EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: TRN on April 03, 2023, 07:45:41 pm
-
This is a follow-up of a SMU200A repair attempt topic I posted on 26 February 2022
I purchased a defect SMU200A on eBay, in late 2021, and when the unit arrived it had several issues.
The fans were not working, which was caused by a blown fuse on the 5.2V supply rail of the motherboard. After replacing the fuse, the unit refused to power up anymore, and after some troubleshooting this was traced back to a faulty Vega 450 PSU. Swapping the primary side of this PSU, with another one resolved the issue, and the unit booted, and remained powered up.
However, after loading the firmware an error message appeared indicating a missing "SSYN module” even though this module was physically installed. This error was caused by a faulty Xilinx - XCR3128XL CPLD, and an Atmel AT25160AN serial EEPROM on the SSYN board which had a missing Vcc pin. I replaced the EEPROM & CPLD, but I didn’t have the binary file to program the new EEPROM.
In August last year, Qw3rtzuiop, who was so friendly to open up his SMU200A, provided me with the EEPROM dump file (thanks again for that), but unfortunately by that time I had already packed-up my electronics lab, because I was relocating from Belgium to Portugal.
Well, to make a long story a bit shorter I only now finished setting up my lab, and unpacking all the gear.
So I programmed the EEPROM, and switched on the SMU200A, and to my surprise … no boot and a black screen |O.
Troubleshooting done so far:
1) Checked all the PSU voltages, and they are within specs
2) Checked all the fuses on the motherboard, all OK
3) The HDD spins, but does not seem to do anything else
4) Bios error beeps varied; most of the time no beeps at all after switching the unit on, but one time I had 4 beeps, another time 2 beeps after applying power
5) After applying power, the unit automatically switches from standby to on, and pressing the on/STB button on the front panel does not have any effect (unit remains ON)
6) The CPU board is an FMR6+/5PCB, and the inverter board is a DAC-05B004
7) I replaced the CMOS battery with a new one
8) I don’t think the CFFL tubes or inverter board is bad, it seems to me that the CPU board does not boot.
9) Does anybody have any suggestions for further troubleshooting, and/or documentation from the FMR6+/5PCB, & DAC-05B004?
Thanks
-
That XCR3128XL CPLD will need programming too -- you'll need to read the old one (and pray it doesn't have the security fuse blown) or somehow get a replacement.
I'd try pulling the SSYN board and see if you get the original fault (the "no SSYN board" error) back. If you do - that points to an issue on the repaired SSYN board, which probably means either your EEPROM contents are garbage or (more likely) the CPLD isn't programmed and isn't doing what it needs to do.
-
Why start a new thread instead of continuing the previous thread with all the context? Or at least link to the old thread?
-
Thanks for the info, which I wasn't aware of,
This is a real bummer because I think that the original XCR3128XL CPLD was fried, so reading it is probably a no go.
The reason I am suspecting this, is (if I remember correctly) that both the CPLD and the EEPROM are powered from the 3,3 V rail via a 10ohm resistor, and the CPLD was sinking ~ 0.3Amps, which dropped the supply voltage to ~0.3 V. changing the CPLD resolved this issue.
I already tried pulling the SSYN board which had no effect at all. As a matter of fact, only the DACIF board is required to be installed for the PSU to switch-on, and the FMR PCB board only requires 3.3 V & 5.2V to power-up and boot if I am not mistaken. I also tried powering the FMR PCB from a 3.3 / 5.2v bench supply, but everything remains dead.
What I don't understand is that when I boxed up the SMU200A 7 months ago, It was booting OK, and only had the missing SSYN board error. It is unclear to me why sitting in the box for 7 months caused this new error
-
Hi,
You are correct.
I only started a new thread, because when I tried to post via the old one, I received a message that this post had been idle for a long time.
Anyway my applolgy for not posting the old link.
Here it is
https://www.eevblog.com/forum/repair/rs-smu200a-repair-question/msg4032550/#msg4032550 (https://www.eevblog.com/forum/repair/rs-smu200a-repair-question/msg4032550/#msg4032550)
Regards
-
I only started a new thread, because when I tried to post via the old one, I received a message that this post had been idle for a long time.
Its just a warning, you can ignore it in the future.
-
O.K. :-+ :-+
-
The FMR5 & 6 CPU PCB's are also used in the CMU200.
Did anybody who owns a CMU200 ever encounter a similar boot-problem?
-
I had a weird problem with a CRTU (similar to CMU200), which would not start past the first BIOS sceen and just beep, shutting down.
Turns out it was a bad memory DIMM. Definitely worth trying that. If I recall correctly, you can run it with just one DIMM and check which might have a fault, but I am not totally sure about it. In the end I used a DIMm from an old laptop. That fixed the problem.
-
8) I don’t think the CFFL tubes or inverter board is bad, it seems to me that the CPU board does not boot.
This can be easily checked with a bright flashlight. Just place it directly at the screen.
iirc there is an adj. voltage regulator on the FMR6 that likes to die. But thats probably not the cause here.
