New R&S CMU200 owner

[Dek]

Hello fellow EEVers,

I have recently received my CMU200, what I think a bargain (because of a non working screen), initially passes all tests while using an external monitor.
After studying the various threads, I was hoping that the screen problem was due to failed CCFCL or inverter.
I'm happy to say that it appears that is the case and a simple fix of the inverter or lamp is required as I can see the LCD working when illuminated by a torch.  I haven't opened it yet as I'm waiting for replacement SSD to swap the HD at the same time as fixing the screen driver (hopefully a 30p SMD fuse :-) )

My unit is fitted with B12 OCXO and additional generator and trying to work out what the rest are from the identity.dat file.
The unit has fairly old 4.35 firmware and I'm wanting to know which base firmware best to update with and which options to enable, but waiting until I've backed up the existing HD and replaced with a 32gb SSD.
Interestingly Demo pack elapse time is only 4,200hrs where as many of the others I have seen on line have +30,000hrs. Does this mean my unit has had low useage?
Photos attached.

/***************************************************************/

GENERAL INFORMATION


Serial number: 10XXXX/002


CPU and memory:

CPU (FMR)       FMR6
Memory         256 MB
BSW/FSW         V4.35

Statistics data:


Firmware versions:

RXTX_1 CP-SW    08.50, 2006-05-02

Available options:

rxtx1                V8.50    02.05.06
PMGR_RX_DSP_1     
PMGR_TX_DSP_1     
AUXTX_1           
CMU-B21Var14     
CMU-B56Var14     
CMU-B54Var14     
COPROC           
COPROC_FULL           3.04
CMU-B83           V2.84q (Var. 12)
CMU-B88           Rev. 0
CMU-B21           
2G_Signalling     
CMU-B66           included in B68
CMU-B76           included in B78
DIG               
DIG_HWC           HWC: 4
DIG_FM           
ADC 1             
WDDC400           
CMU-U65           
AUC 1             
OCXO 2           
NETCLK 1         
NETCLK 2         
REF_EEP           
RXTX1_EEP         
CORR1_EEP         
DIG_EEP           
AUC1_EEP         
ADC1_EEP         
AUXTX_EEP         
RXTX1_DG          DG: 13
RXIF3_1           
CMU-B11/B12       B12
Harddisk         
PCMCIA           
CPU               Intel Pentium III
CRTU-B1/B2        B2
FE (1100.3120.02)
CMUDrv           
CMU-B95           
CMU               
K21               
K22               
K23               
K24               
K27               
K28               
K29               
K42               
K43               
K59               
K61               
K62               
K63               
K65               
K66               
K67               
K68               
K69               
K83               
K84               
K85               
K86               
K88               
K92               
Audio             
RF               
GSM 850 MS       
GSM 900 MS       
GSM 1800 MS       
GSM 1900 MS       
IS136 800 MS     
IS136 1900 MS     
WCDMA Signalling B
WCDMA Signalling B
WCDMA Signalling B
WCDMA Signalling B
WCDMA HSUPA       
WCDMA HSDPA 10Mbps
WCDMA Signalling B
WCDMA Signalling B
WCDMA Signalling B
WCDMA HSDPA 3.6   
WCDMA TX Tests   
WCDMA DL Generator
WCDMA Signalling B
WCDMA Signalling B
WCDMA Signalling B
WCDMA Application
WCDMA WB AMR     
cdma2000 450 MS   
cdma2000 800 MS   
cdma2000 1900 MS 
cdma2000 2200 MS 
1xEV-DO           
EVDO_DSP   


A great forum, I hope you don't mind a 'beginner' picking your brains.

Dek

[Ice-Tea]

If its only 4000h, that's low indeed. But its not consistent with the failed CCFL tube. They fail often, but they get dim first and fail after, what, 20k hours? Pretty sure there's a running hour counter somewhere else but I can't remember where

When you have her open consider:

- replacing the CR2032 battery. Fails sooner or later.
- replacing the small fan for the "computer" module. If you are dismounting everything anyway to get to the CCFL tube it won't be too much of an additional effort.

[richnormand]

When I got my CMU200 a dead display was one of the issues I faced.
Turned out to be the fuse in the inverter section driving the CCFL tube.
Similarly the small fan for the computer section was frozen. I was able to dismantle it and re-oil.
Still OK but I also did the same treatment to the main unit fan on the side.

I'll go locate my original post and edit this response.

OK, here it is:
https://www.eevblog.com/forum/repair/rs-crtu-need-help-removing-the-front-panel/msg1127400/#msg1127400

Look at posts: 4, 8 and 13 but I suggest reading the whole thread. It might save you a lot of grief.

Also, in addition to the fan check the BIOS battery as previously suggested (mine was almost dead and it is not that fun to get in there)

Best of luck with it.

[Dek]

Hello Ice-Tea and Rich,

Thanks, I'll check the fans and bios back up battery.
First option is to solder a new fuse, if that doesn't do the trick and the inverter is dead then I have led strips on standby :-)

Cheers Dek.

[richnormand]

Have a look at the edited post just above for the details in
https://www.eevblog.com/forum/repair/rs-crtu-need-help-removing-the-front-panel/msg1127400/#msg1127400

Also take notes on hidden screws and connectors  as you take it appart. Easy to miss one a week later when you re-assemble. Ask me how I know....

