Can I connect two 3 phase transformers in parallel to increase KVA rating?

Status
Not open for further replies.
dmanda24

dmanda24

Senior Member
Hello, I am doing a job for a printing company, where they need to install a particular used printing press, to make it easy I will list the facts (numbered)and issues(lettered):

1) The machine runs on 3 phase 480 V
a) There is only 3 phase 120/208V coming in to the building, transformer needed.

2) The main panel for this used machine was fed with size 500 wire with a 400 A disconnect
b) Given the size of the wire the maximum load should be 252KVA so the use of a 300 KVA 240/480 D/Y transformer is required, pretty expensive.

3) The owner has two identical 3 phase 240/480 D/Y 150 KVA transformers

4) Can these two transformers be connected in parallel to achieve the needed KVA for the equipment.
If so are there any special considerations?
How would the overcurrent protection be for these parallel transformers?
Is the KVA rating just simply additive?
Any special way to ground the two transformers?

Any help would be greatly appreciated.

Daniel.
 
This is a far more complicated problem than you might think. My first question is whether the equipment can handle a voltage supply that is 13% low. If you start with 208 volts, and use a 240/480 step up transformer, you will only get a voltage of 416. That is 87% of nominal, or 13% low. I doubt that you will have tap settings that can correct for that low a voltage supply.

The most important issue, in my opinion, is whether your power distribution system can handle the fault current that would be supplied by two parallel transformers. You will recall, perhaps, that paralleling identical resistors will give you a net resistance of half the value of either one alone. That means that the two parallel transformers will have a net resistance, as seen from the rest of the system, and as measured at their output terminals, of half the resistance of either transformer alone. That means they can supply twice the total fault current. In order to verify that your system can handle that level of a fault, you need to perform (or have someone perform for you) a fault calculation.

The answer to your question in line #4 is yes, the KVA that will be supplied is additive. But the impact of a trip of one transformer, with the other remaining on line supplying only half the needs of the equipment you are trying to supply, might be to destroy the equipment.

I don?t know enough to advise you on overcurrent protection or grounding. But you are right to ask about them. These too, will not be easy tasks to handle.
 
How many motors are in this machine? Are they all fed from VFD's? Seems like it would be at least worth investigating replacing the motors and control transformers. Perhaps you could contact the manufacturer and get a list of parts that differ between the 480 and 208 versions.

Mike
 
correction

correction

Charlie, I am sorry, the 2 150 KVA transformers are also 208/480 V, and as far as the overcurrent is there some sort of disconnect that would trip both transformers at the same time?
thaks
Mike, these machine is huge is like an assembly line maybe 100 ft long with a ton of motors and compresors, everything is already hocked up to a main panel which is what I'm trying to power up, messing with the machinery is out of the question. Thanks
 
dmanda24 said:
. . . as far as the overcurrent is there some sort of disconnect that would trip both transformers at the same time?
Click to expand...
I think that what you need is called a "network protection scheme." Having said that, I have exhausted what I know on the topic. But I do think it is worth your while to bring in the services of an engineer licensed in your area and familiar with this type of design.
 
In your first post, you said that you had a pair of 240V to 480V delta to wye transformers; in a later post you corrected yourself and said that you have 208V to 480V transformers.

I would strongly suggest that you confirm that you have 208V _delta_ to 480V wye transformers, and not the far more common 480V delta to 208V wye transformers. The latter sort is used to provide 120V circuits when you have 480V distributed in a building, and is very common. You can wire such transformers 'in reverse' to get 480V out with a 208V supply, but there are significant grounding issues that you must account for.

-Jon
 
"winnie: In your first post, you said that you had a pair of 240V to 480V delta to wye transformers; in a later post you corrected yourself and said that you have 208V to 480V transformers."

transformers are step up transformers 208/480 D/Y they were being used to power other equipment that the current machine is replacing.
thanks
 
As far as fault current - 2 x 150 KVA transformers in parallel with an impedance of 3.5%, as seen on the 480V side of the transformer, would only be around 10,400A of total fault current contribution.

Check the actual fault current through each transformer for exact values but it should be well within 480V equipment.

Additional concern is when you say identical transformers, you need to make sure the impedances are nearly identical, because even a slight difference in the transformer impedance will cause circulating currents, resulting in additional heating.

You will need to make sure that if one circuit trips, you will want to trip the other one. This can easily be accomplished by using contactors that are interlocked.

Buyer beware, that by the time you get done putting together all the parts and pieces to use the 2 x 150KVA Xfmrs, it may be cheaper to get the 300KVA transformer.
 
Contact your local utility provider. I have had to upgrade building services to 480 VAC when installing machinery with less KVA than you are describing. Maybe a seperate service for the printing line.
 
