Rohde & Schwarz RTO1024 upgrade ??

[jjoonathan]

Message sent

[myhobo]

I want to install the firmware in the windows XP system, but I can't install it sucessfully. The 634H error is reported

[jjoonathan]

Sorry, I've never seen error 634H before and don't know what it means. Once I got my scope upgraded to Windows 7 I didn't play around much with XP.

[dc101]

I guess a good reason to upgrade. I'm still on XP firmware 2.80. I'm pulling the DLL up in IDA to see if I can find something close to that function.

I wouldn't mind upgrading but this thing has gremlins, like others have mentioned I think this thing has heat issues. The scope seems to display traces fine, but sometimes I get RTC1 RAM errors during a self test, and the next time there's none but some SPI trimming register failure    I just don't know if I want to sink any money into this thing. I've had it for years but it's only seen light use.

[dc101]

It's almost like it heard me talking about it. No sooner did I post that previous message when it started doing this.

Updated: So I turned it off, left it off for about 20 minutes, powered it up and now the display is fine? Any thoughts on what would cause this? Overheating is my first thought, but my temps looked ok earlier. And even with the screen going berserk, the selftest it was running at the time somehow passed???  I'm not sure what to make with all these random errors that seem to come and go. Ground issue? Dirty board connections? It's never been taken apart as far as I know.

[jjoonathan]

The RTC RAM and screen are very far from each other in the architecture (by my count, at least 5 busses and 6 chips) so it seems much more likely that the problem is a common cause: power, heat, or maybe bad CPU. You've checked heat, so we move on to power: if the RTO1000 works like the RTO2000, just about everything is powered from 12V. The easiest way to get at 12V is probably by soldering wires to probe the CPU board. I have one photo of a RTO1000 CPU board in which I was probing the other side of the buck converters, but IIRC they all feed from 12V so just solder to the other side of the switching chip.

RTO1000 CPU Power: https://www.dropbox.com/scl/fi/hzdjx42optghtmkxzavpq/RTO1000-Computer.pdf?rlkey=rq22a6p4xuclm8vz6fyzeb1as&st=fr8zz9s9&dl=0

As for bad CPU, the RTO can boot without the mainboard / frontend installed because these just appear as a PCIe peripheral as far as the CPU is concerned. You could use this to segment responsibility between the two, but honestly it's probably safe to assume the mainboard didn't cause the screen glitches and if power isn't responsible the likely culprit is CPU/PCH. These aren't impossible to replace, but as always it's a question of whether or not it's worth the effort.

[dc101]

Thanks for taking time to write back. Good call on checking the power, and removing subsystems. I'm going to have to clean off my desk so I can take this thing apart and see what's going on. It would be nice if it was a "simple" power supply issue, but life doesn't ever seem that easy does it

Right now all the issues are very intermittent. I just started using it again after about a year of non-use and everything was fine for about a week. I hadn't noticed any issues, but it had been a while since I last did a self cal so I figured why not. That's when I got the RTC1 RAM walking one/zero test errors. I believe it reported 4294967295 errors for both the one and zero test as well as 32767 errors on the ADR test. Both numbers suspiciously 0xFFFFFFFF and 0x7FFF which makes sense because max unsigned 16bit.

After that I decided to do an alignment. From what I could tell everything was fine? I regret not taking a screenshot though, but I did tab through each channel.

I can't recall if I restarted it after the alignment or not, but I ran another selftest after the alignment. This time I received no RTC errors at all, but got errors on SPI Channel 1 Trimming Register failed, No. 29 Value: 42 != 42. And the same error but different value for SPI Channel 3 or 4, I can't remember which one it was.

So then I  saw someone mentioned, maybe you? to feed in a full range saw tooth wave. So I rebooted and used my Pico's AWG output to feed in a sawtooth one channel at a time on all four channels and it looked fine to me. No crazy distortion or missing bits, or noise. So I start the selftest again and that's when the screen started going haywire. And then yeah like I said, after shutting it down and waiting. The screen was fine and then I could finally read the results from the selftest that had been running and it was fine.

I will report back with my findings once I'm able to make some room and do some proper power measurements and a visual inspection.

[jjoonathan]

RTC DRAM tests are going to be much more sensitive than looking at the waveform because the visible waveform doesn't usually use all (or most, or even a significant fraction of) the DRAM. That said, the RTC cycles round-robin between 14 DRAM chips (they each run at 400MHz DDR so it needs to round robin a lot of them to store 10GS/s), so if there's a problem with a chip or connection it shows up as a periodic waveform distortion so long as your waveform exercises all of the bits. The fact that you saw nothing suggests that the issue is sparse across time rather than sparse across components of the RTC system, once again pointing to a heat/power/cpu issue.