Can you check which beep codes you get?
http://www.bioscentral.com/beepcodes/awardbeep.htm (http://www.bioscentral.com/beepcodes/awardbeep.htm)
-
I had a weird problem with a CRTU (similar to CMU200), which would not start past the first BIOS sceen and just beep, shutting down.
Turns out it was a bad memory DIMM. Definitely worth trying that. If I recall correctly, you can run it with just one DIMM and check which might have a fault, but I am not totally sure about it. In the end I used a DIMm from an old laptop. That fixed the problem.
Unfortunately the FMR6 only has one memory slot with a 256MB 133MHz SO-DIMM module, and I don't have a spare one, but I will try to source aspare one
-
8) I don’t think the CFFL tubes or inverter board is bad, it seems to me that the CPU board does not boot.
This can be easily checked with a bright flashlight. Just place it directly at the screen.
iirc there is an adj. voltage regulator on the FMR6 that likes to die. But thats probably not the cause here.
Can you check which beep codes you get?
http://www.bioscentral.com/beepcodes/awardbeep.htm (http://www.bioscentral.com/beepcodes/awardbeep.htm)
Thanks for the tip, and I tried the bright flashlight, but no luck everything is black. This confirms more or less that the CFFL tubes are not the problem.
I also measure no input voltage at the DAC board, which I presume needs 5VDC to output ~ 1.5KVAC required to light-up the CFFL tubes.
Thanks for the BIOS error code page link, but the majority of the times there are no error beeps at all. I heard only once 2 beeps in succession, and another time 4 beeps, but now everything is silent.
Do you know If these CPU boards were manufactured by R&S, or some other company, and if there is some documentation available?
The adjustable regulator you mentioned I presume is the LMS1587 CS-ADJ?
I measure the following voltages:
Input: +3.31VDC
Output: +1.89VDC
ADJ/GND: )+0.639VDC
I pulled the heat-sink from the CPU, which is an Intel Mobile Celeron SLB8, of which Vcc core = 1.10V, so Vout from the regulator seems to be O.K
-
This morning when I flipped the switch at the back of the unit, The orange STB light lit and remained on.
(All other times the unit would automatically go from STB to Run affer flipping the switch).
So my hopes went up, and when I pushed the STB/Run switch, the unit actually booted into XP :-+.
But while loading the firmware, it switched into STB again, and after renewed pressing of the STB/Run button, the green light lit, but the screen remained black again :palm:
-
The adjustable regulator you mentioned I presume is the LMS1587 CS-ADJ?
Its this regulator https://www.eevblog.com/forum/repair/rohde-schwartz-universal-radio-communication-tester-cmu200-psu-issues/msg989219/#msg989219 (https://www.eevblog.com/forum/repair/rohde-schwartz-universal-radio-communication-tester-cmu200-psu-issues/msg989219/#msg989219)
The FMR6p has soldered RAM and a SO-DIMM slot. You should be able to boot without a RAM stick in the slot.
I s the voltage of the cmos battery still okay?
-
Thanks for the link, and I checked the LM2651MCT-ADJ regulator, and by doing so must have shorted two or more pins, because it went up in smoke, so now it is definitely fried!
This will halt any further troubleshooting, until I have sourced a new one :palm:
On a side note, the only companies which seem to stock these are Mouser, Farnell, and Digikey, but they all ship from the USA, which means expensive shipping cost plus import duties, which further racks up the bill.
Anyway, before frying the chip, I had three more or less successful boots, but while loading the firmware the component server error popped-up, so it seems I need to reinstall the firmware after the hardware errors are fixed. closing the error message triggered the shutdown script.
After the shutdown, a renewed pressing of the STB/Run switch would give the following result; 1) the orange LED would cease, and the green LED would lit, and remain green for 2 or three second, then the orange LED would lit again briefly for less than 1 second, and the green LED would lit again, but all resulting in a black screen. Only after flipping the main power switch, and leaving the unit switched off for a limited time, the unit would boot again, but would switch itself into STB while loading the firmware (even without closing down the component server error message
-
On a side note, the only companies which seem to stock these are Mouser, Farnell, and Digikey, but they all ship from the USA, which means expensive shipping cost plus import duties, which further racks up the bill.
If anybody knows a European supplier based in Germany, The Netherlands, Belgium or France which stocks the LM2651MCT-ADJ, please let me know, so I can save some money for a few beers ;D
-
I have one in my parts drawer. A leftover from my CMU200 repair some years ago. But the cheapest shipping to portugal seems to be 5,99€ for me.
-
On a side note, the only companies which seem to stock these are Mouser, Farnell, and Digikey, but they all ship from the USA, which means expensive shipping cost plus import duties, which further racks up the bill.
Don't both Mouser (https://eu.mouser.com) and Digikey (https://www.digikey.pt) ship to Portugal DDP, so duty pre-paid paid (but not VAT)? That still leaves the shipping unless you are able to combine orders to qualify for free shipping (that's getting easier with the increased component prices :P), but you shouldn't have to pay any duties or import fees other than VAT.
