MV transformer in equipment

[patm72]

Hi, I'm looking for some guidance on how to integrate a MV/LV transformer in a machine. Looking for MV machines on the web, well, only MV motors pop up. If you search for MV equipment with embedded MV transformer, you invariably hit switchgear apparatus. What I have is a packaged machine comprising a PMSM with the drive and controls that normally operate at 480V, to be now offered as a 4160V "package" by simply introducing a MV/LV transformer in an adjacent enclosure.

I want to ensure the equipment can be made safe to use. Maintenance is another thing but appropriate measures can be taken to service it safely, so it is more "straightforward".

To me, it is the sheer factor of having a MV voltage transformer next to the controls that makes me wondering if there are specific considerations and measures to be taken. I don't know that such an equipment would not allow for operators to come close to it (and this will happen if it is designed that way). Think about three consecutive enclosures, each one bolted to the next:
1- Enclosure (WHD) of 48"x72"x24" for control components
2- Enclosure of 80"x96"x48" for VFD's and power components
3- Enclosure of 80"x96"x48" for transformer 1200kVA, dry type, phase-shifting 12-pulse, 4160V to 480V

I attached an image of the enclosure train for your reference.

As always, thank you for your comments.
Patrick

 

[waross]

I don't think that you will find what you want "Off the shelf".
Contact transformer manufacturers for a custom design.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[patm72]

Waross, done, already have quotes for core and coil transformers to put into the enclosure. But they are not designing my machine. We asked about cooling and connections, but as far as the safety around our enclosure goes and how this is operated (installed even), they do not care. That's what I'm missing. What are the "rules". And I read the 450 section of the code but none of that seems intended at transformers ON equipment.

Regards.

 

[waross]

I misunderstood your question.
I will step aside and hope some of the gurus with panel building experience step up.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

It’s a really bad plan. I know you are thinking this is “just” a transformer, but it needs a primary disconnect and fuses at a bare minimum. MV transformers need a LOT of room to maintain safety and the switching devices are larger, need mechanical interlocking to keep anyone from being able to open any MV comportment when energized, etc. etc. etc. LV is only dangerous if you touch it, MV will come looking for you if you are just in the vicinity! You also cannot mix the wiring in any way, so totally separate compartments. The reason you don’t see any standards for what you are planning is because it isn’t done. Keep the MV totally separate from the LV equipment.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[waross]

Some years ago I had the opportunity to tour a SAGD pad. (Steam Assisted Gravity Drainage. A method of recovering oil sands tar.)
The heavy oil or tar is pumped by submersible pumps controlled by VFDs.
The pump is typically hundreds of feet down the well.
This gives rise to two issues.
Voltage drop due to the long lead lengths.
High reflected harmonic voltages due to the long lead lengths.
One solution to both issues was to use a 480:4160 Volt transformer on the output of the drive.
This reduces the current by a factor of 480/4160 = 0.115
Further, the voltage drop in Volts is now a percentage of 4169 Volts rather than 480 Volts with an advantage of over 8.6 times.
The transformer also acts as an effective harmonic filter.
The drive provides isolation and protection.
With bolted covers on the transformer, some jurisdictions will accept suitable warning labels in place of interlocks.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

In your scheme, it appears that the desire is to accomplish the 12 pulse power source and voltage change in one swell foop. While it CAN be done, you end up with a totally custom and long lead time transformer if anything goes wrong and/or whenever you want to sell to someone with a different MV voltage distribution system, such as 2300V here or 3800V overseas, it means a new design of that transformer and probable delays in attaining one.

Were it me, I would use a relatively standard/common 12 pulse LV auto transformer for VFDs and if the end user wants an MV feed, let them or your sales team source a standard off the shelf MV transformer and requisite switchgear for it. I find that typically people who use MV will like to have the same gear when they can, it keeps their Safety and maintenance procedures consistent.

Also: 12 pulse? That is used a lot overseas, but most large drive users in North America demand 18 pulse in order to help meet IEEE 519 requirements. You might also consider the newer Active Front End Low Harmonic drive options out there, several of them are capable of driving PMSM motors now. Just a suggestion.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[patm72]

Thanks Jraef for your feedback.

I have the same concern for MV close to operable equipment. That doesn't mean that it is not achievable though, as pointed out by Bill. To be investigated further.

As for your suggestions, let me give you some background of where I'm coming from. The PMSM is supplied by a European company, and so is the VFD (proprietary), for a niche equipment tuned to work together. So you see that at this size (1MW), we already have our toes into the world of potentially long lead times should something go wrong, a choice that was knowingly made by our management. For the transformers, the manufacturers I approached have lead times of 8-9 weeks, which is acceptable and there is room for discussion since we are planning for a certain volume of these orders, so having one "stock" could be a possibility.

Your suggestion for 12-pulse LV autotransformer is interesting as this is exactly what came with the first system. The customer was able to provide for 480V to the unit, so this was more or less plug an play and needed no MV transformer. It is only to serve future customers better, which have already some MV equipment that we could aim to replace (a quite common situation), that we came up with the solution of an included MV transformer in the unit. This way, the existing power feed could presumably be reused without the hassle of having to install MV-to-LV-switchgear-somewhere-that-would-steal-so-much-of-the-precious-footprint (footprint is extremely important to salesmen and managers as you know). But since we were at it, while stepping down the voltage, why not mitigate harmonics at the same time? The result is that we can spare (remove) the two autotransformers in the VFD enclosure that were doing the phase-shifting for the four drives, and their corresponding weight. We drop in the process from 24-pulse reduction to 12-pulse, but I came to learn that the gain was marginal, as we would still be well within the IEEE-519 limits.

For the switch gear needed on the transformer (or the full unit now that the transformer would be within its adjacent enclosure), it would be left for the customer to take care of on the primary side. The secondary side (LV), we take care of within the VFD panel, equipped with two electrically interlocked circuit breakers.

Safety is always the first concern that I have and being aware of arc flash, I am really careful to no go ahead too quickly on this. This is why I'm seeking some advice on it. If you would know of any resources I could reach out to in order to validate this design, I'm listening.

Patrick

 

[waross]

An old rule of thumb is that when the motor HP exceeds the voltage it is time to seriously consider using a higher voltage.
Another conservative rule of thumb is that 1 HP = 1 KVA.
500 HP, or 500 KVA on 480, you can bend the rules of thumb that much.
At 1 MW, there is justification to suggest building a 4160 Volt version of the machine and use a small internal transformer for controls and any small motors.

Bill
--------------------
"Why not the best?"
Jimmy Carter