Checking the power is still important, but another thing you could do would be run prime95 / occt / cinebench to test the CPU and grab RWEverything to look at PCIe error counters: bad DLLP counts, bad TLP counts, replay counts, etc.

[dc101]

Well, now I'm not so sure about temps. Here's a look at temps with the scope just at idle, one trace on the screen. I can feel the difference in case temp with my hands, the left side feels much cooler than the right side. Not sure if that's normal? Screenshot of temps attached. The ADCs seem a little on the high side. My lab/room is probably around 74F.

I took a look at RW Everything and I honestly have no idea what I'm looking at I mean I get it's reading memory space and register, but I didn't see anything specifically calling out hw errors.

I've got prime95 running now while I monitor the motherboard temps with HWInfo. CPU temps look pretty good so far though, hovering around 42-43.

Hoping to have some time over Thanksgiving weekend to take things apart and look at power. Thanks again for the help so far!

Edit: Here's the latest selftest log

Code: [Select]

22.11.2025 01:57:22
DAB
Step 1 of 1 :  Passed.
RTC 0 DAB_EYE: 14
RTC 1 DAB_EYE: 20

DAF
Step 1 of 1 :  Passed.
RTC 0 DAF_EYE_CLK: 12
RTC 0 DAF_EYE_DATA: 9
RTC 1 DAF_EYE_CLK: 13
RTC 1 DAF_EYE_DATA: 10

TrgBus
Step 1 of 1 :  Passed.
RTC 0 TIF_EYE_CLK: 10
RTC 0 TIF_EYE_DATA: 11
RTC 1 TIF_EYE_CLK: 12
RTC 1 TIF_EYE_DATA: 11

TrgRtc
Step 1 of 1 :  Passed.
RTC2 TRG_RTC_PRN_CLK :11
RTC2 TRG_RTC_PRN_DATA :10
RTC1 TRG_RTC_PRN_CLK :10
RTC1 TRG_RTC_PRN_DATA :9

FecBus
Step 1 of 1 :  Passed.

RTC_SDRAM
Step 1 of 1 :  Passed.
RTC 1 RAM_DATA_WALKING_ONE_TEST: 0 ERRORS
RTC 1 RAM_DATA_WALKING_ZERO_TEST: 0 ERRORS
RTC 1 SDRAM_ADR_TEST: 0 ERRORS
RTC 1 SDRAM_RANDOM_PATTERN_TEST: 0 ERRORS

RTC 2 RAM_DATA_WALKING_ONE_TEST: 0 ERRORS
RTC 2 RAM_DATA_WALKING_ZERO_TEST: 0 ERRORS
RTC 2 SDRAM_ADR_TEST: 0 ERRORS
RTC 2 SDRAM_RANDOM_PATTERN_TEST: 0 ERRORS

RTC MCO-RAM
Step 1 of 1 :  Passed.
Spike test based on mask test with 0 Divs offset
        Test passed, No Spikes found in mask test of any channel
Spike test based on measurements with 0 Div position
        Channel: 1 Measured: 0.00592885 Volt, max.limit: 0.05 Volt
        Channel: 2 Measured: 0.00790514 Volt, max.limit: 0.05 Volt
        Channel: 3 Measured: 0.00790514 Volt, max.limit: 0.05 Volt
        Channel: 4 Measured: 0.00790514 Volt, max.limit: 0.05 Volt
Spike test based on measurements with 4 Div position
        Channel: 1Measured: 0.00013834 Volt, max.limit: 0.001 Volt
        Channel: 2Measured: 0.000167984 Volt, max.limit: 0.001 Volt
        Channel: 3Measured: 0.000123518 Volt, max.limit: 0.001 Volt
        Channel: 4Measured: 0.000167984 Volt, max.limit: 0.001 Volt
Spike test based on measurements with 5 Div position (overflow)
        Test passed, No Spikes found in mask test of any channel

SPI Channel: 1
Step 1 of 1 :  Passed.

Cal Relay Channel: 1
Step 1 of 1 :  Passed.
        DC 1 MOhm difference = 1.17963%
        DC 1 MOhm min/max detector difference = 1.14286%
        DC 50 Ohm difference = 0.35767%
        DC 50 Ohm min/max detector difference = 0%
        AC difference = 0.287055%
        AC min/max detector difference = 0%

SPI Channel: 2
Step 1 of 1 :  Passed.

