Delta/Delta vs Delta/Wye Isolation Transformer

[Lbartowski]

I'll try and keep this short and simple.

4 isolation transformer with a 480v primary and 480v secondary was specified on a project I'm working on. It's fed from a 250 amp breaker and the secondary goes to a 150HP VFD which controls a large cooling tower. The riser diagram shows a 3-wire primary and a 3 wire secondary...but it was symbolically depicted on the riser (with fancy delta and Y symbols) as a delta in and wye out.

What my supplier submitted were delta/delta transformers. Neither myself, the customer, or the engineer noticed it, and the submittal was approved and the transformers were ordered. Not until they were on-site did the customer notice this.

Now I have 30K worth of transformers sitting on site that the factory won't accept as a return (says they're custom made to order) and the customer refuses to accept.

Can anyone explain to me the customers point of view? Is a corner grounded delta less safe than a wye? It only feeds the one load which is straight 3-wire 480v cooling tower?

 

[winnie]

The key is that the transformers feed VFDs. The VFD my require a grounded wye source, or the VFD may have to be configured differently for an ungrounded or corner grounded delta source.

480V wye is also more common and may have more value down the road.

-Jon

 

[jim dungar]

So for this year I have have been involved in at least three of these situations.

We end up supplying an artificial neutral and sometimes an HRG.
The most common artificial neutral is probably a zig-zag transformer, although I just recently put in a wye-delta system as it was all we could find on short notice.

 

[Lbartowski]

What did you need the neutral for? In this case the transformers is only feeding one load? Through the VFD (VFD manufacturer is OK with delta/delta) and to the cooling unit, so I don't understand 1. why they would need a neutral and 2. why are they saying it's unsafe?

 

[Jraef]

What did you need the neutral for? In this case the transformers is only feeding one load? Through the VFD (VFD manufacturer is OK with delta/delta) and to the cooling unit, so I don't understand 1. why they would need a neutral and 2. why are they saying it's unsafe?

You are not going to USE the neutral wire, it is just bonded at the transformer. The entire PURPOSE of the Delta-Wye isolation transformer is to reduce the voltage reference to ground that is feeding the VFD, because the components inside of VFDs are designed based on that lower ground reference voltage.

On a Delta system, your reference voltage to ground is the line voltage, so 480V in your case. With a Wye system, it is only 277V. The components used in the drive are expecting 277V max, line to ground. If you feed them 480V referenced to ground, they can be damaged the first time there is a ground fault ANYWHERE upstream in the system, then your drives are toast shortly thereafter. Some drives, such as AB, allow you to easily remove the ground reference inside of the drive to make it floating along with the system, others, such as Siemens, make you completely disassemble the front end of the drive to get to that connection to clip the wire. But even if you do that, you are basically just prolonging the inevitable and it removes the path for common mode noise that is created in the drive from getting to ground, which causes a different set of problems.

So ordering a custom Delta-Delta isolation transformer was, to be blunt, a complete waste of time and money. Sorry.

 

[Strathead]

But I must say, I would hold my supplier responsible for it. Getting incorrect submittals approved without pointing out the discrepancies on the cover sheet, is not an acceptable excuse to say they provided what was asked for, and it isn't your responsibility or the Engineer's to catch it.

 

[mike_kilroy]

Once you get past the sky is falling, you are screwed, it can't work, give up, quit, etc., you can recover from this IMHO.....

You can either scream the sky is falling, all parts will fail quickly, and go with replacing these transformers, or you can try to argue logically to keep them and why that is ok.....

First you need to fully understand WHY the arguments against delta...

VFDs today typically have either 1) MOVs from each phase to ground, and/or 2) RC networks from each phase to ground.

For #1, the MOVs are usually rated for 460v wye, or 277v to ground; so when you supply a corner grounded delta supply, the 460v to ground often pops them - if not disconnected as Jraef said. So - if this is the case, just disconnect 'em and you are good to go.

