Sizing frequency converter transformer correctly

[powerpete69]

I haven't posted in a while, but I'm back.
We have a small extruder from Europe that needs a three phase 400V, 50 hz with a 4 wire plus ground feed.
At our plant, we have three phase 480v, 60 HZ, 3 wire plus ground available.
The schematics don't show the breaker size, but it does show a main disconnect switch rated at 63 amps with a 6mm^2 wires (a hair bigger than 10 guage)
Its an older unit, does not show main breaker size unfortunately.

The extruder has a few motors that add up to 8 KVA (normal motors, no VFD's)
There is another 4 KVA in heaters.

So I looked online at ATO frequency converters and Advanced power and controls. If you have experience with either manufacturer, please let it be know.

At any rate, if you look at a typical motor chart, it says to roughly double the KVA of your motor total to size your transformer. (I know, there are many variations on this)
If this is the case, I need 8KVA x 2 = 16KVA. Add in my heaters and I need a transformer at 21 KVA .....or a 30 KVA would prob be a good guess.


When I talk to the manufacturer, they say this:
Ok, just note that we size the converter based on motor inrush current at starting period, Not rated current/power, and the motor inrush current/power is approx 6-7 times its rated current/power, so recommended frequency converter model: HZ-50-3360, 60kVA power capacity, input: 3ph, Delta (L1,L2,L3 + G), 480V (+/- 10%), 50/60Hz and output: 3ph, Wye/Delta (L1,L2,L3 + N,G), adjustable (0-520V), adjustable (40-120Hz), 83.2A.

So they literally took my 8 KVA and multiplied it by 7 to get 56KVA plus 4 KVA in heaters.

My question is, why the discrepancy between the manufacturer and the motor chart? Are they trying to up sell or is it legit?
I (we) need to UNDERSTAND why !!?? Perhaps they are being ultra conservative to avoid problems? Prob not a bad idea.

 

[winnie]

In this case you are 'misusing' the frequency converter, in the sense that you are using a device which can supply variable frequency, but then using it to 'across the line start' some motors. So the frequency converter is going to have to be able to supply the starting current of the motors.

Motor starting current is 6-7x running current, and you need a frequency converter that can supply this transient. But this capability will only be used for a very small fraction of the time.

Frequency converters have some transient overload capability, so I guess that the manufacturer is being ultra conservative, but I can't really tell you by how much.

The transformers where the rule is 2x motor kVA have significantly greater overload capability than a solid state frequency converter. If the transformer is run at 3x rating for a couple of second there is no problem.

If you want to muck with modifications internal to the extruder, you could use a standard VFD to run the motors (probably 15 or 20 Hp) and a simple step down autotransformer to supply 400V 60Hz to heating elements.

Jon

 

[synchro]

If you want to muck with modifications internal to the extruder, you could use a standard VFD to run the motors (probably 15 or 20 Hp) and a simple step down autotransformer to supply 400V 60Hz to heating elements.

I agree that that should be the most efficient and lowest cost alternative if it can be done.
If there are other constraints like controls, etc. that need 50 Hz, perhaps soft starters could be used on the motors to minimize their starting current and allow a significantly smaller frequency converter than would otherwise be needed.

 

[jim dungar]

Do all of your motors start at the same time? As others mentioned it is the amount of starting current that impacts your sizing. And I also vote for keeping the heaters on a transformer and not a frequency convertor.

 

[powerpete69]

Of course its never simple! That's no fun!
There's (1) 7 KVA motor and the 5 other motors add up to the last 1 kva. So you might as well say that all run at the same time, but obviously the 7 KVA motor has the biggest "short circuit" effect.
One of the company reps mentioned the type of switching in the heaters may also pull extra power. Maybe he said SCR switches? Might be off about the "SCR" part.
I'll get more info this week. Thanks for the intelligent thoughts.
Can I take a "normal" motor and run it off a VFD?

 

[Besoeker3]

Can I take a "normal" motor and run it off a VFD?

Some "normal" motors are not designed to operate on VFDs. We had some that didn't. The problem was the winding type. - random vs mesh. The ones that had a problem with were random type. They failed after about a month. The manufacturer had to replace them mesh type. A very costly mistake considering they were 760kW (over 1,000 hp)!

