Dionex 3000 Pump - air bubble problem

[bunnahabhain]

Hello folks

I am experiencing a severe problem with pressure ripples in my Dionex Ultimate 3000 pump. We do have to change running buffers rather frequently but channel D is always 100 % acetonitrile. Each time when I purge any solvent line I get into severe trouble with the D channel. (and ONLY the D channel). I have pressure ripples of ca. 5 bar at operationg pressures of ca. 40 bar and 1 ml/min flow. All other channels are OK.
What I tried to remove this air bubble:
- I flushed line D with Isopropanol -> water -> ACN. This helps sometimes. Today it did not.
- I tried to run the pump at 100 % D at high pressure (150 bar) for 20 min. Did not help
- I tried to knock on the pump head with opened purge valve. Did not help.

Has anyone an idea how to get rid of this ripples? The strange thing is that on line B which is pumped by the same pump head the pressure is stable.

Thanks a lot,
joerg

[Gaetan Glauser]

Line D filter or degaser membrane are the two things that come to my mind.

[John Lim]

Dear Joerg,

The suggestion by Gaetan can be tried as the most facile one by first replacing the in line filter in the CH D eluent bottle. If the problem persists and a clogged degasser membrane is suspected then perhaps isolate the in line degasser by temporarily bypassing the degasser CH D and priming with manually degassed acetonitrile. I would have expected an error fault indication but maybe it's not severe enough to trigger the degasser sensor. If you are not already using an in-line filter this would be a good idea to install one for each eluent line channel.

If the above actions do not work then your D channel solenoid proportioning valve may be faulty. Most straightforward way is to perform a gradient proportioning valve step & ripple tests to directly isolate and verify the issue. Your local service engineer can quickly run the tests to verify performance to factory specs and if the gradient step tests fail, then the LPG valve assembly will require replacement.

You also mention buffers for the other channels so if you are mixing channel D containing acetonitrile with your buffer be diligent to avoid in line precipitation at higher [ACN] depending on the buffer being used. The resultant precipitates may damage the LPG valve, possibly the pump head check valves and injection valve rotor seals downstream. Pump pulsation issues are also frequently associated with bad pump check valves. Could also be a port/fitting leak or insufficient back pressure on your detector flow cell.

I recommend following the basic rules of troubleshooting by performing only one test variable at a time until the source of the problem is identified.

Again, a service call is recommended to quickly address the issue to get your system running to maximize your uptime. Good luck!

[bunnahabhain]

OK Thank you John for these instructions.

What I did so far is:
- I cleaned the pump check valves in ultrasonic bath => No effect
- I used channel D without the degasser. => Pressure is OK (I still have 0.5 bar ripple, which is too much for my taste but it is within the pump specification).
- At the moment I run the degasser channel in backwards direction with water, I will then try out forwards again.

I'll give you an update tomorrow.
Joerg

[unmgvar]

it looks like you have a high pressure pump
the degasser you have is probably a 4 channel
inmost cases the dionex technicians use the degaser channel of D in order to connect the solvent line used by the syringe in order to have a good degassed solvent there. for better accuracy and greater reproducability.

so it is normal if that channel tends to be more affected by air troubles if you use it for work. because D is not normally used, it is the reason it's degasser channel is given to the sampler solvent.

now has suggested maybe you get more bubbles now because of the filter in the channel
most of the time the D bottle simply sits there and many time it is not replaced too often,
so stuff accumulates in the bottle and then clogs the filter, which can also result in increasing bubbles going out of the solvent

anyway anything that is within +- 1 bar is fine and should not be of matter unless you actually see that the chromatogram is not stable enough

[bunnahabhain]

it looks like you have a high pressure pump
the degasser you have is probably a 4 channel
inmost cases the dionex technicians use the degaser channel of D in order to connect the solvent line used by the syringe in order to have a good degassed solvent there. for better accuracy and greater reproducability.

This is correct. I have ACN in channel D and the tube from degasser to pump is split by a tee-piece to pump and syringe wash. I have checked these fittings and they are finger-tightened by strong fingers

anyway anything that is within +- 1 bar is fine and should not be of matter unless you actually see that the chromatogram is not stable enough

This is true, but with 3 bars I experienced peak wandering of +- 0.5 min which is not OK for my needs. I now tend to pre-mix eluents for isocratic runs and use one of the other channels to avoid this.

I'm just new to this HPLC system and I never experienced this kind of trouble before (with Waters and Agilent equipment).

Greetings,
Joerg

[unmgvar]

if you have a T it means that D bottle is also connected to the degasser and then split to the sampler
so you should not get too much bubbles then in D. i thought D was not connected at all to the degasser.
3 bars, maybe it has to do with the amount of ACN?
are you using a small amount?
in high pressure mixing each pump give the ratio of solvent by giving the ratio of the flow
so if you give 3% acn, yo uare working only at 0.03ml/min on the pump.
in gradient profile it is ok, but in isocratic mode the pump cannot work well at such low flows.
if you have a big ratio, of 40% 60% then you should not get a problem, if you do it means that you have a problem somewhere that needs looked into
in most cases, the bubbles are coming from the water solvent, so actually the problem might be there, and maybe you have a compression problem.

and yes you are correct for isocratic solvents, it is best to do pre-mixing. it is more reliable