Starting Current wth Soft Starter 8

[jwilson3]

I'm in a disagreement with an Engineer with a major electrical equipment manufacturer over the basics of a thyristor controlled soft starter. I am trying to calculate the voltage drop profile on the supply system during starting.

The Engineer's analysis shows the initial starting voltage at 30% and the motor current at 330A with 20%PF and ramps up to 598A as the voltage is raised. That's the range of current values I say needs to be used in the voltage drop calc. The Engineer says the current to use is the current on the source side of the starter, which would be 30% of the current on the motor side at start and would rise to equal the motor side current as the voltage is ramped up.

I say he's wrong, that the current is virtually the same on either side of the starter.

 

[itsmoked]

Howdy aolalde! You're right! I was in too big a hurry. I thought Saftronics only did softstarters and that was their "here's how they work page"..

Here's the pics from Siemens "starters" page.

These seem to be showing that they're switching at nonzero voltage points - on AND off. ???

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[waross]

The magnitude of the available voltage at the input is imaterial during that part of the cycle that the current control devices are not conducting.
The input voltage waveform during the time that the SCRs are conducting, will be the same as the output voltage wave form. When the switch is closed, the voltage is the same on both sides of it.
If the effective value of this waveform is used instead of the effective value of the input voltage sin wave, the equations will balance.
Think of a flashing light that is on 50% of the time. The output voltage equals the input voltage 50% of the time.
Similarly, the soft starter input voltage equals the output voltage only during the time that the SCRs are conducting.
respectfully

 

[jwilson3]

Thanks, waross. You stated it better than I did.

 

[rbulsara]

A few deserving stars.

Can I blame this temporary blockage of my understanding on my personal termoils going on in my house?

Cheers

 

[Skogsgurra]

Sorry to get in so late. I had an interesting experience in May this year. A large fan for a boiler, 1200 kW, had a failure in its Siemens soft starter.

I made some recordings. This is what it looked like when we finally got it working:

Voltage in upper trace and current in lower trace. We are at about 50 percent of end speed. Current scale is 3000 A/division.

It may be interesting to see what happened when we tried to start with the fault still in the controller (one firing pulse missing). It looked like this:

The scale factor is now 6000 A/division. As you can see, the missing firing pulse produces a DC component, which not only drives motor current high - it also saturated the transformer feeding the system so that it had a very high overcurrent which made a protective relay fail. It did that after about ten minutes. There were probably other damages to the system as well. But I left when the soft starter was OK. So I do not know what else had happened.




Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Lotta amps there.. Whew! You do put a lot of time on your data recorders don't you Gunnar..

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Ya know what? My customers pay me to do that. And as a side effect, I get some RL data for you.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[stevenal]

Perhaps it would help to state that real and reactive power are conserved quantities, but that distortion power is not. And in general, apparent power is not either. I confess I'm not entirely sure if reactive power is conserved in the general case (help me out if you know), but I thinks it's fair to say that distortion and apparent power are not.

 

[Skogsgurra]

Introducing concepts like distorsion power is more confusing than helpful. It obscures the fact that the same current is flowing in the supply wires as in the motor wires after the soft-starter.

Distortion power was used as a way of explaining what harmonics do to the apparent power. Remember the 3D Pythagoras rule used by some educators and salespeople that actually had no idea what they were talking about? Nor had their listeners, I think. They just got confused. That's why you do not see this concept any more.

The soft-starter just lets a smaller part of the sine voltage through. The smaller the part, the smaller the RMS voltage - and hence the current. It is no more complicated than that.

jwilson3 is right. And so are the guys supporting his view. I jumped in only because I thought that my recordings are illustrative. The soft-starter has a very limited ability to manipulate the voltage (Can only fire thyristors, has to take into consideration that the phase/phases returning the current must be "on" when thyristor is fired. It must also take the inductive nature of the load into account. As you can see, the thyristor conducts long after zero voltage crossing) All these circumstances complicate things considerably. But it does not change the fact that Current In = Current Out.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[jraef]

Keith,
You are batting 1000 buddy!
That 2nd waveform posting is not a wave form for a true 3 phase controller. Siemens (and a few others) have a "low cost" version of their soft starters that only control 2 of the 3 phases in an attempt to reduce the component count. One phase is just a piece of bus bar. While this works OK on low inertia loads, it is somewhat controversial and IMHO should not be used where long acceleration times are expected. That waveform shown is for their 3RW - 2 phase controller only, and even then it is a little "simplistic" to say the least. They publish this sort of stuff because people who don't understand it get excited from comments from competitors about negative sequence currents and DC components imposed on the sine wave which can theoretically damage the motor windings. They are trying to simplify the concept for the masses.

