There is no straightforward answer for this because it is not a recommend procedure. I have provided various answers in relation to this topic:

Microarray and RNA-seq

• Combine microarray and RNA-seq gene expression for differentially expression analysis
• Compare Microarray with RNA-Seq

Microarray and microarray

• A: How to integrate multiple data sets from microarray platform prior meta-analysis
• How to merge different microarray data sets from different platforms (not for differential expression analysis)?

Your best option is to process them independently and then perform a 'meta analysis' of the results. You can also do things like correlate the fold-changes, as I did here with RNA-seq and proteomics:

You could also perform gene enrichment and pathway analysis on the results and then cross-compare.

Kevin