Cal Relay Channel: 2
Step 1 of 1 :  Passed.
        DC 1 MOhm difference = 1.18231%
        DC 1 MOhm min/max detector difference = 1.68539%
        DC 50 Ohm difference = 0.721127%
        DC 50 Ohm min/max detector difference = 0%
        AC difference = 0.0360921%
        AC min/max detector difference = 0.568182%

SPI Channel: 3
Step 1 of 1 :  Passed.

Cal Relay Channel: 3
Step 1 of 1 :  Passed.
        DC 1 MOhm difference = 1.06195%
        DC 1 MOhm min/max detector difference = 1.14286%
        DC 50 Ohm difference = 0.486977%
        DC 50 Ohm min/max detector difference = 0%
        AC difference = 0.0228102%
        AC min/max detector difference = 0.578035%

SPI Channel: 4
Step 1 of 1 :  Passed.

Cal Relay Channel: 4
Step 1 of 1 :  Passed.
        DC 1 MOhm difference = 0.732534%
        DC 1 MOhm min/max detector difference = 0.574713%
        DC 50 Ohm difference = 0.288921%
        DC 50 Ohm min/max detector difference = 0%
        AC difference = 0.269875%
        AC min/max detector difference = 0.578035%

[dc101]

This works for XP as well for anyone interested. Follow the same steps as mentioned, just need to look for the config.ini here: C:\Documents and Settings\All Users\Application Data\Rohde-Schwarz\RTO\EEPROM\Frontend

If anyone else worries about how hot R&S runs their unobtainium ADCs, here's how to run the fans a little faster and thereby obtain 10°C cooler running temperatures. If common knowledge is to be trusted, this should ~double the lifetime.

1. File > Maintenance > Service, enter "16052008f" into the Password field, and press the enter key. If successful, the text should disappear from the "Password" box.

2. File > Service 2 > Frontend Temps. Tick "GetHwTemperatures." Write down "FAN1 adjust" and "FAN2 adjust" (91 and 89 on mine). You can observe the thermal regulation by waiting for an ADC to wander above 100°C, watch "FAN1 adjust" briefly increase, followed by the ADC wandering back below 100°C.

3. Edit C:\ProgramData\Rohde-Schwarz\RTO\EEPROM\Frontend\Config.ini, changing ADCTEMP_OFFSET_CH* from -41 to -31.

4.  File > Maintenance > Frontend. This is where shit gets real and I must pause for a moment to remind you that I am describing a procedure that worked on *my* scope, not selling you a free insurance policy for modifying *your* scope. Once you decide to accept the risk, press "Flash frontend EEPROM" and "OK." A dialog should pop up "Flashing device in progress, please wait. Do not turn off the device!" In about 5 seconds, the dialog should disappear and you should see "Flashing successfully finished." in the upper right corner.

5. File > Exit

6. Wait for a few seconds and restart RTx.exe. If you did not wait long enough, it will launch/exit, and you should wait a bit longer before trying again.

7. Once the scope app has started, the new settings should be active.

Verification 1: in a quiet room, you should be able to hear difference in fan speed.

Verification 2: File > Maintenance > Frontend > (scroll to bottom) > ADC temp offset should be the new value -31 instead of the old value -41.

Verification 3: File > Maintenance > Service > Password: 16052008f, (press enter, text disappears), File > Service 2 > Frontend Temps, (tick GetHwTemperatures), FAN1 adjust should now be consistently higher than the value you wrote down in step 2. On my RTO, under the ambient temperatures of my bench, the steady state FAN1 adjust increases from ~91 to ~115 after applying this change.


General background information: the EEPROM gets read from the frontend and written to C:\ProgramData\Rohde-Schwarz\RTO\EEPROM\Frontend every time RTX.exe starts. When you modify a ini or dat file in that folder and choose "Flash frontend EEPROM," the files in C:\ProgramData\Rohde-Schwarz\RTO\EEPROM\Frontend get written to EEPROM. Same goes for the mainboard etc.

Happy hacking!

[myhobo]

hi guys.Who has the R&S®ScopeSuite of RTO1000. i can't download it from the official websit of R&S.

[jjoonathan]

Here you go! https://www.dropbox.com/scl/fi/lv1446aozaudcbvbrusmb/RSScopeSuiteSetupV3.6.0.exe?rlkey=eztzh2qgkqle113wn8cup6421&st=46iqvzxa&dl=0

[sergeyklenov]

Options for this device more interesting than passwords

[a390323160]

Hello, friend. My RTO1012 has a GB Receiver issue. Could you please help me?

