How to know whether my choice design is orthogonal? (Validity of choice experime

[#1]

For my master thesis I made a Discrete Choice Experminent in JMP. The following design roled out of JMP (with 4 attributes, 3 levels each, and 16 choice sets) 

When I checked the validity of the design I made a correlation matrix of dummy variables. The attributes Gar2/Gar3 (warranty) and EEO 450,525 and 600 (efficiency of energy conversion) correlates with each other. Is this correlation to high to be a orthogonal design or does it not reach a specific threshold/ boundary? In other words, do I need to make a different design? 

[#2]

The design is not orthogonal when there is even a single non-zero correlation. The question is not actually about the design matrix. It is about the model matrix that is used to estimate the parameters.

The question, "is the correlation too high," is still valid. The answer is that it depends on the variance of the response. The correlation of the estimates will inflate the variance of the estimates. This inflation reduces the power of the tests for significant estimates. It will increase the uncertainty in the predictions of the model. If your response exhibits slight variation, you can tolerate more considerable inflation due to correlation. If your response shows a substantial variation, you might be unable to take any inflation.