Inside the delta vs DOL with soft start. 4

[wrsharper]

Not too long ago we had a star-delta motor controller blow apart but did not hurt the motor, at least from megging and ohms measurements. I read this forum with enthusiasm and followed your collective advise and asked the boss to order a soft start. He ordered a soft start controller to handle the motor FLA. He also wanted to connect the soft start in "inside the delta" configuration. I said that this was not neccessary since we could DOL with the soft start since it was rated for motor HP. I also told him that engineers at this site generally frowned on that configuration. Now for my question, the motor is a 12 lead motor that was being used in star-delta. I have been installing the soft start and intend to wire the motor to the soft start controller just like it would be if it was connected to M1 and M2 in the old star-delta configuration during run. In that:
Output soft start controller,
Phase 1 tied to T1, T7, T6 and T12.
Phase 2 tied to T2, T8, T4 and T10.
Phase 3 tied to T3, T9, T5 and T11.
That is how the motor was connected to the line when both M1 and M2 are energized and the motor is up to speed and in run. This is a little late in asking but is there anything wrong with my design?

 

[aolalde]

Your connection is correct, when a soft starter is applied the motor is connected with the running configuration. For the lower rated voltage, delta 2 circuits (as for DOL starting). Se diagram below.

 

[wrsharper]

Many thanks for the advise and diagram. I have always maintained that when the motor was in delta and run that the windings were configured as two deltas in parallel. Happy Holidays.

 

[wrsharper]

I hope I can ask four more questions concerning the conversion from star/delta to soft starter.
1.) By a full size soft starter does that mean that if the motor has a FLA of 560A and the soft starter has a rating of 570A that we should be able to use the standard delta connection and not "inside the delta"?
2.) Is delta 2 with two parallel windings the same as one delta winding during start?
3.)In the old star/delta the transistion time was set to 6 seconds. I have set the new soft starter to 6 seconds also. The default start time of the new ABB soft starter with built in bypass contactor was 10 seconds. Do I need more time when the compressor is connected?

I have started the new ABB 570A soft starter with the 500HP (480VAC) high speed uncoupled motor and the max current it pulls is 890A before the current starts falling rapidly and then the bypass contactor pulls in. In the old star/delta connection current would ramp up and pull 1200A when in star and still pull 1200A in delta before current started falling. Coupled or uncoupled. 4.) Have I forgoten anything?

 

[Mobius44]

I suggest you read your ABB documentation. If you don't have any then try this ABB link to softstarter's:

http://www.abb.com/Product/seitp329...960020aebf.aspx?productLanguage=us&country=US

1)
When I looked at the ABB documentation I found that the 570A softstarter had dual ratings:

315kW (~400HP) for "in-line" connected
540kW (~725HP) for "inside delta" connected

The different HP rating's are probably why your boss is asking you to wire your 500HP motor "inside delta" instead of "in-line".

2)
The ABB literature also illustrates "in-line" is the same as the "delta2" wiring shown above, and for "inside-delta" one of the two windings is connected to the soft-start while the other winding bypasses the softstarter (it's connected to the line voltage upstream of the softstarter).

3)
The factory settings are appropriate starting point, then you can experiment to find what works best for you. The timings are adjustable so that you can tweak it to fit your conditions.

4)
It is perfectly normal and expected for a softstarter to significantly reduce starting currents.

However, you should wire the softstarter to the motor "inside-delta" to avoid exceeding the softstarter's ratings.

Best of Luck!

 

[jraef]

Wiring a soft starter inside the delta provides NO BENEFIT other than being able to use a smaller (translate: cheaper) soft starter. It does however add significant risk of catastrophic failure unless additional hardware is added, which would of course negate the cost savings issue and increase the complexity, thus the number of things that could go wrong. It also increases the motor heating during starting and an incorrect connection can result in motor damage.

When a soft starter is wired normally (DOL as you refer to it), if one SCR shorts, there is no immediate risk of damage because there is no return path for current flow through the motor windings. You need at least 2 shorted SCRs in opposite phases to present a risk. In a soft starter wired inside the delta, if ANY one SCR shorts, you can lose the motor in short order because 1/2 of the current path through the windings is always present in that configuration. So the prudent course is to use what is referred to as a "fault contactor" in the circuit that opens up if there is any kind of fault on the circuit, i.e. shorted SCR, thus isolating the motor windings. Now the cost and space savings of using the soft starter rated for 58% of the motor FLC is offset by the cost and space required for the Fault Contactor and additional wiring.

