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

[KungFuJosh]

This weekend I got an R&S SML03 to play with. So far, I like it:




I cleaned it up a bit, and it looks a lot nicer now. I opened it up and dusted it with some compressed air, and attacked the fan also. That helped reduce the fan noise a little, but it's still a loud ass fan.

Signal quality looks good, and it's pretty darn close considering it hasn't been calibrated since March of 2007. A 500MHz signal shows 499.9995MHz on my SDS2504X HD. 100MHz shows as 99.99990MHz. 10MHz shows as 9.999989MHz, which is about what the internal reference showed over an extended period of time. I may try to do a small adjustment.

Here's some useful info for the series:

Stock Fan: 622N EBM-PAPST
12V DC, 160mA (SML service manual specifies 12V, 170mA max)

Passwords (use at your own risk):
LEVEL 1 Password: 123456
LEVEL 2 Password: 261339
LEVEL 3 Password: 710845
LEVEL 4 Password: 593748

From the manual:

Password 1 deactivates the lock for the calibration of Main Loop, Level Preset, LFGen Level, Leve,
IF Filter, Harm Filter, Mult Filterl.
Password 2 deactivates the lock for the calibration of RefOsc.
Password 3 factory internal
Password 4 factory internal

ETA: I've attached the final firmware version 2.51.15 (with permission) for the SML series. USE AT YOUR OWN RISK!!! It requires a 32bit Windows version to be installed. It took 24 minutes to install for me.

ETA2: Power supply model LPQ154 datasheet added.

ETA3: Maintenance manual section including error codes added.

Thanks,
Josh

[KungFuJosh]

The only defect thus far appears to be a couple horizontal lines that aren't supposed to be there. This appears to be common with these.

Anybody know any tricks to fix those lines without replacing the screen?

I don't know what causes them, but there's a 3rd line that only appears occasionally, and I haven't seen it yet since dusting this thing.

Thanks,
Josh

ETA: Service manuals attached.

[KungFuJosh]

I'm not a fan of the stock fan, and I'm considering replacing it. It's a simple two wire connection.

Here's the specs:
Stock Fan: 622N EBM-PAPST
12V DC, 160mA (SML service manual specifies 12V, 170mA max)

Operating voltage: 8-15V DC
Fan dimensions: 60x60x25mm
Fan efficiency: 40m³/h (~23.5CFM)
Noise level: 35dBA
Rotational rate/speed: 6100rpm

https://www.ebmpapst.com/us/en/products/compact-fans/axial-compact-fans/p/622N.html

What do you think about their air performance curve chart? I wish it specified the voltages on the chart too. Is 40m³/h happening at 12V or 15V?


I was thinking about using this as a replacement:
Noctua NF-A6x25 FLX
12V DC, 80mA

Operating voltage: 12V DC
Fan dimensions: 60x60x25mm
Speed: 3000 RPM
Airflow: 29.2 m³/h (~17.7CFM)
Noise level: 19.3 dBA

The CFM difference is less than 6. Is 6CFM enough to worry about? The significant noise drop would be nice.

Thanks,
Josh

[KungFuJosh]

I think it's cool that the SML can do two three modulations at once.

Here's a 500MHz sine:


Here it is with FM modulation:


Here it is with AM modulation:


Here it is with both:


Thanks,
Josh

[tautech]

Amplitude modulation sweep too ?

SSG3021X can do and you can set up some eye catching effects for exhibitions.
You have to do a video to really show that effect.

[KungFuJosh]

Amplitude modulation sweep too ?

SSG3021X can do and you can set up some eye catching effects for exhibitions.
You have to do a video to really show that effect.

I'd love an SSG3021X. Send me one! 😉

[KungFuJosh]

Holy Shit, Batman! I love how easy it was to adjust the calibration with this. 🤯

You activate the Adjustment State by setting it on, and then modify the Frequency Adjust setting. That's it. It was set to zero. I connected it to the counter on my SDG2122X and adjusted it there:





Now the waveform counters are all spot on:
Adjusted reference output from SML03:


10MHz RF Output:


500MHz RF Output:



Thanks,
Josh

[KeBeNe]

HI,

try to install an original fan or at least a fan with the same values.
The R&S devices get very warm internally, the Noctua fans are nowhere near the static pressure.
The components usually die slowly due to the heat (in my experience)
You can also look at Sanyo Denki, they also make good fans.
I like to use them and if possible with more thickness, they can build up more pressure with the same noise.

the SSG3021x have a bad phase noise and only go down to -110dbm level, the SML goes down to -140dbm, the phase noise is ok so far, not high end, but normal for the class. (this was/is the “beginner low budget” class)

If you can find the calibration software (service kit) somewhere, I would be grateful if you could pass it on.

