5975C - Random terrible spiking/baseline noise

[BerkeleyPhD]

Dear All,

I have a problem where the baseline of my mass spectrum (top chromatogram in Google drive photos) randomly goes nuts. I know it's a problem downstream of the column because my FID chromatogram (2nd chromatogram from top) always looks good. When I look at the m/z in the noisy part of the baseline, it appears to be uniformly distributed among all m/z values. For reference, the two peaks between 2.4 and 2.8 are the only peaks that are supposed to be there.

I have made no changes to the instrument recently that could have caused this. Please let me know if you have any idea what could be wrong. Thanks in advance for the help!

Link to photos:
https://drive.google.com/drive/folders/ ... sp=sharing

Sincerely,
Frustrated graduate student

[Steve Reimer]

Noise across all masses suggests a problem with electronics. If you don't know how to trouble shoot this, get help.
If you try an autotune, what do you see displayed during the tune process?

[BerkeleyPhD]

The autotune is not much help because it aborts whenever the spiking is occurring and looks fine otherwise.

Unfortunately my advisor is too cheap to pay for Agilent to come service the instrument and I'm stuck troubleshooting on my own.

Is it possible that noise across all masses is caused by ion source contamination or electron multiplier contamination?

Thanks,
Haefa

[James_Ball]

One source of problems can be a contaminated HED. This it the small silver button looking thing at the end of the MS just before the electron multiplier horn, it can be cleaned with a Kimwipe and some Isopropyl alcohol. If it has the triple axis detector it will look more like a rod than a button I think, not sure since I haven't used one with that.

It can also be from a overload by a solvent peak that elutes after your targets of interest. What are the analytes that elude in the 2 minute range and what is your solvent? Water will do this since it is normally below the scanning range of the instrument and the quads can't reject all of it if it is too large an amount so it appears as hits all across the mass range.

The third problem is something Agilent would most likely need to fix, since it would be a problem with the mass filter electronics. If it is doing this while in autotune and no analytes are passing through the column, then something with the quad driver board is probably to blame. That is the board on the side of the analyzer, more often called the Side Board by Agilent techs. It could be as simple as the quads need balancing, but if you have never seen it done, probably not something to try on your own. You can inspect the board by powering down and removing the black plastic covers and looking to see if there are any fried components on the board. Otherwise probably a call to Agilent will be the only solution.

[BerkeleyPhD]

Thanks for the input! It's definitely not an elution issue because it occurs even when I'm not injecting anything.

So are you saying that it isn't possible for ion source contamination or electron multiplier contamination to cause this particular issue?

Thanks,
Haefa

[James_Ball]

Thanks for the input! It's definitely not an elution issue because it occurs even when I'm not injecting anything.

So are you saying that it isn't possible for ion source contamination or electron multiplier contamination to cause this particular issue?

Thanks,
Haefa

Any place that can have contamination that causes unstable electrical conductance can cause noise, including the source and multiplier. Cleaning the source and making sure all of the connector posts are tight is the first check, if that doesn't fix the problem, then the other places should be checked.

The strange thing about the first two chromatograms is how the baseline is flat until about 4 minutes then it goes crazy. What is the target analyte and what is the solvent in those two runs?

[BerkeleyPhD]

Thanks again! I was able to take apart the detector and clean it myself (against Agilent's recommendation), and this fixed the issue!

[cjm]

I just recently had this same exact problem...for the second time. The first time it occurred I cleaned the source, changed the filaments, ran an HED clean test, new liner, clipped the column, and it seemed a temporary fix...until it showed itself again in a few days. Then I was told to replace the column. Seriously. I said there's no way this noise with all m/z is coming from the column. I just changed a column yesterday after this problem reoccurred. Signal is nice and smooth. Problem gone. I can't explain it.

If anyone can tell me why in the world the only fix I had for this is a column change, I'd love to hear it.

[BerkeleyPhD]

I just recently had this same exact problem...for the second time. The first time it occurred I cleaned the source, changed the filaments, ran an HED clean test, new liner, clipped the column, and it seemed a temporary fix...until it showed itself again in a few days. Then I was told to replace the column. Seriously. I said there's no way this noise with all m/z is coming from the column. I just changed a column yesterday after this problem reoccurred. Signal is nice and smooth. Problem gone. I can't explain it.

If anyone can tell me why in the world the only fix I had for this is a column change, I'd love to hear it.

That's interesting. Are you sure the problem has actually been permanently resolved? It's really hard to determine what does the trick when the symptoms are intermittent anyway. I'm wondering if my "fix" is only temporary as well.

[cjm]

I'll be very curious to hear if you have fixed it permanently. I struggled with it for a couple weeks after taking the MS down several times, and the problem kept coming back randomly. Some runs had noise everywhere, others had random time segments that were contaminated with noise and then the baseline would reset to normal. I changed that column out on November 3 and the problem went away permanently until just a few days ago. 3 months is about my expected column lifetime with the amount of extracts I run, so I didn't hesitate to change out for a new column. I'm using Rtx-5Sil MS 30m/0.25mm/0.25um by the way..

[BerkeleyPhD]

I'll be very curious to hear if you have fixed it permanently. I struggled with it for a couple weeks after taking the MS down several times, and the problem kept coming back randomly. Some runs had noise everywhere, others had random time segments that were contaminated with noise and then the baseline would reset to normal. I changed that column out on November 3 and the problem went away permanently until just a few days ago. 3 months is about my expected column lifetime with the amount of extracts I run, so I didn't hesitate to change out for a new column. I'm using Rtx-5Sil MS 30m/0.25mm/0.25um by the way..

Yeah, I'll let you know! My issue was similar; some runs had no noise at all, some had only partial segments that were noisy, and some runs were all noise. My column is a HP-1. I'll consider changing it if the problem comes back after the HED cleaning, but so far so good!

[cjm]

Another test to try: cap the transfer line and run an autotune. Somewhere on the printout it will tell you the total number of peaks. If it's less than about 250 it's not the MS and most likely the column. If it's 400 or more, it may be time to replace the EM. This test is the simplest way to isolate the column from the MS and test the MS only.

Also, here's a way to test MS for cleanliness (found here on the forum): In manual tune in the command line type: Macro "\MSDChem\diag\uphonics.mac" (then hit enter, then type): u_phonic (then hit enter)

it should give you printout of EM noise at three voltages. Standard deviation for all three should be single digits and all about the same. If the one at 2000v is high I believe it's the HED that may need cleaning.

Anyone else use this?

[aldehyde]

Thanks again! I was able to take apart the detector and clean it myself (against Agilent's recommendation), and this fixed the issue!

Actually Agilent does recommend cleaning this part, you just need to be careful. The MS detector is very simple (solid state electronics) but can become contaminated with dust. Cleaning it with IPA, a lint free cloth, and clean/dry nitrogen is the recommendation when this occurs.

The likely cause of your problem is that at some point when you vented some dust got onto the HED surface. The HED is set to 10 KV during operation, and that piece of dust became an electron emitter. As the quadrupole scanned and filtered the piece of dust was constantly shooting electrons up into the electron multiplier, which causes you to detect response at every mass.

When the HED power supply gives problems the symptoms are different -- total loss of signal, the run or tune aborting with an HED fault, etc. That requires replacement, otherwise the problem can almost always be solved with careful maintenance.