There is also a theoretical need to use SCRs rated for a higher PIV rating when used inside the delta compared to a standard line connection because the potential peak voltage is higher. This is ignored by most soft starter manufacturers who promote this configuration because it would add cost, but technically using standard rated SCRs in starters wired inside the delta make them MORE prone to shorting. This argument has been ongoing in the industry, unbeknown to many smaller manufacturers who simply copy what other people do.

Harmonics during soft starting are generally considered insignificant, but when the soft starter is connected inside the delta, the non-sinusoidal nature of the current is exaggerated and THD is TWICE that of a standard connection. So for the same amount of starting torque, more of the current is being wasted as heat in the motor or conversely you get less torque-per-amp during start. So if your application requires a long start-up ramp or frequent starts, using an inside the delta configuration will result in more thermal stress in your motor.

The connection pattern of a standard line (delta) connection is somewhat simple, just get the rotation correct. On the other hand, connecting inside the delta is extremely specific, the pattern must be followed exactly, otherwise the motor windings can be damaged almost immediately upon energization. There is little room for error, yet almost nobody is familiar with exactly how to do it (other than following the written instructions explicitly). For the average end user, this can present a significant risk of future problems during an emergency repair.

Bottom line, inside the delta is not worth the risk IMHO.

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

 

[wrsharper]

Mobius44, thank you for the feed back. This is all new to me and I do not want to make any mistakes. I have worked in electrical/electronics for 36 years on all types of equipment but large motors are a new area for me.

I just opened the book that came with the ABB PSTB 570-600-70 to page 3.6 and in the description it states that this soft starter is rated for 500HP "in line" connection and 700HP "inside the delta" connection. (you scared me a little because I can no longer connect inside the delta due to physical restraints) I was at home when I wrote the last post without any documention and was concerned because the last time I started the uncoupled 500HP motor, as it was coming up to speed the soft starter was vibrating some.

In the motors application it will either be left on a long time or shut off a long time. The boss proveded me with 3/0 cable to make soft starter connections. The terminals on the soft starter can only accept three cables per phase. Is this enough copper? I've checked the code and it seems on the edge to me.

 

[jraef]

It will depend on the type of wire and the way you run your conduits. If you split up your runs into 3 separate conduits with 3 conductors in each, you may be fine as long as your wire is rated 75degC or higher. If you pull all of your conductors in one conduit, you must derate the 3/0 to 70% of its rating at 3 conductors per raceway. So if for instance even if you have 90degC rated wire, it is good for 225A for up to 3 conductors per conduit, but only 157.5A if all nine conductors are in one. So 157.5 x 3/phase is only 472.5A.

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

 

[wrsharper]

Thanks jraef, the cable is AWG TYPE MTW OR THHN OR THWN OR GASOLINE & OIL RESISTANT II VW-1 600V FOR CT USE. The three 3/0 cables per phase are only being used in the old star/delta controller cabinet so I should be OK there like you said. There are 12 4/0 cables and 2 2/0 ground cables that run to the motor peckerhead in two 3" conduits with a 10' length. There are seven cables in each conduit with 2 A phase, 2 B phase, 2 C phase and one ground cable in each condut. I was hoping to only use three 4/0 cables per phase because of the difficulty in connecting the 12 lead motor "in line". From your last post it looks like 3 4/0 cables when derated 70% will only give me 546A which is not enough since the motor FLA is 560A and when the motor is running it will be at full load most of the time when its warm outside. Do you agree that I will be forced to use four 4/0 cables per phase. If that is the case could you make a suggestion for tap/line bonding in the peckerhead other than the lug, bolt, nut and tape method. I thought about cable blocks (most likely not the right name). I have three mounted in the old star/delta cabinet. They are retangular in shape and have aluminum blocks with holes for the cables and an allen tightening screws. What I am concerned about with using this method is the motor tap cable is flexible with fine steel strands. I have seen problems with another installation using this method because of the motor tap leads not making good contact in the cable block leading to burning.

 

[davidbeach]

If you have two conduits and two conductors per phase per conduit and you want to reuse the existing, there is no permissible way of using only three conductors per phase as you will not be able to comply with the requirement that all conductors of a given phase have the same impedance. The impedance of each of two conductors in the same conduit will not be the same as the impedance of a conductor in a conduit when there are not other conductors of that phase in the conduit. You could use one, two (two different ways) or four conductors per phase but three is off limits.

