Agilent 5975C bad baseline noise

[ajpope85]

Hi,

We're having a problem with one of our instrument's baseline



I've opened the instrument up and cleaned the source and sprayed off the quads and detector to clean it of any lint that may have fallen into it. I've also made sure that the leads are all fairly snug. This looks like some sort of electronic failure to me, but I'm not very experienced with MSDs.

My 69 peak is around 800k count with my EMV @ ~1500. I have roughly 150 peaks in manual scan. The scan threshold is set to 75, but this noise is well above that. All my peaks look fairly normal until you get to the tail, then this noise starts to present.

Could this be a problem with a power supply or lead? Any help is appreciated.

Thanks

edit: I discussed this with a colleague and she seems to think it's vibrational noise. Normally, I would agree but I feel like the magnitude of the jumps is too regular to be vibrational.

[James_Ball]

What ions are present at that point in the chromatogram?

Can you post a pic of the spectra at around 12 minutes?

[Steve Reimer]

Was this present before you opened up the MS and cleaned it?
I have seen this before in my 5975 and it was a connection problem, but I don't remember which connection.
When checking wires and plugs don't forget the ribbon cables (prone to breaks) and the high voltage connector (likes like the coil wire on your car and can burn if loose). I've had both fail on 5975/5973s.

[kubowicz.tomasz]

Hello

First of all run autotune - it is best diagnostic tool (you can paste report here).
It looks like dirty dynode in MSD - you can clean it with IPA (normally it helps).

Regards

Tomasz Kubowicz

[ajpope85]

Hey, sorry for the delay in response. I recleaned the source and it turns out the repeller ceramics had cracked, so I had to replace them and let it pump down over the weekend. It did not fix the problem.

In response to the first question, this is the spectrum at ~12 minutes.



Yes, we saw this problem before we had cleaned the source, and blew out the quads and detector with duster.

Here is the autotune report that I did today. The instrument PC automatically prints to PDF, and I don't know how to change that so I can't just copy/paste it.

https://drive.google.com/file/d/0B8xUQV ... sp=sharing

The bit of noise in the autotune spectrum is probably our hydrogen carrier gas still interacting with the new ceramics. When I pumped the instrument down, I had to leave and forgot to apply the heater settings, so the source and quad only set at 100 degC over the weekend. Normally, you don't see all those tiny low mass peaks.

I may try to clean the detector with IPA today if I get a chance. When you talk about cleaning the detector, are you talking about cleaning everything past the quad?

Thanks again for your help.

[DReggio]

Pull out the column and plug the detector. That way you can definitely rule out anything before the detector. Did this happen all of the sudden or after a source cleaning? What do you use to clean your source?

[James_Ball]

Hey, sorry for the delay in response. I recleaned the source and it turns out the repeller ceramics had cracked, so I had to replace them and let it pump down over the weekend. It did not fix the problem.

In response to the first question, this is the spectrum at ~12 minutes.



Yes, we saw this problem before we had cleaned the source, and blew out the quads and detector with duster.

Here is the autotune report that I did today. The instrument PC automatically prints to PDF, and I don't know how to change that so I can't just copy/paste it.

https://drive.google.com/file/d/0B8xUQV ... sp=sharing

The bit of noise in the autotune spectrum is probably our hydrogen carrier gas still interacting with the new ceramics. When I pumped the instrument down, I had to leave and forgot to apply the heater settings, so the source and quad only set at 100 degC over the weekend. Normally, you don't see all those tiny low mass peaks.

I may try to clean the detector with IPA today if I get a chance. When you talk about cleaning the detector, are you talking about cleaning everything past the quad?

Thanks again for your help.

Hydrogen carrier definitely explains what looks like hydrocarbons in the spectra in both the analysis run and the autotune.

You don't want to clean the electron multiplier itself. What you clean is the HED which it the silver button about the size of a 2ml autosampler vial cap and is mounted to a ceramic piece between the end of the quad and the electron multiplier. If the 5975 is like the 5973, you just have to unplug the heavy silver wire that connects it to the side plate and remove the two allen screws that hold the ceramic piece on. Just pull it out and clean the metal part with IPA, really easy to do. While I have it out, I usually go ahead and swab down the inside of the metal mounting bracket where you pull the HED out just for good measure.

Usually lint on the quads will cause much larger and more random spikes to occur, at least it always has for me.

With the spectra not having some unusual ion pattern, there are only a couple other things I can think of that would cause the noise, which would be an electron multiplier acting up near the end of its life, or the signal amp located on the side board(I think it is on the side board instead of the main board) giving problems. First is easy to diagnose and fix if it is the problem just by replacing the electron multiplier, the other is a little more difficult unless you have a spare board.

Also make sure your grounds are good because that can cause all sorts of problems. I just had an ICPMS that was giving very unsteady signals and was spiking like it had static discharges in the analyzer. After about a week I finally looked at the plug in the wall socket, it was not pushed all the way in and locked, once I cleaned it up and installed it properly it worked great.

[Gizmo]

The spectrum you showed above for your baseline noise looks very much like a second problem I've developed while troubleshooting for a different random noise issue in another thread. It showed up right after I cleaned the source. I took it apart and re-did the sonicating solvent rinses and it is still there. I capped the MSD to rule out contamination from the column/gc and it is still there.

Were you able to determine the source of this problem?

[ajpope85]

Hi, sorry for the very late reply. Been really busy.

No, we never figured out what exactly was causing the noise. The only thing that removed it was increasing the sampling rate to 4 and upping the threshold to 250, while changing the low end of the mass acquisition from 25 amu to 34 amu. A big part of the noise, which wasn't shown on the images I posted before, because I hadn't isolated what was causing the spikes, was mass 29.1. It was spiking all over the place, causing that sawtooth wave pattern that you see in the TIC. I'm assuming it's some sort of electronics failure to get an aberrant peak like that so consistently. We're experimenting right now to see if there's any combination of lower threshold and sampling rate that will avoid all the noise, because we are uncomfortable with the amount of data points over our peaks.

Thanks for all ya'll's (that's an awkward looking word) help.

[James_Ball]

Hi, sorry for the very late reply. Been really busy.

No, we never figured out what exactly was causing the noise. The only thing that removed it was increasing the sampling rate to 4 and upping the threshold to 250, while changing the low end of the mass acquisition from 25 amu to 34 amu. A big part of the noise, which wasn't shown on the images I posted before, because I hadn't isolated what was causing the spikes, was mass 29.1. It was spiking all over the place, causing that sawtooth wave pattern that you see in the TIC. I'm assuming it's some sort of electronics failure to get an aberrant peak like that so consistently. We're experimenting right now to see if there's any combination of lower threshold and sampling rate that will avoid all the noise, because we are uncomfortable with the amount of data points over our peaks.

Thanks for all ya'll's (that's an awkward looking word) help.

m/z 29.1, that is strange.

If you are still running hydrogen carrier I wonder if it could be some kind of interaction between the hydrogen and trace nitrogen in the system. N2H would be a strange molecule for certain. COH would be another strange molecule, but who knows what could happen when you get chemical ionization with hydrogen in the source.