Page images
PDF
EPUB

each, by means of a suitable key. The distance apart of the pole tips can be read off on a scale C fixed to the body of the machine.

Note. The pole tips must never be closer together than the two zero scale divisions which will be termed their normal position in what follows, and must

always be left at this distance after the test is over. To increase this distance turn the screw clock-wise. It will be noticed that the initial slopes of the curves in (a) are determined by the air gap, also that the air gap causes the curve to bend over.

All lubricators must feed properly before the

chinery is started.

Observations. а -

ma

- (1)

Connect up as shown in

[blocks in formation]

Fig. 69, and adjust the pointers of all the instruments to zero. (2) Set the pole tips at exactly the normal distance apart and adjust the speed so that with the maximum excitation allowable in the F.M. coils 50% above normal, the E.M.F. can be read off

on v.

(3) With air gap and speed constant, adjust the excitation to about of the maximum allowable. Note this reading A and that on (v) viz. E.

(4) Repeat 3 for about eight ascending equal increments of current to about 50% above the normal excitation.

(5) Repeat 3 and 4 for the pole tips half-way and the farthest apart.

(6) Repeat 3-5 for the same current values descending.

(7) Plot curves in each case with M.M.F. as abscissæ and Nas ordinates.

B-(1) Adjust the exciting current to the normal value and the speed so that the E.M.F. can be read off on v.

(2) With M.M.F. (i. e. A) and speed constant and the pole tips at exactly the normal distance apart, note the reading (E) on v.

(3) Repeat 2 for eight different distances increasing by at a time to the maximum possible.

(4) Repeat 2 and 3 for a return set of distances to the minimum (normal).

(5) Plot curves in each case with distances between iron of armature and pole face as abscissæ and N as ordinates.

7-(1) Adjust the excitation to that a suitable low reading of, say,

on v.

maximum and the speed so maximum voltage is obtained

(2) With N (i. e. E) and speed constant and the pole tips at exactly the normal distance apart, note this distance (d) and the exciting current A.

(3) Repeat 2 for eight values of (d) rising by of the maximum at a time to the maximum, noting A, at each position, which is necessary to keep E constant.

(4) Repeat 2 and 3 for a return set of distances to the minimum (normal).

(5) Plot curves in each case with M.M.F. as abscissæ and (d) as ordinates.

[merged small][merged small][merged small][merged small][ocr errors][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

Deductions.

State very clearly all the inferences which you

can draw from your results and point out their bearing on dynamo design.

(82) Localization of Faults in Magnetizing

Coils. (Induction-Ballistic Method.)

Introduction. When a magnetizing coil of insulated wire is wound on a metallic bobbin, the latter is usually insulated on the inside by a thin strata of insulating material before winding on

the covered wire. Notwithstanding this, it may and does sometimes happen that the wire core becomes "shorted" to the metalwork of the bobbin, through the covering and insulation of the bobbin. This is particularly liable to be the case in shunt coils of dynamos which are wound on metal “formers,” insulated with vulcanized fibre tissue before winding.

Such a fault, through poor contact of, in many cases, a very uncertain nature, gives trouble in the ordinary methods of testing for its position, by giving unsteady readings. Thus the ordinary resistance methods are extremely liable to be vitiated by variable contact resistance at the fault. The following method for localizing the position of the fault by means of induced currents, measured ballistically, is often a more convenient and reliable one for the purpose.

Apparatus.-Metallic bobbin or former F to be tested, wound with the magnetizing coil (m) which is "shorted" to frame at the point (f); high resist

[merged small][merged small][ocr errors][merged small][subsumed][merged small]

be tested; rheostat R; known high resistance box r.

N.B.-It will be noticed that, as represented in Fig. 70, the fault (f) is on the first layer of turns next to the frame F, and we will suppose that the turn at (ƒ) is making contact there with the metallic frame (F). Thus it will be seen that the point (ƒ) divides the total number of turns on the whole bobbin into two parts between the leading out wires TT of the coil, so that total turns turns between T1 and ƒ + turns between T, and ƒ. Observations. (1) Connect up as in Fig. 70, and adjust A and G to zero, the temporary coil PP having been previously wound on and a wire soldered to any point (p) on the metallic bobbin frame F.

=

2

2

(2) With R full in, close S and adjust the current on A to some convenient amount. Next also close K to stud 1 and adjust r to such a value as will give, say, or scale deflection d1 on G when S is opened suddenly. Repeat two or three times with the same constant current, both made and broken in P.

(3) Close K to stud 2 and repeat 2 above with the same current, noting the new resistance out in r to give a suitable first throw on G.

(4) Repeat 2 or 3 for about four or five current strengths A so as to obtain finally different throws on G which will check one another, and calculate the position of the fault (f), or the number of turns to be unwound, to reach it, from the relation turns between T, and ƒ__ mean 1st throw d 1

N

[ocr errors]

X approx.

[ocr errors]

N2 turns between T2 and ƒ ̄mean 1st throw da 2 where r12 are the total resistances of r+G when obtaining d1 and d, respectively, and which are assumed to be very large compared with the contact resistance at ƒ and also the resistance of the turns between ƒ and both 7 and 7. If the resistance of the coil (m) is from 5 to 20 ohms then (+G) should if possible be at least 10,000 ohms.

1

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small]

N.B. It will be noticed from the formula in 4 that if r is adjusted so that d1 = d2, then

[ocr errors][ocr errors][merged small][ocr errors][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small]

If G is insensitive an iron core may be inserted in the coil to form a closed circuit if possible; this will increase the flux for a given current made or broken in PP, and therefore also the first throws d d on G.

2

2

This has the further advantage that N1 and N2 will now enclose the same number of lines of force, which is only approximately true if there is no iron core and the coil long.

It should be observed in passing that even a simpler method still than the one described above, for finding the position of the fault (ƒ), would be to employ a slide wire or meter bridge or other convenient form of potential divider in the following manner. Connect the ends T1 T2, Fig. 70, of the faulty field coil to the extremities of a meter bridge wire and also to two or three Leclanché cells; connect the galvanometer G, which need not now be ballistic, but which must be sensitive, between the metallic former at Ρ and the slider key of the bridge wire. Now move the key such that on tapping it G does not deflect. Then the lengths T1f and fT, of the faulty coil are in the proportion of the corresponding lengths of the stretched wire either side of the K, and are therefore known if the gauge of winding and its resistance (which can be measured in the ordinary way) are known.

2

(83) Efficiency of Direct Current Dynamos. (Swinburne's Electrical Method.)

Introduction. This method, due to Mr. James Swinburne, has the advantage, firstly, in point of accuracy, of being solely an electrical one, and therefore far more accurate than a dynamometer method in which the power required to drive is measured mechanically; secondly, of not requiring another similar machine for coupling to it, in addition to the one tested. The method, which is often termed the "Stray Power" method, is consequently very suitable for employment in workshop determinations, where usually no good transmission dynamometer is available for measuring the H.P. used in driving the generator under test. The principle of the present and all similar methods is based on the following, namely, that the total power put in = total power given out total power lost internally or in symbols

W1 = Wo+ WL

where the suffixes I, O and L denote the input, output, and total losses in Watts (W) respectively.

N

« PreviousContinue »