It's been so long since I used XP that I forgot how often it crashes. I remember it was not unusual for Windows 3.0/3.1 to freeze as I had half a dozen freeze a day.
But an instrument just sitting, outputting a steady tone and locking up, that concerned me a bit. Hopefully it's just XP.
But an instrument just sitting, outputting a steady tone and locking up, that concerned me a bit. Hopefully it's just XP.
Given the age and unknown provenance of the system may be you should run some memory and hardware diagnostics.
I dealt with similar issues with a Tek DSA running internal XP. Teks answer was junk it and buy the new one running Win 7. No path for software or MB upgrades because the hardware was too interdependent. My reaction was that for the price of a car you can't make this as reliable as an off brand pc??
I dealt with similar issues with a Tek DSA running internal XP. Teks answer was junk it and buy the new one running Win 7. No path for software or MB upgrades because the hardware was too interdependent. My reaction was that for the price of a car you can't make this as reliable as an off brand pc??
I did have occasional issues with XP on desktop computers back in the day so I was somewhat apprehensive when test equipment started being based on Windows - but you don't have much choice.
I have ended up with several Windows based instruments: R&S FSP, R&S ZVB, R&S AMU, R&S PSL, Tektronix RSA3303A, and Agilent CXA.
I have never had XP completely lock up on any of them and I have left some of them on for extended periods. I had the spectrum analyzers running for more than a month unattended collecting spectrum data at radio sites and transferring to a connected PC.
That said, I NEVER install anything on a test instrument that didn't come on it from the manufacturer.
If I were in your shoes - right about now I would be getting a little concerned that I had bought a Franken-lyzer.
I have ended up with several Windows based instruments: R&S FSP, R&S ZVB, R&S AMU, R&S PSL, Tektronix RSA3303A, and Agilent CXA.
I have never had XP completely lock up on any of them and I have left some of them on for extended periods. I had the spectrum analyzers running for more than a month unattended collecting spectrum data at radio sites and transferring to a connected PC.
That said, I NEVER install anything on a test instrument that didn't come on it from the manufacturer.
If I were in your shoes - right about now I would be getting a little concerned that I had bought a Franken-lyzer.
I turned on the UPV this morning.. it was on for about 45 mins waiting for use and when I went to use it for the first time, it was frozen and would not respond to any input. Had to power cycle. This is worrisome, as it had been on for less than an hour and not doing any active testing.
Not sure if it is possible, but might it be an idea if a mod could split this into 2 threads, I.e. UPL and UPV? Mods probably have enough on their plates, so please forgive me if this is an imposition.
ampexperts: that does look like hardware problem. The CPU & RAM are soldered, elyt caps look polymer Rohde & Schwarz UPV Audio Analyzer - teardown and repair - Page 1
You get a crash after something warms up. I would start with the AT(X) PSU, replacing all elyt caps first. It should be quite simple and non-destructive with any desoldering station.
You get a crash after something warms up. I would start with the AT(X) PSU, replacing all elyt caps first. It should be quite simple and non-destructive with any desoldering station.
Good WinXP installations would run unattended for months and months and months, even in public-facing server-service, until the Updates got to be frequent.
A WinXP machine here has been running continuously for over 2 years.
Since XP stole the unix TCP stack, an unstable WinXP/7 machine suggests bad hardware. One time I found a loose screw under the mobo. Another time I turned the machine slots-down, beat it, and metal scraps fell out.
A WinXP machine here has been running continuously for over 2 years.
Since XP stole the unix TCP stack, an unstable WinXP/7 machine suggests bad hardware. One time I found a loose screw under the mobo. Another time I turned the machine slots-down, beat it, and metal scraps fell out.
I have several metrology systems that are still using Win XP in our lab. IME the manufacturers software that is loaded onto Win XP is one of the biggest factors in the lockups.
We bought a new KLA P-6 profilometer in 2013. It came with a Dell workstation and XP SP3 on it and from day one it would lock up randomly. I have an external customer that came in a few years ago and they own the same profilometer and their engineers asked me if our system locked up frequently. My answer was daily... Then earlier this year a grad student from another university wanted to use this system because they use the same system in their lab. When I was training them on the use of our system it locked up and required a reboot, the student said "of yeah, ours does that all the time too!"
