Some fundamentals:
These were not widely realised by the grouting fraternity, back in 1962. The extensive testing described here was carried out at that time in order to demonstrate and evaluate these fundamentals. This could be regarded as an advance in the wider general understanding of grout behaviour and continues to be the most comprehensive evaluation of the subject, to this author's knowledge. It has been the basis for much subsequent development of grouting practice.
Pairs of reinforced concrete slabs were held in clamps, braced against deflection and spaced apart with shims to form cracks of various widths, which were grouted. At the centre of the upper slab of each pair, grout was applied through standpipe fittings identical with those on regular grout holes, and then it travelled out from the hole After the grout had set the slabs were separated and the quality of the filling was examined. Bleed voids and discontinuous filling were found to be dominant apart from tests where thick grout had been used. Various crack sizes were tested using grouts of differing water : cement ratios, mixing methods and sometimes additives.
The tests were given particular validity by the use of full size equipment operated by regular grouting crews on a large grouting site. There was no scaling down such as can be necessary in laboratory testing. Supervision and direction of the research was by site engineers and by the author who were all employees of a State Government organisation: the Water Conservation and Irrigation Commision, NSW, Australia.
A full report has not previously been publicly available although extracts were included in the 1985 paper "Cement grouting: water minimising practices." by A.C.Houlsby, Issues in Dam Grouting. American Society of Civil Engineers, Denver, pages 34-75.
The cracks between slabs tapered from 0.06 inch [1.5 mm] or 0.12 inch [0.3mm] to zero. The water : cement ratios were by volume.
This photo shows examples of grouting found when upper slabs were lifted off. The crack width in the two tests at mid-photo were open 0.06 in [1.5 mm] at the near side and zero at the far side. The grout was 3 : 1 starting mix (Tests 4 and 5).

Poor grouting produced by very thin grout. This was 8 : 1. The floor of the crack has only a paint-thickness of hard cement on it; grout reached up to the roof of the crack in only a few isolated places. The matchbox gives scale (Test 8).
Good crack-filling produced by 1 : 1 grout. A few pockets of bleed water were found. The crack had been filled to the top. (Test12) The soft rubber used to close off the perimeter of the crack can be seen.
Assessment of the success (or otherwise) of the grouting was made in terms of the degree of filling, leading to these conclusions:
The tests were carried out during foundation grouting at Burrendong Dam, near Wellington New South Wales, Australia.