Inserting data from SQL Server to Postgres - UTF8 0x00 error - sql-server

Question

I am using PySpark to read data from SQL Server and write it to a Postgres DB on AWS.
The INSERT to Postgres fails with this error
ERROR: invalid byte sequence for encoding "UTF8": 0x00
I have been searching for a fix but no luck. That statements I have used to find char(0) say there are none in any of the columns. I'm using this
cast(ExportName AS varchar) like '%' + char(0) +'%'
REPLACE doesn't work.
I tried changing the encoding in pyspark script with
import sys
reload(sys)
sys.setdefaultencoding('utf-8')
but still getting the error.
Any help is greatly appreciated.
Edit: Here is the full pyspark code
# pyspark --jars /deployment/mssql-jdbc-9.2.0.jre8.jar,/deployment/postgresql-42.2.11.jar --num-executors 3 --executor-cores 8 --executor-memory 16g
import sys
reload(sys)
sys.setdefaultencoding('utf-8')
from pyspark import SparkConf, SparkContext
from pyspark.sql import SparkSession
conf = SparkConf()
conf.setMaster("local").setAppName("My app")
sc = SparkContext.getOrCreate(conf=conf)
spark = SparkSession(sc)
ms_url = "jdbc:sqlserver://hostname;instanceName=instance;databaseName=database;"
ms_user = "username"
ms_password = "password"
mssql_query = ''' (SELECT *
FROM DBO.TBL_SENSORS
WHERE TYPE IN ('S','C')) query'''
jdbcDF = spark.read.format("jdbc") \
.option("url", ms_url) \
.option("dbtable", mssql_query) \
.option("user", ms_user) \
.option("password", ms_password) \
.option("driver", "com.microsoft.sqlserver.jdbc.SQLServerDriver") \
.load()
pg_url = "jdbc:postgresql://hostname:5432/database"
pg_user = "pg_username"
pg_password = "pg_password"
pg_table = "SCO.TBL_SENSORS"
jdbcDF.write.format("jdbc") \
.option("driver", "org.postgresql.Driver") \
.option("url", pg_url)\
.option("user", pg_user) \
.option("password", pg_password) \
.option("dbtable", pg_table) \
.mode("append") \
.save()