In contrast I have some other systems with 10 - 15 year old Dell desktops & laptops and Win XP SP2 or SP3 on them and they are on 24/7 for months on end and they never lock up.
These systems are all in a Class 100 cleanroom, so no dust or dirt and ambient temp is a constant 68F. We also do not allow connection to the network because of the old OS so needless to say its hard to blame any external factors.
I ran MEMTest 86 on it yesterday for about 5 hours. No errors found. However, when I hit <esc> to end the test, it didn't return to DOS. The twiddler was still spinning in the lower part of the screen which indicates testing still going on.
I don't know what to make of that.
I don't know what to make of that.
Have you tried using the event viewer to look for clues as to what's happening?
Microsoft says:
"In Windows XP, an event is any significant occurrence in the system or in a program that requires users to be notified, or an entry added to a log. The Event Log Service records application, security, and system events in Event Viewer.
With the event logs in Event Viewer, you can obtain information about your hardware, software, and system components, and monitor security events on a local or remote computer. Event logs can help you identify and diagnose the source of current system problems, or help you predict potential system problems."
Microsoft says:
"In Windows XP, an event is any significant occurrence in the system or in a program that requires users to be notified, or an entry added to a log. The Event Log Service records application, security, and system events in Event Viewer.
With the event logs in Event Viewer, you can obtain information about your hardware, software, and system components, and monitor security events on a local or remote computer. Event logs can help you identify and diagnose the source of current system problems, or help you predict potential system problems."
Today I ran MEMtest 86 again all morning. No errors.
Then I downloaded a 32 bit version of Prime95 and let that run for about an hour. Again, no errors.
So maybe it's just lousy firmware coding that is causing the lockups?
Then I downloaded a 32 bit version of Prime95 and let that run for about an hour. Again, no errors.
So maybe it's just lousy firmware coding that is causing the lockups?
It also could be the drive. My Tivo was intermitantly doing weird things until the drive crashed. The problems looked like bad reception. . .
Interesting.. when I got the unit, the D: partition was corrupted such that I could not format it or even delete it. C: was okay however. Drive is SSD Samsung drive, which are similar to other drives I own and have been very reliable and error-free. But how D: got corrupted, of if the seller somehow used an improper method of creating the partition, I do not know. I managed to fix all of that, but the one time I tried to start Windows in Safe Mode, it would BSOD. So I booted from a Windows install CD and used Recovery Console to delete D: partition and create a new one.
I ran a disc check on C: and it found no errors.
The system has 2GB RAM, so I don't use a swap file. Maybe I should set that up again?
I ran a disc check on C: and it found no errors.
The system has 2GB RAM, so I don't use a swap file. Maybe I should set that up again?
Drives are so cheap and loosing that drive would be a real PITA. I would get another and copy the drive. Even if it doesn't fix it you would have a ready to go backup. And even Samsung SSD's fail. With an SSD a swap drive would make a big difference in XP's performance.
2GB is not particularly a lot. I would definitely setup the swap. With all memory used the system can/will become unstable as most softwares are not properly tested for such condition.
As Demian says - make a copy of your drive, especially when experiencing errors. Do not wait for a failure.
I did make a Macrium Reflect image backup of C: drive.
I will create a swap file and see if that improves reliability. Thanks!
I will create a swap file and see if that improves reliability. Thanks!
It locked up again today, while doing a file transfer over the network. I opened a share folder on the UPV from a desktop PC and started copying files to the PC over the LAN. It froze halfway through the transfer. Went into the lab and sure enough the UPV was unresponsive again. And this is with a swap file created.
Maybe something wrong with the network stack?
Maybe something wrong with the network stack?
The network stack would not use any of the special software. Your hardware (with the possible exception of the drive) seems solid. If it were possible I would start with a fresh drive and install of the correct version of XP on that hardware without the R&S stuff and make sure the XP is stable. Also Prime 95 does not use the drive. These things can be a real PITA to troubleshoot, especially on older/slower hardware. I just installed win 10 on an Atom dual core. It took two days to fully install and update. Time I should have used elsewhere.