For #2, I have never seen even the cheapest VFD with RC noise networks with less than 600v caps, so this is not an issue.

Even the cheapest VFDs will use 1200v minimum input rectifier diodes, so there is no issue here against your delta supply.

So knowing to just disconnect the MOVs via the internal jumper if required, you can continue your argument to your customer that THERE IS NO MEASURABLE DIFFERENCE BETWEEN A CORNER GROUNDED DELTA AND A WYE as far as the vfd is concerned!

The issue of a NON GROUNDED vs any kind of grounded supply for the VFD is this: If not referenced to ground SOMEHOW, then if a voltage spike comes in from say lightning, ALL 3 PHASE will rise equally above ground; as long as you have a wye or a corner grounded delta, all 3 phases have a low impedance path to ground. The issue with non grounded, is that 10,000v spike raises all 3 phases above ground 10,000v - the PCB traces inside the vfd input section cannot survive that and will blow off the substrate. Remember those 600v noise caps from each input to ground? They will pop. Damage Damage Damage.

But tie wye or a corner of delta to ground, and you have 2-3 ohms impedance to ground - that potential 10,000v spike will not be a 10,000v spike into the VFD - so not much measurable difference between either grounded system now.

You should be able to win this argument with your customer if attacked in this way, and in the end not have to replace those expensive transformers.

In addition, know most all VFD manufactures will accept delta power sources - ask your vfd mfgrer and if they tell you it is fine, share that with your customer!

In the future, pay better attention to the BOM as a D-Y IS better choice so you don't have to worry about the 277v vs 460v to ground, as well as it DOES cancel odd harmonic noise.....

 

[junkhound]

as well as it DOES cancel odd harmonic noise.....

Bingo --cancels the triplets is likely the customer's reason.

Of course, if they did not size the custom transformer primary to take that circulating current and dissipate as heat, then there will be failed transformers.

 

[mike_kilroy]

as well as it DOES cancel odd harmonic noise.....

Bingo --cancels the triplets is likely the customer's reason.

Of course, if they did not size the custom transformer primary to take that circulating current and dissipate as heat, then there will be failed transformers.

OP did not say kva size of xfmr so no idea if undersized.

OP said "What did you need the neutral for? In this case the transformers is only feeding one load? Through the VFD (VFD manufacturer is OK with delta/delta) and to the cooling unit, so I don't understand 1. why they would need a neutral and 2. why are they saying it's unsafe?"

They did not say anything about noise. A lot of vfd drives are run directly off the power source without a transformer today, so noise may not be a major concern. If it were me in this situation, I would not be giving them more ammunition to shoot me down.

 

[kwired]

What did you need the neutral for? In this case the transformers is only feeding one load? Through the VFD (VFD manufacturer is OK with delta/delta) and to the cooling unit, so I don't understand 1. why they would need a neutral and 2. why are they saying it's unsafe?

OP did say the VFD is ok with delta-delta though - so what he has will work with the VFD.

Owner still doesn't have to accept something outside of what was specified if they don't want to, even if it will otherwise work.

Supplier may be somewhat liable, may depend if OP was presented a quote that showed the wrong item yet the OP accepted it. Can more less turn into a legal battle from there.

 

[Jraef]

It's not that the drive will not work, it will, and plenty of drives work every day on delta power systems. But it IS at a higher risk for damage to the front end components, which is WHY the spec would have called for the delta-wye transformer in the first place. It will reduce some of the non-triplen odd order harmonics (triplens already cancel each other out in 3 phase), about the same as a line reactor or DC bus choke would, either of which would have been a lot cheaper than the transformer, so that's not the point of specifying a delta-wye transformer ahead of a drive. You go to that extra expense for the express purpose of creating the wye secondary for feeding the drive, then because it also does what a reactor would do, you don't need that too.

