SP 92-4

Assessment and Repair of Fire-Damaged

Concrete Structures-an Update
by A. K. Tovey

Synopsis: The purpose of this paper is to outline some of the

proposed changes to the Concrete Society Technical Report 15 which
is the most comprehensive UK document dealing with assessment and
repair of fire-damaged concrete structures. The publication is
being updated to take into account changes of practice and methods
of assessment since 1978. Repair techniques other than Gunite are
to be included with particular comment in respect to resin
repairs.

Keywords: buildings; concrete construction; damage; estimates;

evaluation; fires; plastics, polymers, and resins; repairs; shotcrete

47
48 Assessment and Repair

A. K. Tovey is Principal Structural Engineer in the Building

Group, Cement and Concrete Association, England. Member of the
Institution of Structural Engineers and Chartered Institute of
Arbitrators. Involved with various British Standards committees
and the Concrete Society Fire Resistance Committee.

IN'rRODUCTION

A number of publications (1-4) within the UK advise on the

assessment of the effect of a fire on concrete structures. The
most comprehensive, with regard to concrete structures, is the
Concrete Society Technical Report 15 (1 ), "Assessment of
fire-damaged concrete structures and repair by Guni te". That
publication is currently being revised by the Fire Resistance
Committee of the Concrete Society UK and this paper outlines the
major rev1s1ons. It is expected that the revised report will be
available mid-1986.

The principal objectives have been to update the information

given in the light of changes of practice and methods of
assessment since 1978 and to include repair techniques other than
Gunite (sprayed concrete). This has led to the change of title to
"Assessment and repair of fire-damaged concrete structures".

The revised Report is split into three main sections -

1. Assessment, 2. Design, 3. Repair methods. The detailed section
on assessment follows that given previously; a general
qualitative assessment followed by a quantitative investigation of
individual members. A flow chart (Figure 1) has been added to
assist in developing a programme for assessment.

QUALITATIVE ASSESSMENT

The general qualitative assessment method gives a simple way

to indicate the degree of damage and likely repair of individual
members and overall structure. Figure 2 shows part of a first
survey on a four-storey warehouse. The number adjacent to each
member indicates the visual degree of damage according to the
classification system given in Technical Report 15 (1) which is to
be standardized to make it simplier to use (Table 1). A detailed
first assessment of this type enables a very good initial picture
to be drawn of the overall condition of the structure and the
likely repair.

An example is given in Figure 3 which shows the whole of the

first and second floors of the part structure used in Figure 2.
 Tovey 49

This clearly shows that the main damage was sustained·in the upper
left-hand quarter of the first floor with the most notably
affected members in the area bounded by columns 5, 8, 19 and 22.
The degree of damage is again clearly shown to reduce away from
this area towards the lower right-hand corner (col. 59). This
corner was the position that the fire started but the major
combustible material was located to the rear of the building
adjoining a single-storey steel framed warehouse which was totally
destroyed. The fire can be seen to have spread to the second
floor but was brought under control before a major fire or
extensive damage had developed. A comprehensive report giving a
good indication of both damage and likely repair can be prepared
in only a few days using the simple visual assessment procedure.
At this stage, any surprising aspect in the performance of a
structure would also be noted.

QUANTITATIVE ASSESSMENT

The detailed quantitative investigation using various tests

is to determine the likely temperatures to which the member has
been subjected and hence the loss, if any, in strength of
reinforcement and concrete. From examination of the building
contents, it is possible to assess the maximum likely fire
compartment temperatures and, from eye witness reports and
evidence of combustible content, to assess the equivalent standard
fire period. This can be compared with the temperature
distribution and strength loss determined from test evidence.
Report 15 ( 1 ) indica ted three main tests which could be carried
out in order to determine strength loss in concrete. These are:
color determination, Schmidt hammer and core testing. The revised
report deals with these and, in addition, gives guidance on the
use and selection of other tests such as ultrasonic pulse
velocity (5), Windsor probe (5), BRE internal fracture (5) and
thermoluminescence tests (6). The information obtained from tests
is then used to assess the strength of the member. An example of
tests carried out on a structure is shown in Figure 4. This
figure shows the position and results of a series of different
tests on a section of wall and floor slab. Each of the results
can be compared together to improve the confidence of the
assessment. The pundit readings on the floor, for example, show
the degree of damage from the maximum occurring at the seat of the
fire (readings around 200) to that where the concrete is undamaged
(readings around 34) and hence indicate the area requiring repair.

