• "We use gradient mobile phase".

I am sorry, but your message does not make any sense. You report a change in system pressure during a gradient analysis. This would be expected, and is normal. If the pressure does not change during the gradient analysis, then something is wrong. All three solvents you list have different physical and chemical properties. Methanol is more viscous so usually increases pressure. This is normal. A change in pressure from 40 bars to 52 bars is 30%, not 50%.

• Also, if you are using UV/VIS detection, be sure to stay above 330nm to stay away from Acetone's major abs peak around 275nm.
• Confirm complete miscibility of all mobile phase mixtures before use.

• "ratio of phases was A:B:C as 7%:50%:43%";

I think there may be a language and/or training issue(s) causing confusion with your post.

Are you running a gradient or isocratic analysis? Maybe you are NOT running a gradient as you stated. Maybe you are using what we refer to as a "ternary mixture" (three different solvents mixed together) in an ISOCRATIC analysis (the same composition during analysis). * A gradient analysis is by definition an analysis where the % COMPOSITION of the mobile phase solutions is changed during the analysis run (or run time). We do not refer to a mixture of liquids as a 'gradient'.

If your analysis is in fact an ISOCRATIC analysis, using Hexane/Methanol/Acetone ( 7%:50%:43% ), then: (a) Confirm complete miscibility of all mobile phase mixtures before use. (b) Confirm that all solvents and additives used are HPLC grade or quality and Filtered. (c) the column used is FRESH, new and clean. Columns are inexpensive compared to the time and materials used, so start with a fresh column. (d) A proper wash method using a column wash solution that is STRONGER than the mobile phase is used after each run. (e) All samples are fully dissolved in the mobile phase and then filtered through a chemically compatible filter before being injected into the HPLC system. *These are all normal procedures used in HPLC analysis.

Please have an experienced chromatographer who is local to you assist you with the method and analysis.

• "Diagnosing & Troubleshooting HPLC Pressure Fluctuation Problems (Unstable Baseline)": https://hplctips.blogspot.com/2014/01/diagnosing-troubleshooting-hplc.html

A common reason for HPLC system pressure to slowly rise during an isocratic run (analysis) is caused by fouling of the column and/or flow path. The above article includes information on this issue.

Thank you for detailed answer. The ratio A:B:C as 7%:50%:43%" is only from 23 minutes (ratio of B and C changes during the analysis, A phase remain unchanged) before that there is no such problem but with increasing of B phase the pressure also get higher. Indeed, it would be better to have an experienced chromatographer, but we do not have such opportunity and you're right it may be that something lost in translation because English is not my native language. Anyway, I'm very appreciate for your help. Спасибо!

Не за что

SO you are running a gradient, not isocratic method. As already stated, pressure MUST change as composition changes (normal). Perhaps you have a problem with your HPLC pump? We can not troubleshoot such items via the web. Run the method with NO SAMPLES and compare results. Record pressure vs % composition change. They should show logical response over time, with delay time for the volume of the HPLC system and the Column. I am still not sure what you are describing. The mixture of solvents alone is very troubling and may imply an improper method. Can you share the actual HPLC method with us?