Dax measure that sums data by week, then returns max value

[h4tt3n]

Hello folks,

I am looking for a DAX measure that returns the max value of a data column summed by week.

I have a table with two columns and millions of rows, with data and timestamp in each column. I aggregate the data so it shows a graph with one column representing a week with the summed data for that week in each column.  (Note that table columns and graph columns are different things). So far so good.

Now I want to scale the data properly, eg. showing columns with large values not as Watt W but kiloWatt kW, by dividing with 1000. In order to show all columns with the same scale, I need to calculate the "global" max value of all week columns and then divide all columns with a factor derived from this max value ( this equation is in place and working).

Example:

I have 24 * 30 hour-based data points in a table together with the timestamp for each data point. ALl data stretches one month (30 days). I am looking for a DAX measure that can do a sum-by-week of the data set, turning it into a table with four rows and two columns, with the week sum in one column and week number in the other column. Then, finally, the Dax measure returns the scalar value of the biggest weekly column.

Ideas and feedback is appreciated.

Cheers, Mike

[Greg_Deckler]

@h4tt3n Not really enough information to go on, please first check if your issue is a common issue listed here: https://community.powerbi.com/t5/Community-Blog/Before-You-Post-Read-This/ba-p/1116882

Also, please see this post regarding How to Get Your Question Answered Quickly: https://community.powerbi.com/t5/Community-Blog/How-to-Get-Your-Question-Answered-Quickly/ba-p/38490

The most important parts are:
1. Sample data as text, use the table tool in the editing bar
2. Expected output from sample data
3. Explanation in words of how to get from 1. to 2.