The strength reduction curves given in the previous report

were based on research on unstressed concrete (7). It is known
that the strength reduction of concrete unstressed at the time of
heating is greater than for concrete under compression. Since the
majority of structural concrete will be stressed (at least under
dead load) at the time of heating, then some modification to the
previous data is appropriate. The strength loss data has been
50 Assessment and Repair

simplified to a straight line plot for consistency with other

documents (8,9) (Figure 5).

The color of concrete can change as a result of heating (10)

and therefore may be used to indicate the maximum temperature
attained and the equivalent fire duration. In some cases at above
3000C a pink discoloration may be readily observed. The onset of
noticeable pink discoloration is important since it coincides
approximately with the onset of significant loss of strength due
to heating. Therefore, any pink discolored concrete should be
regarded as being suspect.

Pink discoloration is due to the presence of ferrous salts

in the aggregate and/ or in the sand, and in some cases these are
not present. Therefore, concrete which has not turned pink is not
necessarily undamaged by fire. The color change to pink tends to
be more prominent with siliceous aggregates, Calcareous and
igneous crushed rock aggregates are less susceptible to this
effect.

The residual strength of reinforcement can be judged from

the assessed temperature of the surrounding concrete or, if
spalling has occurred, from the period of exposure and likely fire
compartment temperature.

The effect upon steel during elevated temperatures and after

subsequent cooling has been researched in detail (11,12).
Significant loss of strength may occur while the steel is at high
temperature and this is usually responsible for any excessive
residual deflections. However, recovery of yield strength after
cooling is generally complete for temperatures up to 450°C for
cold worked steel and 600°C for hot rolled steel. Above these
temperatures, there will be a loss in yield strength after
cooling. The actual loss in strength depends upon the heating
condi tiona and type of steel but the simplified values given in
Figure 6 will be sufficient for most purposes. Where this aspect
is critical to the assessment, the matter should be discussed with
the reinforcement manufacturer if known or, alternatively, tests
carried out on samples taken from the member. Loss in ductility
may occur after exposure to particularly high temperatures.
Buckling of reinforcing bars often occurs as a result of
compressive stress induced at high temperatures by restraint
against thermal expansion.

A damage factor may be determined for each member although

normally it will only be necessary to consider Class 3 and Class 4
damaged members in detail. The damage factor is determined from
Figure 5 using the estimated average temperature within the
compression block obtained from temperature profile curves as, for
example, are given in Figure 7.

The general process for this is to record the depth of the

layer which has reached a temperature of 300°C by reference, where
appropriate, to the depth of the pink zone. Estimate the increase
 Tovey 51

beyond this depth to reach a zone which has not exceeded 1000C by
reference to Figure 7. The average damage factor may then be
determined for the compression block taking a factor of 1 .0 for
all concrete subjected to temperatures less than 1oooc, and a
factor of 0.85 for concrete within the 3000C to 100°C zone.

It may also be necessary to determine a damage factor to

allow for reduction in bond or anchorage of steel reinforcement.
A number of papers have been written (13,14,15,16). These
indicate that bond is influenced not only by temperature but also
by bar diameter, type of aggregate, section size and the
compressive strength of the concrete. The presence of stirrups or
other confining steel is also likely to influence bond. A
conservative damage factor for bond of 0.7 may be applied to
reinforcement within the 3000C to 1000C zone although a value of
0.8 or better might be considered for lower temperatures, small
diameter bars (12 mm or less), -concrete of lower compressive
strength (25-30 N/mm2), concrete containing calcareous or
lightweight aggregates and for reinforcement contained by
stirrups.

