Exploding Transformers?

Wow. I am posting here mainly to see what others, who know more about this than I do, will say. Especially if by "re-tapping" you mean simply moving a wire from one tap to another existing (but currently unused) tap, I can't imagine that there would be any problem whatsoever. The operating temperature could change slightly, but for a 10% change in voltage, it seems very improbable there'd be anything remotely close to a problem. In many settings, a similar temperature change might be experienced from day-to-day weather changes.

But I can't be your official myth buster, I'm not a transformer expert.

Ken, this has got me scratching my...head. Re-tapping is just as you said: moving a wire connection to change the number of windings in the flux circuit. I can't imagine that insulation breakdown of the windings could be accelerated to a catastrophe by adding in a few more windings. However, while I have never heard of a transformer doing this from a tap change, I live in the ComED electrical service area and in the bad old days, transformer vaults would catch on fire and explode in the summer loads. This had to do with oil-cooled/insulated transformers, not plain old air-cooled.

Grousing aside, our facility has pretty good power for a third-world country.

Dave Meador

I'm no transformer expert also but this is what I know.

A transformer provides power (in Watts). Watts is equal to Voltage times Current. If your transformer puts out 196 volts, it will give out a certain amount of current that would be equal to or less than the maximum allowed by the transformer rating. If you re-tap your transformer to a higher voltage, the maximum current that the transformer can provide will go down. In short, you can't get more power out of a transformer that what it is built for.

This being the case, what happens if you do re-tap to a higher voltage?

If the current is not at maximum, it will run fine. However, your margin has shrunk. If the power goes out, your UPS will take over, and when power comes back on, the UPS will draw more current than usual (to recharge the batteries). If that current exceeds the maximum capacity of the transformer, your voltage will drop. In addition to that, your transformer will heat up. Explode? It might, if the current is way, way high or if the high temperature continues for several hours until the insulation fails. If the UPS has an auto disconnect, the transformer may be saved but your precious data may be lost.

Analyzing a little further, from your description, I can't tell if your transformer has a rated output of 208V but is giving out 196V. Might your supply on the primary side be too low? Or your transformer might be 115V to 208V but your actual supply is 110V?

Another possibility is that your transformer may have some shorted windings resulting in the lower voltage. In this case, your transformer may run a little hotter than normal.

Or, are you saying that the transformer's rated output is 196V which is below the required 208V.

In any case, my recommendation is to replace the thing with what your system needs.

My first suggestion would be to check the primary voltage and determine that it is within the "normal" range,

Next, determine the secondary load, (KVA) and determine if the transformer KW rating is not being overloaded. This may happen over time with increased loads being brought on line. If this is the case install a transformer with greater capacity.

Transformers can explode, this can happen when a short circuit occurs in either in the primary or secondary windings. Overload or over current protection may not prevent the explosion from happening.

Did ya ever think of modifying your UPS to be happy at 196? It's unlikely that the ole transformer there is supplying only your UPS/data center. There may be gobs of LIFE preserving equipement on that transformer, along with your really possibly irrelevant archieves, which you might just have to store redundantly, as the hospital's money maker/funding/profit is life saving, your precious (to you) data may be totally not a concern to the hospital operation. Tell your radiologists, that they need to add a constant voltage supply system to protect your ups. And maybe the sensible thing to do is to merely add a bucking transformer to your VOLTAGE SENSOR LINE to bring your 196 to 208. Or would that make too much sense? Or maybe be something you could do rather than shutting down a hospital for several hours while the primaries are shut down, old corroded bolte removed, hopefully, and new ones installed on another set of taps. Since it appears that your data is secure at the 196, I think you just have to buck up, buckaroo, and make your equipment work on what's available. I might just side with those stupid electricians on this one, even if they are trying to be nice and tell you fibs to get you off of their backs, when the world and the answer are there in your hands. Try toplan the work with the maintenance wupervisor if you can't do it yourself.

