HTTPS connections can time out, so to download large files, I recommend using "Range Retrieval Requests" to download in pieces. For example, issuing such a request to download the first 1000 bytes of a file using /2/files/download would look like:
curl -X POST https://content.dropboxapi.com/2/files/download \
--header "Authorization: Bearer <ACCESS_TOKEN>" \
--header "Dropbox-API-Arg: {\"path\": \"<FILE_PATH>\"}" \
--header "Range: bytes=0-999"