Uneven retention times (HPSEC method)

[michbrrt]

Just forgot a detail: the flow is now 1.006 mL/min, as it should have been all along. Remember: no leaks were ever detected on the pump.

[unmgvar]

HW what i tried to explain was the correlation between the change in the flow to the peak area. not all changes in the flow rate will affect the peak area like they all affect the RT.
the flow variation will affect the peak area only if it occurs when the sample is going throu the cell.
we can show this easily.
lets say we run something that elutes in 5 minutes at a flow of 1 ml/min.we will get a certain peak area.
lets play and run it now with a fake leak. we run the first 4 minutes at a flow of 0.8 ml/min and bring it back to 1 ml/min for the remaining of the run.
the RT will shift but the peak area should be about the same because our compound will go throu the flow cell at 1 ml/min. the overall shift in RT will make it look as of the flow rate on average was a value between 0.8-1 ml/min.
now if the change in the flow rate is constant then it might mean one thing, for example a leak, and if it shifts then it will mean somethig else for example bubbles, or clogging due to dirty samples.

[HW Mueller]

unmgvar, OK, I misunderstood your statement somewhat, probably because I would correct a flow rate undulation of a certain size, no matter how good the areas look.
So you did have a correlation between pressure and flow/rt. Don´t quite understand how, but if it is working . . . . good luck.

[unmgvar]

the final result counts, it works fine now.
like HW, i think that there has to be a connection between flow/rt in your problem and it most probably showed has a pressure fluctuation somewhere. it was probably dismissed because your overall pressure is very low to begin with.
from one of my experiences when the tubing between the needle and column inlet is blocked then we see a serious fluctuation in the first few minutes of the run and then it generally stabilises back to normal.