Surge Caps and Lighting Arrestors on 5kV motor 2

[rockman7892]

I have an installation of a new 350hp 4.16kV motor in which the motor termination box contining the surge capacitors and lighting arrestors is too large to fit in the location next to the motor where it was intended.

Because of this several people here have proposed some solutions.

Solution 1 was to eliminate the Surge Caps and LA's all together and therefore eliminate the large terminatation box thus allowing a smaller termination box to be used. I dont feel as if it is a good idea to get rid of these components but dont have enough knowledge to defend my thought. Can someone help me understand why they are needed and point me to some info to back my thought.

Solution 2 was to keep the Surge Caps and LA's in this large box, however mount it elsewhere away from the motor where there is room. We would then come from this box to a much smaller box next to the motor itself to pick up the motor leads. My qustion and concern is, what is the effect of locating the surge caps and LA's away from the motor itself. If this is allowable, what is a safe or acceptable distance or location?

 

[electricpete]

Thanks for the comments edison. Good point about motors failing on start rather than stop. Also glad you took the plunge to order the Nailen book. I'll bet you won't be disappointed.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

Yeah, I think that book will be worth the price even with additional $ 23 for the shipping.

 

[Skogsgurra]

I will have to sit down and sort through these postings. But some comments can be made immidiately:


First. Bill's simple model is very valid when it comes to overvoltage when breaking an inductive circuit. The need for protection wasn't really a need until the vacuum breakers appeared. And the OP was about vacuum breakers. The simple model does, however, not take the transmission line effect on the wavefront into account. I still maintain, and I have got good reasons for it, that such effects are a minor problem and that no protection is needed because of them.

Second. Edison's observation "As a rewinder, I agree with you on many motors failing during switching on rather than on switching off or during running. Based on my clients' feedbacks, I would roughly categorize the percentage failures as 80% switching on, 18% during running and only 2% on switching off" is a very common false conclusion. A motor that was destroyed by inductive kick-back at switch-off will remain undetected until energized again. It will then blow fuses or trip breakers. The observation that it failed when switching on is, therefore, one of the most common false conclusions in industry.

Third. "And I have never seen a surge protector located away from the motor or generator" As a rewinder, you probably only see the motors. And then you will see protection at the motor, if there is any. What you do not see is all protection left in the cabinets. I have seen them - many of them.

Pete. All references have wordings like " as close as possible to the motor". Of course! But, in the OP's case, "as close as possible" is obviously a bit away from the motor - but still rather close. It is also very natural that a producer of protective equipment recommends mounting that makes his equipment perform as good as possible. Also, I do not think that you have given us a quantative answer as to how bad more remote (say 10 or 20 feet, but still connected directly to the motor cable) mounting really is. Does it mean that the stresses on the motor winding will be 1 percent more? 10 percent? 100 percent? If the answer is 1 or 10 percent - even 20 percent, then I think that most of this thread has been totally unnecessary.






Gunnar Englund

http://www.gke.org

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

 

[edison123]

Gunnar

I rewind not only motors but lotsa utility generators up to 250 MW, both at my shop and at sites. Like you, I also visit various sites in many countries. I have never come across surge arrestors located away from the terminals in my nearly quarter century in this line. None. Never.

Your suggestion the motors could have failed unnoticed during switch-off goes not gibe with many of my clients telling me that they had tested (at least an IR) the motors/generators before energizing.

 

[Skogsgurra]

We have different experiences here. Why did they test a motor before energizing? OK if it was a newly installed motor. But such motors do not fail very often.

Most motors that fail do so unexpectedly. It is not very common to test a motor before every routine start.

I am a wee bit reluctant to accept such talk.

The different experiences regarding installation of protection may be because of different parts of the world? You know the saying about Heidelberg and Jena?

I will be seeing ABB Motors in Västerås within short. I am not so sure they know about this either, but shall talk to the designers directly.

