If measured under identical running conditions (same exact flow rate and mobile phase composition), a back pressure increase from 1500 to 1900 psi usually indicates column fouling (very common with ion chromatographs due to lack of cleaning of the main column, suppressor column (if used) and system after several runs). Fouling over time obstructs flow through the column resulting in a rise in back-pressure over time. This is very different than an obstruction or 'plug' which will result in the pressure hitting the max limit. Plugs of this type require removal and evaluation of the major in-line components to find the area which is clogged. Sometimes parts (or the column) must be replaced as removing or backflushing a plug out of the system is not always possible. Partial obstructions as you describe are far more likely to be found in the column(s) used or possibly in the injector, if it has not been properly flushed and maintained. These areas should be addressed first.

Please clean the ion column using the manufacturer's recommended procedure, then recheck. If there are any concerns or you are unable to restore normal operation of the column, then please replace it with a new one (and include proper washing steps in future methods to reduce the chances of it happening again). Columns are consumable items and do require replacement.

BTW: The main column can be removed and a restriction capillary (or union) can be placed in-line to clean the system and not risk damaging the column. This should be covered in ion chromatography LC training as well as mentioned in the system operating manual.

Additional general information on why system backpressure can rise over time in HPLC systems can be found in this linked article: "Diagnosing & Troubleshooting HPLC Pressure Fluctuation Problems (Unstable Baseline)".

http://hplctips.blogspot.com/2014/01/diagnosing-troubleshooting-hplc.html

In addition to the above wines do have organic components like tannins which may foul your guard column unless you take sample clean-up measures like SPE.

Keeping flow at a rate of 1ml per minute, sequentially disconnect in-line items from the point that fluids exit the system. If pressure drops after disconnecting a part, this part is where the clog/obstruction  is. It is important to disconnect one piece at a time in order to pinpoint where the problem is. Once you find it,  look for any frits or filters that may have become clogged in your system. Or, if you have spare columns or guards, you can swap them out one at a time to rule out column fouling as a problem.

Instead of trying to break your head on such things, just call up the application engineer to help you solve this problem, while you concentrate on some important aspects of your research problem. 

Bill has linked a very good information to attend such problems. But Dr. Ghosh is also right, we can not resolve all technical faults. Better to consult engineer.