REPAIR TECHNIQUES

In addition to reviewing the sprayed concrete process

(Gunite) which is the usual method of repairing major fire-damaged
concrete structures, the working group has considered and included
in the revision of Report 15 other repair methods such as resins,
polymer modified and cement mortars, plaster, sprayed mineral
preparations, as well as the use of alternative supports.

It is essential that a repair must restore any loss of

strength, maintain durability and fire protection. In situations
where, following a fire, there is still sufficient strength and
cover for durability then a thin hand or spray applied material
could be used to restore a loss in fire protection. Equivalent
thickness of concrete cover for various materials may be found
from manufacturers. Some information is given in reference 21 and
in BS 8110 (22) the revision to CP 110 (8).

Resin Repairs

As a result of the study, it has been necessary to make

particular comment on resin repairs. These are commonly used to
overcome problems of reinforcement corrosion but may not provide
the necessary protection in the event of a subsequent fire.

Resin repairs may consist of a variety of configurations of

patch or infill of Epoxy, Polyester and Acrylic mortar. Resins
are often used for repairs to lightly spalled areas and, though
they may perform quite satisfactorily in normal service, there is
no comprehensive information on the performance of such repairs or
that of the materials when subject to heat or an actual fire test.
52 Assessment and Repair

What information does exist, including some published

papers (17,18,19), indicates that these materials may soften at
relatively low temperatures (sooc). As a consequence, it is
possible that some resin repairs may not provide adequate fire
protection to the reinforcement and may fail to retain adequacy in
compression zones. Accordingly it is recommended that resin
repairs only be used when either:

(i) performance data can be supplied to show that the

particular formulation has adequate fire resistance and
retains its structural properties under the envisaged fire
condition, or

(ii) the material is adequately fire protected by other

materials and retains its structural properties at the
expected fire temperatures at the relevant depth in the
section, or

(iii) loss of strength or other properties of the material will

not cause unacceptable loss of structural section or fire
resistance.

The designer should refer to the manufacturers' literature

for details of the performance of the various materials available.
A wide variety of materials exists and their specifications are
liable to changes at relatively short notice. In general, the
materials will be capable of providing good bond and compressive
strengths, the flexural and tensile strengths may exceed that of
concrete but the thermal expansion is considerably larger than
concrete and this may be a point to be considered where the
temperature range is large. For further information, the designer
should refer to the Concrete Society Technical Report No. 26,
"Repair of concrete damaged by reinforcement corrosion" (20).

Polymer Modified Mortars

In many instances, hand repairs of small areas can be

effective by use of polymer-modified cementitious mortars. These
repairs will generally be to areas or patches of between 12 mm and
30 mm depth. In particular, styrene butadiene rubber (SBR)
modified mortars appear sui table. There is limited test
information on such mortars but it is expected that they will be
satisfactory in a fire as they should behave as a cementi tious
product. It is to be expected that the use of a small sized
aggregate will improve the performance of mortars compared to
concretes due to a lesser tendency for damage caused by aggregate
splitting in the event of a subsequent fire. In other respects,
these mortars are also described in the Concrete Society Technical
Report No.26, "Repair of concrete damaged by reinforcement
corrosion" (20) to which the designer should refer.
 Tovey 53

Cement Mortars

These may be hand applied to damaged areas but great care in

surface preparation is necessary in order to ensure adequate
adhesion. Generally, mortars will be applicable to well-defined
areas placed in layers using good rendering practice, up to a
total 30 mm thickness.

Plaster

This may be readily applied to both plain and roughened

concrete surfaces. It can restore a degree of fire resistance but
will not replace cover requirements for durability.

Sprayed Mineral Preparations

This technique will not assist where strength restoration or

replace cover requirements for durability are required. Where
internal repairs of a minor nature are necessary, these systems
will restore fire resistance and shape to damaged members. For
particulars, the designer should consult specialist contractors.

Alternative Supports

The designer can consider the use of alternative supports

such as further columns or new beams to sub-divide floor spans.
Such schemes may well prove economic as they may allow lesser
restoration to damaged members. Ideally, new supports would be in
reinforced concrete but there is no reason why steel, timber or
masonry should not be used providing the required strength, fire
resistance, durability and appearance are achieved.