BTW, the protection in Narvik (see previous post) was installed in the cabinet. Not at the motor site. Has worked ever since. A recent generator in a co-gen plant was destroyed because of vacuum breakers (after 840 starts) There was no transient protection in the cabinet or the motor. The company that deleivered the motor control panel installed protection in the panel. Works very well. There was no discussion as to where to put the protection. Have we been doing this wrong for decades? And why doesn't it show?

Gunnar Englund

http://www.gke.org

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

 

[electricpete]

I will have to agree that there are probably many factors that contribute to the 80% failures during start. Motors might become moist or contaminated during shutdown period under non-ideal conditions. High mechanical and magnetic forces during start. High temperautres in parts of the motor. But I do believe that starting surges are one factor that can attack weak motor turn insulation and push it over the edge to failure.

=================
One small quote which may or may not be relevant. "Electrical Transients in Power Systems", 2nd ed by Greenwood ISBN 0-471-62058-0, page 541

There is an issue which must be discussed in this application and that is the need to place the arrester as close as possible to the transformer it is protecting. Indeed, installing all protectors as close as possible to the protected object is a cardinal rule of protection. Let us see why this is so...

What follows are some figures and analysis that I don't understand. Looks like waves reflecting back and forth from the terminals of the protected device. Will try to study it when I get more time.

============================
Here is another very interesting thing:

http://books.google.com/books?id=V1...aining&sig=ACfU3U3aubBWByzKaQLravBCrtWcFcyX6Q

Note
The length of the interconnecting cables plays a vital role in containing or enhancing the severity of the incidence wave. After the interrupter, the surge enters the cable and propagates ahead. As it propagates, it rises in amplitude, at a rate of Vt/t, (Figure 17.15 [at the bottom of this page]) until it reaches the far end of the cable. The longer the cable, the higher will become the amplitude of the incidence wave which will be more severe for the terminal equipment (refer to protective distances, Section 18.6.2). The length of the interconnecting cable is therefore recommended to be as short as possible. The manufacturer of the interrupting device can suggest a safe length for different sizes of cables, depending upon the voltage of the system and the equipment it is feeding.

I think they are saying that the voltage spikes apparently grow in magnitude as they travel down the cable, so shorter cable is better.

That is just werid. It does ring a bell from the discussions on VFD cable length. I vaguely remember for VFD's that too short a cable length or too long a cable length can cause the higher spikes at the motor. Is it a similar type phenomenon? I am not sure how to interpret that chart at the bottom of the page which shows voltage increasing then decreasing with time where time related to length.

Even though I don't understand it, I suspect it is probably right. It doesn't agree wtih my circuit analysis, but that doesn't mean much - I have said all along that my circuit analysis didn't include wave effects.

If it is true that the voltage spike gets bigger and bigger as it goes down the line, then we have found a very good reason that the protective device should be located at the motor end. Certainly for an arrester – a spike which was below the threshhold for conduction at the breaker end might grow much bigger at the motor end. For a cap, the implications are not as clear but still seems to make sense to put the protection at the location you want protected if magnitudes can increase going along the line.

==============================

Also, I do not think that you have given us a quantative answer as to how bad more remote (say 10 or 20 feet, but still connected directly to the motor cable) mounting really is.

I did provide in my attachment 23 Aug 08 17:32 results showing that for one type of protection, the motor surge increased from 1.5 pu. to 3.0 pu when we moved the protection from the motor to the panel. Take it for what it's worth. In my part of the world, we don't need detailed justification/analysis to follow standard and manufacturer recommendations, we need justification/analysis when we deviate from them.

=============================

22 Aug 08 16:14
I simply wanted to tell rockman that, if he wanted to move the protective elements away from the motor, he could do so. And that is regardless if the transients are from lighning, closing or breaking the motor current. The resulting discussion has surprised me.

24 Aug 08 3:16
then I think that most of this thread has been totally unnecessary.

You have advised me when you didn't like my tone inappropriate and I will do the same. On the surface, it sounds as if you are suggesting that you had provided the definitive input in your very first post, and no-one else should have dared to provide any discussion after yours. I don't really think that's what you meant and I chalk it up to heat of the moment.
If you can shed any light on my questions about the thumbrules for cable length between VFD /motor and how/whether it might shed some light on the passage showing longer cable creates higher surge, I'd be interested to hear.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Skogsgurra]

Pete.

