help..tensile fracture behaviour on tempered steel
No, I would not expect a quench delay to be the cause of the strange results you describe. What a tempering delay could do is give time for quench cracks to form. If the stress within the part is high enough, untempered martensite will crack, and the stress to cause cracking will be less if you have more time. (I've had a load of cutting tools crack because they sat overnight without tempering. You really remember your mistakes!)
Does the fracture surface of the one tempered at the lowest temperature look brittle? That is, lots of flat fracture, perhaps a granular appearance? If so, then perhaps cracking while waiting played a role, although I would have expected it to show up in all the test samples.