I wish to present here the benefits of utilization and operation of MV/LV parallel transformers in electrical distribution. Having such experience, i will be interested to get feedbacks from other engineers.
2 MV/LV parallel transformers distribution network
For all my experience, I’ve never met a working electrical installation, where operation of parallel transformers was used. Once, we used this solution in the design of welding shop, mainly to reduce the voltage drop in the network and maintain a stable level of short-circuit current. But the project was stopped by the Employer, and the idea was gone. Below is the list of benefits of such solution, as it is seen. It would be great if interested engineers commented and shared experiences on this issue.
Generally, application of parallel transformers allows achieving the following benefits:
1) Reducing the total capacity of electrical transformers (as compared to separate their work). The decrease of total installed capacity is reached:
- by lowering the overall demand load to the diversity of loads connected to different transformers
- by using a higher load rate of parallel transformers
- less required backup in case of electrical transformer failure
2) Reduction of electricity losses in electrical transformers due to a possible disconnection of unloaded transformers
3) Improving the power quality due to the stable level of short circuit current throughout the network
4) Increasing the reliability of operation of protective devices in the case of phase-to-earth short circuits in the network.
5) Possibility of placing electrical transformers in operation phase-by-phase
Parallel transformers are allowed, provided that none of the windings will be loaded by current exceeding allowable current for that winding.
Of course, there are limitations for using electrical transformers connected in parallel. For instance, Russian Federation “Rules for technical maintenance of electrical consumers” allows operation of parallel transformers under the following conditions:
- windings connections of transformers have the same vector groups
- the ratio of transformers capacities of less than 1.3
- rates of transformation differ by no more than +/-0.5%
- short circuit voltages do not vary more than +/-10%
- transformers have same phase polarity
Artem Kropachev
This post is very useful. Parallel Distribution network information with diagrammatic representation is very helpful and nice.
thanks for providing such information. it helps us to clear our queries.
This post is very useful.
Good,Helpfull
Parallel transformer is quite interesting. It will help to decrease transformer faults rate.
For parallel operation of the Transformers, they should have same short circuit current level.
And what do you do with your short circuit current increase in your installations !!!
Yes it’s a good Idea for the System reliability. I have seen it in refinaries and GasPlants (K.N.P.C-Kuwait) But only the problem is the short cirtcuit level will be higher and you have to go for a high sizing of Bus Bars and Switchgears ( economically burden is there).
I have come across huge problems in the past with this arrangement.
Whilst parallel transformers increase redundancy the prospective fault level doubles.
Also if there is a fault on one incomer where the TFers are utilized more than 50% each the overall system will be lost due to overload. Capital cost / expenditure for clients is also increased.
Would not recommend it.
Using two (or more) transformers in parallel might cause problems. The voltage level of the transformers need to be exactly the same, from no load to full load. If not, then there will be blind current running from one transformer to the other. Running with generators is about the same but then you can change the voltage charcteristics (voltage droop). With transformers you do not have any possibilities to change the voltage charcteristics.
I would not recommend it.
In my field of application I do not apply such configuration. However some time I found it already implemented in some installation without the proper protection arrangment. It is extremely important to implement the directional overcurrent protections otherwise the configuration is useless and dangerous. Other important points are the short circuit level that shall be consdered.
Speaking on my behalf this type of distribution is quite common.
Not just in refineries but also steel mills, glass production, pharmaceutical, ciment factories and even in some cases supplying tertiary installations.
The comments made earlier regarding this type of distribution reflect the aspects with which this type of distribution has been evaluated.
Cost, efficiency, redundancy, reliability, network quality, loss of production etc.
There are cases where this type of distribution is obligatory by law (Hospitals).
The “rules” mentioned for the use of parallel transformers by the author are also used in Belgium although they are not officially included in the electrical regulations, they are considered however to be good practise.
So proper -technically correct- design is mostly dependant on the way the design engineers conceive the whole distribution.
Just as in any case if something is not done properly then it should not be expected to perform well.
So yes, when dimensioning such a network all the afore mentioned factors (short-circuits levels, busbars cross-sections) should be taken into account to correct implement this type of distribution.
what about the circulating current between the two transformers in case of both are loaded or in case of now load moments ,we have noticed in one of the cases where we have energized two transformers in parallel that the measuring instruments and specially the KWHM was recording some values in one of the transformers although there was no load connected to the parallel busbar , does this mean we have losses due to this parallel operation or due to this circulating current between the two transformers specially when we are sure that there are no 100% matching impedance’s between the two transformers , this case happened when each Transformer was provided with its separate KWHM either on the LV or the MV side, do you have suggestions on how to overcome this problem
The main parameter regarding circulating current is the transformers voltage ratio. They shall be equal as far as possible. For transformer with off load tap changer it’s important to check that the two transformers taps are in the same position.
