As well as straightforward repairs, we also develop modifications to allow the fitment of alternative parts where the original is no longer available.
One such item is our replacement ovened crystal reference oscillator module for the 2955. This replaces a failed original and drops straight into the original footprint. The only modification needed is to replace a resistor with a wire link, or simply bridge it out.
One of the other issues with the 2955 is that the line output transformer (LOPT) fails and is no longer available. The LOPT supplies the EHT to the CRT as well as various auxiliary supplies and (in the case of the 2955) the vertical scan drive.
Whilst others have successfully transplanted an LCD into the 2955, it is recognised that some prefer to keep the original CRT look.
We have managed to graft in various LOPTS from other small monitors, such as those used in the older ATM's, but none were available in sufficient quantity for it to be worth developing past the experimental stage, and with the decline of CRT displays, they are only going to become harder to find.
As a result, we are currently working on a complete replacement AC1 board for the 2955 series. This separates the scan and power supply requirements with a custom wound transformer running at a much higher frequency than the original to generate the auxiliary supplies and an EHT overwind and tripler to generate the final anode supply. The vertical scan is driven by separate circuitry. 
This is still a work in progress, but is looking promising.
Update - We have managed to secure a quantity of new old stock LOPT's of a type previously tried and tested as a replacement for the original black or blue Eldor LOPTs and are currently awaiting the arrival of adaptor PCB's. We have a number of 2955's in stock requiring LOPT's and will fully test the modification on these before making the modification kit available for general sale early 2025.
We are also working on an step attenuator repair kit for the Marconi 2965 series. We currently have a 2965 in stock with a cracked pad in the signal generator attenuator, needless to say, these are no longer available. As a result, we have designed a PCB comprising four pads made from SMD resistors with gold plated contacts for the switch leaves. This fits into the centre channel of the existing attenuator body in place of the original pads. We are again awaiting delivery of the PCB's. If successful, we'll expand the range to cover the input attenuator and possibly the 2955. Unlike the 2965, there is no specific calibration for the attenuator in the 2955, so it all depends on the accuracy and frequency response of the final result.