It should be a straightforward changeout, and you would most likely be able to use your same support structures (especially if you go with the much lighter optical units).
I have some 230-kV 1050-kV BIL combined units (PT and CT in the same device) on order at about $50K each, to give you a rough idea of the cost.
My personal recommendation would be to go with a CCVT over an optical unit, from almost every perspective, including cost, reliability, flexibility, etc...
Also, at 230 kV - 500 kV, CCVTs will be a fraction of the cost of PTs.
A bit of advice...if using for revenue metering, but the extra-high capacitance design. This should be around 4500 pF to 5400 pF or so. Reason being is that you want the CCVT capacitance to dominate the line capacitance so you don't get a voltage dividing effect, that could slightly affect accuracy.
Thanks for the advice.
Does anybody have any reliability and performance comparison between EHV CCVT and PT ? Is it utilities trend to use only CCVTs for protection, control and metering in EHV substations with microprocessor-based relays?
Both PT and CCVTs at 500kV are mature products and quite stable.
One potentially big issue with EHV PTs is that ferro-resonance becomes a bigger concern. You don't have that worry with CCVTs.
Strictly from a failure mode, PTs failure mode is much much much more violent than a CCVT. CCVTs will also often indicate an insulation problem (via the output voltage) prior to failure, although the indication is typically more pronounced on lower voltage units.
In US it is definitely the utility trend to use CCVTs over PTs at EHV level.
The only performance concern with CCVTs for protection is transient response. However, with the low burden that modern relays present, transient response is not nearly the issue that it used to be. Rarely comes up now.