Yes, you did get me right when you think that I thought I had delivered a definitive answer.

The whole thing is about the kick-back when opening an inductive circuit. And there, I think that my answer was very valid.

What you and several others have discussed are transients coming from the outside. Transients that have a very low source impedance, i.e. same as the cable wave impedance. I have, obviously in vane and before a completely blank auditorium, tried to explain the difference and why there is a difference.

Sorry that this seems to have developped into something personal. I did not intend that.

Gunnar Englund

http://www.gke.org

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

 

[electricpete]

To summarize one point: There are differences in surge withstand capability of motors, both in the specification world and the design world.

In the specification world, we can test to 2.0 pu or 3.5 pu (0.1-0.2 usec) for NEMA large motors. I'd hope you'd agree that's a pretty big difference.

In the design world, I would group it into three categories of turn insulation from worst to best:
Worst - no dedicated turn insulation. No mica strand insulation. Turn insulation funciton is performed simply by film on the strands.
Middle - no dedicated turn insulation. Strand insulation includes Mica tape.
Best - Dedicated mica-paper tape turn insulatio.

You will find all of these varieties in service. I did failure analysis of a motor with shorted turns (tripped during start...I'm not making any claim when it failed). We segregated the non-failed portion of the winding and surge tested it - some sections didn't even come close to meeting the reduced maintenance level surge test requirement. We cold stripped one full top/bottom coil section of the windings (to avoid damaging them in a burnout) and did an inspection. There was no dedicated turn insulation and the strand insulation was just film. You would be amazed how brittle and fragile that strand insulation was. I have a video somewhere showing how it crumbled in my hands if anyone is interested.

The point being, there are big differences among motors. Do you know the specification or construction of the turn insulation of the motors whose service history you are citing Gunnar? If not, there is reason for caution because the genralization may not apply to motors with lower strenght turn insulation.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

What you and several others have discussed are transients coming from the outside.

That is not what I intended to address. If there is a particular statement or argument you think applies exclusively to external surges, please let me know. Note he attachment to my post 23 Aug 08 17:32 is labeled "The Effect of Surge Suppression When Switching Motors..." Clearly, this particular reference addresses the motor switching transients. And in one table, it describes a fair number of the items we have been talking about (vacuum vs air breakers, opening vs closing, protection at motor vs switchgear).

If you are inclined to agree to disagree, that's fine with me. I think if you read my comments carefully, you will find that I have never asserted a strong conclusion... just brought additional references, analysis, and questions into the discussion.

But there is still a question I am really interested in.

The link in my post 24 Aug 08 13:55 to the document that described increasing surge voltage magnitude as the cable gets longer.

I had several questions about that including how it relates to vfd cable lenght which is somethign I remember you and the others on this board have discussed (it's not something I'm familiar with... we don't have vfd's).

What are the thumbrules for selecting vfd cable length?
Is there any explanation for those thumbrules?
Does it shed any light on the idea that surges can grow larger as they travel down the cable?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

My post 24 Aug 08 14:42 - the description of the middle category was a little weak and narrow. There can be a variety of strand insulation which incoporates Mica or some new materials whose names I don't know which perform a lot better than simple film on strand described in the weakest category. All of these would be included in the middle category.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Skogsgurra]

Pete.

I am not in any mood to continue this thread. There are several sites where you can find the data you are asking about. One very early paper is the classical "Riding the Reflected Wave" by the guys at Allen Bradley (now Rockwell automation). It can be accessed at

http://ieeexplore.ieee.org/iel3/4210/12267/00564866.pdf?tp=&arnumber=564866&isnumber=12267

if you have an IEEE account, which I believe you have.

There are several installation guides where cable length is discussed. The series of ABB Technical Guides, notably #5 is also highly recommended.

