CAST and CONVERT rounded a float by passing it to varchar - SQL SERVER 2014 - Microsoft SQL Server Management Studio version 12.0.2269.0

The main problem is that the type of data you want to convert is float , which is an approximate numeric data type. Since you are using SQL Server 2014, you have the FORMAT function available:

DECLARE @flo float = -34.6919581;

SELECT FORMAT(@flo,'G');

The result is -34.6919581 in string format.

Here is a demo of this.

As Lamak says, the float is an approximate data if you need to maintain a precision you should modify the type to decimal for example a decimal(15,8) . To complement, you have some other options. Let's see some:

• CONVERT()

This function can be configured with the parameter style of float a 128

CONVERT(VARCHAR, @flo, 128)

Important : Do not use this form, I only indicate it for information purposes, according to the documentation 126, 128, 129 Included for legacy reasons and might be deprecated in a future release.

• CAST previous to decimal

You can previously "cast" the float to decimal with the accuracy you are looking for, for example:

CAST(CAST(@flo AS DECIMAL(15,8)) AS VARCHAR(30))

Surely this option is the best for old versions of SQL server , it is not your case.

• STR

This function is available from version 2008 and gives better control over the representation in characters of a float since it allows to establish length and precision:

STR(@flo, 15,8)

Of all the forms I am still staying with the proposal of Lamak since you do not add 0 to the left, example:

DECLARE @flo FLOAT = -34.6919581

SELECT  CONVERT(VARCHAR, @flo, 128),
    CAST(CAST(@flo AS DECIMAL(15,8)) AS VARCHAR(30)),
    STR(@flo, 15,8), 
    FORMAT(@flo,'G')

CONVERT       CAST DECIMAL  STR             FORMAT
------------ -------------- --------------- -----------
-34.6919581    -34.69195810    -34.69195810 -34.6919581