ACKNOWLEDGEMENTS

In addition to the author, members of the current working group

revising Report 15 are:

Mr R H Jackson Andrews, Kent & Stone

Mr C D Jones Freeman Fox (Wales) Ltd
Mr E Mellor De Leuw Chadwick 0 hEocha

REFERENCES

1. The Concrete Society., Assessment of fire-damaged concrete

structures and repair by gunite, Report of a Concrete Society
Working Party, London, The Concrete Society, 1978, 28 pp.

2. Green, J.K., Technical Study : reinstatement of concrete

structures after fire. The Architects' Journal. Vol.141, No.2,
14 July 1971, pp.93-99, No.3, 21 July 1971. pp.151-155.
54 Assessment and Repair

3· Smith, L.M., The assessment of fire damage to concrete

structures, Thesis for PhD degree, Paisley College of Technology,
Paisley, Scotland, September 1983. (Bibliography contains 259
relevant references)

4. Tucker, D.M. and Read, R.E.H., Assessment of fire-damaged

structures, Garston, Building Research Establishment, November
1981, 4 pp, IP 24/81.

5. Keiller, A.P., A preliminary investigation of test methods

for assessment of strength of in situ concrete, Wexham Springs,
Cement and Concrete Association, 1981, 36 pp, Technical
Report 42.551.

6. Placido, F., Thermoluminescence test for fire-damaged

concrete, Magazine of Concrete Research, Vol.32, No.111,
June 1980, pp.112-116.

7. Malhotra, H.L., The effect of temperature on the compressive

strength of concrete, Magazine of Concrete Research, Vol.8, No.23,
August 1956, pp.85-94.

8. British Standards Institution, CP 110 : Part 1 : 1972, The

structural use of concrete, Part 1 : Design, materials and
workmanship, London, 156 pp.

9. Design and detailing of concrete structures for fire

resistance, Interim guidance by a Joint Committee of the
Institution of Structural Engineers and The Concrete Society,
London, The Institution of Structural Engineers, 1978, 59 pp.

10. Bessey, G.E., Investigations on building fires, Part 2 : The

visible changes in concrete or mortar exposed to high
temperatures, London, H.M. Stationery Office, 1950, pp.6-18,
National Building Studies Technical Paper No.4.

11. Stevens, R.F., Contribution to discussion on: Steel

reinforcement, by R.I. Lancaster, Structural Concrete, Vol.3,
No.4, July/August 1966, pp.184-185.

12. Holmes, M., Anchor, R.D., Cook, G.M.E., and Crook, R.N., The
effects of elevated temperature on the strength properties of
reinforcing and prestressing steels, The Structural Engineer,
Vol.60B, No.1, March 1982, pp.7-13.

13. Morley, P. D., and Royles, R., The influence of high

temperatures on the bond in reinforced concrete. Fire Safety
Journal, 2(1979/80), pp.243-255.

14. Sager, H., and Rostasy, F.S., The effect of elevated

temperatures on the bond behaviour of embedded reinforcing bars,
Bond in Concrete : Proceedings of the International Conference,
Paisley, 14-16 June 1982, London, Applied Science Publishers,
1982, pp.206-216.
 'lbvey 55

15. Royles, R., Morley, P.D., and Khan, M.R., The behaviour
reinforced concrete at elevated temperatures with particular
reference to bond strength, Bond in Concrete : Proceedings of the
International Conference, Paisley, 14-16 June 1982, London,
Applied Science Publishers, 1982, pp.217-228.

16. Morley, P. D., and Royles, R., Response of the bond in

reinforced concrete to high temperatures, Mazagine of Concrete
Research, Vol.35, No.123, June 1983, pp.67-74.

17. Levitt, M., The fire resistance of resin jointed concrete,

Proceedings of Conference on Plastics in Building Structures
(1965), Plastics Institute, Pergamon Press, 1966, Paper 12,
pp.77-81.

18. Plecnik, J,M., Bresler, R., Chjan, H.M., Pham, M., and Choa,
J,, Epoxy repaired concrete walls under fire exposure, Proceedings
of the American Society of Civil Engineers, Structural Division,
Vol.108, No.ST8, August 1982, pp.1894-1908.

