The Sound of Bells

Home Page Back to list

Manchester Town Hall, 23, 42-2-25 in B

Recorded: WAH 27/3/02
Analysed: WAH 28/3/02

This installation is unusual; it is a carillon of 23 bells, of which 13 (a twelve with a sharp second) are hung for ringing. The tenor of the ringing peal is also the bass bell of the carillon. The twelve are rightly held in high regard; the sound from outside is excellent. They go very well, helped by the almost complete absence of tower movement (the tower, though high, is supported by the surrounding buildings). The Town Hall complex, by Alfred Waterhouse, is an exciting set of buildings if you like Victorian gothic with a continental air. At Christmas, the belltower is usually graced by an enormous inflatable Santa climbing up the side. The clock bell at the top of the tower, Great Abel, is named after Abel Heywood who was mayor of Manchester when the town hall was completed in 1887 at a cost of a million pounds.

When the tower was first built it contained a 21 bell carillon by Taylors of 1877. The heaviest bell of the carillon was 162 cwt 3 qtr 3 lbs or 8,269kg and the total weight of bell metal was just less than 35 tons. Ten of the carillon bells were hung for ringing with a tenor of 52 cwt. All the bells apart from Great Abel were recast by Taylors in 1936 to create the current installation. I had some trouble in getting a good recording of the twelve being rung - they sound best from Albert Square but there is usually a lot of traffic noise. This recording was taken through an open window in an upstairs corridor and is actually of the echo from the gable-end of the Great Hall. From the floor above the bells the sound is a little less distinct.

I am grateful to Jeff and Stephen Brannan for permission to record the bells and for help with the recordings. Unfortunately between us we omitted to record the smallest bell of the carillon. I am also grateful to David Bryant who first pointed out that the smaller non-ringing bells in the carillon were cast to a lighter scale than the ringing bells, a point I return to below. David supplied, from Chris Pickford, weights for all the bells.

Bell Founder Tuning
1 - 23 Taylors 1936 none since

Tuning of main partials

Tenor nominal: 505.3 Hz. The two left columns in the table below give the position of each bell in the carillon, and the position in the ringing peal.

Bell Ring Hum Prime Tierce Quint Nom'l S'quint O'nom. Weight (kg).
1 - - - - - - - - 84.8
2 - -2409 -1201 -901 -493 2300 685 1225 100.2
3 - -2403 -1205 -893 -508 2200 680 - 126.6
4 - -2401 -1203 -895 -500 2100 679 1213 153.3
5 - -2401 -1202 -890 -507 1999 679 1213 181.4
6 1 -2403 -1207 -896 -504 1902 667 1191 272.6
7 2 sharp -2395 -1202 -892 -504 1799 671 1198 278.5
8 2 -2407 -1204 -906 -509 1703 685 1221 302.1
9 3 -2402 -1204 -901 -501 1600 682 1217 313.9
10 - -2404 -1202 -906 -504 1502 697 1247 268.5
11 4 -2407 -1203 -898 -508 1403 678 1216 377.4
12 - -2397 -1202 -902 -506 1298 696 - 318.0
13 5 -2398 -1202 -894 -501 1196 691 1239 395.1
14 6 -2393 -1208 -902 -513 1101 699 1254 436.8
15 - -2387 -1198 -886 - 995 688 - 466.3
16 7 -2398 -1193 -899 -493 894 705 1267 535.7
17 - -2395 -1203 -888 -501 800 691 1245 635.0
18 8 -2397 -1198 -897 -495 697 698 1253 694.5
19 - -2400 -1199 -893 -496 598 697 1256 840.1
20 9 -2384 -1201 -885 -508 498 684 1230 920.3
21 10 -2390 -1204 -896 -506 398 703 1264 1164.8
22 11 -2401 -1203 -895 -500 200 704 1266 1512.3
23 12 -2419 -1203 -894 -489 0 700 1259 2170.4

(The figures in this table are all given in cents. For all partials except the nominal, the partials are given from the nominals of the bell. Cents of the nominals are relative to the tenor. Pairs of values indicate a doublet. Frequencies for the quint are often not given, especially if inaudible. The links in the first column provide recordings of all the bells.)


The tuning of the Town Hall bells shows a considerable departure from Taylor's approach of twenty or thirty years before (as described elsewhere). The nominals of these bells show very exact tuning in equal temperament (none is more than 5 cents out), with no stretch whatsoever. Taylor's earlier true-harmonic peals were tuned in just temperament. This tuning in ET with no stretch is no doubt motivated by the fact that the bells were designed to be used as a carillon. By 1962 (Tewkesbury) Taylors were back to tuning stretch again - but that is another story. The tuning of the other low partials in these bells is also quite precise, though clearly some sacrifice has been made to get the nominals spot on. The tierces are all close to 900 cents below the nominals, an equal tempered minor third above the strike. This again is a departure from previous practice; until the mid 1920s the tierces would have been sharper, approximately a just minor third. (This partial was probably not explicitly tuned, Taylors by repute relied on design to position this partial. It is said that their tierce tuning changes over the years due to wear on the strickle used to form the mould from which the bell is cast.)

This is really all that needs to be said about the low partials - these bells are probably as closely tuned as one is likely to get.

As will be clear from the weights given in the table above, which do not go up monotonically, the smaller chiming bells are rather lighter than the ringing bells. Taylors (possibly learning from Gillett and Johnston) discovered that they could cast the small bells of a chime to lighter weights than those of ringing peals, thereby saving both metal and cost. It is clear from the tuning table above that this did not affect Taylor's ability to design successfully for the low partials which are consistent across both chiming and ringing bells. However, the plot below of the superquints and octave nominals of these bells shows some interesting effects.

Ringing and chiming bells

This plot probably requires some explanation. The axis along the bottom is the number of the bell - the smallest bell is on the left, the largest on the right. The lowest (blue) line is a representation of the weight of each bell. The quantity plotted is the percentage by which the actual weight of the bell is less than a theoretical bell with the same frequency. (For the technically minded, this theoretical weight was calculated from a regression fit of w = A*f^n to these bells, where w is the weight and f the nominal frequency). The higher the weight plot line, the proportionately lighter the bell. The smaller chiming bells can clearly be seen (bells 2, 3, 4, 5, 10 and 12) and are more than 20% lighter than the corresponding ringing bells. The heavier chiming bells (15, 17 and 19) have weights which match the ringing bells.

The superquint and octave nominal frequencies are given in cents from the nominal (with 640 cents arbitrarily deducted from the superquint and 1140 cents from the octave nominal to bring the values to a good place on the graph). Ringing and chiming bells are plotted as two separate lines in each case for clarity. In the smaller bells, there is a definite relationship between the tuning of these partials and the relative weight of the bell. The heavier scale ringing bells have clearly flatter upper partials. Beyond bell 13, when the scales between ringing and chiming bells become compatible, this relationship disappears.

This relationship between tuning of higher partials and weight or thickness of the bell is very common - small thick bells usually have much flatter versions of these partials. In fact, it is the smaller chiming bells which maintain the design of the tenors; the smaller ringing bells are cast heavier to give them a higher rotational inertia which makes them easier to ring together with the bigger bells. David Bryant gives other examples of Taylor's output where ringing and chiming bells are cast to different scales.

I have not bothered to present the intensity profiles of the bells as part of this write-up. They show very little variation across the peal and between ringing and chiming bells. Despite the differences in scale, the bells have a similar sound throughout the chime.


Last updated April 19, 2002. Site created by Bill Hibbert, Great Bookham, Surrey