The the way I understand your model, you take an external market in USD, apply market share and discounting to arrive at PQ in USD. Then you divide it by V to arrive at M in USD. This looks like the incorrect application as shown in Weber’s post — equation 9. The issue arises from the fact that you are using an external price per GB into the UT economy and not a UT price per GB, which would be the correct approach. If you do that, then you get both sides of the equation in UT (that would be a correct usage as per equation 5 in Weber’s post). From there you can use a simple PPP and divide the external USD price per GB by the UT price per GB and come up with the exchange rate, i.e. your UT/USD price.