VLAN mystery

Okay guys/gals, I have a good one for you. I'm having a very interesting problem, if you consider it that, because everything works...Ha. Theoretically, I don't think is possible. It's a complete violation of 802.1Q. Anyways, I'll do my best to explain it fully and clearly. My test machine(HP EliteDesk | Win10 | 64bit) is plugged into an access switch(Cisco small business SGE2010P) via ethernet cable. The port it's plugged into is an access port with VLAN 20 untagged. The access switch has a trunk port going directly to our core layer 2/3 infrastructure(Juniper VCF consisting of QFX 5100's and EX4300's). The trunk link has VLAN 1U, 20T, and 22T. U for untagged and T for tagged. If I do an ipconfig /release and /renew on this machine, sometimes I get a .20 address.....sometimes I get a .22 address. No matter what address I get, .20 or .22, everything works fine. I even tried reserving an IP in DHCP on the .20, since that's what VLAN the port is on, /release /renew, get a .22. What in the world is going on here? How is it possible to first, have 100% functionality with a .22 IP address on a VLAN 20 port, and second, get a .22 IP address from DHCP on a VLAN 20 port. I'm really confused. I appreciate you taking the time to read this and help me out. Let me know any thoughts, comments, suggestions.