1. Whenever compaction of pavement structure is presented, it is always a good idea to identify CLEARLY (a) the type of material (i.e., crushed stone aggregate, river gravel, etc) (b) the compaction method used (i.e., standard or modified Proctor (light or heavy tamping for our British friends). This establishes facts that anyone making a reply will want to know - and lessens confusion.
2. Normally, in my experience, base courses are compacted to 98 to 100% of the maximum dry density (MDD) Modified Proctor.
3. GeoPaveTraffic indentifed several aspects that are important.
4. 7% OMC (optimum moisture content) seems a bit on the high side for pavement base courses I have used. It would suggest that you are not using crushed stone aggregate and that your base course material is a bit on the sandy side - or, forbid, the base course has significant fines that might be clayey in nature (high PI).
5. 85% is very much on the low side - especially since end dumping into a pile will be in the order of 85% (standard proctor). So any passes you are making should improve considerably this number. Sands I have used in the past would gain 95% standard Proctor under 4 passes.
6. I would doubly check your proctor values if after all those passes you are getting such low results. I might suggest that you carry out a relative compaction check using the old Ontario's Ministry of Transportation and Communication (MTC)-or whatever they call themselves these days. I've explained it fully in previous threads - but basically, you dig a hole, use a sand cone or rubber balloon to determine the volume of the hole. you take ALL the material from the hole and compact, on site, into a proctor mould at "about" the optimum moisture content(depends on how good your field tech is - but if they do this a lot, they would have a good feel). Adjustments in the number of blows per layer are made for holes that are not of the "standard" volume. You then determine the volume of the compacted material in the mould. Compare the volume in the dug hole with the volume in the compacted mould to get the approximate relative compaction. It is a good test - takes time but as a check on the lab proctor values would be reasonable.
Sorry it is long reply . . . Cheers