Getting values out of a device is the easy half. The half that wastes an afternoon is a CSV that Excel opens as one column, or a decimal comma that turns 230.4 V into two cells. This page covers the export format, the locale traps, and how to lay the file out so a pivot table works on the first try.
Modbus RTU, ASCII and TCP · Windows 10/11 · One-time license
A commissioning report, an energy audit and a warranty claim have one thing in common: the person receiving it opens a spreadsheet, not your historian.
ModbusManager Pro records values to a local database, and any series over any time range can be written out as CSV. That file is the handover artefact — the customer keeps it, the auditor reads it, and nobody needs a licence or a login to open it in five years.
The rest of this page is about the details that decide whether that file is usable or annoying.
This is the single most common complaint about Modbus CSV export, and it is not the exporter’s fault. It is a mismatch between the file and the machine reading it.
Excel does not read a separator from the file. On Windows it uses the list separator from your regional settings. In the US and UK that is a comma; across most of continental Europe it is a semicolon, because the comma is already the decimal separator.
So the same file behaves differently on two desks:
| File contains | Excel with comma locale | Excel with semicolon locale |
|---|---|---|
| time,volts 10:00:01,230.4 |
Correct: two columns. | One column — nothing is split. |
| time;volts 10:00:01;230,4 |
One column. | Correct: two columns, 230,4 is a number. |
Do not double-click the file. In Excel use Data → From Text/CSV, which asks for the delimiter and the decimal separator and shows a preview before importing. It takes ten seconds and removes the guesswork on any machine.
If a file has to be double-clickable on a specific customer’s PC, match that PC: comma-separated with a dot decimal for an English locale, semicolon-separated with a comma decimal for most of Europe.
Excel guesses a type for every cell. On industrial data those guesses are sometimes wrong in ways you only notice much later.
| In the file | What Excel may show | Avoid it by |
|---|---|---|
| 0012 | 12 — leading zeros gone, so a padded unit ID or serial stops matching. | Import that column as Text in the From Text/CSV dialog. |
| 4294967295 | 4.29E+09 — still correct underneath, but unreadable in a report screenshot. | Widen the column and set a number format with no scientific notation. |
| 1-2 | 2 Jan — a register range or a tag name read as a date. | Import as Text, or avoid bare number-dash-number in tag names. |
| Non-ASCII names | Mojibake in headers if the file is read as ANSI rather than UTF-8. | Use the import dialog and set the encoding to UTF-8 explicitly. |
The same data can leave as one column per signal or one row per reading. Which one you want depends entirely on what happens to the file next.
Best for charting and for handing to someone who will just look at it. Every row is one moment in time, so a line chart is two clicks. Falls apart when signals have different intervals — you get gaps.
Best for pivot tables, databases and mixed intervals. Nothing needs to line up, and adding a signal does not change the column layout. Harder to eyeball without pivoting first.
A practical rule: if the file is going to a person, export wide. If it is going into another system, export long. Deadband makes this decision sharper — once values are only written on change, signals no longer share timestamps, and the wide layout starts filling with blanks.
If the devices sit on a shared serial line, the interval you can achieve is limited by the line rather than the software — see Modbus RTU data logger over RS-485. For the logging features themselves, timestamps and database location, see Modbus data logger software.
Full Pro edition, no credit card. Poll a device, log it, export the range and open the file — the whole round trip in a few minutes.
Download Pro trial — free 14 daysWindows 10/11 · 14-day trial · Pro from $119 one-time
We use Google Analytics cookies to see how the site is used. You can accept or decline — declining keeps analytics off. See our Privacy Policy.