Rohde & Schwarz SML Series Signal Generators - SML01 SML02 SML03 SMV03

[KungFuJosh]

Please look under utilities, there you will find the item Install for SME and SMIQ.

This is the SML. There is no item Install anywhere on any of the menus. At least not that I could find. See attached. Maybe a hidden menu? That is possible, but doesn't seem likely. Maybe it's related to passwords 3 and 4 that they don't give us any details about. Still seems unlikely though since B3 showed up on the other mainboard.

Yes, if you install the B1, you do not need a key, this is the case for all devices, but it is for PM, even if the HW is installed, for the other RF generators this is the option B11(HW). this also includes software keys.

B3 PM module shows up on either mainboard. It gave an error once on the mainboard that didn't come with it, but that was more likely me doing weird settings (or an issue with the hardware?). It shouldn't have shown up, or allowed me to do anything if an install key was needed. I could be wrong, I dunno. I'm not plugging back in the other mainboard to find out at this point. 🤷 I will eventually screw around with the B3 module and see if I get that same error again.

Calibration, the self-calibration has nothing to do with the factory calibration.
The factory calibration is carried out over the entire RF chain and sets a reference point; this reference point is used as the target for the user calibration.

Yes, I understand that. The self-calibration does what it's supposed to do, and corrects the levels, among all the other cal stuff it does. It's a great feature so that you don't have to ship it out to a cal lab every 3 months. 😉

Thanks,
Josh

[KungFuJosh]

I was able to recreate the error with the motherboard that came with the B3 option. It's odd. The Pulse mod comes up in the menu, but it doesn't seem to do anything.


Pulse Period, Width, and Delay all say Option Absent.


Selecting one of them shows "parameter not available."


But then it shows the current frequency there for some reason. Maybe a 20 year old bug?


Then when I turned on the Double Pulse State option, it gave the error (same as other mainboard).


So I restarted the SML holding down the 1 key (which seems to clear errors), now the options for Pulse show up, and can be edited without issue.


I set the values randomly and pretty high, so now when I turn on Double Pulse State I get an error.

[KungFuJosh]

I RTFM and it didn't help. 

Thankfully, Paul has an awesome video (as usual):


This one too: video (embedding is disabled in that one).



Now I understand it a little better, and got rid of the error.

Thanks,
Josh

[KungFuJosh]

Aaaaaaaaaaaand the Pulse generator works fine...now that I know how to use it. 🤣

Thanks for the awesome videos, Paul!

[KungFuJosh]

Well, I got a 3rd one. 🤷🤣

This one had a dead screen, so I took the screen from unit 1, and poof, it's awesome. I think this one was more recently calibrated than the other 2 units.




I'm going to sell this one, and unit 1 for parts and keep unit 2 with the B1/B3 options. Or maybe keep this, and just sell the parts unit. 🤔 TEAS is real.

Thanks,
Josh

[KungFuJosh]

I dunno if this is helpful / useful to anybody, but, here's a few images of what the firmware update process looks like. This other unit was faster by a whopping 19 seconds!

I also grabbed the temp folder from the firmware updater app. Might be easier to update on a modern computer that way. 🤷

[KeBeNe]

Hi,

As I said, I don't have an SML.
But you have re-initialized the device, which must have re-read the assemblies, good advice, maybe I will come across an SML03 at a reasonable price.
I look forward to hearing more about your experiences.

[KungFuJosh]

As I said, I don't have an SML.
But you have re-initialized the device, which must have re-read the assemblies, good advice, maybe I will come across an SML03 at a reasonable price.
I look forward to hearing more about your experiences.

Regardless, your input is appreciated! I think you'll like the SML if you find a good deal. It seems good at what it does.

Despite its large footprint, I also like the design choices R&S made. The footprint was obviously to make it fit in a rack. The modular design and layout of everything is brilliant. It still stands up pretty well 20 years later.

Thanks,
Josh

[KungFuJosh]

Yesterday I was playing with the pulse generator "video out" which directly outputs the set pulse. Today, I played with the pulse modulation.



It gets weirder when you combine Pulse + AM + FM modulation.



Messing with these settings also helped improve my triggering skills, which was a nice bonus. 😉

Thanks,
Josh

[tautech]

Yesterday I was playing with the pulse generator "video out" which directly outputs the set pulse. Today, I played with the pulse modulation.



Messing with these settings also helped improve my triggering skills, which was a nice bonus. 😉

1st grade simple stuff ! 
When you need a Zone trigger or 2 things are really getting interesting..... 

[KungFuJosh]

Messing with these settings also helped improve my triggering skills, which was a nice bonus. 😉

1st grade simple stuff ! 
When you need a Zone trigger or 2 things are really getting interesting..... 

If you knew what I was actually referring to, you'd call it kindergarten. 😉

[KungFuJosh]

Is it correct that those RF switches marked 4227T are actually SSW-124?

[KungFuJosh]

More modulation fun.



SML03 settings:
100kHz -10dBm
AM Modulation: 80%
Phase Modulation: 1.800 rad
LFGen Freq: 100Hz

[KeBeNe]

Hi,

Yes, that is correct.
You can check if the switching voltage is +/-6V, R&S has sometimes operated them with up to +/-8V to improve the isolation, which is outside the specifications.
An exchange for HMC270 would be conceivable, but you would then need the calibration software

[KungFuJosh]

Yes, that is correct.
You can check if the switching voltage is +/-6V, R&S has sometimes operated them with up to +/-8V to improve the isolation, which is outside the specifications.
An exchange for HMC270 would be conceivable, but you would then need the calibration software

I found a source for the 124s, but they're $6 a piece...which isn't terribly terrible.