 

[powerpete69]

Some "normal" motors are not designed to operate on VFDs. We had some that didn't. The problem was the winding type. - random vs mesh. The ones that had a problem with were random type. They failed after about a month. The manufacturer had to replace them mesh type. A very costly mistake considering they were 760kW (over 1,000 hp)!

That's what I'm used to from the consulting days. VFD motors and VFD feeder wires were always spec'd for the VFD's.....or visa versa.

 

[powerpete69]

Hey all,
I attached the wiring diagram.


In addition to what I said, the first unit is listed at 11 KW with all the things mentioned.

Also, there is a 30 amp plug that goes to another 10 KW machine that only has one small .7 KVA motor.



So we have 8 KVA of motors and 5 KVA of heaters on first unit listed at 11 KW.

The 2nd unit plugs into the first one and is listed at 10 KW and has a small motor of .7 KVA.


After reviewing the electric diagram, can you see if that 5.5 KW motor is on a VFD? Probably not. Someone also mentioned that heater switches. What do you see there if anything.

What size KVA frequency converter would you recommend??
Also getting advice from manufacturers. Will make intelligent decision after all opinions are stated.

You brilliant minds might just see something that I am missing. Wouldn't be the first time!

 

[powerpete69]

Ahhh....the missing link. I found the VFD drive for the 5.5 KW motor. Inertek drive. 14 amp adjustable output.
This makes sense. An extruder drive needs to have adjustable speed.
This should lower my KVA requirements signifigantly.

22 KW, including 7 KVA VFD motor....plus roughly 1 kva normal motors........45 KVA is prob the ticket.
Be back w more info when i get it.

 

[powerpete69]

Found two more VFD's for the couple remaining 1/3 hp motors.
Amazing how the answer usually lies within YOU if you give it a chance.
30 KVA is right on the limit per my opinion....I think 45 KVA is safe bet with no hassles.

 

[winnie]

More for you to chew on.

A VFD and a solid state frequency converter are at their core the same sort of device. Input AC is rectified to DC, then transistors are used to synthesize the desired output frequency. Where they differ is in the sort of output filtering that they use.

A VFD for a motor is using the motor itself as the low pass filter; the VFD supplies a very ugly high frequency modulated square wave to the motor, the motor's inductance filters this, and the net result is a reasonably smooth current flowing through the motor. When Beo mentioned that some motors don't work with VFDs in post 6, what he meant was that some motors either don't do a good job of acting like filters, or the motor insulation actually gets damaged by high voltage spikes caused by the ugly modulated square wave. Some types of small motors use capacitors and pretty much depend upon operating at a single fixed frequency.

A 'frequency converter' will often have the necessary filtering built in, to be able to supply clean sine waves to loads that need them. One problem is that the output filtering might not do well when supplying a load like the input rectifier of a VFD. You may need a 'higher rated' frequency converter to be able to deal with the harmonics that the VFD presents as a load.

So now that you have more specific data on the machine, I'd suggest going back to the frequency converter supplier to see if their recommendation changes.

You should also go to the equipment supplier. The VFDs built into the extruder should be just fine with a 60Hz input, and so should the resistance heaters and their SSR controllers. The only things in that machine that are not 'frequency agnostic' are the line connected fans and possibly the controller. A simple step down transformer supplying 60Hz 400V might be all you need.

-Jon

 

[powerpete69]

Thanks for additional info Winnie. I will contact equipment supplier as well.


I did give the manufacturer the updated info.
Their reply:
"In this case, 60kVA converter would be sufficient, as the converter is the most advanced solid state type "static", so using larger converter capacity model would be more reliable, and longer life span."

This could certainly be a sufficient route.
This converter rates its harmonic distortion at: Pure sin Wave ≤2%

 

[drcampbell]

Ahhh....the missing link. I found the VFD drive for the 5.5 KW motor. Inertek drive. 14 amp adjustable output. ...

Searching for "ODP-34055", I found this, which can be run directly on 480v 60Hz:

ODP-2-34055-3KF42-MN: Optidrive P2 Variable Frequency Drive Datasheet | Invertek Drives

5.5 kW, 14 A, 380-480 V, 3PH IP20 Variable Frequency Drive with EMC Filter and TFT Display

www.invertekdrives.com

My recommendation is to evaluate the system and each component, and see if frequency conversion is even necessary.
Likewise voltage conversion. The heaters seem to be thermostatically controlled, so if the thermostats can handle 480v, there's no need to convert either frequency or voltage for the heaters.
My guess is that it is not necessary, or can be made unnecessary with only a few minor changes.

