analyzers bnc connectors. What kind of cable should i use? 50ohms or 75ohms? Do i need the hz22 adapter (50ohms terminating resistor) to connect them? Thanks

[goaty]

Now i want to connect the scope directly to my audio analyzers bnc connectors. What kind of cable should i use? 50ohms or 75ohms? Do i need the hz22 adapter (50ohms terminating resistor) to connect them? Thanks

For audio frequencies, the impedance mismatch is of no concern, but for measurements, the level of the signal is relevant.
What does the analyzer manual say ? It could be it tells you "into 50Ohms", then you need the inline terminators (or build some yourself).

[salomonander]

Thanks goaty,
It doesnt say much. There is a analyzer bnc, a input bnc, and a trigger out.
Regarding the trigger out, the manual states that it gives a constant amplitude of 1.0V RMS and that the source impedance for this signal is 680 Ohms. It is LSTTL compatible - whatever that means

Regrading the input and analyzer outputs, these are buffered versions of the input of the analyzer before and after filters. The amplitude is typically between 0 - 2.83 Vpp. In my scenario im feeding the units own generator to the inputs which has an output impedance of 40 ohms.

Here is a link to the manual, but it’s extensive:

https://manuals.repeater-builder.com/te-files/AUDIO%20PRECISION/AUDIO%20PRECISION%20Portable%20One%20Plus%20Access%20User.pdf

[2N3055]

Pro audio usually specifies 600 Ω as a standard loading...

If you need that , it is easy enough to make at audio frequencies.

[egonotto]

Hello,

if you need to buy a new BNC cable, I would recommend a 50 Ohm cable.
You can connect the BNC cable directly. I don't see any problem with the RTB 2 input being 1 MOhm.

Best regards
egonotto

[goaty]

Regrading the input and analyzer outputs, these are buffered versions of the input of the analyzer before and after filters. The amplitude is typically between 0 - 2.83 Vpp. In my scenario im feeding the units own generator to the inputs which has an output impedance of 40 ohms.

I guess from quickly reading the manual, you can only monitor the waveform (but not measure a level in some kind) on the rear Monitor BNC Outputs.
For that, you can connect it to the scope directly, as the level ist of no concern.

[salomonander]

Thanks, it worked great!

[Darkover]

I found a bug! 

When I connect my current probe and I am measuring mA oder uA, why it can not show 0mA? Okay 0A is probably mathematical correct, but it disturb my sense of aesthetics.



Olaf

[2N3055]

I found a bug! 

When I connect my current probe and I am measuring mA oder uA, why it can not show 0mA? Okay 0A is probably mathematical correct, but it disturb my sense of aesthetics.

(Attachment Link)

Olaf

Is it a bug?
It is functionally correct.
It IS mathematically correct.
Reading is correct.

It is your sense of aesthetics that is offended.
Do not get me wrong, it bothers me too. I agree it would look nicer if units were the same.

But I would not call it a bug.
We can call it GUI quirk.

[nctnico]

I found a bug! 

When I connect my current probe and I am measuring mA oder uA, why it can not show 0mA? Okay 0A is probably mathematical correct, but it disturb my sense of aesthetics.

(Attachment Link)

Olaf

Is it a bug?
It is functionally correct.
It IS mathematically correct.
Reading is correct.

I don't think it is correct given the scale of the channel. 0A can be any reading between -0.499999999999A and 0.499999999999A when rounded to a single significant digit.

[voltsandjolts]

It's not a reading, it's a tick marker on the scale. 0 means 0.

[tv84]

It's not a reading, it's a tick marker on the scale. 0 means 0.

Naahhh... I think 0 mA is more zero than 0 A... The math realm is long gone. We're now into "aesthetics".

[2N3055]

I found a bug! 

When I connect my current probe and I am measuring mA oder uA, why it can not show 0mA? Okay 0A is probably mathematical correct, but it disturb my sense of aesthetics.

(Attachment Link)

Olaf

Is it a bug?
It is functionally correct.
It IS mathematically correct.
Reading is correct.

I don't think it is correct given the scale of the channel. 0A can be any reading between -0.499999999999A and 0.499999999999A when rounded to a single significant digit.

Amazing how you both ignored mathematics  (0 x 10e69 = 0, so any suffix is irrelevant) and excluded my statement that I would also like for it to be consistent because it would look prettier.

It is right up there ...

Did you get confused that you cannot accuse me that I shill for Siglent when advocating for proper use of word bug in R&S topic... 

Jokes on the side, these are not printouts for measured values. Those are legends for tick marks.  In which case a tick mark has stated value.

As for why this is correct, when you make 0 value mark, 0 means no value at all. Hence there cannot be suffix for exponent, it is zero..

A suffix in this case does not carry a value, but signifies that scale is expressing current, hence  A.

There would be also a valid option to just write 0 without a current signifier.  You could see it is mA from other values ...

But this is now a polemic on aesthetics of mathematical graph, like Tv84 says...
But it is not a BUG in my book.

[bayjelly]

Oh, philosophy.

It is not incorrect, but it may still be a UI bug. A "bug" is usually something that does not work as designed or expected.

For example, if a button is a few pixels too far to one side, making it the only button that's awkwardly hugging a frame, even if the button still works and remains readable, then it still may be obvious to most that this layout is neither intended nor expected.

If R&S did not design for nor expect that the tick mark shows A instead of mA, and if users don't expect it either and get distracted by it, you may well consider it a UI bug. You (and especially R&S) don't *have* to, but you/they *can*.

(There is no question that mathematically and semantically, "0 A" is not wrong. Lots of things are "not wrong" and still make for bad behavior in a product.)

[salomonander]

I have another question: is there a way to use the rtb for stability tests? I have an audio unit that had drop outs from time to time. I recapped it and now i would like to just run a sinus through it for some hours and have the rtb tell me, if there were any drop outs during the test. Is there such a feature? Thanks!

[Kean]

I have another question: is there a way to use the rtb for stability tests? I have an audio unit that had drop outs from time to time. I recapped it and now i would like to just run a sinus through it for some hours and have the rtb tell me, if there were any drop outs during the test. Is there such a feature? Thanks!

Should be possible with mask testing
https://www.testandmeasurementtips.com/mask-and-limit-testing-on-oscilloscopes-faq/

[salomonander]

Thanks Kean! Ill give it a try