Capacitor hook up for star delta motor starter 1

[testtech]

What is the proper method of connecting power factor correction capacitors to a star delta motor starter? Is there any NEC reference for this (I could not find anything in NEC 460)? The motor in question is 3 phase, 208V, about 500 HP.

 

[jraef]

Jeeze Louise, that's a big motor for a 208V service!

Because of the potential for a major contribution to a VERY nasty transient when switching from Star to Delta, the best method is to have a separate contactor in front of the capacitors and only switch them on after the start sequence is complete. Tie the output of the contactor to the line side of the starter ahead of the point where the power splits off for the two circuits. Anything else is asking for trouble.


"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

 

[testtech]

It is a big motor. I measured 1200 amps at full load. The nameplate requires circuit ampacity of 1700 amps. The bus duct that powers the motor is rated 1600 amps. The story gets better--a salesman convinced them to install "energy saving devices" to reduce their energy use by up to 50%! These turned out to be capacitors. It does raise the power factor from 89% to 98%. However, the local utility does not charge for reactive power--no power factor penalty. The capacitors were originally installed at the starter incoming feeds. No protection or disconnecting means. I objected to this as a violation of NEC 460. They then moved the capacitors to the load side of the M2 contactor.

Can you elaborate a bit on the reason for a large potential transient during the transistion. Should I be able to measure this with a disturbance analyzer?

 

[jraef]

You can search this forum (use the "Keyword" search above) for Star Delta Switching transients (or Wye Delta) to see where this has been discussed in the past, several times if I'm not mistaken. Reader's Digest version:

When the starter switches from the Star to the Delta pattern, it must open the circuit to avoid a short (let's not get into Closed Transition here). In doing so, the motor is left un-powered, if even for a second. During that time there is no torque output, so the load will slow down. This alone means that under the best of circumstances, the motor flux penetration is lost in transition and must be re-established for the Delta run operation, so there is usually as much inrush current, again, as there would have been without starting in Star. but it gets worse. Large motors have significant residual magnetism, so when the motor is re-connected in Delta, it is often still acting as a generator. Now you are connecting a generator to a line that is out of synch! So depending upon where in the sine waves the re-connect is made, there is either a huge current spike or a huge voltage spike, or sometimes one followed by the other. The torque transient that this can create has been known to shear the shafts off of the motor. the voltage transient is capable of shorting out SCRs in other parts of the facility.

Add capacitors to that mix and you are essentially GUARANTEEING that the motor field is being maintained during transition. Most likely those "energy saver" caps were blown the first time they were energized. Not matter though, they were not doing squat towards saving energy anyway, your people were scammed by that salesman.


"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

 

[LionelHutz]

Is the M2 (2M) contactor the one that transitions to full-voltage? The output side if it would be shorted until it the transition when it is closed. This means the caps would be shorted until the transition and then they get connected to line power.

The typical installation would be to connect the caps to 1M because it closes at start and then stays energized during the transition. The caps can still see some transient voltages during the transition as jraef describes.

Raising the power factor will lower the line current draw which will lower the conductor heating a little bit. But, you will never see anything close to 50% savings, not even 5% most likely.

 

[testtech]

LionelHutz-If the capacitor is on load side of 2M, then the capacitor is out of the circuit until transition is complete, as you pointed out. Would not this hook-up eliminate the capacitor's potential contribution to a high transient on start-up? It appears to me that the reason the capacitor is dangerous, when hooked up to 1M, is that it maintains the magnetizing field in the motor windings during the transition. When hooked to 2M, I would expect the transition forces to be the same experienced without the capacitor. I think this conclusion combines what you and jraef have expressed. If this statement is incorrect, perhaps you or jraef can set me straight.

 

[waross]

This depends on the connection. I just looked up Wye/Delta starting in an old text book and I don't want to assume where in the circuit M1 and or M2 are. Capacitors may be safely connected to the line at any time. If the capacitor is connected to the line there should be no problem.
The problem will be if both open transition is used and the capacitor is connected to the motor terminals.
Open transition generates severe transients on transition. A capacitor on the motor terminals will make these transients more likely and more severe.
To add to jraef's explanation, when the motor circuit goes open during transition, the frequency of the voltage generated by the motor will always drop to a frequency determined by the motor speed. With an induction motor in normal operation this frequency will always be less than line frequency. The window of opportunity for a reclosing in sync is very small, and even if the voltages are momentarily in sync at the instant of closing, the lower frequency will give rise to torque and current transients. Capacitors will make a bad situation worse.
Capacitors on the line side will reduce the starting current slightly, but so little that although there have been papers written on the subject most of us have never seen capacitors used to reduce starting currents. If your capacitors are properly applied they may reduce the starting current by about 7% or 8%.

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

 

[testtech]

Thanks Bill. The issue is whether the transients are impacted by the specific location of the capacitor connection. In this case, the capacitors are terminated to load side of the M2 contactor. Thus, they become part of the circuit at the completion of the transition. This is an open transition starter. In terms of transient potential, is termination to load side of M2 functionally different than termination to load side of M1?

 

[jraef]

I see LionelHutz's point. In a NEMA diagram, the load side of 2M will be shorted during start, so even though the caps are connected to the line, the shorting through the "S" contactor will prevent them from charging.

BUT...
Take this diagram showing how PFC caps are connected internally and add it (mentally) to the diagram above, connecting the caps between 2M and S.

