Rohde & Schwarz CMU200

[tech255]

The model number ending in .30 is a production line model normally banked in fours and is used to program and calibrate mobile phones as they move along the production line there is something embedded in the software that stops a lot of the options from working believe me i have tried.
After purchasing several scrap units i found one with a hard drive which was a .02 model this one is the best for modifications i now have every conceivable extra added and all of the latest software v5.22 i also have 50w input on channel 1 and +13dbm output from channel 3.
The audio board option B41 does not give you audio from the speaker its not available its for future development which is never going to happen, took me a while to work this out but it is possible to use the rf and audio together to get sinad and distortion readings but you need the Amps boards and software.
If anyone needs the latest software i can send it to you, i dont have a website you can obtain it from Rohde & Schwartz sent it to me after i contacted them about a fault.

[Alfons]

The module arrived today. For some time now, the shipping time from China has been significantly reduced. It usually takes about 2 weeks for parts that are a little more expensive. Only small parts take longer.

As described, the part is used, which is also clearly recognizable. But it looks undamaged, attempts of repair are also not visible. AR electronique, the manufacturer of the OCXO, is a French company, that specializes in the manufacture of quartz resonators, oscillators and piezoelectric filters. At least that sounds trustworthy.

If I have installed it, I will report. Next week I have a GPSDO here for a few days, so I can also check and adjust the frequency.

[Alfons]

Well, the installation was done in five minutes. The device recognized the module as B12 during booting, no further steps were necessary.

It seems to work. After a quarter of an hour warm-up, the signal shows an improvement. I still have to see, if I have comparison images, but the phase noise is obviously lower. Before that, I had a crater-shaped collar at the top of the bell. This has disappeared and the phase noise has decreased by about 7-10dBm (but only from my memory). I am really amazed to see that. In addition, the frequency display is now more accurate and the selective power meter shows the frequency in the last places much more quietly and precisely. So, it looks like the purchase was really worth it.

Edit: Since I was a little too quick, here is a Correction: the measurement accuracy has actually improved, but the phase noise is at best 3 dBm better, not 10 dBm. But I have to check it more closely. At least I can say that the module works well.

[heavenway]

Hi everybody!

new to this forum and really apriciate so much enthusiasm and competence, was wondering if some of you use the CMU200 in ham radio lab or for receivers/transmitters troubleshooting and align: will be glad for some advices in the case.

I am in the willing to buy one but... the only (and not small) limit it has is the spectrum scope that starts at 10mhz, and its a big limit for me.

cheers

[Ringmodulator]

Hi,

there is a solution for this:

Look on Ebay for CMU200 10MHz Spectrum Analyzer Expansion for Radio Communications Test.

It is just a simple mixer. Works for me.

Chris

[heavenway]

thank u Chris, seems to be upconvert dc-10mhz signals or am i wrong? still looking for a good unit to my purposes .02 of course 

[Ringmodulator]

Yes, it uses the CMU200 RF out as LO at 2GHz.
0-10MHz will then be upconverted to 2.0-2.01 GHz and you can display the spectrum there.

There are similar mixers available on ebay with a shielded housing, which will also do the job.
Then you need cables and adaptors seperatly.

Regards
Chris

[heavenway]

need te CMU200 first 

[balage]

I think the Ebay mixers have the HMC213 IC from Hittie/Analog Devices. Can anybody confirm this?

I am building my own, with a self-designed PCB. I think I will sell out the remaining PCBs.

[Ringmodulator]

Hi,

mine is on a BG7TBL board and uses a HMC213.
There is nothing else beside this one and the SMAs on the board.

Chis

[Ringmodulator]

Hi balage,

you can get them ready made with HMC213 for under 20$ delivered.
https://www.ebay.com/itm/HMC213-RF-Passive-Double-Balanced-Mixer-Diode-Frequency-Conversion-Module/264330247742

Chris

[balage]

Hi Ringmodulator,

Oh, if I would have found the enclosed earlier, I would bought one. Especially HMC213 costs 15 bucks. 

Anyway, it was fun to design my own board...

[Sas]

@tech255 I have two .02 CMU, one older with PS2 and newer with USB. Both units are removed from the production line, have a generator +13dBm RF3 OUT and only 2W RF1 input. They also have a second generator, B41 and all GSM modules.

@Alfons B12 looks good. I don't know this Pulsar 510 oscillator. I think it is low phase noise ocxo. I calibrate my ocxo once a year, the error is less than 0.1Hz @10MHz.

I use the Anzac (macom) MDC-162 mixer. I amplify the IF from RX / TX module with a 40dB @2-4GHz old cascade amplifier mounted inside and powered from the CMU. At the amplifier output I have about +10dBm, good level to control the LO input of the mixer. Mixer conversion losses around 8dB, RF3 OUT generator output +13dBm level is enough to get 0dBm. I miss the AGC, because of course CMU has no normalization function for tracking generator.

[Alfons]

Have now tried the B12-OCXO to compare with the GPSDO as a reference. After the GPSDO had been in operation for three days and nights continuously, I tried to take the phase curve on the oscilloscope to cover. Even without adjustment, the B12-OCXO signal was quite close to the GSPDO and only migrated very slowly.

The best thing I can set is that the signals move against each other at around 20ns per hour. If I turn the spindle trimmer just a mini-mini-tick, the direction changes, I can't find a sync. So there is nothing more. I might be able to get slightly better results by changing the value of the spindle trimmer, but I don't think that will work for long time. I guess everything will be slightly shifted again after restarting the GPSDO.
If I connect the GPSDO to the input of the CMU200, the integrated counter shows me the 10MHz down to 9 digits, the tenth digit oscillates between 1 and 0, the 11 digit fidgets. That is stable.

