_ (.

) al Acc:eleraton and Power Elec:tronics

'Ii"ansformer Temperature Rise Test

'l'nmlformen ue ~abject to a temperature rile te1t cllll'iDJ Factory Aceeptuu Tettl to verify the akulated
temperatare rile (.&.T). The teet il perCormecl iD two atepl. l'he "11.0-load" teet nnltl wtemperature rile clu to tile
hyltlftllllouaiD the lllq1lttk eore. Tile "tlulrt-clmdt" tilt renltllll temperature rile du to tbe thenuli.OIIet oa
the wllld!ql.

Otller IIUIIUI: Also bown u "heat nm tat".

Objed~Ye: to wrify the wont ~ue averaae temperature rise in the traufonner windinp.

The tat invoms two indepaldall testa; an open-ein:uit (DO-load) test and a short-clreuit test. The pocedum ~
s\IIIIIIWDecl below and f.O!!fimn to IEC60Q76-2 (paiagraph S) and using tho proc:oclure dcsc:ribod in IEC60076-ll (parapph
23) for Illy-typo 'liansfoimcrs.

Using the "simnl!!!ed load method" described in ptll'23.2.1 (IECii0076-ll). The Older oftho two teats (open-c:in:uit md
shmt-c:in:uil) can bo cii!Cided by the Dl8llllfactun!r 1

Temperature Rise Limits

Far dry type 11:111W'01:111et11 these are de1iDed in IBC60076-l1 (page 27, 18ble2). For eumple in a componmt with illaulation
system tempe:ratute c1us F the mflrimnm cempenture (hot-spot) is 155"C ad the lempellltlll'C rise limit is 100"K. At
Cern a 70"K temperature riJe ie often apeeified. (max ambient tempetature of 40" C)

Test Procedure
Before the tests the de w.indina resi.stan= is memared1 and noted toaetbcr with the ambient trmpemture. To check tho
~variation in the compcment, ICDSOl'S ~ pJaced imide tho windings (in the upper half of the windinp).

In tltme-pluut! trtmafoi'IIID'! tcmpersture is meuured in the caller column and one oftbe oumr oobunna windinp. The
msiallml:e islliC!BliSliRid across a tonniDal of a winding in tho cllllter limb and a taminal ofa winding at the outer limb (all
priDwy/soc:cmdaries).

Open-dn:alt test
Tat Objtdift: To lllCUlllC the lllmafomler windinp tcmpeJatun: rise clue to magnetic: 4lOR hyatcmria loesea.
CODiltCtio:u: Gle aeoonMry wiDding8 are lell opea.-c:in:uited while the primary ia cODDeC1ed to a IIOUICC that npplica the
nomiDa11rauforme:r voltage at liiODiinal D:cque:Dcy.
Tat MqueDCe:

the initial ambieat 1aDpel8tU!e ~ in the lett filcility is measured and DOted.
a tempeJitUie ~Je~~Bor il placed in the windiDga and core u deecribecl above.
the initial wiDdinp raistance ~ il meuurcd before tho heal-up il81al'!ed.
the teat ~mw (for several hours) uati1 stculy state tempelaliD'C is reachccf
the ambicmtcmpamurc T. at ~is mcorded.
a chnmometer is set at fhe inltaDce of supply imerrupCioD.
the windinp timelreaiatance il reconlm replarly.
Pott-pnlllftliq: UJins tho eeistanco-timo pairs ono can plot the msiatance cbK:ruse in time and can projocl to find tho
msistllllee valuo R. at time 0 (lima of disc:cmnedion ofthe pow. supply). This m~istance a1aog with the initial n!Sistmco
Be ~~~:~~applied in tho followin! fummla to find tho can1ributian 1!10~ of the core. This llOIIJWly bas a IIDIIID influence in tho
fimll ae,., calculation (dcecribed in IE.C60076-2 puagraph 5.4).

ao~ =
B.-Be
~ (Kt+7i)+n -T.

where Kt u22S in~ ofaluminum ICOI!ductoll and 235 in~ ofcopper COllductore, ~ is the iDitia1 ambient
taDperature aDd T. il tho ambiall taDperature at clilcoJme.ctio11.

112
31612017 Technical note: Transformer Temperature Rise Test

Short-circuit test
Test Objective: To measure the transformer windings temperature rise due to full current in the windings.
Connections: the secondary windings are short-circuited while the primary is connected to a source that supplies the
nominal transformer current at nominal frequency. When heating is complete a resistance measurement setup is connected to
2 terminals on each winding (primary-secondaries).
Test sequence:

the initial ambient temperature Ti in the test facility is measured and noted.
a temperature sensor is placed in the windings and core as described above.
the initial windings resistance ~ is measured before the heat-up is started.
the heating-up can be stopped by disconnecting the supply when the temperature has been stable4
the ambient temperature T 8 at disconnection is recorded.
a chronometer is set at the instance of supply interruption.
the windings time/resistance is recorded regularly (up to 10 measurements usually).

Post-processing: using the resistance-time pairs one can plot the resistance decrease in time and can project to find the
resistance valueRs at time 0 (time of disconnection of the power supply). Rs along with the initial resistance~ are
inserted in the following formula to find the windings temperature rise !::!i..Oc during the short-circuit test. This test normally
has a major influence in the final l::!i.Ototal calculation and is (described in IEC60076-2 paragraph 5.4).

where Kt is 225 in case of aluminum conductors and 235 in case of copper conductors, Ti is the initial ambient
temperature and T 8 is the ambient temperature at disconnection.

Calculation of the temperature rise

Using the f:1.()E and f:1.()c we can calculate the total temperature rise l::!i.Ototal

where, l::!i.()E is the winding temperature rise during the no-load test, l::!i.()c is the winding temperature rise during the short-
circuit test and K 1 is 0.8 for natural air-cooling and 0.9 for forced air cooling.

1It is often wise to start with the no-load test where the transformer core is heated up (slowest process) and continue with the short-circuit test (where the magnetic core flux is low hence it is heated

by induction from the windings).

2Usiog a special instrument that injects enough current in the winding e.g. SA to measure a voltage drop accuralely.

\n the windings and magnetic core (this is when temperature changes less than 1 K per hour ocross all sensors).
4 Stabilisation ofthe temperature has been reached when a variation of less than 1 K per hour is observed for 3 hours.

Contents page I return to the top of the page I edit article

Ull(' XHTML
n.J.... 1.0

http://www.kpap.neVnotes/item.php?id=TempRiseFAT 2/2