Sandberg was one of the first to start using Ground Penetrating Radar (GPR) for chimney flue location. Sandberg is probably responsible for popularising the method as a non-destructive means of chimney flue location.

In this article, I look back at the early days of GPR flue location surveys and how things have changed to what they are today. A trip down memory lane.

Background Information

The accepted method of locating flues before GPR was by drilling. If you hit a flue, you would usually use a borescope to work out the general direction of the flue. Alternatively, you could remove bricks from the wall. So, using a non-intrusive survey method was groundbreaking!

I remember undertaking the first chimney flue location surveys for Sandberg back in 1998, or maybe a little earlier. GPR was an archaeological survey method; it was extremely rare to use for construction industry applications. Consequently, the equipment was big and geared to obtaining deep penetration depths.

Historic GPR equipment

The first GPR control unit I used, was a GSSI SIR-2. Things I remember the most: Firstly, it didn’t have an internal battery. It had to be plugged into the mains, or use an external battery. On-site, power was supplied by a big car battery. Secondly, the car battery meant that the system was not very portable. Flexibility to scan was achieved by using a 10 m or 20 m cable from the control unit to the antenna. Thirdly, it ran on DOS. No Windows operating system!

ImagGSSI SIR-2 GPR control uniteAlt
GSSI SIR-2 GPR control unit

The most common antennas we used then had a central frequency of 900 MHz or 400 MHz. They were big, heavy, cumbersome, and low-resolution compared to what we use for chimney flue location now. High-frequency antennas were just starting to be introduced.

Historic GPR data

Software to process the GPR data was available (just!) but ran on DOS (not Windows). It was not user-friendly, and you also needed a computer!

The generally accepted method of recording data at the time was to print the raw data on thermal paper using a special printer. This was done on-site, – also powered by the car battery! The output was in black and white. The paper was on rolls about 125 mm wide and many metres long. Markers were added to the data to record position by manually clicking a switch, typically at 1 m intervals as the antenna was passed along the wall. The print-out was then interpreted on or off-site, and the approximate position of any detected flues could be interpolated.

GPR data from a chimney flue survey printed on thermal paper
ImageGPR data printed on thermal paper – note the dashed distance markersCaption

Present-day chimney flue surveys

Nowadays, GPR surveys are much easier and offer better resolution and higher positional accuracy. The control units are powered by internal batteries, and the antennas are much smaller, utilising a higher frequency (typically 1.5 GHz). This gives a much better resolution.

The antennas incorporate a survey wheel that regulates the scan rate so that scans have a meaningful and constant proportion. The data is saved electronically for off-site processing, which typically includes surface position correction, gain adjustment, and migration. The GPR data files are then visually inspected, analysed, and detected flue positions plotted in CAD.

Results from a modern chimney flue location survey presented on a CAD elevation drawing with detected flues marked-up.
Results from a modern flue location survey presented on a CAD elevation drawing with detected flues marked-up (click image for larger version)

Further information

To find out more about chimney flue location surveys offered by Sandberg, please visit our web page about Chimney flue location. If you want to find out how GPR works, see our How does Ground Penetrating Radar work?