I would recommend a bus section normally open switch connected as a 2 out of 3 autoselective scheme. Generally this is the approach in oil and gas electrical design. Switchgear is designed to include the high short circuit not for the two transformers operating in parrallel but for the contibution to the fault from the motors. In oil and gas installations there are generally duty and standby motors and with all A motors connected to the A bus and B connected to the B bus, the loss of any one of the 2 transformers would not effect items connected to the busbars; ie 1 transformer is sized for the total load. Note in normal operation there are 2 transformers carrying each 50% of the load. This is considered to be a very reliable solution and does not incur the increase of fault levels and circulating currents when parallelling transformers. To reduce the size of transformers forced cooling could be employed as one transformer carrying 100% of the load would not be a normal running condition; although the cost of forced cooling fans may be a similar cost of the difference in the kVA rating!
As commented by other posts paralleling of transformers results in higher fault levels, circulating currents and if sized as 2×50% machines, possible overload of one transformer should one transformer be out of commission.
Utility networks companies do not generally parallel transformers in the distribution of electrical energy; providing resillience of supply by alternative switching. I personally would not parallel transformers I think any benefits are outweighed by the disadvantages. Other apparent advantages mentioned in the text can be achieved in other ways.
Hope this helps
Kindest regards
Stephen Alty
In case of parralleling transfomers , the switchborad prospective fault rating will have to at least equal to the fault 2 * transfomer fault current. Even when the load is shed, when a fault occur at the main bus it will replicate fault situation on both transfomers. the fault current will realy on the short circuit volatge of each transformer minus the volatge drop at any impedance between the tranformers and the fault position. Hence it will be mandatory to sizs the switcborad ( combining both tranfomers secondaries) to 2 the fault capacity of s ingle transfomer.
Very useful information – thanks – for my teaching – Electrical Power. Some books even don’t go into too much detail on paralleling T/Xs, only the usual sam clock number, same rating, etc. Thanks!
Very useful information.
Correct me if Im wrong, but will the use of 2 transfomers in parallel result in higher recative losses as opposed to using a single transformer with combined KVA capacity. This may be significant when the transfomers are not fully loaded.
According to statiistics, winding faults come to 50% and tap-changer faults come
to 17% in power transformers. If one runs n power transformers of identical
design in parallel, will there be n tap-changers? This could be reduced using
magnetic shunts which give a much smoother regulation but this is not done
in power transformers. I have no working experience with power transformers.
So I don’t understand why these are not replaced by toroidal power transformers
where magnetic shunts could be readily implemented. Could some experts
comment on this?
Parallel operation of transformers results in increase of fault level on the downstreamside of transformers and resultant increase in cost of downstream side ciircuit breakers, panels etc may nullify the benefits of parallel operation. Perhaps this might be the reason why it is not practised widely.
This is a very important consideration which affects the installation cost. It does not affect the MVA rating of the incoming breaker from the transformer; But the outgoing breakers from the bus with more than one transformer feeding it will have to have a MVA rating corresponding to the feed from all transformers in parallel. If outgoing are many, it will blow up the cost. Designers prefer to put normal operation scheme without paralleling and provide buscouplers to have flexibility and transfer of load without break and also to obtainin or realize most of the advantages of two transformers in paralled
I have a friend who has been a very experienced designer of power transformers up to 1000 MVA. When I asked him why the core must be of laminated designs
and not toroidal designs, his reply was that one cannot build large toroidal core
transformers greater than 3KVA. I thought this may be a mindset since
conventional designs had been around for more than a centuary. I believe that
far more benefits could be accrued from large toroidal power transformers to
the utilities industries. Toroidal transformers could be assembled by full
automation … one cannot do that with laminated power transformers. Coreloss
is far less in toroidal transformers. Do you realise that you household consumers of electricity are paying for the coreloss which you have no control over?
As far as I’m concerned, I have experience of parallel transformer in two types of installation.
***First in installation where high short circuit power is required for the well operation of the process. This is the case for example for welding process where the quality of the arc welding is directly related to the limitation of the voltage drop in the upstream installation. In this type of installation the two solutions to reach a high short circuit power are:
- use of single low ukr transformer,
- use of parallel transformers.
In that case parallel transformers allow the use of standard transformer, and this point is interesting in terms of failure, it’s easier to replace a standard transformer.
It has to be noted that sometimes:
- the transformers are not coupled directly but via a busbar trunking system that can reach 100 meters (the busbar trunking system is used to distribute the energy on large areas)
- up to 4 transformer can be coupled to reach the required power and short circuit power.
The typical applications of these solutions are automotive industry or similar industry using welding process.
***Secondly, parallel transformers can also been used for standard application but with the following requirements:
- high power to deliver, but necessity to use small transformer due to limitation of the access to the technical room,
- redundant solution to limit the impact on the application in case of a transformer failure.