All of this technical knowledge is extremely self -concerting and admirable, but the fact remains, if those transformers have even 1/2% impedance difference you will overheat one and the other will have to carry most of the load which it cannot do. If that happens you might as well take them both to the nearest recycling yard. This whole idea is extremely dangerous.
 
Jim Carlson said:
All of this technical knowledge is extremely self -concerting and admirable, but the fact remains, if those transformers have even 1/2% impedance difference you will overheat one and the other will have to carry most of the load which it cannot do. If that happens you might as well take them both to the nearest recycling yard. This whole idea is extremely dangerous.
Click to expand...

I totally agree. Not worth the risk IMHO. Not only do you need identical impedances, but a slight difference in tightening torque on a connector or a slightly longer cable (more resistance) could start the ball rolling as well. And once one of the transformers starts to heat up, its impedance will change, accelerating that meltdown effect somewhat rapidly. Even utilities don't take this risk.
 
Jraef said:
I totally agree. Not worth the risk IMHO. Not only do you need identical impedances, but a slight difference in tightening torque on a connector or a slightly longer cable (more resistance) could start the ball rolling as well. And once one of the transformers starts to heat up, its impedance will change, accelerating that meltdown effect somewhat rapidly. Even utilities don't take this risk.
Click to expand...
Although the concerns you mention are real, and design issues must be considered, this is not a dooms-day scenario.

There are many instances where transformers much, much larger (e.g. 10, 20 MVA and even larger) are paralleled without having problems. As far as Utilities not taking this risk, apparently you are not aware that in Europe they are paralleling 67% generator step-up transformers so that in the event one is out, they can still produce 67% of the generator power.
 
Utilities parallel transformers ALL THE TIME.

I have Many, Many, Many, Many, subs with parallel transformers....., typically 5-30 MVa

While I agree they need to be matched as well as possible, to say that 0.5 % Imp. diff is a killer is not true..... both transformers would contribute, yes one will carry more load- based on the %Z, so the limiting factor for total load is the max load at which the lower imp. transformer is at full kva.

I have one sub that has one 7.5 mva @ 8%, two 7.5's @ 7.25%, and a 10 mva @ 10.5%, all connected in parallel.........for 12 years now. (138 kv to 23 kv).

I tend to agree that for a 300 kva service at 480 one trans is more practicle, but to say that 2 150's that are very similar won't work is not true.

A college actually hooked up 6 single-phase units up with the primaries in parallel and the secondaries in series---then through a tap-changing voltage regulator (to get proper voltage)... to get a him out of a bind due to long delivery times. Odd yes, but it worked.............
 
I have done many jobs like this where we try to save the clients money and only create headaches and liabilites for ourselves. I believe common sense is your best ally and whomever is putting in this big bad machine is:

A) Going to want it to work
B) Going to want it to work right the first time and many times afterwards
C) Is going to be extremely upset over downtime/going back to fix the original problem and probably come after you with everything he has got to make you pay for this new installation

It is truly in the client's best interest for you to provide a new 300 KVA transformer such that no problems arise. I don't think you need a consultant, special contact switches, funky parallel configurations...what your trying to do by saving the client money is only going to come back and bite you in the behind. I understand that the client does not want to spend money and wants to reuse the existing transformers - there is probably a way to do it. However, it is extremely risky and just a bad, bad, bad idea all the way through. Spend the money, do it right, forget about it, be done with it and have a happy client who wants to return to you for future projects instead of take you to court. This is my two cents, but I see this everyday with jobs that we do...when a client is putting in a data center he doesn't mind spending an extra few bucks here and there on stuff...he wants a solid job and that is what we deliver.
 
If I understand correctly. You have 2 seperatly derived systems, your phases may not be synchronized between systems on the secondary even though your primary is. I could be off, but it's worth checking. I never ran into this situation before.
 
If the transformers are identical models, they should be well matched. I don't see a problem with a parallel connection.

And yes, since amps add, volt-amps also add.
 
Jim Carlson said:
All of this technical knowledge is extremely self -concerting and admirable, but the fact remains, if those transformers have even 1/2% impedance difference you will overheat one and the other will have to carry most of the load which it cannot do. If that happens you might as well take them both to the nearest recycling yard. This whole idea is extremely dangerous.
Click to expand...

I agree. Practically speaking this is too complicated. As jim says, impedance matching would be very important to assure the load split evenly between the two transformers. This would be mitigated a bit if the two transformers in total were oversized but still risky. I know this is tempting since you already have the two transformers and it would be kinda fun to do, but too uncoventional. Unless this is some dire situation, get yourself a 300 KVA transformer.
 
Last edited:
Status
Not open for further replies.
Top