[Dek]

Imaged the hard disc today and fitted a new SSD - re-booted the CMU and all looks good.
The display inverter fuse tested OK, so it's the inverter or the CCFCL tube that's faulty. I might just bite the bullet and fit LED strips I have waiting. Just wondering where to get an internal 12vDC supply for the LM317 Ill install to drive the led strips?

Both the main fan and cpu fans look good, in fact very clean. photos below show their condition before cleaning, so won't fiddle with these other than a quick 'de-dust'

Dek.

[richnormand]

Considering the low hours it is surprising the lamp would be bad. Did you check dc power to the CCFL board (you might solder two wires and partially reassemble to monitor that indeed the board does get power)?
Might not be any output from the board either. When you get to the lamp is it blacken at the ends (or cracked)?

I do remember someone that did a led conversion and found a suitable 12V supply point. Don't remember if it was EEVblog or the Yahoo R&S group. Will post here if I find it back.

[Bicurico]

You can still buy the inverter as a replacement part. I bought one about a year ago and it was not too expensive. Just google for it and search on AliExpres, too. Take note that there are different versions, so make sure you get the exact reference.

Are you sure the fuse is not dead? I had to replace the fuse on one of my CRTU.

The firmware is easy to upgrade, if you have a PCMCIA card. I bought a 512MB CF card and an adapter for PCMCIA. It contains the latest firmware upgrades. You just boot into the verm (Version Manager) and install each module from the PCMCIA card.

And of course, consider getting a GPIB-USB interface (the cloned ones from China seem to work OK, at least for me).

Regards,
Vitor

[Dek]

Hello Rich,

Thanks. I haven't opened up the back of the display to inspect the tube or tested the supply to the inverter board. - run out of time yesterday.
After a bit of rummaging about I found a JST? connector I can use to power the inverter board from my bench supply. Google came up with the data sheet (below) which shows the Pin out - pin 1 the black lead 5v   pin 2 0v and the other two blue leads appear to be display enable and variable brightness. I understand that the display is normally held off so the BIOS startup screen is not shown....so could also be that signal that's missing. I'm guessing that the 0-3v variable brightness is just set to max.

Vitor, thanks for the reply. Fuse measures 0.2 -ish so it looks OK. I've found some suppliers of the inverter on ebay, unfortunately all slooow boat from China and nothing local in the UK.
If the inverter is dead (or the CCFL), I'm tempted just to go ahead with the led mod so don't have to wait a couple of weeks while the inverter replacement arrives.
Once I've tested the inverter and looked at the tube I'll know which way to go.

Dek.

[Dek]

Some progress:-

I inspected the lamp, very slight blackening one end - about 2mm in length.

Powered the inverter from the bench supply, with the enable linked to 5v and brightness input fed from a potential divider to give about 3v.
 "you beauty" it fired up 

So I put the CMU back together 'loosely' and tested the supply and signals at the connector. 5V ok but no enable or brightness signal...hummm.

Is there a setting that disables the LCD?
I found a setting for the screen saver time out and also a 'screen off' tick box.
Fiddled with these and it does turn off the external monitor but no effect on the LCD control signals.
I suspicious that this is not a hardware fault but some setting somewhere???

More play time tomorrow

Dek

[Dek]

Success.....sort of.

For whatever reason the CCFL driver wasn't getting the enable or brightness signals.
So rather than have the CMU scattered in more bits to trace the origin of those signals, I decided to hack the CCFL driver board.

Cut the tracks from pin 3 enable and pin 4 brightness.
Linked enable trace to a convenient 5v point near the fuse.
Linked brightness on the underside to a 0v via.

CMU reassembled with working screen.
Dek

[Skallywampus]

Hello...another new cmu-200 new user. Bought w/o display working.....
replaced 2A fuse on ccfl board, and installed new ccfl bulb ...display now working.
Next problem is attempting to access system self tests in maintenance.. I cannot move highlight down to other options such as System Identity etc....is there a secret?
Also any attempt to access the spectrum analyzer results in the following error message...dsp.cpp line 2701  see pic. any suggestions appreciated!

[Ice-Tea]

Re error message: check all ribbon cables inside the chassis (not just the stuff connecting the disaply module). They tend to get semi-loose when you take it apart.

[Bicurico]

IMHO you have the worst of all possible problems, as no DIY fix is known.

The error message implies that your DSP module on the DIGITAL BOARD is broken. This is a PITA, because this module is essentially composed of FPGA and once they break you can only replace the whole module. The problem is: where to get them from? You will need to find a donor device/DIGITAL BOARD.

As a tip: the CRTU has two such DSP modules inside the DIGITAL BOARD, because each control one of both RX/TX modules.

However, to use the CRTU as a spectrum analyser, you only need the first RX/TX board and thus the first DSP module, too.

So, try to buy a working CRTU and it will allow you to get the CMU200 running while not voiding the DOS use of the CRTU (the Windows use is not useful, anyway).

Regards,
Vitor

[Skallywampus]

thank you for the reply. ok...more work to do....more to learn...I will digest that and read more....I solved one problem today disassembling the front again because the wire must have jiggled loose the first time and the variation knob wasn't working....re seated the cable and all is good..... this time I tried booting the thing before reassembly!

I read where the cmu200 will create a file for errors at cst.err  I am wondering if a keyboard may enable me to look at the error files to help me determine if this is truly a software issue as opposed to a hardware issue?
Would running chk dsk  be of any possible help?
Would installing an updated software perhaps help? (currently 5.13)
all self tests pass except internal loop RF test path......
thank you for your thoughts!