In my opinion whomever made the mistake needs to bear the cost to fix it. Stuff happens, the measure of integrity is in how you deal with it. Arguing about the necessity of the spec is one way, but sucking it up (by the person determined to be the proper source of the mistake) is the high road.

 

[kwired]

In my opinion whomever made the mistake needs to bear the cost to fix it. Stuff happens, the measure of integrity is in how you deal with it. Arguing about the necessity of the spec is one way, but sucking it up (by the person determined to be the proper source of the mistake) is the high road.

Determining who made the mistake isn't necessarily simple though.

If I sent plans to supplier and they send me a BOM pending my approval before ordering, then if the wrong thing was listed on BOM and I approved it - though I wouldn't like it, I see it as I am as much at fault as whoever placed wrong item on the BOM? If I haven't paid the supplier yet I may have more leverage on a compromise, but still comes down to at least two people being wrong at some level. Unfortunately if you turn this into legal battle - you now have problems with future business with the supplier, at least for a year or two, then maybe they treat you right again. Maybe not a problem for those that are in metro areas - there are many suppliers to choose from, where I am the local suppliers are still 40-50 miles away, and next closest get into 100-150 miles away - and some of those are just a different branch of same company that is closer.

 

[Strathead]

Determining who made the mistake isn't necessarily simple though.

If I sent plans to supplier and they send me a BOM pending my approval before ordering, then if the wrong thing was listed on BOM and I approved it - though I wouldn't like it, I see it as I am as much at fault as whoever placed wrong item on the BOM? If I haven't paid the supplier yet I may have more leverage on a compromise, but still comes down to at least two people being wrong at some level. Unfortunately if you turn this into legal battle - you now have problems with future business with the supplier, at least for a year or two, then maybe they treat you right again. Maybe not a problem for those that are in metro areas - there are many suppliers to choose from, where I am the local suppliers are still 40-50 miles away, and next closest get into 100-150 miles away - and some of those are just a different branch of same company that is closer.

Not the way I see it. It is very common for the specs on submittals to state that approval doesn't constitute acceptance of deviation from the specifications and some wording to the effect that any deviation must be identified on the cover page of the submittal. On gear and fixtures, I don't do the submittals, my supplier does. My BOM request states, "per plans and specifications provided" I have emails that prove what is provided. I also have verbal conversation with my suppliers about this very issue. I agree that legally it is gray area, but morally it is 100% the responsibility of the supplier, and/or his vendor. I do my best to review what they send me, and I have caught things many times, but my suppliers know that if they mess up like the situation described here, I am going to place the responsibility on them, period. I have yet for one of them not to live up. I will admit that one time the lighting vendor, didn't make good, however, my supplier did. I didn't use that lighting vendor for 2 years after.

There is a flip side though. That is treating my suppliers with respect, and wanting them to make a profit. I don't beat them up on price, I don't share their numbers until after a project has been awarded. And when something comes along that gives a better opportunity for profit, I select and work with those who have served me the best. Luckily, I haven't had a situation where a supplier has felt the need to get really mean. Let's face it, even here, I would again, work with my supplier. I would eat the labor, a new transformer is probably around $2,000 from say Square D and then the only other cost is freight. Your relationship with the vendor should be better than $2,000

 

[kwired]

Not the way I see it. It is very common for the specs on submittals to state that approval doesn't constitute acceptance of deviation from the specifications and some wording to the effect that any deviation must be identified on the cover page of the submittal. On gear and fixtures, I don't do the submittals, my supplier does. My BOM request states, "per plans and specifications provided" I have emails that prove what is provided. I also have verbal conversation with my suppliers about this very issue. I agree that legally it is gray area, but morally it is 100% the responsibility of the supplier, and/or his vendor. I do my best to review what they send me, and I have caught things many times, but my suppliers know that if they mess up like the situation described here, I am going to place the responsibility on them, period. I have yet for one of them not to live up. I will admit that one time the lighting vendor, didn't make good, however, my supplier did. I didn't use that lighting vendor for 2 years after.

