The thermocouple inputs on the EMS are pretty solid. I don't know of a case where there was something wrong with the actual EMS in these cases of "incorrect" temperature reports.
Case in point - we recently helped an OEM partner determine that their much more expensive certified EMS that they were checking their Dynon EMS against was much less accurate than the Dynon when they were suspecting the Dynon was giving them CHTs that were too high (it wasn't - in fact it WAS the engine overheating as the Dynon was pointing towards).
Some things to check:
-is there any temperature change across any of the junctions in your thermocouple wire? For example, if you have a junction at the firewall (where there's a temperature change for sure), that's bad, and will cause temperature inaccuracies.
-you can learn a lot about what's going on by moving probes into different ports, and also by connecting different physical probes to different harness leads.
-consider the possibility that you actually do have some temperature variance in your engine that the EMS is correctly reporting. Our EMS does some pretty advanced things, such as cold junction temperature compensation, which in non-technical terms means that our EGT/CHT readings WILL often be different than an analog gauge, but because the analog gauge is simply wrong except under very specific (non-real world) conditions.