Many moons ago, the late Iain Rice suggested reducing flangeways to 0.9 mm or even 0.8mm in EM to improve appearances. I use a strip of 0.9mm thick aluminium in preference to the EMGS standard stainless steel gauge of 1mm. This works quite well provided one uses finescale wheels from Gibson etc. RTR wheels were not a good idea at the time (I agreed with Joe Brook-Smith who upset the EMGS hierarchy) but I have not tried the new finer profiles now being produced by Hornby et al.
If you have some 1mm and 0.9mm strip to hand, try running them through the new EMGS B6 turnouts. If they are like mine, the flangeways are nearer to 0.9 mm wide than the EMGS official standard! My new Finetrax double slip (yet to be used in anger) has 1mm flangeways; I wonder if Wayne can be persuaded to produce EM-SF versions? Somehow I doubt it.
Bernard Weller promoted a wider BTB and Exactoscale sold a back-to-back gauge of 16.7mm for EM which I use in preference to the EM standard of 16.5mm. In practice I find that pushing wheels in to fit the gauge usually gives a 16.6mm BTB; it seems the plastic wheel centres bounce back? Anything between 16.5 and 16.7mm is acceptable to me; I refuse to believe the P4 guys can measure accurately and consistently to 2 decimal places of a millimetre! As Geoff says above, George Watts sells a variable BTB gauge which can be set to 16.7mm. My experience is that the wider BTB gives more reliable running. One result is that if your handbuilt point work is dodgy like mine, the wider BTB will quickly reveal anything narrow to gauge. My recommendation is to slightly widen the track gauge through any turnouts (even if you stick with the proper EMGS standards); be wary of relying solely on triangular gauges when doing this.
No doubt Martin will be along shortly to tell me the check gauge needs to be adjusted accordingly.