Strange noisy baseline

[ScottHorn]

I'm trying to nail down some semi-regular periodic noise that recently appeared on my analytical HPLC SEC system. First off, my setup:

Pump: LabAlliance/Chromtech Series 1+, single piston
Injector: Rheodyne 9725, 5ul loop
Column: Phenomenex BioSep SEC S 3000, 5um, 4.6mm x 300mm
Detector: SpectraSystem UV2000, 280nm
Flow rate: 0.35 ml/min
Average backpressure: 790 PSI +/- 10 PSI (250 PSI BPR after detector)

So, yesterday, I swapped this column for another (SEC, larger pore size), ran a few samples, and swapped back. I believe the noise showed up after the first swap, but stayed after I went back to my original column. The interesting thing is the periodicity of the noise. I'll try to post a picture if necessary, but in short, the amplitude is fairly consistent at 0.6 mAU, the period is usually 0.4 minutes, and this pattern will stay constant for up to ten minutes at a time, then the period will increase to around 1.5 minutes with the same amplitude, and stay that way for 2 minutes, up to ten minutes, before going back to the shorter period. Sometimes the noise will disappear altogether for 5 minutes at a time, before returning. I've primed the pump, made sure all my fittings were tightened well, pulled buffer through the flow cell, and added a backpressure regulator after the flow cell (thinking I had a trapped bubble). None of these actions seemed to change the baseline noise in any way. I think I can rule out electrical interference, as no changes were made around the time I swapped columns that would cause interference. I'm crossing my fingers that the pump isn't about to bite the dust. Does anyone have any ideas? I'll try anything.

Scott Horn

[mjordan]

Hi Scott,
It does sound like you may have an air bubble. I have found that flushing the system with isopropyl alcohol has been very effective at removing air bubbles. Is this something you have tried?

[Gerhard Kratz]

Hi Scott.
Maybe Mark is right. I also would check the piston seals. For me it looks like a leackage. Or the puls damper is gone. Your column is ok. Good luck.

[ScottHorn]

I tried flushing the detector itself with isopropanol. The noise is still there, so I'll try flushing everything with isopropanol, from inlet to waste. I'm concerned about the column though. I don't know if 100% isopropanol will do any damage, and the manufacturers website doesn't mention isopropanol. It does indicate that up to 100% acetonitrile is ok. I'd rather try the easier solutions before I tear apart the pump. Is there a way to test the integrity of the piston seals without getting inside the pump?

Thanks for the assistance.

[Gerhard Kratz]

I would recommend to build in a long tubing between your pump and the injector to see if the pulsation is gone. Please ask Phenomenex about 100% Isopropanol on the BioSep SEC column. Lets see if their tech support is as good as they claim all the times. If you don't get an answer from Phenomenex just ask Showa Denko. Check with a piece of filter paper if there is a leackage on your pump head. Don't be affraid to work on the pump, the piston seal respectively. In HPLC we all are "Plummers" (I instantly hope my English is good enough).

[ScottHorn]

Ok, I'm still having this problem. I've tried the suggestions mentioned so far. Flushing with isopropanol did nothing, I swapped the pump out for a nearly new one, still no change. I've noticed something that leads me to believe it may not be a bubble, and in fact may be some electrical issue. I noticed that stopping the pump does nothing to change the noise. There is still a steady, nearly sinusoidal change in the absorbance with the same period and magnitude as when the pump is running. However, if I disconnect the column outlet from the detector, and block the detector outlet, the noise disappears. If I hook the column outlet back to the detector (whether the pump is running or not) it re-appears. Even if loosely place the fitting from the detector into the column outlet, without screwing it in, the noise is there. It almost seems as if the buffer itself is conducting some sort of electrical signal which causes the response in the detector. Is this possible? Has anyone seen something like this before?

[HW Mueller]

Did you also reopen the detetor outlet when you reconnected the column? Anyway, clearly appears like a massive gas/air problem to me.

[carls]

A ground loop or faulty ground is a possiblity. Can you plug your detector into an oulet on a different electrical circuit than the pump?

[ym3142]

what is the mobile phase? What does it look like? Or can you see any bubbles there or inside the tube?

[ScottHorn]

I really think I can rule out air bubbles. Mobile phase is phosphate buffered saline, w/ sodium azide and some other additives. It's 0.2 micron filtered and I have an inline degasser. I've purged the flow cell w/ isopropanol, water, detergents, etc. I also have a 250 psi backpressure regulator after the detector. The noise is the same with the pump on or off, no change whatsoever. As I stated above, it's nearly sinusoidal with a period that varies between 1-2 minutes. It disappears for a few minutes at a time, but always comes back. It doesn't appear on any of my other detectors, and remains even if I switch this detector to a different channel on the A/D card. This all leads me to believe that there is an electrical problem with the detector itself. I'm interested in the previous post about a ground loop, or faulty ground. Is there some way I could test this theory? I'm not able to move the detector to a different circuit. All of the outlets on the lab benches near it are on the same circuit. Is there some sort of power conditioning device that might solve this problem? And could someone recommend some brands for me to check out?

Scott Horn

[bisnettrj2]

If the other detectors are substantially similar, and are all on the same circuit, shouldn't they all show the same effects of a poor power source?

Can you remove the flow cell from the lightpath and monitor the baseline? This will give you a better idea if the problem is in the detector optics/electronics and not in the cell/pump. If the noise continues, then you have pinned down the problem to the lamp or the optics or the boards in the detector. If it goes away, you're back to an issue with the flowcell, lamp, pump, column, etc...

[HW Mueller]

ScottHorn, are you abandoning your observation of what happened after removing the column? How would removing the column ground your detector electrics, or whatever? What happens if you remove the column and force a liquid through the cell via a syringe?