Rohde & Schwarz CMU200

[Alfons]

I also bought a CM200 now. I only need the device for a few measurements on filters and to repair an old spectrum analyzer, otherwise I have really little to do with RF. I had looked around for alternatives, but good, functional single devices would have come much more expensive. In this case, I did not want any repair equipment, because I have no proper measuring equipment available for repairs of RF equipment. And the game with the whole China gadgets has annoyed me increasingly.

So I bought a CMU200 for 715 € (inc. Purchase), a lot of money (for a pure base unit, as I later discovered ). However, the seller has assured me, that the device is faultless. Only after the purchase, I noticed that it is a 1100.0008.30, so a basic device without any hardware options. The seller had agreed to a cancellation certainly, but then let it come. So it has only the generator, the analyzer and the power meter (saves electricity costs:)).

I then found that it works well and is completely sufficient for my purposes. I could not find any errors, all self-tests are "passed", the values ​​really good. The buttons work like the first day and the display is clear and bright. I'm impressed. It is easy to operate and even if the device is not a specialist in any of the disciplines, the performance for a relatively "new" device at the price is very good in my view. OK, the right ham radio hobbyists have more knowledge and will probably prefer single units. After the repairs, I'll probably sell it again. Maybe I will keep it too. Have fun with the device.

[Bicurico]

The other options, which you don't have, are completely useless, unless you want to specifically measure GSM/GPRS/Edge mobile communications.

You don't want to do these, because you should not transmit in the respective frequencies to begin with.

So be happy that you purchased a fully working CMU200. The price you paid is a great price, so no worries here, either.

The only option that might be somehow interesting (INHO) would be the Audio Measurement option. But again, you could probably do them in a more versatile way with a good PC audio card.

Check my blog for my "VMA Simple Spectrum Analyzer" software, which is available for the CMU200 & CRTU, too.

Regards,
Vitor

[Alfons]

Yes, I already know your software and will try it out in the next few days. OK, so the price for the device was OK. Actually, I wanted to update with the latest software, but the device don't accepts either the card reader or the CF card. Is formatted in FAT. Although I can see the files in the file manager, the version manager does not want to read the card.

Edit: Did not know that you have to extract the files. Now the update is running ...

[usagi]

as davey would put it, for the money the cmu200 is a real bobby dazzler 

even barebones it is impressive. but then it should be considering they originally ran anywhere from $20k-$100k depending on options.

[markus_jlrb]

@Alfons,

check out in addition the link in the german MC forum too in order
to find usefull infos concerning the CMU200.

https://www.mikrocontroller.net/forum/all?filter=CMU200*&x=0&y=0

Markus

[Alfons]

@Alfons,

check out in addition the link in the german MC forum too in order
to find usefull infos concerning the CMU200.

https://www.mikrocontroller.net/forum/all?filter=CMU200*&x=0&y=0

Markus

Of course, I have already read everything there about the device. And since I was waiting for the device, i already read everything worth about the device here on EEVblog and in the MC-forum So I was able to serve the device relatively quickly. I have just tried FreRes: really great, very useful.

What I've noticed is that you really need to have reasonable cables and connectors for the instrument to deliver accurate results. I'm not fortunate enough to have a few very high quality cables shipped from China some time ago. With my old cable, I could not perform the self-test error-free. Yes, I am new to RF.

[Alfons]

Had posted the question elsewhere, but I'll ask here:

In my MCU200 (1100.0008.30) the option cmu-U99 RF1 = RF2 limits the output power of RF1 to 2W. There is no second signal board installed, which usually sits where the heat sink is installed on the front end. The frontend in my Device is the "thick-one" with heat sink and fan. So there must be another reason why U99 is active and the RF1 / RF2 is limited to 2W. I do not think you can disable the cmu-U99 option. Or is that possible?
The 2Watt are also OK, you just have to be more careful with what you connect.