 

[jraef]

Actually, your derate is only 80% for 6 current carrying conductors per raceway (don't count the grounds), so it appears you have 90degC rated conductors, 260A each, x .8 = 208A x 3 = 624A. This doesn't include any voltage drop issues however, so you still need to factor that in but if it's only 10 feet, no problem if you need to do that.

You are right to question a mechanical lug on the extra flex cables at the motor, that is a concern to be sure. I always opt for high compression crimp lugs on the cables. But you can still use a connection block, just use one with studs instead of mechanical lugs like this (appropriately sized of course);

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

 

[wrsharper]

davidbeach, What you write about is something new to me that I would have never have thought about. How about, if in one conduit I have three cables phase A and three cables phase B and in the other conduit I have three cables phase C. Would this meet the impedance requirement? If this does not meet the impedance requirement what would be the negative effects? The motor remains running for months at a time and off for months at a time as different compressors work duty is rotated.

jraef, Thank you for the picture.

 

[wrsharper]

davidbeach, one conduit I have two A phases, two B phases and two C phases. In the other conduit I have one A phase, one B phase and one C phase. On the surface this seems to be balanced to me. Would this meet the impedance requirement?

 

[DickDV]

jraef, thanks for the perspective on "inside delta" softstart risks.

I've always done softstarting in the main motor leads but have been tempted to try inside delta once or twice. Not any more!

Thanks again, DickDV

 

[davidbeach]

wrsharper, three phase conductors of the same phase would work in three conduits provided there is no ferrous metal that encircles any group less than all conductors. Inside, connecting to a motor, you are undoubtedly using steel conduit, so single phase per conduit won't work.

Two conductors per phase in one conduit and one conductor per phase in a second conduit will not have the same impedance per conductor. The impedance imbalance will result in current imbalance.

 

[jraef]

That's right, sorry I didn't think that part through thoroghly.

What davidbeach is referring to is that in ferrous metal conduit, the fields around the conductors interact with the conduit itself to create impedance in the conductor. In the conduit with 2 cables per phase, you will have more interaction than in the conduit with only one per phase. So in each phase you will have 2 cables with slightly more impedance that the 3rd. Remember that current flows in the path of least resistance, and impedance is the same as resistance in this case, so more current will flow in the 3rd (single) cable and overload it. If you did this with PVC conduit it would not be an issue, but metallic conduit is a problem. You can read the NEC, article 300.3.B.1 and 310.4, but to be honest, this issue is not well covered IMHO (I'm looking at the 2002 NEC, maybe they amended it in the 2005). This paper gives a better perspective however.

http://www.ul.com/regulators/ode/0703.pdf

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

 

[wrsharper]

Thank you all for the enlightenment. Now I know what to look for with the impedance reqirement. All of our motors exhibit some phase imbalance in the 10% area even with the conductors properly installed in the conduits. In the past I have thought about rotating the phases to find a good match but in all cases this would be work intensive and the arguement would be given "It has always worked before". When I started the new soft starter controled uncoupled motor I did not see any phase imbalance out of the ordinary and so did not pay it any mind. I also know from reading about phase imbalance here and other places that the imbalance diminishes as the motor gets more and more load. I will watch for this and in this installation it is easy to rotate the phases to find a happy match. If the phase imbalance is over 10% at any time coupled or uncoupled I will ask for a conductor and conduit refab. Thanks again.

 

[waross]

Before changing conductors and conduits check the incoming voltages. I have found unbalanced supply voltages to be the most common cause of voltage unbalance.
respectfully

 

[wrsharper]

Thank you waross but our 480 supply voltage has only a 1% unbalance. That was one of the first things I looked at for the cause of the current unbalance of 10% on all of the compressors.

 

[wrsharper]

Mr. davidbeach and Mr. jraef, You are 100% right about the impedance mismatch. I just made current measurements on the individual cables with the uncoupled motor. The way I have the motor temporarily wired for testing is two sets of three phase cables in one conduit and one set of three cables in the other conduit. The current in the three phase cables in the conduit with only one set is 20%-30% higher than the three phase currents in the conduit with two pairs of three phase cables. My thinking is that when the motor is connected to the load and operating at FLA that as long as the current the conduit with one set of cables is in code then I should be OK. If not then I will have to connect all four wires. Even if I had noticed this without consulting this forum I would't have a clue to what was wrong except that there might be a connection problem. KUDOS and Bravo Zulu