RichH

I'm a journeyman electrician. One thing you need to look at is how far away your sub panel is from the transformer. The transformer may be tapped correct and have the appropriate voltage at the transformer. You may have voltage drop problems, if your circuits are fed from a sub-panel that is considerable distance from the electrical room. Voltage drop is a factor of load in amps, size of and conductivity of the wire used, distance and whether the load is single phase or three phase. If the transformer is feeding more than one sub-panel, changing the taps to correct voltage at a distant sub-panel could result in higher voltages at closer sub-panels. If the transformer feeds no other sub-panels changing the taps should not be a problem as long as the transformer is operating with in its KVA rating in the ambient temperature stated on the label which is usually 40 degrees celcius. If the transformer is maxed out and they are truely concerned about burning it up, you are right, it should be changed out. Check your voltage at the sub-panel which provides the branch circuit for your equipment if it is low the problem is in the feeder to the sub panel, if it is ok the problem could be the wire is to small on the branch circuit to your equipment.

Voltage drop problems are fixed by installing larger wire. This is not rocket science, any competent electrician should be able to diagnose a voltage drop problem and calculate the appropriate size of wire needed to either avoid the problem in the first place or to fix it.

Another option you have if the voltage drop problem is in the feeder to the sub-panel, is using buck/boost transformers on the branch circuit to your equipment. Mounted close to the load utilizing power these small transformers can increase or decrease your voltage usually 12 or 24 volts depending on how they are wired. Talk to your electricians about this. If you get dumbfounded looks when inquiring about these issues, and are offerd no solutions perhaps you need to hire another company to do your electrical work.

As a field service engineer in the medical equipment industry for over twenty years, I have seen what Dave is experiencing many times. He's working with plant facilities people in a VA (government) institution. Most of these folks aren't known for wanting to get things done, and if you find one that will help you, the government's procurement process will kill any chance of you getting what you need.

The archived data on a PACS system is required to be kept there by Federal Law under HIPPA. Since the modern world has done away with film in x-ray departments, the medical imaging industry has to maintain all that imaging data electronically.

The best chance he may have is getting them to help him find a new feeder with more capacity already on it. The radiology manager may have more political clout to get him some help.

With the possible exceptions of a few fans, a few monitors, and probably Air Conditioning units, all of the rack units are powered by self adjusting transformers/rectifiers. The 5 % low supply voltage probably meets their requirements/design. Large fans may be at or under their design. The AC unit(s) almost certainly are at the max of their design limits. If the only problem is burning up batteries, I think your cut-over switch sensors could be the only thing that needs bucking transformers. I'm sure that you have 3 leg sensors, and a several minute minimum cycle time to prevent chatter. Your split AC system compressors are probably on the roof. I'd also guess that you have a redundant AC system, since that is probably the most critical element of the works. System shutdown should only lose info being transfered in in the event of complete failure. Error-correction raid storage means that your data is at least secure. You probably have off-site continuous Ø storage, and additional daily, or weekly back-ups to protected storage.

The powersupply for the Dell laptop that I am using now is rated 100-240 VAC. Your 196 V should equate to 113 V/leg, which should fall into the self correcting range of all of your 3.5, 5, and 12 V DC power supplies. If the racks are ventilated, those vent fans are probably line voltage. The fan in the evaporator unit in your data center is certainly line voltage, maybe even 208 V, either 1 Ø or 3 Ø. This is probably less than a 1 hp fan, but even a 3 hp 3 Ø will be less than a 15 A draw and is a fine candidate for bucking if it needed it. I doubt it. The rack fans, if individual, are almost certainly 120 V, and 113 V should be satisfactory for their operation. If they use a system ventilation, that could potentially be a 3 Ø fan, and checking draw would tell you if you are within operating range. If not, most 3 Ø fans are belt driven, and if not equiped with adjustable sheaves, should be, and can be brought into range easily, unless your air flow is already minimum. Rarely, at full rated draw. A 3 Ø motor will not run underspeed if it runs at all. Overloading will occur at as little as 2% phase slippage, and vibration will make you aware of the difficulty pretty quick.

That mouthful said, I really doubt that anything other than your line voltage sensing circuits are suffering, and there is a strong possibility that you are crying wolf for a mouse. Try bucking your line voltage sensors only to 208 V, and that should solve the only problem you have described, which is frying your UPS batteries from over-sensitivity at too near their set point. That, of course would be based on non-adjustable set points, which is unlikely.

By the way, is your battery room exhausted to prevent Hydrogen accumulation?

RichH

I have been having a think about your problem and agree with NoSciFi that 196 V phase to phase is not that low for a 208 V system. This is equivalent to 113 V instead of 120 V you would normally expect. While this is over the usual specification of ±5% in most countries it's not an unexpected voltage drop.