 

[powerpete69]

Searching for "ODP-34055", I found this, which can be run directly on 480v 60Hz:

ODP-2-34055-3KF42-MN: Optidrive P2 Variable Frequency Drive Datasheet | Invertek Drives

5.5 kW, 14 A, 380-480 V, 3PH IP20 Variable Frequency Drive with EMC Filter and TFT Display

www.invertekdrives.com

My recommendation is to evaluate the system and each component, and see if frequency conversion is even necessary.
Likewise voltage conversion. The heaters seem to be thermostatically controlled, so if the thermostats can handle 480v, there's no need to convert either frequency or voltage for the heaters.
My guess is that it is not necessary, or can be made unnecessary with only a few minor changes.

Yes, I did notice that the VFD takes 380 to 480 volts and 50 or 60 hz.
My obvious concern is something might fry if not done right.
And who knows what the controls and switching might do at the "wrong" frequency.

One other concern is that the electrical drawing clearly doesn't say use 400 or 480 V and use 50 or 60 hz. It states 400V at 50 hz.
I will need at least a transformer. We have 3 wire with no neutral. They need 4 wire with neutral. So changing to 400V at least.

But all good points. I got to keep researching.
I will attempt to contact the manufacturer today. European manufacturer....good luck to me.

Special shout out to Winne or Jon above. Outstanding explanation on how motors can act as filters. Bravo....very well put.

 

[drcampbell]

... The electrical drawing clearly doesn't say use 400 or 480 V and use 50 or 60 hz. It states 400V at 50 hz. ...

That did not escape my attention. "Evaluate the system and each component" should not be interpreted as, "Casually hook it up to 480v/60Hz and see if it melts down". I'm thinking of completely reverse-engineering, then re-engineering, the machine (on paper), then deciding what the best course of action is.

A big-ass frequency converter like that is what, $15k? Plus maybe another $5k for the 480:400v transformers? Seems like a few hours of serious engineering work would be a worthwhile investment.

Ever plug a 120V vacuum cleaner into 240V by accident? It starts on fire immediately!

No, but having seen this factlet, I will never be able to unsee it. Next vacuum cleaner I encounter that's beyond repair is going to get a Viking funeral. (on a broad, non-flammable driveway with a LONG extension cord)

 

[powerpete69]

Yep, about $20,000 for complete converter and transformer at 60 KVA.
Here's the thing. What's the lead time on a 480V 3 wire to a 400V 4 wire transformer at 45 KVA or so?
If it's a year like I'm thinking, might as well just get the frequency converter in 6 to 12 days.
No factory wants a machine a year from now....that simply want it now.
But, lets see what Eaton and Schneider have to say on lead times.

 

[synchro]

... Special shout out to Winne or Jon above. Outstanding explanation on how motors can act as filters. Bravo....very well put.

I agree. The output switching transistors of the inverter that's within a VFD or frequency converter needs to feed into an inductance, whether that is coming from a motor or an inductive input filter. In contrast, a capacitor would be a low impedance at switching frequencies which would cause high peak currents that could damage the inverter output transistors.

 

[powerpete69]

Suprisingly, 3 to 6 weeks delivery on a standard 3 wire 480V to 4 wire 400 volt transformer. Looks like about $3,000 compared to $20,000.
All the VFD's and motors say 50 or 60 hz input.
Could I hurt anything by putting 60 hz across the controls or heaters with the correct voltage?
I does say the control wiring can be 50 or 60 hz on the cover page. Does not mention the main power of course.

Just found the email of the manufacturer......I shall email them.

 

[winnie]

Clearly lots to plan out. $20K for a converter might be cheap compared to dealing with warranty issues for a machine that is needed up and running _now_.

On the transformer lead time: there might be a long lead for a 480V delta to 400V wye transformer, but 480V to 240V single phase transformers are off the shelf items, and a bank of three of these gives you 480V delta to 416/240V wye. The 2.5% tap or the 5% tap will get you pretty close to 400V

-Jon

 

[powerpete69]

My next logical move is to wait for the manufacturer of the extruder machine to reply with their input about the 50 hz vs 60 hz.
It's in Thailand so response wont be quick due to the time change alone.

The machine is from 2012. It was "free" to us and their is no warranty.