19. Plecnik, J,M., Bresler, B., Cunningham, J.D., and !ding, R.,
Temperature effects on epoxy adhesives, Proceedings of the
American Society of Civil Engineers, Structural Division, Vol.106,
No.ST1, January 1980, pp.99-113.

20. The Concrete Society., Repair of concrete damaged by

reinforcement corrosion, London, 1984, Technical Report No.26.

21. FIP/CEB Report on methods of assessment of the fire

resistance of concrete structural members, Wexham Springs, Slough,
Federation Internationale de la Precontrainte, 1978.

22. British Standards Institution., BS 8110:1985, The structural

use of concrete, Part 2: Recommendations for use in special
circumstances, London.
 TABLE 1 - VISUAL DAMAGE CLASSIFICATION FOR REINFORCED CONCRETE ELEMENTS c.n
0\
Surface Appearance Structural Appearance
Class Element Plaster/ Color Crazing Spalling Condition of Cracks Deflection
>
Finish
I main reinforcement 1:rJ
1:rJ
~
1:rJ
0 Any Unaffected or beyond extent of fire 1:rJ

1 Column some normal slight minor none exposed none none

3~

Floor
peeling
.. .. .. .. .. .. .. .. =
......
~

Beam .. .. .. minor very minor exposure . .. =

0...

~
"'0
2 Column substantial
loss
..
pink

..
noticeable

..
localized
to corners
up to 257..
none buckled
none

..
none

..
=·
'"l

Floor localized up to 107..

to patches all adhering

Beam .. .. .. localized up to 257.. .. ..

to corners none buckled
minor to
soffit

3 Column total buff/ extensive considerable up to 507.. minor none

loss friable to corners not more than
one buckled

Floor .. .. .. considerable up to 20%. small not significant

to soffit generally adhering

Beam .. .. .. considerable up to SO%. . .. .

to corners not more than
sides one buckled
soffit
----
 'lbvey 57

1-
z
w
>
w

CARRY OUT DAMAGE

CLASSIFICATION SUrlVEY u.
0
1-
z
w
::i
if)
if)
w
if)
if)
<(

NO

PRELIMINAFW
DESIGN FOf1
MAJOH

YES z
~
if)
w
D

NO

-'
<(
>
0
a:
Q_
Q.
<(

1-
0
<(
a:
1-
z
0
0

Fig, !-Assessment procedure for fire damaged structures

 c.n
co
2 3 3 4 3 5 3 6 3 7 3

)
1
I 2
1 3
3
3 2
2
1 2 2
3
3
3 3
3
3 3
3
3
3 >
00
00
1 2 1 1 3 2 3 3 3 3 3 2
1L 2 3 1 3 3 3 3 ('l)
00

15
1
I 2
2
16
2
3-
3
17
2
1
1
18 2 3
2
19 3 3
3
20 3 3
3
21 2
3
3
-
00
a
('l)
2 3 1 1 3 3 2 2 :::::
3 3 1 2 2 3 .......
2 3 2 2 1 1 1 3 2 2 1
p;
3 1 2
3 3 1 2 1 2 :::::
3
a..
30 31 2 3 32 1 33 1 1 34 1 35 1 36 1
~
3 1 2 2 2
)
2 1 1 1 1 1 '
2 2/_ 3 1 2
1
1
1
1
1
1
1
1
1
1 '"i:l
2 2:.
L 1

1
2
1 1
1
1 1
1
1 1
1
1 1
1
1 0
0 ""
°

-
44 ot,5 1 46 1 47 1 1 48 1 1 49 1 1 50 1 1 51 0
. 1
)
0

Ref. Classification factors - Table 1

E
4

Col
---r
'

'

.
I

Member Ref

Fig. 2--Member classification (Grd - lst)

