Unfortunately recent projects have shown that installation costs are high, and tend to run out of control, and the supporting works, bridge, tunnel and platform canopy changes add further cost, including issues which only come to light once you start. OHLE may be the gold standard, and on routes where line speeds are 100mph+ (including an allowance for reasonable upgrades to 100mph+) then its also the right answer. However a cheaper and less invasive solution is needed for a lot of secondary routes if diesel operation is to be phased out. Batteries are another option, but add weight, and still have limitations. Realistically there are large sections of the network which will never see OLHE
Bottom contact exists in other countries, so why are we starting from scratch... Take a current design and use it. Go for a contact rail that has protection on 3 sides, possibly some form of extruded insulation, and make a kit of standard parts, with say 60, 30 and 10ft lengths with standard lead in and out ramps which bolt together. Mass produce these, with the mounts. Design a standard 3 phase HV AC / DC grid fed converter packaged in a container that can sit at the side every few miles, and in areas where grid feed is tricky link with a high voltage cable and maybe have a standard static battery module to provide peak power where grid supply is limited/unreliable. Reduce the 'custom' design element to a minimum. Think meccano kit. For the first phase dont try and mix with the current DC top contact system, there are huge parts of the network which are nowhere near current 3rd rail territory.
Use a higher voltage, (1000-1500v) dc to reduce losses, and set a realistic max current draw, we are talking max speed 90mph on rural and suburban services and in a lot of cases 75mph, probably 8 car max, and frequently 3/4 car. Have small batteries to avoid gapping and provide additional acceleration as well as short (5 miles at half power?) moves off the juice. This also means that complex conductor rail configurations at junctions can be avoided.
Set up a trial on a (mainly self contained) route to iron out issues. Then rollout on an area basis. Spec all new EMUs as dual voltage, modern electronics means this is no longer a problem. Long rural routes could have bigger batteries and powered islands typically in and around stations and maybe on long climbs, with less visual intrusion through sensitive areas. This solution could in my opinion be much cheaper to install and could deliver a way of dealing with the large number of routes off the main Inter City and big city metro systems. If the sparks effect revitalises a route then consider upgrade to OLHE and transfer the DC kit into a pool of spares to maintain and extend on other routes at some point in the future.
There will be limitations, but properly developed and delivered it could give a low cost quick to install solution. Set some very clear goals to avoid over engineering and scope creep.
In terms of disruption a custom installation train could probably lay significant amounts during overnight possesions, the routes that would be upgraded probably have no overnight service.
