Missing phase detection failure 3

[hughkaza]

Recently, a field engineer complains that several elevators made by our company have missing phase detection failure. It is said that the elevator could run with R phase or T phase cut out. But this does not happen on the test machine in our factory.
So I went to the site and did some measurement. Surprisingly, when R phase was cut out (we directly took off the wire from the very beginning input terminal), the voltage waveform was almost the same as it was when R phase was on. The elevator was not running when I did the measurement to limit the places cause this failure. I think this malfunction is not caused by the inverter and the missing phase detection circuit on the PCB woks normally stand-alone.
Since the control cabinet of the elevator is quite complicated, I only draw the main circuit involved in the attachment. I will give more details if needed. I think the line filter or the transformer caused the malfunction. The transformer is a “V”-connection transformer.
One assumption is that the capacitor in the filter caused this but I just don’t get it.

 

[itsmoked]

Welcome to Eng-Tips hughkaza.

I give you a star for a complete and clear question.

The motor generates the missing leg. You would find current imbalance and the motor would likely fail very quickly in seconds possibly if under a load. But the voltage would look pretty good, just as you are seeing. Also since the motor is now generating the missing phase it is displaced to where it should be for the motor to be creating it.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[hughkaza]

Thank you for your reply, itsmoked.

As I mentioned, when I did the measurement, the motor was not running.
Do you mean that the coil in the motor generates the missing leg? I don't understand. You know, as the elevator is not running, the IGBTs in the driving unit are shut off. I see no connection between the motor and the transformer.

 

[itsmoked]

I was thinking you disconnected the motor mechanically!

I see, it was electrically removed. Well the same thing can happen with the transformers you have.

If you think it's the filter why not disconnect everything from its output and then check its output directly.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Kieth, I think that the OP may be using an VFD, and the VFD is charging the dc bus from the available phases.
Another possibility if there is no VFD is that the line filters are generating enough of a "false" phase to start the motor. It is extremely rare for a simple DOL motor to start on single phase, particularly if the motor is loaded. Oce the motor is running, your comments are correct, but without either an electrical kick or a mechanical kick three phase motors generally just HUM on single phase.

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

 

[LionelHutz]

I'd also guess the capacitors in the filter are causing the problem by creating a third phase feeding into the transformer. Google "static phase converter" and you'll see how to use capacitors to basically create a 90 degree offset third phase (instead of the true 120 degree third phase) so you can run a 3-phase motor on a single phase supply. Not quite the same as your filter but very likely the same effect is involved.

 

[PHovnanian]

Its either a phantom phase caused by a capacitor, as LionelHutz suggests. Or you have a power supply with a transformer that has a delta secondary or tertiary winding somewhere downstream of the opened phase. With essentially no load on the system (no motors running), either of these could be causing what you see.

As you stated that the cabinet is quite complicated and you have omitted some details, look for other power supplies connected downstream of the open point that differ from your factory configuration (which doesn't suffer this phenomenon).

 

[REDDOG]

Is their a Code requirement for a phase reversal relay?

 

[waross]

Hello hughkaza;
Can youtell us where in relation to the filters the phase failure relay is located, and where you disconnected the wire?

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

 

[hughkaza]

Thank you guys for your suggestions.

To waross:
Yes?we do use a VFD.
We disconnected the R phase wire from the power distribution box to our elevator and leave other things what it was. I suppose this simulates the missing phase failure.
We do not use a phase failure relay. Instead, we detect the DC voltage after a three phase rectifier. I got almost the same voltage before and after I disconnect R phase since the waveform is almost the same.
Another test was done yesterday.

To PHovnanian:
I see no other power supplies.

I read this

http://www.eng-tips.com/viewthread.cfm?qid=244885

May be something wrong with the transformer?

 

[waross]

Don't be mislead by the thread that you were reading. That is a method of developing three phase from two transformers. The connection is called An "Open Delta" connection in North America and, I believe, a "V" connection in Europe. Such a connection will give you three phase power but if you were to remove one wire as you have done, you would still remove one phase from the load.
A bit about VFDs. A VFD recifies the AC to supply DC to the DC bus. In smaller sizes (about 5 HP and less) VFD are sometimes used to drive a three phase motor from single phase.
Single phase supply power will adequately supply the DC bus but if this is done intentionally the VFD is oversized.
It is no surprise that the VFD will drive the motor with an incoming phase missing.
However the issue now is: "Why does it do this in the field and not in the shop?"
There are several possible reasons:
First, at no load, one or two phases will charge the bus but under load the bus voltage may be a little lower.
> The voltage in the shop may be a little lower than the voltage in the field.
> The VFD in the field may have the DC bus monitoring threshold set a little lower than the test VFD in the shop.
> If a neutral connection is used in the field but not in the shop test hat may make a difference. Without a neutral connection the the DC bus is being charged by one phase out of three. With a neutral connection, if the recifiers use the neutral, the DC bus will be charged by two phases out of three. Are you using a 220 VAC Delta supply in the shop and a 380 VAC Star supply in the field??
> In North America buildings with elevators often have very large air handling fans for HVAC. If a large motor is running when a phase fails it may act as an induction generator and supply the missing phase. Once it is stopped it will not restart with a phase missing.
Bottom line:
I don't believe that DC bus voltage is a reliable indication of phase loss. For phase loss protection you may be better to monitor the incoming power.
There are two possibilities, phase loss and voltage or current unbalance. Some installations use both.
Phase loss protection may be used to prevent equipment from starting with a phase missing.
Unbalance protection may allow equipment to be connected with a phase missing, but running or attempting to run with a phase missing will cause unbalances that will cause the unbalance monitor to operate.
In your instance, I would suggest either a dedicated phase loss monitor or a current unbalance monitor.
I would then consider a control philosophy where loss of a phase does not stop the elevator but rather sends a signal to the VFD that limits the output (motor speed and load) to a value that may be safely handled by the VFD when it s derated for single phase operation.




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

 

[hughkaza]

Thank you, waross.

I don't think it is a reliable phase loss detector, either. I believe it was designed to lower the cost.
And in this case, as the waveform is almost the same, it doesn't work.

With DC voltage high enough, VFD can work perfectly. In this case, the AC current of the remain two phases is about twice of what it is when run with three phase. This may do some damage.

I agree with you that the problem now is why it happened to this particular machine.

Another test was done in the field yesterday. Since one of the engineers believes that the filter causes the phantom phase, we added a capacitor assembly to the filter. To my surprise, this works.

See the attachment, guys.
We have a solution now and I need an explanation....

 

[jraef]

Excellent response Bill.

hughkaza,
If cost is an issue, I would think a proper line monitor relay as Bill suggested would be far less expensive than adding more capacitors and risking resonance problems, failures etc.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[waross]

Thanks Jeff. Much of what I know about drives I learned from you. (And a lot of others here too numerous to mention.) Thanks all.


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

 

[ScottyUK]

Ain't that the best reason for being a member here - the chance to learn from others? I've learned a fair bit about diesel generation from you Bill (and CatServEng too), so don't underestimate how much you contribute!


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

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