Problems
In 1980 problems with materials used in the Oxford Method System were beginning to emerge. The two most prominent being the cladding panels and the use of asbestos. A third was the penetration of water. A problem with the panels had got to a situation whereby an urgent remedy had to be found. The PVC coating to the metal panel was breaking down and a report was commissioned to find a solution.
The following extract gives a clear indication of what was going wrong:
INFORMATION
We were informed that
- The external cladding was Stelvetite marketed by the British Steel Corporation.
- The cladding to the 30 bed ward at Wokingham Hospital had been in service since 1968 and the twin ward unit since 1970.
- The cladding was used as part of a system known as the Oxford Method buildings and had been used in many hospitals in the Oxford Regional Health Authority area.
- The PVC coating applied to the cladding had developed various forms of deterioration.
We were requested to examine the cladding panels and carry out laboratory tests to determine the cause of the defects, and to advise on
- The treatment necessary to render good the defects.
- The maintenance needed to preserve the panelling for the life of the building.
- The probability of unaffected panels becoming defective, including the treatment required to avoid breakdown and preservation of panels for the life of the building.
Wokingham Hospital
General Observations
Generally the PVC coating was of a pale grey colour with marked variations ranging from near white to a dirty grey.
The near white panels were found to be exhibiting varying degrees of chalking from slight to severe and in some cases had resulted in a chalk deposit on glass surfaces below the PVC coated panels.
The darker panels were found to be exhibiting varying degrees of surface soiling.
These features gave a contrasting colour appearance to adjacent panels and panel frame sections.
The mitre joints were usually found to have been filled with a sealant which showed varying degrees of degradation often by splitting or cracking and appearing white in colour in contrast to the adjoining PVC surfaces.
A few examples were observed where red rust staining was present in the region of the mitre joint accompanied by some lifting of the PVC coating away from the metal substrate on either side of the joint.
A common mode of failure was by small scale cracking, flaking and lifting of the PVC coating along the upper horizontal edge of the rails. Frequently these defects were accompanied by traces of red rust corrosion of the steel substrate in the region of the bend.
In some cases the PVC coating along the top of these rails appeared to be thin relative to the thickness of the coating elsewhere.
The adhesion of the coating to the galvanised steel surface was found to be variable. Where the coating was sound, the adhesion was frequently judged to be good. It was usually possible to extend an area of detached coating peeling away from a bend by inserting a scalpel blade and enlarging the defect. In these cases, the adhesion of the coating was judged to be poor to zero.
Isolated examples of mechanical damage were observed and in most cases these had become readily visible because of red rust corrosion of the steel substrate.
Examination under low power magnification (x 10 hand lens) showed that the coating was of variable appearance exhibiting checking and cracking in some places or showing deposits of chalk lying in the depressions of the textured surface of coating in others.
Two panels dating from 1973/74 and understood to have been taken from
Churchill Hospital, Oxford were provided and received on 22nd September
1980.
( We were informed that these panels had not been fixed but had been kept
in the open against the end of the building.)
Laboratory Work
- Specimens cut from the samples of coating flakes were mounted in resin, sectioned, polished and examined under the microscope to determine the thickness of the coating.
- Three further specimens were extracted with diethyl ether by a Soxhlet extraction procedure to determine the plasticiser content of the coating samples.
- The coating from one of the two panels from Churchill Hospital was similarly tested and the plasticiser from this sample was examined by infrared spectroscopy.
- The zinc coating thickness on the plastic coated side of the Churchill Hospital samples was also determined and a simple test for evidence of coating type was carried out.
- Photomicrographs were taken of the exposed face of the coating flake samples from site.
Site observations and measurements
From our site observations it is evident that several types of defects are contributing to the deterioration and breakdown of the PVC coating applied to the galvanised steel panels.
Severe chalking was a prominent feature and panels exhibiting this defect were noticeably white in colour compared with those where chalking was much less severe and the coating had become discoloured with surface soiling deposits.
The coating on the vulnerable top surface of window rails and sills was often found to be severely defective exhibiting a degraded, embrittled film with little or no adhesion to the zinc surface and with white and red rust corrosion spots from the zinc and steel substrates.
It was found that the adhesion of the coating to the zinc remote from obvious defects was frequently good but closer to breaks in the coating it was often possible to extend the damage by peeling away the coating from the metal surface.
During sampling, it was observed that the PVC film appear to be of a
variable build being relatively thin on sills for example compared with
major flat surfaces.
The severe chalking observed on panel faces results from degradation of the
PVC polymer resin with release of the pigment and filler components of the
coating.
The panels show a marked variation in the extent of this form of degradation.
The suspected low build on the top face of sills is probably due to the washing erosion effects of rain which have carried the "chalk" away and left white deposits on windows beneath the rails and sills.
Before considering the results of the laboratory work, reference should be made to the manufacturing specification details for Stelvetite cladding.
The literature made available to us by yourselves is not very detailed but states that Stelvetite 'G' which we understand was used for Oxford Method Buildings includes PVC on one side of hot dipped galvanised steel.
From the range of protective coating types and thicknesses available, it is not clear which grade of material was used at Wokingham Hospital.
The general properties quoted for "PVC laminated 250 microns" are "will withstand severe forming, is readily welded, and has good chemical and abrasion resistance. Wide range of woodgrain, print and pattern effects and good selection of colour for external applications. Excellent weather properties. Temperature limitation 80oC."
We were informed verbally by yourselves that it was thought that the steel was hot dipped galvanised and coated with a 0.012" to 0.014" plastics coating.
Our findings show that the coating is a plasticised PVC applied to galvanised steel which has not withstood severe forming and has shown poor weathering properties.
The zinc coating thickness measurements from site were not markedly dissimilar to the values found on one of the panels from Churchill Hospital.
