Enforcing Data Integrity Across Three Columns in SQL

I have a bit of a question regarding data integrity on a table I'm creating. For reference, here is the code to create the table:

CREATE TABLE tbl_Course_ExamModules
	(
		CourseExamModuleID INT IDENTITY(1,1) NOT NULL,
		CourseCode VARCHAR(3) NOT NULL,
		ExamModule VARCHAR(10) NOT NULL,
		CashInCode BIT NULL,
		AcademicYear INT NOT NULL,
		StudentLevel INT NOT NULL,
		CourseModuleCount INT NOT NULL
	)
	
	
ALTER TABLE tbl_Course_ExamModules
	ADD CONSTRAINT pk_Course_ExamModules_CourseExamModuleID 
		PRIMARY KEY CLUSTERED (CourseExamModuleID)

CREATE UNIQUE NONCLUSTERED INDEX idx_CourseExamModules_Code_Module_Year
	ON tbl_Course_ExamModules (CourseCode, ExamModule, AcademicYear)

Select allOpen in new window

Which is all fine and fairly straight forward. Each module can only be associated with each course once for each academic year.
Then I start to struggle. For each course there is a module which is called the "Cash-In Code" Each course can only have one cash-in code for each Student Level. There can be multiple modules which are not cash-in codes for each course within an academic year and student level. So, I can't place a unique index on the CashInCode, CourseCode, AcademicYear, StudentLevel columns the CashInCode column can have the value 0 more than once.
I hope that makes sense!
I looked at adding a WITH CHECK Constraint, but I don't think they can be applied the way I want (or can they?). Is there a solution to this without using a trigger?