delete repeated values sql server

Basically, you want to stay with only one reading per minute, for the example you show, if you have two rows for the same minute, you want to keep the first one of them chronologically speaking.

A simple way to solve it is to generate a numerator for all rows by ROW_NUMBER() , with the particularity that will restart every minute: PARTITION BY LEFT(CONVERT(VARCHAR, maq_fechahora, 121), 16) , so we finally put everything in a subquery and only recover the rows where this counter is 1.

SELECT  *
    FROM (
        SELECT  n1,
            n2,
            ROW_NUMBER() OVER (PARTITION BY LEFT(CONVERT(VARCHAR, maq_fechahora, 121), 16) ORDER BY maq_fechahora) RN
            FROM    maq_equipos 
            WHERE   maq_id = '020' 
                AND CAST(maq_fechahora AS DATE) = '2018-10-17' 
                AND maq_tip = 2
    ) T
    WHERE RN = 1

If you do not care which data is displayed for the other columns, you can group by the column that shows hh: mm and apply some function that returns a value on the others, such as min() or max() , for example:

SELECT   min(n1) n1
       , min(n2) n2
       , min(convert(char(10), maq_fechahora, 108)) AS maq_fechahora
       , convert(char(5), maq_fechahora, 108) AS minu 
  FROM maq_equipos 
 WHERE maq_id = '020' 
   AND CAST (maq_fechahora AS DATE) = '20181017' 
   AND maq_tip = 2 
 group by convert(char(5), maq_fechahora, 108) 
 ORDER BY convert(char(5), maq_fechahora, 108) ASC

In this case, it is clear that I am not deleting entire rows, but showing some value for them, this implies that a value of the first row could be shown in a column, and a value of the second row in the other column. This is because I am using the min() function that will simply return the smallest value. You can also use max() , or others like avg() , etc.