Most of the "waveforms" published for soft starter outputs are simplified graphical representations of one phase, partly because of exactly what Gunnar's post shows; the complexity of a scope readout. Remember, a soft starter is never controlling just one phase at a time, it is controlling 6 SCRs which are almost carrying current from more than one line phase 2/3 of the time. The effect on scope traces of the output voltage waveform in any one phase becomes very complex to interpret.



JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[jraef]

By the way waross, nice addition to what I had said earlier, I need to remember that way of stating it.

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[stevenal]

I wasn't aware that distortion power was a dated concept. It's in my latest IEEE dictionary. Made sense to me. Just thought it might help those thinking input V*I=output V*I understand that this only true for sinusoidal quantities.

 

[Skogsgurra]

The way it is (or was) abused is not seen any more. Of course, it can be used - but only where it makes sense.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Yeah jraef... Teach me to be superficially helpful!

Skog's scope pictures are quite a bit better. Nasty complex especially in the one phase of three aspect.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[TomG33]

Hey guys, again Like a few others I have leapt into this a little late, But I want to point out to itsmoked something he missed in his post containing the Waveforms and included comments about the non zero Voltage switching.
With SCR's the switching is done at zero current, not zero Voltage, and from the manufacturers diagram of the current waveform, you can see that indeed it is switching at zero AMps.IF you look at the opeation of a SCR Rectifyer operating in the regeneration region, the switching is always done at quite high voltages, both on and off.

Like Gunnar, I spend enormous amounts of my customers money investigating problems with these things. Have literally CD's full of high speed recordings of all sorts of things, except, sadly, Thyristor starters.


Stars to jreff and others for some very good points

 

[waross]

Thanks for the comments jraef and jwilson3;
I have a lot of respect for the knowledge and experience of both of you. Your remarks are more flattering than quite a few LPSs.
Thanks
Respectfully

 

[Marke]

Just came across this thread, and I agree with the conclusions made. After many years experience in designing and working with soft starters I can confirm that the current waveform is a double humped waveform as shown above with the forst hump being the current flowing from phase A to phase C and the second hump being the current flowing from phase A to phase B. As you reduce the conduction angle, the OFF time increases and the humps become more prominant. As the conduction time is increased, the humps blend into a single hump and at 180 degree conduction angle we have the traditional sinewave we are so familiar with.
When the SCR is conducting, we have the same voltage input and output minus tha voltage drop across the SCRs which is typically around 1 - 1.2 volts per SCR. The current is the same, so at each instant that the SCR is ON, the VA input equals the VA output minus the conduction losses of the SCR. For each instant that the SCR is OFF, the VA input equals the VA output, so the laws of conservation etc are fully met.
The difficulty comes when considering discontinuous waveforms. At that time, you can not take the average voltage times the average current to get the average power. This is a stunt commonly pulled by the promoters of energy savers!!

Unfortunately, much of the technical literature about SCR control of induction motos shows waveforms that are true for an SCR based controller with a resistive load. This shows the current waverform to be very different from reality and the commutation occuring at 0 and 180 degrees.
The reality is that the current is smoothly rounded and takes on the double hump, and it commutates at an angle which reflects the power factor of the motor, hence the commutation angle changes with slip.

During the non conduction period, the voltage on the motor terminals does not fall to zero as logic would suggest. The spinning motor acts as a generator and creates it's own voltage profile to fill in the gap. The actual voltage seen is dependant on the voltage generated, but can make the waveform "applied" to the motor seem prety wicked at times. The rality is that the waveform applied to the motor is only the portion where the SCRs are conducting.

So in calculating power and VA, we need to take the integral of the products of the instanteous values to get meaningful results.
Returning to the OP, for voltage drop calculations, is is usually sufficient to consider the average start current to get reasonable results. The maximum start current flows around the voltage zero crossing at zero shaft speed and increases higher up the sinewave as the motor reaches full speed. This is the same for a standard induction motor starter. The major difference is that there is a non conduction period before the voltage crest that can at times include the voltage crest.
So treat the soft starter like a full voltage starter with a reduced average start current. - somewhat like a primary resistance starter with a very small power loss.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[itsmoked]

TomG33 and Mark thanks for the further clarification.

I have learned a truck-load on this one.

Looking at a truly 3ph system by looking at only one of the ph's really adds to the confusion.

Rather like looking at an elephant thru a soda straw at night.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Why would you do that, Smoked? It's dark at night, you know.

I think that MarkE's comment is the final word. He is THE Soft-Starter Authority. He, and Jeff, I meant to say.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[jraef]

And a lot of what I have learned in the past few years has come from Marke!

Mark, you need to get the Google tool bar spell checker!

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376