480V Transformer in same room as IT Rack

[tortuga]

Yeah, but then you have boards that go too far in enforcement and need to be slapped down by the courts: https://ij.org/press-release/oregon-engineer-wins-traffic-light-timing-lawsuit/
I do have a problem with the titles "architect" and "engineer" being applied in a slapdash manner just to make the job sound more important (the late George Carlin had a great sketch about this), but I would argue that "architect" is a very good description, and "engineer" an acceptable one, of what a lot of networking teams do. Just because they're designing data links instead of buildings doesn't diminish the amount of learning and thought necessary.

If I read it right, the difference between the Oregon Traffic light guy Mats and the Paul Tapple case in Washington, is Mats never said he was an 'engineer' in his job title, he was just challenging a math formula in court, and the engineering board was trying to gag that, saying nobody can do math in court unless your a PE. LOL it was a gross overreach of the state!
The case in WA was Paul Tappele, a professional Civil Engineer, and his firm Fisheries Engineers, Inc. sued the State of WA because it was allowing state employees to use the title “Engineer” even though they were not licensed or registered as engineers, and some did not even have an engineering degree at all.

It does not unless you completely ignore the context and language construction. The person controlling a railroad locomotive is called an "engineer" but is not a PE (although they may have an operator's permit for the machine), nor are many "chemical engineers" licensed PE's, either.

That depends on the laws in your state, in Washington State where the Paul Tapple case was
Revised Code Of Washington (RCW) Chapter 18.43 defines “engineer” to mean a “professional engineer” and provides in part that “In order to safeguard life, health, and property, and to promote the public welfare…it shall be unlawful for any person…to use in connection with his or her name or otherwise assume, use, or advertise any title or description tending to convey the impression that he or she is a professional engineer or land surveyor, unless such person has been duly registered under the provisions of this act.”
That ruling in WA had effects on a local power company here, the power company would put "Distribution Engineer" on the business cards of people then send out to look at new services.
However the job did not even require an engineering degree, so the cards got changed and anyone whom did not have a engineering degree got the title 'engineer' removed from their stationary and is now a 'distribution technician'

 

[zbang]

I'm not sure how that squares with all the software engineers, test engineers, sound engineers, broadcast engineers, train engineers, etc etc etc out there in Washington state; it would be a very heavy lift to assume a "software engineer" is a licensed PE even though they are a professional dealing with software. Even Washington State Univ grants a "Bachelor of Science in Software Engineering" it's on their web site.

Again, context is important.

 

[Besoeker3]

BTW, in the UK the word engineer often equates to the US word machinist, and our locomotive engineer is the UK's driver.

Yes, some people do here. But I, and others, are professional engineers. In my case an Electrical Engineer.

 

[jim dungar]

convey the impression that he or she is a professional engineer

This is the key part of the ruling.
Only licensed Professional Engineers can represent themselves to the public as professional Engineers. Internal to a customer's business anyone can have a title of engineer. But for many companies it was cheaper to change titles that run the risk of a lawsuit.

If there is no licensing available then the public image is a little greyer.

 

[ModbusMan]

If I read it right, the difference between the Oregon Traffic light guy Mats and the Paul Tapple case in Washington, is Mats never said he was an 'engineer' in his job title, he was just challenging a math formula in court, and the engineering board was trying to gag that, saying nobody can do math in court unless your a PE. LOL it was a gross overreach of the state!

The problem with Mats (the traffic light guy) was that he does in fact hold an EE degree and works in technical positions. But he doesn't hold a PE license, and the Oregon bureau wasn't going to stand for allowing some upstart tell them how to time their red light cameras.

 

[winnie]

Back to the original issue:

Transformers do create magnetic fields, that is how they work, after all.

Most of the flux is contained in the core, but sine certainly escapes, especially during startup inrush.

The simplest way to reduce the flux that escapes the core is to use more core. If you get a transformer rated for the same current but higher voltage (implying higher kVA) then the core will operate at lower flux density and there will be less flux leakage. There will also be less startup inrush. The transformer will be physically larger, more expensive and possibly less efficient, but everything is a trade.

A quick search shows that 'low emf' transformers are a thing, but I have no idea how much one pays for that 'feature'.

This is all low frequency stuff, and will not be significantly attenuated by an ordinary Faraday cage. The standard steel enclosure will provide most of the magnetic shielding.

