I only scanned that Rane article, but it looks like it's got some good information. I didn't want to muddy things that much, so I just said it will probably work.
There are two main differences between the AES professional standard and the spdif consumer standard. One is that the different standards use different electrical standards for transmitting the data. This is all the Rane article gets into. The other big difference is the data requirements of the standards -- each standard has slightly different requirements as to what the data word that characterizes the audio must have. Among other things, this includes how to set the professional vs consumer bit of the data word. It is this latter difference I think might be causing the problem, so that is what I'd like the OP to test by using the AES output.
In terms of the electrical difference, the main difference I think is the 110 ohm impedence of the AES standard vs 75 ohm for spdif, and that the AES signal is about 5v, while the spdif signal is 0.5v. The Rane article gets into how to properly convert the electrical side of these two digital outputs.
To do it right, you should at least use a 110>75 ohm impedence transformer that will change the impedence of the AES signal and drop its voltage. Something like this:
http://www.canare.com/index.cfm?objectid=DCBD1A35-3048-7098-AFF302A4C46790F8The Rane article describes other ways to get to the same result, rather than a storebought transformer.
When I had my Lunatec V3, I used a Canare transformer, and it worked great.
I also used a simple XLR>RCA cable I built using Canare 110 ohm digital (
http://www.canare.com/index.cfm?objectid=ED839FCB-3048-7098-AFECCDD67DCC2B4C). It was just 110 ohm, with no impedence transformation, and it also worked great. Many people on ts.com have used this type of set up to use their AES output.
The only reason I said it "should work for testing purposes" is that the cable I used was a true 110 ohm digi cable. If you built a cable using random microphone cable, it probably wouldn't be the best for use in your actual rig, but I'm pretty confident it would at least work to see if using the AES output (and getting the AES data stream instead of the spdif data stream) would solve the problem. If it does, it'd be best to rig up things right, get an impedence transformer, etc. If it doesn't, well, back to the drawing board.....