I’m using a new Balance 20 that I just put online, replacing one that bit the dust. My main local network device is a Kyocera KR2 (yeah, I know, not the greatest) that I’m using for three wired ethernet connections, and a fourth wired ethernet connection to lan port 4 on the balance 20. At any given time, I’ll have up to 10 wifi devices connected to the KR2. However, I am using the DHCP server in the Balance 20. For a wan device, I’m using a Pantech UML295 connect to the Balance 20 usb port, with a yagi antenna on the roof pointed at the cell tower 1.08 miles away… Verizon is my ISP.
Here’s the weird part. As this network is configured, the mobile internet speed hums right along. I usually get 15 to 17 Mbps DL speed with my desktop PC hard-wired to one of the ethernet ports on the KR2 router, which pulls the internet connection from the usb port on the Balance 20 . It literally runs as fast as plugging the Pantech UML295 directly into a usb.port on the PC.
However, if I unplug the PCs’ ethernet cable from the KR2 router and plug it directly into the Balance 20 eliminating the KR2, the internet speed drops to less than half. I can even shut down the KR2 router alltogether and just utilize the Balance 20, and the drop in speed is the same. I would have thought that eliminating the KR2 router out of the chain would make the internet connection faster, since it would be one less device in the path. But, this is not the case. The only thing I can figure is that the KR2 is doing some traffic shaping that the Balance 20/usb wan likes. I have fine-tuned my MTU on the Balance 20 to 1428, as this seems to be the optimum size.
I utilize the KR2 router for the local ethernet anyway, and the Balance20/wan internet speed seems to love it. So, this isn’t an issue. But. just off the top of my head, it seems kinda weird that it would work this way.