It would be easier to help if you posted actual code.
If you are using a waveform chart, you have essentially 3 pieces of data. 1. The start time (T0) , the time interval between samples (dT), and a 1-D array of samples (Y) that assumes that they are evenly spaced in time. You can add 1 more data items to the chart at a time, and that data just gets appended to the end of the Y array (also known as the history).
If you set T0 and dT right, then it will look like you have your real time progressing along the X axis. But if you are putting in items at 1 per second, but then skip a minute before putting in a new item, that new item will look like it is 1 second more recent on the chart rather than 1 minute more recent. Your times will fall behind.
You should use an XY graph. It allows you to put in data where the X values are not evenly spaced. You feed it a 1-D array of X values and a 1-D array of Y values. (Check the examples for XY graphs because there are other ways to build the data to feed to the graph using clusters.) The disadvantage is that you will have to maintain your own history by storing the data arrays in shift registers and appending the new data yourself using build array.
One other alternative using a waveform chart is that during your pause time, you feed the constant NaN to the chart at the same rate you would have been feeding actual data. It will essential create blank points in the chart, but the time intervals between points will remain consistent.