I might experiment with the parts unit with the dead output module (attenuator).

That calibration software doesn't seem to exist anymore. I really hope somebody finds it and shares it with us...but I won't hold my breath. 😉

Thanks,
Josh

[KungFuJosh]

Should I be concerned about this intensity fluctuation at all? Do I need to improve filtering or something in the SML? I see similar movement from the SDG, but the intensity color grading is generally consistent.

[kmoonwalker]

you can check SML output with some spectrum analyser to see if there isnt something faint coming out or some tiny instability of main frequency

[pdenisowski]

you can check SML output with some spectrum analyser to see if there isnt something faint coming out or some tiny instability of main frequency

Agreed.  Does the scope have an FFT?

[KungFuJosh]

Of course! Siglent has a good FFT, far better than my skills of utilizing it.

Great ideas, thanks. Now to figure out the test setup, and avoid temptation to buy better SA... 🤔

[KungFuJosh]

How's this look? Did I set it up correctly?

SML03:



SDG2122 for comparison:



Thanks,
Josh

[kmoonwalker]

hmm you written some time ago about it being able to modulate the waveform etc - did you disabled all tricks from settings after playing with fancy signals?

other than that something maybe broken (or cable or something on signal path is not good enough and scope reacts to that)


also thing that I noticed (but it maybe due to not some huuge signal amplitude and scope approximation) - the amplitude is fluctuating a bit - from my university time old Rohde generator was so steady we tested all new ones against it and they usually failed ^^

[KungFuJosh]

hmm you written some time ago about it being able to modulate the waveform etc - did you disabled all

😉

also thing that I noticed (but it maybe due to not some huuge signal amplitude and scope approximation) - the amplitude is fluctuating a bit - from my university time old Rohde generator was so steady we tested all new ones against it and they usually failed ^^

I think you hit the answer right there. I connected both gens with 50MHz 0dBm signals to my scope. Look at the Vrms stats, that looks pretty clear. A small difference, but a noticeable one, and not in the direction your experience would suggest.



It's probably obvious from this conversation that C1 is the SML03, and C3 is the SDG2122X.

Now, the question is, how can I stabilize that voltage further?

Thanks,
Josh

[kmoonwalker]

now this is more like Rohde I remember - glad you have it pinpointed

I did asked about setting due to fact that often with layered menu we do not see everything like on old equipment with knobs etc on front panel so check was a quick look - nowadays it is a good practice to twice/triple check everything in case of crucial measurements with expected result

[kmoonwalker]

I would need to dig into complete specs of that generator but keep in mind that there are several factors:
- not so big amplitude of signal so lot of things can impact it and will impact it
- results are calculated so oscilloscope noise may play bigger impact than one could think for result
- scope has limitations on approximation and thus calculations
- your cabling may receive something even from your body, some EM, etc
- something maybe coming thru filtering of generator/scope both from mains...
Generator is old so filtering in PSU maybe maybe not great nowadays (but it should).

There is some crosstalk between scope channels - you have some 2000x HD model if I remember correctly so it has a nice dampening compared to most but this can also play a factor in some cases.

There is a reason why we use one and "best" channel in scope in a room with stabilized mains power, controlled EM etc to make small values measurements (thou usually smaller than this)

but with this we are getting into metrology area with all its oddities, fancies etc and sometimes the question is if you really wanna know or can achieve it

You may also develop an sickness/need for top notch equipment just to be satisfied but it may not play a role at all on daily tasks.


For example I would not have paid any attention to such small variation on my old 1054Z that I use for usuall stuff and it is more than enough for it - but its resolution is to small to see things. If I would shown you side by side for example HD314MSO with 14bits we could estimate which scope does a better job or second unit like yours to see if this is an expected behaviour for this ADC and maths conversion. Thing is that you have a very nice scope but diggin too much you may stop liking it. This applies to every scope on planet if you find a proper angle

[KungFuJosh]

I would need to dig into complete specs of that generator but keep in mind that there are several factors:
- not so big amplitude of signal so lot of things can impact it and will impact it
- scope has limitations on approximation and thus calculations
- your cabling may receive something even from your body, some EM, etc
- something maybe coming thru filtering of generator/scope both from mains...
Generator is old so filtering in PSU maybe maybe not great nowadays (but it should).

M&P connectors and Hyperflex 5 cable used for the tests.

The scope's voltages, etc. have all been accurate, and in agreement with the V3500A RF power meter.

I tried two different PSUs, and that didn't affect the fluctuation at all. Both looked generally the same, though were also around the same age.

There is some crosstalk between scope channels - you have some 2000x HD model if I remember correctly so it has a nice dampening compared to most but this can also play a factor in some cases.

There is a reason why we use one and "best" channel in scope in a room with stabilized mains power, controlled EM etc to make small values measurements (thou usually smaller than this)

All tests prior to the Vrms comparison were done on C1 with nothing else plugged in. The Vrms comparison test is using C1 and C3 as they don't share the front end toys.

but with this we are getting into metrology area with all its oddities, fancies etc and sometimes the question is if you really wanna know or can achieve it

Let's stay just shy of that rabbit hole. 😉


Thanks,
Josh

PS. You can see in the attached image that it's very clear on this series when modulation is active. It shows that modulation is on, and which types are active (up to 3 at once).