The type of application in that case can be hospitals, big tertiary buildings, hypermarket in certain countries.
also, you can read my own article in this Blog about parallel-transformer
this great opportunity for the youth
With the unusual bad weather in Europe, heavy snow and flooding, and failed undersea rail lines
between UK and Europe, I guess that there must be a lot of breakdown of power transformers.
I heard that Insurers would have to pay quite a sum for the downtime in electrical supply.
This is also happening in Asia and South America. I feel that transformer technology must be
upgraded. I agree that there is great opportunity for the youth who wants “change”. This is
technological change.
good topic.finding solutions to the disadvantages will be interesting.
I have a friend who acts as transformer consultant to insurers. He is kept so
busy with investigating failed transformers especially in manufacturing plants
which have been having a good run for tens of years but now suddenly one
of the paralleled transformer failed. Imagine his surprise that even the original
designers and manufacturers were no more found on planet Earth. Repairing
means transporting from a remote factory to a nearby repair centre which will
take up to 6 months. Meantime one line is idle and the workers … I don’t know
what had happened. .. probably retrenched for the time being. If I am running
such a plant I make sure that I have four transformers instead of two but that
needs deep pocket. Insurance in this business must be very expensive.
ok….its good but more expenditure is required……….!
i am only a student at the moment but i thought paralleling tranformers was only used for redundancy. as in that a large factory that cant afford a large amount of down time would have a medium voltage ring and if one transformer went down the other would be switched in.
…if one tr went down, the other would be switched in.” —– This is not transformers in parallel operation. This is a case of a standby being brought on line following a failure.
am a student and i need a help to become electrical engineer so i can create a factor or contribute with my talent efficient run in electrical programs.Am pleading in the of God,if there is God sent one out there who can help me get to my destiny.Thanks
I probably had not been explicit when citing that example of a factory
going down on one production line because one paralleled transformer failed.
That factory did have a spare transformer but of the wrong primary
configuation. Furthermore, the control circuit was burnt out also leaving
no option whatever on-site. Maybe a well managed factory would have
seen to that but generally maintenance crews are not that well educated and
experienced. I do think young bright electrical engineers should look into
revolutionising the designs of power transformers. A hundred years of
traditional power transformers design and manufactduring need some new thinking. Power transformers should be easily transportable and assembled
with flexibility on sites. There must be a new way of designing such
transformers but I cannot leak out due to patent infringement. In any case
one would have to fight against mindsets in professions.
Very good and helpful information
I work for Caterpillar Co; we usually connect our all generators in parallel those they output LV and then the Genset LV out put as each generator has its own MV transformer rating 13.8KV, then all transformers are connected to one buss-bar (we call it grid) through fs switch gear, mean each circuit has its own protection that is connected to grid. Furthermore, the advantage of raising the voltage is great as that decrease the diameter of supply conductor plus reduction of LV switch gears, as an example, we supply 10M Watts power at the distance of 200 meters through 4 500 mm2 cable that is almost 75% less copper conductor is used. Nevertheless, all that configuration required perfect design by professionals. As an engineer, that is one part of my activity as well. CAT trained their engineer and they have proper tool to train. I shall welcome any question if some body wants to explore it.
sir,
i am an electrical engineer , work for saudi electricity company, i read your comment against parallel transformer benifits, it was very usefull to me , let me know the scematic diagram of the parallel connection of the transformer, then what is fs switchgear, i will be thankfull to you if you will reply to me
hi,
this is murugan , iam an elctrical enginner , i have been working as a project enginneer in transmission and distribution field , you gave some good needs, i want to know the parallel conection of transformer with single circuit line . i will be thankfull to you if you will give information about the above.
I am ending my participation on this blog. This is because
I shall be shortly going for major surgery in a local hospital.
Not being an expert, my remarks are basically armaturish.
Best regards to all of you. Please do not contact me by
email. Thanks.
thanks for your contributions and wish you the best for your surgery
Laurent M, moderator
Operating identical (or nearly identical or equivalent) transformers does not pose problems. That is the transformers should have the same (1)voltage ratio, (2)same percentage impedance, so that load sharing is even at different loads, (3) protection to avoid flow of current from one to the other either as a fault current or a circulating current due to mismatch of internal voltage (terminal voltage will have to be the same as they are connected). Protection scheme will have to have reverse power relay.
Generally designers prefer to avoid parallel operation as operating personnel and maintenance personnel can create problems itf not adequately equiped. If parallel operation is provided to realize some of the advantages, then specific operating & maintenance instructions should also be dispalyed.
Transformer losses partly come down with parallel operation (Iron loss is not affected; But copper loss in comes down as it is proportional to the square of the current.)
In addition to the three points mentioned, there are two more obvious points – (1) same phase sequence and (2) same vector group.