[excel] Multiple axis line chart in excel

I'm looking for a multiple axis line chart similar to the one in the image below, (which is a javascript chart made by amcharts).

Does excel have an option to draw charts line these? Note there are 3 Y axes, and 3 line charts allowing you to compare data.

Is it possible to get more than 3 data points, each with unique axis on one chart ?

enter image description here

This question is related to excel charts

The answer is


There is a way of displaying 3 Y axis see here.

Excel supports Secondary Axis, i.e. only 2 Y axis. Other way would be to chart the 3rd one separately, and overlay on top of the main chart.


The picture you showd in the question is actually a chart made using JavaScript. It is actually very easy to plot multi-axis chart using JavaScript with the help of 3rd party libraries like HighChart.js or D3.js. Here I propose to use the Funfun Excel add-in which allows you to use JavaScript directly in Excel so you could plot chart like you've showed easily in Excel. Here I made an example using Funfun in Excel.

enter image description here

You could see in this chart you have one axis of Rainfall at the left side while two axis of Temperature and Sea-pressure level at the right side. This is also a combination of line chart and bar chart for different datasets. In this example, with the help of the Funfun add-in, I used HighChart.js to plot this chart.

Funfun also has an online editor in which you could test your JavaScript code with you data. You could check the detailed code of this example on the link below.

https://www.funfun.io/1/#/edit/5a43b416b848f771fbcdee2c

Edit: The content on the previous link has been changed so I posted a new link here. The link below is the original link https://www.funfun.io/1/#/edit/5a55dc978dfd67466879eb24

If you are satisfied with the result you achieved in the online editor, you could easily load the result into you Excel using the URL above. Of couse first you need to insert the Funfun add-in from Insert - My add-ins. Here are some screenshots showing how you could do this.

enter image description here

enter image description here

enter image description here

Disclosure: I'm a developer of Funfun


Taking the answer above as guidance;

I made an extra graph for "hours worked by month", then copy/special-pasted it as a 'linked picture' for use under my other graphs. in other words, I copy pasted my existing graphs over the linked picture made from my new graph with the new axis.. And because it is a linked picture it always updates.

Make it easy on yourself though, make sure you copy an existing graph to build your 'picture' graph - then delete the series or change the data source to what you need as an extra axis. That way you won't have to mess around resizing.

The results were not too bad considering what I wanted to achieve; basically a list of incident frequency bar graph, with a performance tread line, and then a solid 'backdrop' of hours worked.

Thanks to the guy above for the idea!


Best and Free ( maybe only) solution for this is google sheets. i don't know whether it plots as u expected or not but certainly you can draw multiple axes.

Regards

keerthan


An alternative is to normalize the data. Below are three sets of data with widely varying ranges. In the top chart you can see the variation in one series clearly, in another not so clearly, and the third not at all.

In the second range, I have adjusted the series names to include the data range, using this formula in cell C15 and copying it to D15:E15

=C2&" ("&MIN(C3:C9)&" to "&MAX(C3:C9)&")"

I have normalized the values in the data range using this formula in C15 and copying it to the entire range C16:E22

=100*(C3-MIN(C$3:C$9))/(MAX(C$3:C$9)-MIN(C$3:C$9))

In the second chart, you can see a pattern: all series have a low in January, rising to a high in March, and dropping to medium-low value in June or July.

You can modify the normalizing formula however you need:

=100*C3/MAX(C$3:C$9)

=C3/MAX(C$3:C$9)

=(C3-AVERAGE(C$3:C$9))/STDEV(C$3:C$9)

etc.

Normalizing Data