If you restrict your efforts to 128B, you have access to upper and lower case characters, numbers and most punctuation. It also saves you from having to write code to shift in and out of A and C symbologies. This makes the code to calculate the checksum really trivial.
Code 128B start character has a value of 104. The stop character for all Code 128 variations has a value of 106, but that value does not figure into the chksum calculation. Let's pick the string "Hello" for a real life exercise. You'll want to make sure you have access to the Code 128 table. All the values I will be discussing are out of that table and not ASCII or UTF-8.
The checksum is calculated by adding the multiple of a character’s value by its position in the barcode with the exception of the start code.
While the start code’s position is ‘1’, so is the position of the first character following the start code. So the start code and the first byte of data (‘H’) are both multiplied by the number 1 ((104 × 1) + (40 × 1) = 144).
The following 4 bytes get an incrementally higher multiplier ((69 x 2) + (76 × 3) + (76 × 4) + (79 × 5) = 1065). Summing it all up together (144 + 1065) we get 1209.
You should be able to use the modulus operator (1209 % 103) in PHP to get 76 for the checksum character for the Code 128B string “Hello” (the 103 is a constant, trust me on that). So the final array of codes to map "Hello" into a Code 128 barcode is:
[104 40 69 76 76 79 76 106].
You'll need lookup tables to convert the string character values to Code 128B character values to whatever your barcode font is expecting. But all you need is a loop, an array index, an accumulator for the sum of the factors and a modulus operator against the constant value of 103.