You will see that, as soon as "S" opens, the capacitors are instantly charged by virtue of the fact that 1M is still closed and the charging current flows through the motor windings to the caps across their internal Delta connection. So the caps are now available to maintain the motor field as it slows down.


"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]

Well, when the S contactor opens, and M2 has not yet closed, the caps will be in series with the motor windings. There will be a division of voltage between the motor winding and the capacitors.
BUT, we will have an inductor in series with a capacitor on a 60 Hz. circuit. The voltage across the motor windings or across the capacitors or possibly both may exceed line voltage. It will depend on the respective reactance of the motor windings and the capacitors.
This may not produce similar transients to the transition, but I would try to avoid it. If you are able to monitor or record the voltages, you may see some interesting voltages across the caps and the motor windings when S opens.
I would put the caps on the load side of M1 and let them supply a little bit of the magnetizing current as the motor starts.
Nice drawings jraef.


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

 

[testtech]

I am going to propose testing to the owner. The company that sold them the "energy saving" capacitors has amended their contract, guaranteeing that no failures will result from their installation for a period of one year! Of course, after the failure occurs, it will be too late to prove that the cause was the capacitor installation. If testing is approveed, I will post back. However, if there are other comments, I will certainly appreciate your insights. Thanks.

 

[jraef]

The thing I hate about contract provisions like that is that the damage incurred is likely to be incremental, so it's entirely possible that the failure will show up later.

I'd be more interested in holding their feet to the fire about the supposed "energy" savings, insisting they measure it in kWh over an exactly measured real operating time under the same conditions, not just Amps or even kWh over a different number of operating hours or conditions as may of them try to do.


"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

 

[itsmoked]

And... They will simply point to the very, very, small print that sez:

1) Your mileage may vary.

2) Savings may vary according to your PoCo's tariff structures.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

Yep, there's the rub. Weasel words. My only wish is that "Managers" would think to hire competent professional engineers to evaluate these kinds of claims before they poor money into some scam artist's pocket. Makles you wonder why they get the big bucks eh?

"Energy" savings is not the same as "cost" savings. Saving energy certainly saves cost, capacitors only save costs if there are tariffs associated with poor power factor. But in this case... "a salesman convinced them to install "energy saving devices" to reduce their energy use by up to 50%!" There is NO WAY that adding capacitors will save 50% energy! Even 50% of their internal transmission losses due to I²R would be a stretch, but those losses are typically so low as to not be worth considering.



"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

 

[itsmoked]

"[green]My only wish is that "Managers" would think to hire competent professional engineers to evaluate these kinds of claims before they poor money into some scam artist's pocket.[/green]"

AMEN brother...




Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

Hook the caps to 1M the motor as long as the 1M contactor does not de-energize during the transition, which is typically the case in most Y-D starters I have seen. This means the caps and the ends of the motor windings connected to the 1M will stay at line voltage and the caps will not contribute to maintaining the field in the motor.

On another note, the transition is so quick that the back emf from the motor has no time to decay (capacitors or not). This is why an open transition Y-D starter has such a high current and torque transient. The problem that I see is when the 2M closes the caps will "dump" their energy into the motor contributing to a higher current and torque transient.

 

[testtech]

Here is another concern--overload protection. The presence of capacitors on load side of the starter will reduce the load sensed by the starter's CT's. The starter will probably no longer detect an overload condition, because the CT's were set up assuming no power factor correction.

 

[testtech]

Addendum to prior thought--If the caps are on load side of the starter, the cap feeds should probably be routed through the CTs. Then, the CT's would sense the load for which the protective circuits were designed.

 

[waross]

I agree with LionelHutz.
The placement of capacitors in relation to the overload sensing is subject to code rules. The caps may sometimes be connected directly to the load side of M1 ahead of the current sensing devices so as to not affect the overload protection. I don't have my code books to hand, but check your code.

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

 

[testtech]

Here is what Trane says in one of their installation manuals:
Rule 2
Size motor overload protection to
account for capacitor-supplied
current.
Overloads are typically set to
measure the total current drawn by
the motor. When PFCCs are used,
they become the source of part of
that current. If the current they
provide isn’t registered by the
overload protectors, potentially
damaging amperage can reach the
motor. The simplest way to ensure
that the overloads “see” all current
supplied to the motor is to position
the PFCCs upstream of the overloads
as shown in Figure 2.
If the capacitor connection points
are downstream of the overload
devices, route the PFCC leads
through the overloads as shown in
Figure 3. This assures that the
overloads register both line and
capacitor-supplied current.

The last paragraph describes the installation at the chiller, except the capacitor leads are not routed through the overloads.

You are correct about the code. NEC 460-9 requires that when capacitors are installed at load side of the overloads, the overloads must be resized to account for the reduced currents.

If the capacitors are installed at line side of the contactors, than the code requires a separate disconnect and overload device. The installers moved the cap terminations from line side of the contactors to load side because they did not want to install the required over current protection and disconnects!

So far, the installers have made two attempts and both contain NEC 460 violations.

However, my concern when starting this thread was to explore the potential for the caps to produce damaging transients. At this point, it appears that if they terminate the caps to load side of the M1 contactor and then run the capacitor leads through the CTs, they will have an installation that meets Code requirements and does not contribute any greater transients than the original system.