[markus_jlrb]

Alfons,

what you are looking for for high RF-Power output is showed on the top left position
of the picture linked below.

https://www.ebay.de/itm/Rohde-und-Schwarz-R-S-CMU200-Universal-Radio-Communication-Tester-Bastelpaket/174039616839?epid=1627651710&hash=item288591cd47:g:LXkAAOSw81Bdh3TT

You have to look for a corresponding spare PA module at Ebay.

Sorry to say.

Markus

[Alfons]

Alfons,

what you are looking for for high RF-Power output is showed on the top left position
of the picture linked below.

https://www.ebay.de/itm/Rohde-und-Schwarz-R-S-CMU200-Universal-Radio-Communication-Tester-Bastelpaket/174039616839?epid=1627651710&hash=item288591cd47:g:LXkAAOSw81Bdh3TT

You have to look for a corresponding spare PA module at Ebay.

Sorry to say.

Markus

I don't understand you: Why should I need a spare part? The front end works and it is the same as the one shown (at least externally identical). It is limited to 2W by the firmware, not by the hardware. At least I think so.  I think that another 50W frontend would not change anything.

And 33dBm should be enough for almost everything.

[Alfons]

Here is a graph from Freres, -30dBm between 100MHz to 2.7GHz. Is this ok? Measured with a 1m coax with adapter N-BNC between RF1 and RF3-Out. With better cables, you can certainly get a little more out and you have to rely on the fact, that the values ​​are correct. I can not compare, because I have no other measuring devices that can RF. As the level increases, the frequency response deteriorates by up to 1dBm, but is still within 2.5dBm. Between 100Mhz and 1Ghz the frequency response is straight.

[Sas]

Here is the graphic of my two cmu200. The cable is 0.5m RG316 N-N between RF2 IN and RF3 OUT.
The first CMU200A calibrated in 07.2017. CMU200A in the middle of 2018 had a deviation of not more than +/- 0.5 dBm. Currently it has worsened and has a deviation of ~ -1.5 dBm.
The second CMU200B was calibrated in 08.2015. I bought a damaged one. Missing dsp module, damaged disk, several cables, gsm modules were damaged. We managed to fix and run everything. The device does not pass the RF Loop Path 1 test correctly. The values are too low from - 2dBm to -5dBm. I can't calibrate it myself.

[Alfons]

Create a correction table yourself. There are instructions for this.

[Sas]

RF Loop Path 1 CMU200 with calibration 07.2017. A year ago it was +/- 0.5 dBm. Now it is +0.8 dBm / -0.9dBm. Correction table does not work in the RF Loop Path 1 test.

[Alfons]

@Sas: Is it possible to use one device to calibrate the other?

[Sas]

@Sas: Is it possible to use one device to calibrate the other?

Not by substituting correction tables for individual modules from another device. You can try to modify the correction tables of individual modules. The editing software should already be used in the device for factory calibration. I don't know any service mode or calibration method.
Correctly you would have to calibrate as R&S does. They probably use a good laboratory spectrum analyzer and generator. Another CMU200 is not a very good reference standard. I guess the calibration is limited to connecting, heating up and using ready-made software via GPIB or COM connector.

[Alfons]

I am an absolute beginner when it comes to SA. The device is great for learning. It is the first time that I am dealing with a (functional) SA and due to the variety of functions you do not need any other devices to try it out. Only slowly am I beginning to understand how an SA works at all, and now do I know what a reasonable spectrum has to look like and how it is also possible to generate a plausible FFT with a DSO. Because with some settings you can get quite good FFTs on a DSO. But extremely incomprehensible compared to the CMU200. Nevertheless, I am amazed to see that my old Infiniium offers a few options-settings, that you would otherwise rarely find on an old DSO for generating spectra (center frequency, span, power and phase and the possibility of setting the sampling-rate independently of the timebase-settings and the memory depth to influence the signal display ...). I only understood that, after I dealt intensively with the CMU. Because of this learning effect alone, it was worth buying the CMU200.

Nevertheless, there are still a lot of questions for me. I'm trying to understand what the I / Q-measuring on the device do. I/Q vs. Slots to measure the Phase-data in a signal-package, which is available in the basic menu and therefore not only to be used for measuring certain GSM etc.. signals?. This is the only option that my device has, which is also available in the basic menu.