Do not expect any further posts from me in this thread.

Gunnar Englund

http://www.gke.org

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

 

[electricpete]

Thanks gunnar. It was an interesting discussion for me anway. Sorry if any of my comments rubbed you the wrong way.

I will take a moment to explain one reason why this thread has really captured my interest.

The most critical motor in our plants are the reactor coolant pump motors. Failure of any one of them would cause a plant transient and would likely cost $10 million (plant must be cooled down and depressurized for replacement).

The motor is located deep inside the reactor coolant building in an area which is completely inaccessible during power operations and still difficult to get to during shutdown conditions (must dress up in yellow anti-contamination clothing, go up and down many ladders, accumulate radiation dose, etc).

The PWR design from this supplier provides the surge caps directly in the motor terminal box. And these surge caps are film type caps, not oil-filled caps (the quantity of oil used in minimized for fire-loading reasons) . They have 51 series elements. The supplier tells us to trend the capacitance reading every 18 months and replace when it shows evidence of a single shorted cap (which we see periodically).

You can imagine, it is a lot of work to go to the inaccessible location every outage to do these measurements. But we do it. And it is even more work when we have to replace those caps since they weigh around 150 pounds and have to be rigged into this location.

How much easier it would be if those capacitors were located in the switchgear which is a very accessible location.

So the question comes.... if it doesn't make any difference, why did the supplier put them on the motor and make our life so difficult? It could be that he just didn't know any better? If that's the case, maybe we can get the moved. It's an idea I'm tossing around as a result of this thread.

Also you have mentioned there is no need for surge suppression in absence of vacuum interrupters. Our motors have air breakers, but yet we still have surge caps on all of our 13.2kv motors, including these reactor coolant pump motors. So maybe another alternative would be to just get rid of those surge capacitors (instead of relocating them) which not only reduces maintenance requirements but eliminate potential failure (cap failure) for a component which MIGHT not be needed. Also we have long cable length which may help attenuate surges but I'm not so sure anymore. At any rate, removal of surge caps if it is an acceptable approach would be a pretty darned easy modification (just lift the leads of the existing caps). Another idea to toss around.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Here is my simplified attempt at a calculation of the effect of cable between arrester and motor. Effect of the associated capacitor is neglected (I don't know how to model that part).

As we know, when the wave first hits the motor, it can almost double, depending on the relative impedances of motor and cable. The ratio is 1 + (Zmotor-Zcable)/(Zmotor+Zcable)
For simplicity, let's assume that the it doubles.

So let's say we have an incoming wave with rise time 0.8 usec and peak value 3 pu.
Slope rho = 3/0.8 = 3.75 pu/usec

The doubling effect means that the rate of increase of voltage at the motor location is 2*rho ie 7.5 pu/usec

Let's say the arrester MCOV is 1.5 pu.
Let's say the speed of the wave in the cable is u = 140,000 km/sec (just under half the speed of light in a vacuum).
Let's say the distance from arrester to cable is L = 28m
Time to travel that distance = T = L / u ~ 0.2 usec

At the arrester, the incoming wave looks like the following:
t=0, Va = 0
t = 0.2, Va = 0.75
t = 0.4, Va = 1.5. At this time, the arrester shorts and also sees reflection from the motor. But none of this will become apparent at the motor until t=0.6

Here's what the motor looks like:
t=0, Vm = 0
t = 0.2, Vm = 0
t = 0.4, Va = 1.5 (remember it's increasing twice as fast)
t = 0.6, Va = 3.0

So with 28 m of cable, the result is that our 1.5 pu arrester alows 3.0 pu voltage at the motor. It is probably overly-conservative based on the fact that we have neglected the capacitor. You could perhaps use a similar calculation as a bounding calcualtion if you are trying to justify a short run of cable.

Anyone please check or comment on my assumptions and calculation.