 Tovey 59

Ref. repair Classification

GROUND -1st i·: :·:-. ·: < >J Class 0 Deco rat ion
Wff~ 1 Superficial
0 Most notably affected members &~~ 2 General repair
~ 3 Principal repair
1 2 3 4 5 6 7 8 9 10 11 12 13 14

r---r ---¥' -¥ -P- -r ~ -r--f· 1i .l';i w- -iH.-01;;,'.97,

rjT·-r·T·T·T·T·T·T:T.-.T:·r··l7\l
'

~--f1·-P~-f~~-T-· ~-
' . ' ' .
3
•. . . . . . . . . . . . '·.· .· ......1

F\>:f.~:~i.,.:/t~<·;.-:J~->~->:k~- 43
~~-f-~?_ha__19_h9_h1~2___li;·.:·:.1~·~.··. :·.:;r~~\··/:l~6~··~~/~~-~j59

Fig. 3-Tnit:ial classification of repair

60 Assessment and Repair

134 133 132 202 126 124

• • 46• • • 11~11
X

ELEVATION
• Pundit
0 BRE
I:B Windsor
x Schmidt
I I
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _ _ _ __J ~
"L..r--------- ----,_J----- -------- -,__r----------- ---.,_,J
X 48 52' 48 X X 47

35 38 52 74 67 85 225 153 216 44

35 37 40 45 43 45 77 123 104 38

35 36 37 38 55 52 47 36

36 34 35 36 35 35

33 33 34

PLAN

_ _ _ _ _ _ _ _ _ _ _ _ _ __r-,__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ......r-,
----------------------------------------·r
!I

Fig. 4-Tests to wall and 1~affle

 1·2r-------------------------------------------;
Typical hot- rolled
1·0 ----.,
O·B

0·6

0·41-- REPORT 15

0·2

0 200 400 600 800 1000

0 200 400 600 BOO 1000 Temp °C
Temp °C

'"~
1·2

'::rt
Typical hot- rolled
O·B

0·6
\
\
\
\
REVISION ~"" ~.. _.,,..
\
0·4 \ REVISION
\
\
\
0·2 \ 0·2
Normal Pmk White grey ,Buff

0 200 400 600 800 1000

0 200 400 600 BOO 1000
Temp °C
o-3
Temp °C
<
('!)
'<

Fig. 5--Compressive strength on cooling Fig. 6--Yield strength of steels at room temperature
after heating to an elevated temperature 0\
.......
 CcJ T ..... ptnh.....iotribvtOooo;., • .,....... .,,_,..,, M ............
0"1
(a) Slabs exposeo ~o hea: oo one face liOO,--,-,---,---,-,,.-,--r-r-r-o-T,;-T-r-l ~

~··· :·.::~
eoo

( ) >
=
000
'.....
::· (fJ
(fJ
('!)
I (fJ
- I OC;

JOO
: :1
Joel
i\imtiLJ I l f"l =t'- 1--.....;,_ "'"'""'I == =r----i- --+==-:....___
1 ~~ ~
i i E
3('l)
:::;
,..-
,.
~
:::;
""""--- I I
c..
2oo! 1 ! ! f"'-- ~ \ ::=t-- --L: :=--r=--9 :zn~
\OQ I .::lVU :-:1m Ul"t uv:.K '>"YI;..;=:; Kl= I =+==-=+- -LJ ~
;:::::;
('l)
v
7~ 100
O•srane~ !rQ<II
1::0::.
~~oos~d T.oee o! si•C·"'"'
(OJ 'hm!'<'falur~ <lo"'robuhun on a d~n·~ concrorT~ rib<>' ~><'•m
=·
>-;

(b) 380 mm columns

~(':''~ Sur!ac~s ~•Po

\0 l1r~
j s~d

300 1 ! i I 1- ... 1'-L r=+ ....d: I T='. :-· I

2001 :;--- -L::;:---.: --L:l ~ 1
,I·:~ 'ool I
I I II -r=r----r1- II IY:r.!
I
o n m n • m ~ •
Dostanc~ !rom surlaceot CCIUMI'I •""" 0 I
o 10 20 Jo .:.o so Go 10 so so :oo 110 120 tJO t.:.o 1so
Monomum distonc• !rom sur!oc• •~<POS•d to lir• (mm}

Fig. 7-Temperature gradients at various times from start of BS furnace test