Chromatography Forum

We have a Waters/Micromass Q-Tof Premier attached to an IBM workstation running MassLynx 4.1 as supplied by Waters. The workstation recently suffered a significant failure which required BIOS reflashing and restoration of Windows XP SP2 and the software.

Before this failure read backs for cone voltage, entrance potential etc in nESI+ were approximately equal to the value set in MassLynx when the instrument was in Operate. Since the repairs we have seen values of -1 in Standby and significantly negative numbers in Operate, even though the voltages are set positive. Also, background signal is not seen in the TOF MS window.

The workstation is connected to the Q-Tof and the EPC is able to login by Telnet with no problems reported.

We had seen this issue before with other (non-supported) PCs too, one running Windows XP SP3 and the other running Windows 7 SP1.

We haven't refitted the MALDI source since the issue appeared but intend to try this next week.

If anyone has any suggestions as to what could be causing this issue it would be greatly appreciated. I feel like it must be something relatively simple which we're not setting up quite right!

Have you set theMCP voltage?

If you re-flashed the BIOS you'll need to set all power management to disable (APM disable), disable onboard audio, disable all hibernate or standby modes, instrument lan should be set to 100/half duplex with no standby or sleep mode and finally check settings for IIS configuration. Watch EPC boot process via hyper-term to insure all units are recognized and loaded. You can also check all required voltages via rio.

Thanks for the replies.

We have the lab powered down until Tuesday 16th but I will double check everything that has been suggested soon after then.

Hi mhr311,

I wondered if you could explain why half duplex is required for the Instrument LAN and what 'rio' is?

We are still seeing the same problem unfortunately.

Thanks a lot.

Duplex refers to if terminals can TX/RX simultaneously or singular path. In short does the instrument send/receive to/from the host PC while the host does the same. On Masslynx systems rio is shorthand for rioMainMenu, a telnet window allowing diagnostics and checks of various system subunits. It allows you to view the units are in fact loaded, initialized and correct voltages/references/readbacks are observed. The first thing to do is exit masslynx, install a 9-pin serial cable between (for example) COM1 on the host PC and COM1 on the EPC (QTOF Premier may have a 9-pin serial port labeled PC Link?). Set up a Hyperterm window with default setting (9600/8/N/1) and connect over COM1. Reboot EPC and watch the actual boot process (Capture if required) to make sure all units are being loaded. Once you're sure the boot is OK, end the hyperterm session, remove the serial cable and reboot the EPC.
To run rio, set up a telnet window, connect to either EPC or whatever ip you use and check units for readbacks. Note you can cause real problems in rio by turning things on/off that Masslynx would not allow so I suggest caution.

I have recorded the boot process as seen in Hyperterminal which I've linked here, just in case you are able to spot anything unusual.

https://www.dropbox.com/s/d3xudozn2hewy ... t.txt?dl=0

I will have a look at using rio too. Caution noted!

Thanks

*Edit*
Just thought I'd ask if you could provide a few instructions on using riomainmenu? I've accessed it through the telnet.

Thanks again.

The EPC boot looks OK although it does note an 8000amu quad front end. I think the base system had a 4000amu quad and an 8000 upgrade required both different link settings and different ini file edits, so a fresh install calling an 8K quad is surprising, but I could be mistaken.
If you haven't used rio before I would be uncomfortable attempting to help you navigate through it. However you can access subsystems, as per the menu, then display readbacks. Do not attempt any writes to DACS or unlocking any blocks since you would need the bit descriptions to do so. Note that you have exited Masslynx when calling rio.

Quick update-

We now have signal once more! It appears on reinstallation MassLynx doesn't recall default TOF tuning settings and so those it starts with are way off what're required. The tune page monitor must not be in High Res mode either.

As far as I can tell it was a combination of abnormal settings rather than a communications problem that has caused our problem.

Thanks for all help and suggestions!