Examination of the samples of coating flakes shown that from ten specimens, the thickness of the textured coating ranged between 70 and 285 microns (0.0027" and 0.0112") with the average thickness values varying between 120 and 235 microns (0.0047" and 0.0096").
Comparison of these figures with the sample from Churchill Hospital which has had an uncertain history but which is believed to be younger than the Wokingham Hospital panels and to have been unfixed and possibly partly protected from the weather, shows that they are all of lower build.
Even allowing for the effects of checking in the case of samples A and G, the average thicknesses are 'well below the figure of 0.012" to 0.014" (305 to 356 microns) specified.
These figures probably indicate the extent of erosion which has taken place during the 10 to 12 years of service.
The plasticiser recovery experiments and infrared work show that the coating material had been plasticised with a di-alkyl o-phthalate type materials.
The percentage plasticiser values vary between 15.8 and 19.8 with the Churchill Hospital sample lying between them at 16.9.
Sample A with a plasticiser content of 15.8% was found to be exhibiting severe checking and cracking and this relatively low value may be associated with these forms of defects.
Plasticiser migration and loss will result in embrittlement of a coating which may lead to splitting of the film as the internal strains produced by stresses resulting from decrease in plasticiser content are released.
Sample B shows a relatively deep relief texture with hair-line checking and
sample 1 shows an eroded-like surface of the same type of texturing.
Samples D, E and F show a markedly different surface, sample F particularly suggesting a chalky eroded surface.
The cases of breakdown in the regions of bends are examples where the coating has split probably as the result of the combined effects of strain imposed by extension of the coating around the bend at the manufacturing stage and the subsequent effects of weathering and possible embrittlement resulting from loss of plasticiser while in service.
Once the coating has ruptured, the entry of moisture is facilitated thereby weakening the adhesive bond between the coating and the zinc surface.
There is evidence from the rear faces of the flakes and illustrated in the case of sample A that the underside of the PVC coating was featured with depressions and had not been totally in contact with the zinc metal surface.
The presence of such voids would in itself be expected to reduce the adhesive bond and would subsequently facilitate the passage of moisture behind the film once the coating had ruptured.
The results of the site observations and laboratory work show the external coatings forming part of the Stelvetite wall panels at Wokingham Hospital, are showing signs of breakdown and deterioration after approximately 10 to 12 years of service.
In our opinion
1. The most common mode of deterioration is chalking of the PVC film by weathering. The areas affected only by this defect could be treated by thorough cleaning and the application of a painting system. There would be no guarantee however that such treatment would be of lasting value. The PVC plasticiser could well cause adhesion problems between the remedial system and the plastic coating surface.
Other areas of coating where adhesion to the zinc surface is suspect would require stripping back to sound edges, removal of white corrosion products from the zinc surface and application of a corrosion inhibiting paint system.
Further areas at present featured with localised regions of red rusting of the steel substrate could be treated by removal of all corrosion products and treatment with a zinc rich paint based coating system.
These treatments would give a very patchly appearance which might be unacceptable from an aesthetic point of view.
2. Such repairs would also certainly require regular inspection and maintenance painting probably every 5 to 10 years depending on the protection scheme adopted.
3. It is highly likely that panels at present apparently sound will develop defects and although it is not possible to predict future behaviour with any accuracy, it seems likely that dirt retentive panels which do not appear to be chalking significantly will provide more lasting protection.
Breakdown on bends is likely to continue but inspection for early signs of failure followed by treatment as outlined above is more likely to be successful than painting over the PVC in situ.
Before carrying out any extensive remedial work it would be advisable to carry out trials on a localised scale.
For further views on all these matters, the advice of the manufacturers
of the panels should be sought."
Harry Stanger Limited - November 1980
With such a detailed and disturbing report it was evident that urgent action
was required.
By 1985 problems were still very evident and another report was commission to look at water penetration through copings at fascia level.
By carrying out a spray test the following results were recorded.
"During the first test, at the head of the fascia panels and coping, only a minimal amount of water passed through to the internal surfaces. This water was observed dripping from the head of the metal column.
The spray was then directed at the underside of the upper flashing on the fascia panel joint. This resulted in a considerable amount of water passing to the internal surfaces and dropping down at both sides of the steel column. Water was also observed to be pouring out of the hole at the base of the wall panel.
This test allowed water to run down the lower wall panels and collect in the recessed corner which failed in the previous test. To determine whether the test at high level also resulted in water coming out of the hole in the base panel, independent of the mitre failure, the recessed corner joints were sealed in stages and the previous series of tests repeated on the lower wall panels.
Water finally stopped dripping from the base panel hole only after the bottom mitre, the completed vertical recess joint and top mitre were sealed.
The fascia panel joint with its sealed gasket remained watertight during a concentrated spray test. This test excluded the end of the gasket abutting the flashings.
Water was then lightly sprayed down the fascia joint and allowed to run over the lower flashing. No water penetration was observed internally. However, daylight could be seen through two holes (approx 40 x 15mm) where the wall and column panels meet the lower flashing.
Water was then sprayed directly at the lower flashing and, as expected, water came straight through the two openings described.
Summing Up
Both series of water tests identified potential points of leakage as follows:
- at the junction between the built up felt and the cladding panels;
- at the gasket joint between the cladding panels;
- at the recessed and mitred joints in the wall panels;
- below the upper flashing over the fascia panels;
- through the openings below the lower flashing at the junctions between the fascia and the wall/column panels."
Bickerdike Allen Partners - April 1985
The use of asbestos tiles had been unfortunate. With the progress of science it became evident that it should not have been used.
As is now the standard procedure, ceiling tiles and column casings are to be removed by a specialist.
©2004 Oxford Method Royalties Management Group