There is a flip side though. That is treating my suppliers with respect, and wanting them to make a profit. I don't beat them up on price, I don't share their numbers until after a project has been awarded. And when something comes along that gives a better opportunity for profit, I select and work with those who have served me the best. Luckily, I haven't had a situation where a supplier has felt the need to get really mean. Let's face it, even here, I would again, work with my supplier. I would eat the labor, a new transformer is probably around $2,000 from say Square D and then the only other cost is freight. Your relationship with the vendor should be better than $2,000

$2000 yes, OP said he has $30,000 of transformers to eat, that changes things a little if supplier won't work with you. For some of us anyway, others that still is sort of pocket change.

 

[Strathead]

$2000 yes, OP said he has $30,000 of transformers to eat, that changes things a little if supplier won't work with you. For some of us anyway, others that still is sort of pocket change.

I didn't see a $30,000 price tag and I looked back through and still didn't find it. I did see where he stated it was 4 transformers. I would expect that the Contractors price on this would be in the $2,500 to $5,000 each. A $7500 transformer would be a very high priced transformer. Either way, take away the supplier markup and the manufacturer mark up, both of whom likely had a hand in the mistake, and the worst case of $5,000 should be no more than $3,000. Plus the labor. Each of course. It is likely that the job wasn't just a few transfomers, there were probably disconnects, breakers and possibly panelboards involved. It seems to me that what should really be at issue here is about $12,000. I totally agree that isn't chump change for a contractor. And if the supply house won't work with you then it would be hurtful if not devastating. I just assume that if you have gotten to the place where you tackle a job that includes (4) 480 volt transformers, you have gotten to the point where the supplier should value your business enough to make good on what is their mistake period.

On the other matter, I wouldn't even be trying to convince the customer that they should settle for the unspecified equipment as I would not, as a customer accept it myself.

 

[kwired]

I didn't see a $30,000 price tag and I looked back through and still didn't find it. I did see where he stated it was 4 transformers. I would expect that the Contractors price on this would be in the $2,500 to $5,000 each. A $7500 transformer would be a very high priced transformer. Either way, take away the supplier markup and the manufacturer mark up, both of whom likely had a hand in the mistake, and the worst case of $5,000 should be no more than $3,000. Plus the labor. Each of course. It is likely that the job wasn't just a few transfomers, there were probably disconnects, breakers and possibly panelboards involved. It seems to me that what should really be at issue here is about $12,000. I totally agree that isn't chump change for a contractor. And if the supply house won't work with you then it would be hurtful if not devastating. I just assume that if you have gotten to the place where you tackle a job that includes (4) 480 volt transformers, you have gotten to the point where the supplier should value your business enough to make good on what is their mistake period.

On the other matter, I wouldn't even be trying to convince the customer that they should settle for the unspecified equipment as I would not, as a customer accept it myself.

I'll try and keep this short and simple.

4 isolation transformer with a 480v primary and 480v secondary was specified on a project I'm working on. It's fed from a 250 amp breaker and the secondary goes to a 150HP VFD which controls a large cooling tower. The riser diagram shows a 3-wire primary and a 3 wire secondary...but it was symbolically depicted on the riser (with fancy delta and Y symbols) as a delta in and wye out.

What my supplier submitted were delta/delta transformers. Neither myself, the customer, or the engineer noticed it, and the submittal was approved and the transformers were ordered. Not until they were on-site did the customer notice this.

Now I have 30K worth of transformers sitting on site that the factory won't accept as a return (says they're custom made to order) and the customer refuses to accept.

Can anyone explain to me the customers point of view? Is a corner grounded delta less safe than a wye? It only feeds the one load which is straight 3-wire 480v cooling tower?

I don't know what KVA capacity is or how much more they should cost over common stock orders - just know that is what OP says.

