480 volt motor quandry 2

[NIDinc]

We are build our own custom 240 vac 3 phase inductive motor in our shop from scratch. This is a 2 hp small motor that is only about 3 1/2" diameter by 8" long. Last week we built a 480 volt version of the same exact motor, and it has drawn very amps.

We have checked and re-checked our windings and connections and are confident that it is correct for 480v. About the only thing I can think of is that we didn't put phase seperators in the windings.

Should be drawing about 1.5 amps and it was at about 8.
Any thoughts or input will be much appreciated.

 

[zlatkodo]

If you expect a specific answer you should give more information: all the winding data and core dimensions.
Also pictures of your motor are welcome.

Zlatkodo

 

[OperaHouse]

Add to that current in each phase.

 

[NIDinc]

You're right. That was pretty sketchy info. Here's more details. I don't see a way to attach photos or I would include some...

This is a two pole motor. The core has twelve slots. There are six coils. For the 240v motor we are using 50 turns of double strands of 22awg spike resistant, inverter duty wire. (these motors are run on a VFD at various rpms). For the 480v motor we made our coil windings 100 turns with one strand of the same wire. The 240v motor has a full load rating of 6.4 amps and draws about 2.6 amps no load. The 480v motor was up around 8 amp running on a vfd. When we hooked directly to our 60 hz 480 volt shop power, the motor fried within a few seconds.

The stator stack length is about 2.150" The I.D. of the stator is 1.612" The air gap is .012- .015" The bare slot area is .135 sq. inch.

The squirrel cage rotor has fourteen rows of double bars,

Let me know if you need more info. THANKS!

 

[electricpete]

The core has twelve slots. There are six coils.

So, it's a single-layer winding, then I guess?

don't see a way to attach photos or I would include some...

Look at the bottom of the edit box where it says "Step 3 Attachment". On the right side click "upload your file to Engineering.com". When all done, don't forget to hit the "insert-link" button (uploading it does not attach it if you don't hit insert link button)



=====================================
(2B)+(2B)' ?

 

[electricpete]

fwiw, I'd think it'd work based on comparison to the 240volt motor (assuming all winding connections are kept the same as the 240 volt motor.... just doubled turns per coil and went from 2-in-hand to 1-in-hand).

The volts per turn stays same so flux density roughly the same as the first motor.

With twice the turns, the magnetizing reactance should be nearly 4 times as much as the first motor, which would give roughly half the no-load current (~magnetizing current) with 480vac applied as the first motor with 240vac applied.

If 240vac worked fine, it's hard to see where the problem is on the 480vac, unless there was a miswiring or a failure.

I haven't done any detailed calcs on whether the original 240vac design was reasonable. Maybe someone else can chime in.

=====================================
(2B)+(2B)' ?

 

[NIDinc]

Here are some photos. We're pretty confident in the 240 volt design. Had it professionally designed by a electrical manufacturing consultant and it has proven itself in the field, in a demanding 24/7 indsutrial application.

 

[NIDinc]

Here's a photo of the rotor components...

 

[NIDinc]

Electric Pete wrote:
"If 240vac worked fine, it's hard to see where the problem is on the 480vac, unless there was a miswiring or a failure."

I agree and this is what has us baffled. We tested two 480 motors and both performed the same. As mentioned, we fried already. The remaining one was run at 208v and worked fine. Only drew about .6 amp. Tomorrow we will try to dig out the connections and do a high pot test between phases at 1000vdc to see if thats where the failure is. As I stated earlier, we didn't install phase seperators/insulation on these prototypes but the inverter duyt wire and our vacume impregnation of the resin before baking should have been sufficient.

Also, note that we ran this 480v motor on a VFD, varying the Hz from 20-120 and it consistently drew about 8 amps before kicking out the overloads. All things point to a wiring error but have not found anything yet...

 

[waross]

Try bringing out the ends of both of the parallel windings from a 240 Volt motor. You should be able to reconnect these in series for 480 Volt operation. I suspect a wiring error.

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

 

[zlatkodo]

Hi, NIDinc,
Need : tooth width, back-iron ( the smallest measured value), span, wye or delta.
Whether the internal connection and wye (or delta) are the same in both cases?
Zlatkodo

 

[ScottyUK]

Any possibility of one winding being wound 'backward' or the winding ends being swapped over?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[petronila]

Hello NIDinc

Try first connecting the motor with out a VFD, and test with no load at 60 Hz.Is the no load amps OK?

 

[waross]

You beat me too it, Scotty. I just logged on to suggest that.
If a winding is reversed, the applied voltage angles will be 120 Deg. The back EMF will be 60 deg, 60 deg and 240 deg. Bound to get hot.

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

 

[zlatkodo]

petronila (Electrical)
Try first connecting the motor with out a VFD, and test with no load at 60 Hz.Is the no load amps OK?

No load current must be to high in both motors at 60 Hz. With such windings they can not operate at 60 Hz at all.
Zlatkodo

 

[electricpete]

Looking at the stator picture posted, there are only 3 leads brought out (correct?). That would seem to rule out external miswiring when connecting to the power supply.

Since the exact same motor performed well at 208vac but not at 480vac (19 Feb 12 18:38), that would seem to rule out internal wiring error of the phase-polarity-swap type. (Although I'm not suggesting to rule out anything.)

Focusing on that single behavior reported 19 Feb 12 18:38, we'd be inclined to think the core is pushed far into saturation at high voltage, but not at the lower voltage. But if wired as stated, the original 220vac motor should have had the same problem.


=====================================
(2B)+(2B)' ?

 

[iop995]

It's 99% core saturation at 480V. Increase slowly voltage and chech current and will see a high current increase at some voltage point. If such behavior, saturation is confirmed. To keep same core, increase winding turns and reduce rotor bars section area or/and keep a single bars winding.

 

[NIDinc]

Well we figured out the problem and it's a little embarassing...
Now that we see it, it seems obvious:

Our motor is a high speed motor designed for a top Hz rating of 250. When we run it at 480v and 250 Hz. the amp draw is fine -about 1.6 (what we expected).
When we run it at 120 Hz, as we did, when we got 8 amps, or 60Hz when we fried it, all makes sense now.

We were able to run our 240v motor at the 120Hz needed for our application without any trouble and so assumed wrongly that we could also run the 480v motor at the same Hz. Big mistake...

So, we switched to a 240v motor and are running it on the 480v VFD with parameters to make it have the needed voltage (about 110) at the 120Hz. Sounds unorthodox but it works fine and allows the use of the 480v line voltage without a transformer.

Thanks again for all help and input...

 

[petronila]

Hello NIDinc

I will add one comment, due to your motor application involves a VFD, your motor windings needs a proper design to support the VFD action, so I will suggest you the use of proper materials like Inverter Duty or Spike duty magnet wire and extra insulation( insulating each of 12 coil) and aplying two varnish cycles or resin to the winding.

This actions will contribute to extend the winding life.

Regards

Carlos

 

[zlatkodo]

Hi, NIDinc,
Thanks for the feedback.
Unfortunately, there is no data for the width of the teeth and the back-iron.
I assume that these values are 4.5 mm and 9 mm. With these inputs I have made a test of flux densities for the case of 480 V, 250 Hz, 100 turns/coil. From the results it can be concluded that it is advisable slightly to increase the width of the teeth and thus reduce the flux density in the tooth. At the same time slot depth can be slightly increased to get the same slot area. In this way we can further reduce the heat in the teeth. Also, I would not choose 14 bars in the rotor. Such combinations could cause noise and cusp. Yes, the cusp can be avoided with skewed rotor bars , but however, increased noise may occur.