[Alfons]

After the signal has moved to the right for approx. 24 hours, it now runs to the left. That commutes back and forth about every 24 hours. The phase shift can hardly be seen with the naked eye. Is this commute normally?

[Sas]

After the signal has moved to the right for approx. 24 hours, it now runs to the left. That commutes back and forth about every 24 hours. The phase shift can hardly be seen with the naked eye. Is this commute normally?

A question without a clear answer. It is not possible to set the frequency perfectly, the frequency deviation is always there.
We can only create wrong theories. Assuming that you set the ocxo generator perfectly, the GPSDO itself will change the phase shift once a day, along with the synchronization of the satellites with the standard time atomic clocks. Phase shift and drift to the left or right would be due to the low DAC resolution in GPSDO, one bit lower and one bit higher. This setting of ocxo alone is impossible, ocxo has too low accuracy. My cesium clock pulls phase shift over 36h and 2.8E-13 stability takes 5 days GPS tracking. I don't know if by chance the GPS update is currently not every 12 hours.
There are different ocxo generators, used after prolonged shutdown, shocks, shocks require several cycles to stabilize again. Frequency may depend on many factors. From temperature, reference voltage standard, supply voltage, output load, interference, altitude above ground, earth's magnetic field, etc. However, it most likely depends on ocxo consumption. Turning ocxo off and on again is a different frequency and never the same. Set with 0.01Hz accuracy at 10MHz. After many turns on and off, it will out of tune anyway 

[Alfons]

@Sas, thanks for the answer.
Some reports from users say, that the signals would run synchronously. But I can't imagine that either, because there will always be a slight drift. I'm still trying to figure out how big the drift is, how I calculate it. I have to familiarize myself with it, just out of interest. I think the accuracy that I have now is enough for the CMU200.

[Ice-Tea]

Was a bit curious, so took a few pics of a "naked" TX/RX board. Figured I'd share. Not scaled. Enjoy!

[Robert763]

- Is it possible to make a print screen save it a usb stick

In my CRTU I bought a PCMCIA to SD adapter to make screen dumps, but all attempts with different card sizes have been unsuccessful, although someone told me it worked in his unit.

This is an old post I know, but I just found it and can give you an answer. The CMU200 CRTU-RU basically runs DOS 6.22 and the PCMCIA interface uses the ATA (IDE) hard disc emulation mode. This mode uses a parallel interface and is not supported by SD cards as they only have a serial interface. Like wise the simple PCMCIA to SD adaptors do not support it either. Compact Flash (CF) cards do support ATA mode so a PCMCIA to CF adaptor will work.

[dcarr]

I have a working CMU200 -> MMC (SD) card setup.  See the pictures below for the PCMCIA adapter and the MMC card I use.  I suspect that the it's important to use a SD (or MMC) card below a certain size...

[ps]

  I suspect that the it's important to use a SD (or MMC) card below a certain size...

At least you can exclude everything above 2 GB because this is the maximum possible size for a SD/MMC card. Anything bigger is SDHC or SDXC.

[lucaspnw]

Hi thread. I just got a CMU200, and after some keyboard / battery repair it works. It's quite old and uses BASE 3.52. It does have B12, B21 (Universal Signaling Unit), B41 (audio), B52 (speech coder), B53 (Bluetooth), and B83.
It does have 256MB RAM and the AMD K6.

It seems to work OK. I'd like to upgrade this unit to do AMPS and Bluetooth and other software features.

Does anyone have the 5.21 software for this unit? I've tried creating a Versions.new and copying BASEV5.21.exe and other files to the PCMCIA, but the versionmanager program does not list the "Install software" option. I've also tried to copy BASEV5.21.exe to C:\INTERNAL\INSTALL, but this also does not cause the install option to appear.
I do know the PCMCIA is working because versionmanager does successfully copy the log files to it when asked.

Thanks in advance!

[_Wim_]

Hi thread. I just got a CMU200, and after some keyboard / battery repair it works. It's quite old and uses BASE 3.52. It does have B12, B21 (Universal Signaling Unit), B41 (audio), B52 (speech coder), B53 (Bluetooth), and B83.
It does have 256MB RAM and the AMD K6.

It seems to work OK. I'd like to upgrade this unit to do AMPS and Bluetooth and other software features.

Does anyone have the 5.21 software for this unit? I've tried creating a Versions.new and copying BASEV5.21.exe and other files to the PCMCIA, but the versionmanager program does not list the "Install software" option. I've also tried to copy BASEV5.21.exe to C:\INTERNAL\INSTALL, but this also does not cause the install option to appear.
I do know the PCMCIA is working because versionmanager does successfully copy the log files to it when asked.

Thanks in advance!

Here you can find what you need:
http://gotroot.ca/cmu200/

Exta info nicely grouped together:
http://vk4zxi.blogspot.com/2018/07/rohde-schwarz-cmu200.html


Edit: I upgraded mine to V5.21, but using an Agilent 82357b GPIB adaptor and the "remoteservicetool" software from R&S

[_Wim_]

This is the remote service tool software I used

[Qw3rtzuiop]

@lucaspnw

You need to execute the BASEV5.21.exe on your computer. It is a self extracting zip file.
You can also open the .exe with 7-Zip to get the files.
Than copy the new BASEV5.21 folder on your CF-card.