I have always wondered when seeing the eclectic collection of vehicles and propulsion NR use for monitoring I have always struggled to work out if it is efficient, efficient in an accountancy sense (keeping stuff going and avoiding a large capital expenditure off the annual books) or just downright a money pit. We have seen class 73 and Mk3 DVT conversions that seem to have beem a waste of money. It's all pretty ancient and the heritage loco's that power some of it must be fuel and maintenance hungry.
I suspect (having just read a book on the matter) it's highly efficient, with drivers and OTTs having lots of experience and knowledge in operating their type of traction/monitoring equipment. Class 73s routinely operate RTTs in Juiceland too, so it's not that much of a waste. In terms of locos, however, it's a case of whatever they can get their mitts on (necessarily diesel).
Would it not be cheaper long term for a few short lengh class 80X's and say a few 195/6/7 units be converted. Think that lot would cover all the high-speed requirement for pathing and also pretty much be a go anywhere fleet with the 19X's. Modern, quiet, fuel efficient and less pollution. Maintenance and spares would also be so much easier.
Route availability is major point here. The HST is a proven train that has higher differential PSRs on not insignificant parts of the network. They also spent much time gathering gauging data for the IETs when they were introduced! Bear in mind that 950001 was also purpose built, essentially what you're suggesting with the 19x series but without needing so much sellotape. They often use locos such as the 67 and 68 (modern), but the reason for the 37s and 43s is the proven design and the need to cover diesel-only loops, junctions and curves. What allows 73s to run in Juiceland is the fact that there is very minimal freight infrastructure, which means everything* has a juice rail. 73s are also bi-mode so that helps, I suppose
*except for freight yards and disused sidings/crossovers