[KungFuJosh]

There's a 25% difference in airflow between the two fans at their stated specs. I'm curious what the stock fan actually runs at though, since it states operating voltage up to 15VDC, and the SML runs it at 12V. I might try a photo tachometer or stroboscope or something to see what the real RPMs of the stock fan are.

Regarding static pressure, I don't really understand that part too well, and the actual airflow/RPMs are still in question with the stock fan. So here's my question: what's the real difference? Stock fan air performance chart says at 30 m³/h, static pressure is under 0.1in H₂O. The internet supplied conversion chart says that's a little bit over 2mm H₂O. The Noctua fan is rated 2.18 mm H₂O at 12V. Seems close to me?

Thanks,
Josh

[Martin72]

Hi,
I am responding to this post here because it fits better here.

https://www.eevblog.com/forum/testgear/siglent-sds2000x-hd-12bit-(published-for-chinese-domestic-market-only)/msg5904862/#msg5904862

10kHz: 208.6mV = -0.603414 dBm
100kHz: 179.4mV = -1.913251 dBm
500kHz: 131.8mV = -4.591392 dBm
1MHz: 122.1mV = -5.255387 dBm
10MHz: 143.36mV = -3.86114 dBm
100MHz: 122.9mV = -5.198662  dBm
500MHz: 98.5mV = -7.120975 dBm

Phew...
Is there another filter active, do you have another filter in between?
There should be a self-test, what does it say?
Which output are you connected to?

[KungFuJosh]

10kHz: 208.6mV = -0.603414 dBm
100kHz: 179.4mV = -1.913251 dBm
500kHz: 131.8mV = -4.591392 dBm
1MHz: 122.1mV = -5.255387 dBm
10MHz: 143.36mV = -3.86114 dBm
100MHz: 122.9mV = -5.198662  dBm
500MHz: 98.5mV = -7.120975 dBm

Phew...
Is there another filter active, do you have another filter in between?
There should be a self-test, what does it say?
Which output are you connected to?

I'm not using any filters, or terminators. 50Ω termination is set in the scope as usual. The SML03 has auto level correction (called ALC in the menu system).

It's my understanding that ALC is supposed to apply the calibration settings for different ranges. It makes little to no difference whether this setting is on or not. Turning off ALC gives the option of either applying a correction Table, or Sample&Hold. Neither improved the level in any interesting kind of way.

ETA: There is an Attenuator listed in the settings. That is either set to auto or fixed range, and the fixed range is preset and can't be edited. Neither option does anything when tested at a fixed frequency.

There is a Ucor (User Correction) setting I want to try out, but I need to be able to accurately measure and correct this at least up to 1GHz for it to be meaningful for the bw test. I obviously can't do that with my 500MHz scope. So how can I test this accurately? My DAQ6510 seems to be the best option, but I'm not sure if that's sufficient either. Do I need a power meter?

Thanks,
Josh

[Martin72]

Hi,

My DAQ6510 seems to be the best option

Its bandwidth will not be sufficient for this.

"Unfortunately", my SML had been functioning properly in this regard, so I am unable to offer any additional assistance beyond general tips such as using the correct output (RF not LF, then switch on RF in the menu as well) and performing a factory reset.

[KungFuJosh]

Hi,

My DAQ6510 seems to be the best option

Its bandwidth will not be sufficient for this.

"Unfortunately", my SML had been functioning properly in this regard, so I am unable to offer any additional assistance beyond general tips such as using the correct output (RF not LF, then switch on RF in the menu as well) and performing a factory reset.

That's what I figured regarding the DAQ. I'm not sure what power meter would be sufficient, but maybe I'll attack eBay...or see how much cal/adjustment would cost.