While it is not likely that this leakage flux will impact properly designed IT equipment, you have a real nocebo problem here: IT equipment has problems all the time, and now you will have a foreign system to blame for the problems. It might pay to invest in a low emf transformer or get something specifically rated for IT environments simply to reduce this nocebo effect (nocebo is the evil twin of placebo, the patient experiences negative side effects from the sugar pill....)

 

[jim dungar]

You are pretty much in a no win situation. You cannot predict what EMF will be seen by the IT cabinet. The best you can do is perform field measurements on an existing transformer, with similar loading. But if the IT guy does not give you an EMF level which you must be under, as @winnie said, you will be blamed for everything that doesn't work.

 

[kwired]

He wants a Faraday cage? Most transformers come with one, it's the bonded/grounded metal enclosure.)

I was going to bring that up if nobody else did.

Even if you had an open unit, I doubt you get much for interference to the data signals from 60 hertz. Untwisted voice pairs you may be able to get some 60 cycle hum in the background while talking depending on how they are routed past anything that may couple with them.

 

[JuliaTruchsess]

The UPS and IT gear do have magnetics in their switching power conversion circuits, but I think it's extremely unlikely that the field from your 45kVA transformer 6 feet away will affect them in any way.

 

[ModbusMan]

(He wants a Faraday cage? Most transformers come with one, it's the bonded/grounded metal enclosure.)

Just had another thought about how ignorant this guy is... As long as they're putting their equipment into a standard IT rack, the rack itself is going to function as a second Faraday cage in addition to the one already surrounding the transformer itself, provided the frame has been grounded.

 

[jim dungar]

provided the frame has been grounded.

And that the enclosure door is kept closed and latched. IT installers are notorious for cutting corners, like it bringing cables through the open door rather than having them properly trained.

 

[winnie]

Remember: A Faraday cage is an electrically conductive cage that block electric fields and blocks radiated EMF. A Faraday cage will not block near field magnetic fields. If you put a magnet on the outside of a Faraday Cage you can deflect a compass inside of the Faraday cage.
(Exception: if your 'Faraday cage' is made of superconducting material, it will work down to DC and block magnetic fields... *grin* )

To block DC and low frequency magnetic fields, you need a magnetically conductive enclosure that shunts the magnetic flux, not some sort of insulator that blocks magnetic flux.

For what it's worth, I've worked with computers in the presence of pretty large AC magnetic fields. This was back in the 90's, and I was running tests on induction motors pushed heavily into saturation. The magnetic fields were strong enough that the images on the CRT (90's!) monitors were wobbling quite significantly, but the data acquisition computers (with their spinning magnetic disc drives (90's) ) had no problem functioning. Modern equipment will have different sensitivity, so this is a barely applicable anecdote.

So far the suggestion that the IT guy provide you with a field strength limit is the best yet. You _cannot_ eliminate all fields. But if you have a number you can go to the transformer manufacturer and find out what it takes to be below that number.

-Jonathan

 

[ModbusMan]

To block DC and low frequency magnetic fields, you need a magnetically conductive enclosure that shunts the magnetic flux, not some sort of insulator that blocks magnetic flux.

Just about any rack on the market is going to be steel, so while it won't block magnetic fields entirely, it will distort/deflect to some degree. The only exception to this I can think of offhand is the Knurr Miracel line of ~20 years ago... those were aluminum frames with steel panels rather than 100% steel.

 

[winnie]

Just about any rack on the market is going to be steel, so while it won't block magnetic fields entirely, it will distort/deflect to some degree. The only exception to this I can think of offhand is the Knurr Miracel line of ~20 years ago... those were aluminum frames with steel panels rather than 100% steel.

Agreed about the steel enclosures, especially the steel transformer enclosure.

I just wanted to make clear that when the IT guy was suggesting enclosing the transformer in a Faraday cage, he was smoking something... When you build a Faraday cage, you spend lots of money on fancy copper mesh, conductive tape, etc. All of which will do essentially nothing for a 60Hz magnetic field.

 

[sgtsparky]

It's not an NEC violation. His servers, UPS, etc will generate far more EMI than the transformer which is inside of a steel enclosure.

Not only is this ok but the ups create harmonics I maintain several 50kva units and they are electrically noisy the transformer won’t be creating any problems for electronics I put my computer on top of transformers all the time some much larger and have never had any issues with WiFi or the hard drive while entering my P.M. data.
In my state my license is way above what most data guys have their LEA license is severely limited as is their knowledge of power circuits.