[jjoonathan]

Look at the power supply chips between the ASICs and the edge connector. These chips tend to die and cause GB Receiver issues.

These are my notes for RTO2000, which is similar to RTO1000: https://www.dropbox.com/scl/fi/3mkb6at33y2vfb0jacpzc/Vmeas.pdf?rlkey=6x8ut0mmuz7fblnpk92lh09ok&st=g8fsy4i2&dl=0

37/20/19/25 and 39/22/21/26 are the ones to check.

[santiall]

coincidentally I fixed mine yesterday, a CH3/4 GB Receiver problem, with the help of Jonathan.
The culprit was indeed one of those regulators under the heatsink, a 1.5V in my case. LTC3026

[a390323160]

Hello, friend. The VGA chip on my RTO front panel is broken. It's that blue ceramic chip. I took it off, but some of the pins fell off. Is there any way to fix it? How can I take it off without the pins falling off?

[TRN]

Hello, friend. The VGA chip on my RTO front panel is broken. It's that blue ceramic chip. I took it off, but some of the pins fell off. Is there any way to fix it? How can I take it off without the pins falling off?

Hi,
If you want forum members to help, you need to be a little more specific.
" That blue ceramic chip" is a pretty vage description
Does the chip have a part number?
Send a picture a picture from the PCB where it was placed.
The pins falling off when removing an SMD chip, mostly happens when you do not heat all the solder pads sufficiently while pulling the chip from the PCB, or in case the chip body was cracked

[Aurelio]

I hope i can get some help/pointers here:
i currently have a RTO1024 on my desk with these messages on the screen:
Can this be related to the powerIC i sawa few posts back, or are the GBs toast?

what would be the best step-by-step approach to get to the rootcause?

Edit: i've opened it up and got the board out. using the file from jjoonathan, i've found the Poweric's. But to power the board, i'd use the connection board; i traced the fuses on the board back to the pads present on the connection board, but i don't know what voltage to supply to it.

Any toughts?

[santiall]

most likely one of these:

[Aurelio]

Hi Santiall,

that is indeed what i want to check. do you know where i need to provide 12V to this board so that i can measure them?

Thx

[santiall]

hmmm, I would have to check... unlike the RTB2000 we don't have the multiplexer to measure the output voltages of those regulators so I soldered some wires to a heather for measuring them with the scope fully assembled. One of those regulators was dead in mine, changed it and new scope
I think the regulators weren't powered from 12V, perhaps from 5V or 3V3?, once I found one dead, I removed it and fed that individual regultated voltage from a lab power supply. The scope worked straight away.

[esi_49]

Hello everyone,

I hope you are all doing well.

I own a Rohde & Schwarz RTO1022 oscilloscope which unfortunately appears to be defective. I found this thread and it seems there are several experts here with deep knowledge of the RTO series, so I am hoping someone might be able to give me a hint.

My unit aborts the Self Alignment procedure with an error message. The self-test also reports a failure. I have attached screenshots of:

The Self Alignment error message

The Self Test results (showing “Failed”)

The scope boots normally and is generally operational, but the alignment cannot be completed.

Has anyone experienced a similar issue? I would greatly appreciate any advice on possible causes or diagnostic steps.

Thank you very much in advance for your time and support.

[jjoonathan]

Hmm, MCO-RAM problems, this rhymes with a problem I had on the RTO2000, which has a very similar frontend and mainboard:

https://www.eevblog.com/forum/testgear/rto2000-ram-artifacts/

I suspect one of your RAM chips might be going dead or you might have broken contact with the RTC chip. If you swing the waveforms through the full scale, can you see stuck bits forming periodic patterns like they did on my RTO2000?

[esi_49]

No, I cannot see any bit errors or periodic patterns.

After booting, the message “Horizontal Alignment External Trigger” appears for a few minutes; it seems to be stuck.

On Channel 1, with 50 Ω coupling and settings below 200 mV/div, there is a large positive offset of the trace. At ≥ 200 mV/div everything looks fine. With 1 MΩ coupling, no fault is visible. The vertical deflection itself seems to be okay — I would estimate it to be within about ±5%. Channel 2 behaves completely normally.

Only the alignment immediately triggers the error message: “AC Cal Signal of CH1 not working correctly.”

When switching from 200 mV/div to 100 mV/div, you can hear a relay clicking inside the device. It seems that a different signal path is being selected at that point. It appears that the issue might be located there. Do you have any experience with whether that section is accessible for inspection or repair?