I was doing some investigation and I may no what is going on. I noticed that the APC Matrix 5000 UPS can be used on 120/208/240 V systems and I suspect that the unit you have is set to operate on a 240 V system. If that is the case the UPS is seeing what it thinks is a voltage drop of 18% and so it switches to the batteries.

I am not familiar with the APC 5000 but a system that is designed to operate over such a wide range of voltages would normally have some sort of setting to allow for the variation. I have a suspicion that your system is actually set to operate at 240 V and this is giving you the problem. If this is the case it should be a relatively simple task to set the system correctly.

Tap changer is regular feature incorporated in the Transformer. This (tap changing ) is required to counteract a voltage drop in the primary or secondary. The drop in the Primary voltage is the result of a fall in the voltage due to overloading of the total system connected to the feeder, while the Output voltage drops due to an overloading in the supply side of the Tansformer.

In both the cases it is corrected by a tap changing. On the first place the tap is to be changed for only a 5% step up correction. There would not be any problem in that. The transformer would not burst for that.

Care should be taken in changing the tap. if it is fitted with an on-line tap changer, the the job can be accomplished immediately, but if it is an off-line tap changer, you would have to shutdown the transformer in the primary side before tap changing. Otherwise it might expolde.

Also take note of your load pattern. If the voltage drop is only in the peak hours of your establishment, please note the swing and the peak voltage when all the loads are turned off. If the voltage without load is correct (230V), tap changing might cause trouble with your equipments (not with a burst transformer of course) with a high voltage on off peak hours and can cause burnt equiments. In such a case of overloading the tarnsformer needs to be replaced with one of larger rating.

In any case a 5% drop in Voltage should not stop the UPS from operating. Please note the name plate on the UPS and see if it is made to work with 230 VAC.

In any case Changing the tap would never cause a transformer to burst, provided care is taken about the online and offline thing.

If his data center is in a strictly administrative or clinical environment, wherein an off-hours tap change could be made, a request for either a contractor or a request for overtime for the electricians should be made through the appropriate channels (usually at least up the organizational chart to the point of commonality, who may still have to carry the request higher). I have worked out there, where services at off-hours were required along with ComEd or your own physical services (I think), and it took months to get authorization.

Fire alarms, security systems, local networks, routers, access points, local servers, maybe building clock systems, phone systems, lots of invisible infrastructure would need to be co-ordinated, and manned because all three taps would have to be interrupted briefly, and each tap changed. The job you need done, even in a clinical or administrative setting of a medical complex, is not as minor as you seem to think. All 3Ø equipment in the building would have to be shut down and re-started to prevent single phasing and very rapid burnout, if not protected by 3 leg sensors, which they probably aren't. Some may not even have 3 leg heaters in the starters.

That's in an administrative setting. A gospital would possibly have enough emergency generation to start up, switch over to, and then be able to work the taps with at least a no load situation. That is much safer than a hot tap swap under load and with arcs as much as 6 " long on removing a loaded hot tap, moving to the next. and then having the tap weld itself before correct attachment, bolting, compressing, or what ever type of transformer you have.

Some parts of Hines are still working off of very old transformers, and very old buried cable.

All of that when the easiest solution is either some dashpot adjustments or a 3 phase or 3 single phase auto sensing bucking transformers that you could probably install hot, in off-hours, with minimal interruption to anything else. Your only concern seems to have been the fried batteries, which means that everything else is handling the low voltage pretty well. And if your batteries are charged, it all could be done without interruption's to any one else. Even if you had to install only one line's per day because of a low capacity of your UPS. The solution is in your hands in most, if not all cases. Additionally, single phase equipment and transformers are usually rated 115 - 125 V anyhow, and your 194 would be 113 per leg, well within a 2% tolerance that could be expected in any environment. Any motor tags for 208V motors that I recall are rated 200 - 240 V, which also results in a 3% below design, which should be within operating tolerance.

In other words, quit blaming the poor tradesmen, and DIY, or at least IMBY.

RichH

To all--

Thank you for all your suggestions. Guest has it pretty much on target about the availability of electrical services here at the Hines VA. Some intelligent beancounter a while back decided that there was money to be saved by laying off the majority of our electricians (in addition to a lot of other trades). The ones left have been increasingly flooded by one crisis after another caused by the lack of electricians. I am in the same boat. The amount of equipment that I cover would be better served by 3 of me.