-
I have one in my parts drawer. A leftover from my CMU200 repair some years ago. But the cheapest shipping to portugal seems to be 5,99€ for me.
That is great!
If you want to sell it to me, I will happily reimburse you the shipping cost
will drop you a pm
-
On a side note, the only companies which seem to stock these are Mouser, Farnell, and Digikey, but they all ship from the USA, which means expensive shipping cost plus import duties, which further racks up the bill.
Don't both Mouser (https://eu.mouser.com) and Digikey (https://www.digikey.pt) ship to Portugal DDP, so duty pre-paid paid (but not VAT)? That still leaves the shipping unless you are able to combine orders to qualify for free shipping (that's getting easier with the increased component prices :P), but you shouldn't have to pay any duties or import fees other than VAT.
Yes they do, but minimum order for free shipping is 50EUR :)
-
Drop some coin and buy grabber hooks for TSSOP packages and apply only with power off, too easy to smoke those when their regulating a rail.
I fried a regulator doing that and since then I realize that unlike most chips, that one has potential of causing catastrophic damage, and I never probe em by hand, even a to220 center pin is risky IMO.
Almost want a package a hazard stickers to put on annoying regulators so I don't accidentally probe them lol
-
Thanks for the tip, and I will do this
-
Yeah I was a little hurt by the price of the smallest grabbers but it takes like 2 seconds to destroy something that could make you very happy.
Just beware of the interface, some of them have square pins, and other round pins, so you need to figure out the hook up your gonna use, you need to be patient with that
and you won't regret having a 'ground point' connection even when you are probing other stuff on those ridiculous boards.
-
So which grabber hooks did you buy, and where did you purchase them?
Regarding the earth point. I have two esd mats on my workbench, and also use a wrist strap
-
No, I mean using a grabber hook for ground when you are probing around so you only need to worry about one probe. This is a benefit of having grabber hooks even on big parts because they make a reliable connection. Alot of people will keep holding a probe with one hand instead of bothering to anchor a connection, then something slips. ESD is important I guess but I mean more for not shorting out the PCB
I shopped from the digikey grabber hooks assortment, I don't have the part numbers anymore. I think its the microclip series (20$ each), and IIRC it has a round pin so you need a collet type connector for it
I measured it, the connection point is a 0.77mm round brass tube. SO you need something that will attach to a 0.77mm tube
-
Thanks for this info, and 20US$ per piece is quite a hefty price tag indeed ;D
There goes all my beer money :popcorn:
-
did you pick a connector to use with it?
You can solder to it too BTW
I had the wrong one so I was just using like clips on it. I think this one will work. I think I also managed to bodge together something that kinda stuck on there but it was sub optimal, IIRC the mating life I had was rather short (like 5 connections) by using maybe a machine pin socket from a DIP or TO socket. But I have not touched a SMD circuit board for a while so they have been dormant. Usually I get away using tektronix grabbers.
I really hate the small parts, last thing is I did a 0201 part on a rasberry pi with thermal tweezers.
https://www.digikey.com/en/products/detail/lemo/EGG-0B-655-ZZM/3792380 (https://www.digikey.com/en/products/detail/lemo/EGG-0B-655-ZZM/3792380)
-
did you pick a connector to use with it?
You can solder to it too BTW
I had the wrong one so I was just using like clips on it. I think this one will work. I think I also managed to bodge together something that kinda stuck on there but it was sub optimal, IIRC the mating life I had was rather short (like 5 connections) by using maybe a machine pin socket from a DIP or TO socket. But I have not touched a SMD circuit board for a while so they have been dormant. Usually I get away using tektronix grabbers.
I really hate the small parts, last thing is I did a 0201 part on a rasberry pi with thermal tweezers.
https://www.digikey.com/en/products/detail/lemo/EGG-0B-655-ZZM/3792380 (https://www.digikey.com/en/products/detail/lemo/EGG-0B-655-ZZM/3792380)
Thanks for the link, and no I did not pick a connector yet.
You are right, that the small parts are a PITA to work with.
What happened to the good old days when transistors and resistors still had wires ;D
-
The connector I linked works, with modificartion. The thing will grip it, but there is like a ring in front of the connector that does not fit the pin.
SO you need to cut that off with ultra flush cutters, then expand and reform the cut edge on a needle of some sort (like a pick) using round nose pliers, then debur with micro OD and ID deburring tools. Then it does fit, but you need to 'pry it open' because there is no taper on the micro grabber, that is push the micrograbber it at an angle to bend the clamp section to slide in the shaft. Then it holds good. If you put some heat shrink over it, it should be even better. And after first insertion, form the very tip (1mm) of the mated pair with round nose pliers.
Aka you need a lego man. I would recommend getting mating pins that you trust of both genders, then solder a pin to the micrograbber, rather then trying to interface the micrograbber with a ill fitting terminal. That way its all proper and only very slightly longer.
But mine survived like 20 insertion cycles, I don't feel like doing it more, so I will call it good enough.
maybe 40 times and it started to bow out and its not gripping well, this is not the right contact for it.
Or just like get their cables, but I am not doing that 8) time to try something else