Another question is: what does an OCXO do compared to the built-in TCXO in General-Purpose Measuring ? A Chinese dealer offers option B12 for around € 30. Does it make sense to retrofit that? I don't think so, because the accuracy of the TCXO is quite high. But what is your opinion?

[Ice-Tea]

The OXCO offers better timebase stability in the long run (will drift less over time) but it also drifts less with temperature variations in you lab etc. Unless you *know* you need it, I wouldn't bother. Accidently, another options is to use an external 10MHz reference. Which can be quite cheap as well.

[Alfons]

Thank you, Ice-Tea, I think I'll leave it that way and run it with the TCXO. At the moment I really don't know what I need an OCXO for. Highest to upgrade the device to get a better sales price. But I think I'll keep it anyway.

Here are a few more pictures: RF1-Out to RF4-IN. It generates and shows even very small signals quite well. But I don't really know if the measurement is OK. I switched off Avarage once, but that doesn't change much.

[Sas]

I don't have B17. I think it is a set of analog inputs and outputs of intermediate frequencies for external modulators and demodulators. There is also a set of IQ modulation inputs and outputs. I think that it is not very useful when working with analog modulations. Without the B17 module, only the 10.7MHz IF output of the receiver. This is the IF3 output after analog RX conversion and the analog input of the DSP module for digital analysis. Sound demodulation with analog modulation should not be a problem.
The TCXO oscillator is sufficient for a spectrum analyzer with RBW 10Hz. With the built-in OCXO, if you need to have a more accurate reading, you'll still use the external time base input for phase synchronization with other instruments.
In my first CMU200 there was already an OCXO generator at the time of purchase. I bought ocxo for the second CMU200 because of the pretty good tx generator with AM, FM, FM Stereo modulations. I also use the generator as a LO or RF signal mixer for the tracking generator. I use CMU200 as a working tool in an electronics workshop, it is not super accurate but universal, convenient interface and everything at hand. I do not regret buying OCXO. The original OCXO R&S module B12 for € 30 is a very good price. If you plan to sell it is probably an unnecessary investment.

[Alfons]

OK, I can try it out. But it has become a little more expensive? Well, 35 € is still OK. This one here: https://de.aliexpress.com/item/4000433902250.html?spm=a2g0s.9042311.0.0.672a4c4dukbieC

[Sas]

https://de.aliexpress.com/item/4000433902250.html?spm=a2g0s.9042311.0.0.672a4c4dukbieC

Wow 499 pieces! It's madness  . Is the choice which item will ship?
Pay attention to this auction for example https://www.ebay.com/itm/123957032489. One of these two B2 / B12 is damaged, the diode or resistor on the pcb is missing. This element is probably responsible for temperature measurement. It is easy to tear off during assembly and disassembly, as it is on the housing line when inserting and removing the ocxo. The fixing plate at the pins will also be useful.

[Alfons]

I often order things in China, mostly small things, but sometimes things that are a little more expensive. I've never had trouble. Defects are very rare and if you do, you can get your money back. I think buying is not a risk. Where did the dealer get 500 of them from? Perhaps R&S has already had production in China and are these remaining stocks? Fakes would not be worth it. I will see... Also: one bad rating and the dealer no longer sells more of this B12.

[Sas]

Perhaps these are unused or removed parts that were sent for recycling to China and went on sale again at a lower price. Write as you receive and it will work.

[Qw3rtzuiop]

These parts are probably from CMU200 that were used in china. A lot of CMU200 are used for cell phone prodution. Just take a look at CMU200 listings on Taobao.

[Alfons]

For 14 € I bought a cable on Ebay, which is equipped with Suhner plugs. Cable is F&G 213/U. Probably comes from military stocks, but in very good condition. With screw caps
I wanted to see, if anything actually happened in the frequency plot, connect with this cables. Yes, the frequency response is actually significantly better, now lies within -1.2dBm at -20dBm. That looks better.