From what has been talked about in this thread, I can see that the UPS problem (it has caught on fire and we just narrowly missed having the sprinkler system put out our PACS archive--Think multi-million dollar problem here) is just the tip of the iceberg. The other 208 systems in radiology have suffered too. FYI--voltage range and stability is a big deal with x-ray systems.

The 208 system here is department wide and some serious work is going to be needed. (Don't worry about things being shut down--happens all the time--we just shovel out the damage and fix what's left.) As Clint Eastwood's characters have said, "A man's got to know his limitations." I have decided that I am going to shoot for hiring a consultant to survey the department and specify a program for upgrading the electrical system. The performance of the electrical system is crucial to the reliability and performance of all of this very expensive medical technology. The electrical system performance directly impacts how many 100s of thousands of dollars we spend in parts and services and patient impact. My job will be to lay out the business and clinical case for getting the work done. Again, thanks! You told me what I need to know.

The importance of getting the power and installation correct can't be emphasized enough. I remember having a running battle with bean counters and admin staff over the quality of the power, air conditioning, line filtering and protection for a critical communications computer I was installing. I won the argument in the end with the whole installation being the way I specified and the system ran for nearly 4 years without being re-booted. The only reason it was rebooted then was that the system designers had never envisaged a system running that long without interruption and the counter that kept track of the time the system had been up over ran. The next tine the system was shut down was when it was decommissioned. All this was in the tropics in an area that gets 1,300 lightning strikes per square kilometer and is prone to tropical cyclones. Get it right and you will never look back.

It the age old adage, you only get what you pay for.

Masu--

You are my long lost brother!

Dave Meador

Any transformer, single phase or three phase can have the taps changed, irregardless of the age. The design of the transformer is just for that reason, to be able to make the connection to the primary source voltage and have the secondary voltage of something close to what you are trying to achieve. Although 196 volts does not sound too low in terms of normal equipment (electric motors) attached to the load, the UPS is a different "beast" and is designed to sense that 196 volts is to low as opposed to 208 volts.

If the APC UPS is configurable, then you should be able to adjust the sensitivity. If not and you cannot get any cooperation from the "on site" electricians in terms of changing the taps on the transformer, have them install a "boost" transformer to correct the low voltage condition. They will also want to add primary and secondary fusing to the boost transformer to protect it from any ground faults, short circuits or overload conditions. By the time that they go through the design stage, the installation stage and the inspection stage, they should be able to figure it out that it would be much simpler to change the taps on the transformer.

Sometimes you can't see the Forest because of the trees.............

Good luck with this project!

Jeff

JLD--

Hmmm. The old information overload trick! This could work. Rather than try to sort out the options, they will try something simple. Diabolical!!

Dave Meador

Davemeador writes: "The question concerns the "reason" that the transformer couldn't be re-tapped. I was told that the transformers in this place have been in place for so many years that they have taken a "set" to the incoming and outgoing voltage and that if it was retapped, the difference in the current flow and voltage stresses could cause the transformer to blow up. It was too dangerous to retap!"

-----

Concerning transformers "taking a set" in this manner is a crock. Maybe the transformer's construction materials have become brittle over time - special paper or fibre sheet can sometimes be found in transformers to separate the windings, and this becomes brittle - and this might be the reason for their reluctance (a pun, btw) to change the tap.

I'd say offhand your utility guys have themselves "taken a set." Your transformer should be replaced outright - and so should your utility guys, seeing as they're feeding you a line.

By the way, is this your transformer?

-e

Thats me! welding!

You da man!

Btw, where is your equipment?

-e

Disapearing in that cloud of smoke.

Ahh. There you are! Almost missed it:

-e

Btw, PlbMac, please feel free to use this pic for your avatar. That ghostly image you have now is cool, but it doesn't much look like a TIG man.

-e

I haven't had dodgy facial hair like that since the eighties!

Fake it. Nobody here will know the difference.

Okay, just to be fair, I'll post a picture of me:

This is what can happen when you fall into the habit of pointing at UFOs.

-e

EU--

I was thinking of declaring a puntest but this thread might explode. The vid is great, but we won't show it to the electricians.

Dave Meador

"I was thinking of declaring a puntest but this thread might explode. The vid is great, but we won't show it to the electricians."

-----

Oh, pleeeeze, let's! Stuff like this is always great for a good laugh cuz it makes 'em run around in their shorts.

-e

EU--

Sadomasochist!

DM

You mean Sad 'ol Masochist. I hang around here don't I?

-e