Maybe you found a good source for your units or purchase enough of them you get a better price. I have had to pay in the $1500 range for a 25 kVA single phase transformer but would think OP has at least 150 kVA three phase units based only on the fact he says they are on a 250 amp breaker. $7500 seems like it would be in a reasonable range to me for at least that size of a unit.

 

[Strathead]

I don't know what KVA capacity is or how much more they should cost over common stock orders - just know that is what OP says.

Maybe you found a good source for your units or purchase enough of them you get a better price. I have had to pay in the $1500 range for a 25 kVA single phase transformer but would think OP has at least 150 kVA three phase units based only on the fact he says they are on a 250 amp breaker. $7500 seems like it would be in a reasonable range to me for at least that size of a unit.

Him showing as 30K made it so i didn't find it thank you. Obviously, there are a wide variety of issues with pricing, copper windings and how common being the most likely two, but in my experience over $5,000 for a 150KVA transformer Contractor cost would be too high. I don't want to exaggerate, but if the manufacturer actually pays $3,000 to produce, then $12,000 isn't a huge amount of money given the issues we are referring to.

 

[mike_kilroy]

Him showing as 30K made it so i didn't find it thank you. Obviously, there are a wide variety of issues with pricing, copper windings and how common being the most likely two, but in my experience over $5,000 for a 150KVA transformer Contractor cost would be too high. I don't want to exaggerate, but if the manufacturer actually pays $3,000 to produce, then $12,000 isn't a huge amount of money given the issues we are referring to.

My wife gets mad at me every time I say I refuse to pay over $10 for a pair of jeans. She says I am living in the 80's...

I have helped design, and sell, custom transformers - both alum and cu - since 1975. Yes, prices for both have gone up a lot.

But to say someone could build a 175kva xfmr (OP never said kva size, and I will assume he bought something like specified here: http://www.hammondpowersolutions.com/files/HPS_Catalog_Drive_Isolation_Section4.pdf) for $3000.00 is ridiculous - even to cheapy me. I can easily see $ 7500.00 each for either alum or cu xfmr today to a reseller. If one went out of their way to find a really really cheap design in alum that is undersized, with way too high flux, with 30x inrush, stacked 10x10 with non grain oriented steel, ya, maybe they could get it for $5k in qty 4. I would never sell such crap. I think the OP paid a very fair price for his listed price.

You want to see crazy prices, check the $ 30-60,000.00 each prices for 150kva from sq D: https://stevenengineering.com/tech_support/PDFs/45175-14.pdf

 

[mike_kilroy]

oops

oops

in strathead's defense, i searched a lot more andd DID find a $6k 175kva xfmr on line....

 

[kwired]

My wife gets mad at me every time I say I refuse to pay over $10 for a pair of jeans. She says I am living in the 80's...

I have helped design, and sell, custom transformers - both alum and cu - since 1975. Yes, prices for both have gone up a lot.

But to say someone could build a 175kva xfmr (OP never said kva size, and I will assume he bought something like specified here: http://www.hammondpowersolutions.com/files/HPS_Catalog_Drive_Isolation_Section4.pdf) for $3000.00 is ridiculous - even to cheapy me. I can easily see $ 7500.00 each for either alum or cu xfmr today to a reseller. If one went out of their way to find a really really cheap design in alum that is undersized, with way too high flux, with 30x inrush, stacked 10x10 with non grain oriented steel, ya, maybe they could get it for $5k in qty 4. I would never sell such crap. I think the OP paid a very fair price for his listed price.

You want to see crazy prices, check the $ 30-60,000.00 each prices for 150kva from sq D: https://stevenengineering.com/tech_support/PDFs/45175-14.pdf

Those $10 jeans in the 80's were still fairly decent quality and yes there were designer stuff that was even higher, I never had any of them though. Today you can still get jeans in the $10 range - but they are nearly a wear once and toss type of thing, unless you don't do physical activity at all while wearing them.