The LF output is DDS that only goes up to 1MHz. I've also done a factory reset.

Thanks,
Josh

[KungFuJosh]

With the output level set to +5 dBm, this is what it looks like on my scope. A level adjustment should take care of things nicely. Last calibration was in 2007, so there's that.

[KungFuJosh]

Interesting! I found the level calibration list that was in use. That's exactly what the problem is. The Level cal table is 18 years old, and the values are stupid. 🤣

[KungFuJosh]

I got the last firmware installed (was 3 versions behind), but it's a 16bit app with a 32bit installer. So WineVDM was useless (it can theoretically run the 16bit app, but can't touch the 32bit installer). I had to find an app to install 32bit Windows on an external drive so I could run it from my wife's old laptop that I'm not allowed to format.

It took 24 minutes to update the firmware, which was hilarious because the SML says FASTLOAD on the screen the whole time. I don't want to know how long SLOWLOAD would have taken! It's a good thing I'm patient sometimes in my old age. 😉

Thanks,
Josh

[pdenisowski]

It took 24 minutes to update the firmware, which was hilarious because the SML says FASTLOAD on the screen the whole time.

 

[KungFuJosh]

After running the firmware install, the internal calibrations were cleared (as expected). What's interesting is that in clearing those level lists, it's already a lot closer to 0dBm, and more linear...except for a few frequencies that put out nothing. 🤔

I haven't run the auto calibrations yet; I'll do that soon.

Right now, I've unlocked Level 2 so I can do the 10MHz ref calibration (without using the user override option). The SML warmed up for over an hour, and it seems stable enough.

Final firmware added to OP.

[KungFuJosh]

Well, that worked!

After installing the new firmware, calibrations cleared, ref adjusted, then running calibrate all- it's MUCH better.

It's pretty stable at 220mV (within -1mV) from 9kHz all the way to 100MHz. Then it drops as follows:
200MHz 213mV
300MHz 201mV
400MHz 191mV
500MHz 176mV
600MHz 136mV

I don't have a proper RF power meter to determine whether that's the scope's limitations or the sig gen, but I assume the former or a combo of both. I might need to get me an RF power meter now lol.

Thanks,
Josh

[Martin72]

Hi,

Congrats...
Now do a sweep....

By the way:

500MHz 176mV
600MHz 136mV

That could be right; the -3dB point of the scope is somewhere between the two frequencies.
Taking 220mV as a basis, that would be 220mV* 0.707, or approximately 155.5 mV.

[Grandchuck]

Have you tried other coax cables ... I find they vary some at the higher frequencies.

[KungFuJosh]

Hi,

Congrats...
Now do a sweep....

Thanks! It's running now. 😉

That could be right; the -3dB point of the scope is somewhere between the two frequencies.
Taking 220mV as a basis, that would be 220mV* 0.707, or approximately 155.5 mV.

That's what I assumed...and why I put an offer on an RF power meter. 😉😉

[KungFuJosh]

Have you tried other coax cables ... I find they vary some at the higher frequencies.

Right now I'm using an RG316 BNC cable with a type N to BNC adapter. I'll also try a cable that's coming this week that's RG142 type N to BNC (no need for adapter), and I'll try an RG58 BNC cable with the N to BNC adapter.

[KungFuJosh]

There's a 25% difference in airflow between the two fans at their stated specs. I'm curious what the stock fan actually runs at though, since it states operating voltage up to 15VDC, and the SML runs it at 12V. I might try a photo tachometer or stroboscope or something to see what the real RPMs of the stock fan are.

Regarding static pressure, I don't really understand that part too well, and the actual airflow/RPMs are still in question with the stock fan. So here's my question: what's the real difference? Stock fan air performance chart says at 30 m³/h, static pressure is under 0.1in H₂O. The internet supplied conversion chart says that's a little bit over 2mm H₂O. The Noctua fan is rated 2.18 mm H₂O at 12V. Seems close to me?

Does anybody else have any thoughts on this? I have an RPM tester coming today. I'll see how fast the stock fan is actually running.

[KungFuJosh]

Stability of the internal ref seems okay.



ETA: Updated photo.