Clark, you are a god. This is a great technique! I can't tell you how much fun I've been having with it.
Here are a couple of things that may help others.
1. For the sine sweep from Voxengo, I reduced it by -18dB and saved that to a new file. Voxengo makes a 0dB sweep by default, which can clip when you run it through the EQ without reducing the level.
2. Based on that -18dB sweep, I made a separate sweep file from it that has 1 second of silence at the end. This way I don't need to add the silence manually each time.
So the matching EQ sweep output is made from the "-18dB with silence added" sweep input. Then in Voxengo the original "-18dB without silence added" sweep is used as the reference sweep. Same as what you are doing, but it saves a couple of steps in the workflow.
I'm confused by something though. I tried bypassing all the sweeps and Voxengo by putting an impulse directly into the EQ. That's a wav file with the first sample at max value (0dB) followed by all silence. In theory it should work. The frequency response looks overall the same, except everything is stretched in gain. Peaks are higher and dips are lower. But otherwise the shape is the same. If this worked it would be convenient because the impulse can be made straight from 1 file without the deconvolver. Not sure what's going on with it otherwise. Hmm..
Great method and I'm getting good results!