Fwiw, I based my calc on my understanding of the attached excerpt from Handbook of Power System Engineering by Yoshihide Hase (he talks about the slope, but doesn't carry the calculation through to find the peak). Note also the discussion of the effect of separation distance upon oscillations on the last page.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

More info on a similar subject

http://books.google.com/books?id=VR...=ACfU3U3PW8F8_XvAEfSqNVOQk8CfAaKERw#PPA332,M1

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Actually, my example problem could have had the incoming wave peak magnitude at 1.5pu (vs 3.0pu) , provided the slope is still maintained at 3.75 pu/usec. In this case, the result is that our lightning arrester does absolutely nothing. It allows the incoming 1.5 pu wave to double exactly as it would have in absence of surge arrester. In contrast if the surge arrester were very close to the motor, the voltage would be limited very close to 1.5 pu.

At least, that's my understanding. I'm open to comments.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

"An Update On Surge Protection Of Medium Voltage Motors:
A Comparison Of The Standards And Applications"
Lanphier, M. Sen, P.K. Nelson, J.P.
NEI Electr. Power Eng. Inc., Arvada;
Electrical and Instrumentation Applications in the Petroleum & Chemical Industry, 2007. PCIC Europe 2007. 4th European Conference on
Publication Date: 13-15 June 2007
On page(s): 1-8
ISBN: 978-3-9523333-0-3
INSPEC Accession Number: 9693240
Digital Object Identifier: 10.1109/PCICEUROPE.2007.4353996
Date Published in Issue: 2007-10-22 11:34:05.0

The rate-of-rise of a surge can be reduced by selecting proper values of (L) and (C) and maximizing the time period (T). The series (L) is usually given by the cable. The shunt capacitance (C) can therefore be changed to limit the risetime. Surge capacitors are most effective when placed as close to the motor as possible. The time constant associated with a capacitor placed at the motor end of the cable is larger than if the same capacitor were placed at the source end of the cable since (ZcC) > (RsC). A common practice is to use a lumped capacitance per the table below.
[Note Zc = Cable Surge Impedance while Rs = source impedance.]

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Here is the article "Riding the Reflected Wave" that Gunnar had mentioned:

http://www.ab.com/drives/techpapers/ieee/pcic.pdf

After reviewing that and some other references, the effect of cable length seems a lot simpler and less mysterious:

A very long cable causes an amplification at the motor upon reflection of the wave.

The amplifiation factor depends on ratio of impedances. Worst case is 2.0 when Zmotor >> Zcable.

Starting with a short cable, as we increase cable length, we transition from no wave effects (no amplification) to full wave effects (amplification based on impedance ratio). The transition point depends on the wavelength of the surge
(shorter rise time => transition at lower cable lengths.)

Knowing this, the passage that I cited 24 Aug 08 13:55 was misleading: "After the interrupter, the surge enters the cable and propagates ahead. As it propagates, it rises in amplitude, at a rate of Vt/t, (Figure 17.15 [at the bottom of this page]) until it reaches the far end of the cable". The surge doesn't increase magnitude "as it propagate" down the cable... we just have the potential to see more amplification due to reflection when it hits the receiving end for longer cables. So the conclusions that I speculated about in that particular post based on that description of growing pulses were wrong. In particular, the doubts I had expressed in that post about using my circuit model ( for predicting effects of changing cap location ) based on that strange description are eliminated.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

I have spent awhile reviewing various references and google. I found numerous additional refernces recommending to put the protection close to the motor or at least analyse the situation: For example:

IEEE Std 141-1993 -" IEEE Recommended Practice for Electric Power Distribution for Industrial Plants: (The Red Book)

6.7.3.9 Rotating machine protection
Incoming surges can be transferred through transformers by electrostatic and electromagnetic coupling. Therefore, surge voltages can be experienced on the transformer secondary as well as the generator terminals as a result of surge-voltage impulse on the transformer primary terminals. This can occur even though the transformer is protected with arresters at the primary terminals.
When high-voltage surges are internally generated, the standard protective circuit for rotating machines consists of arrester and capacitor located near the machine terminals. The function of the arrester is to limit the magnitude of the voltage to ground, while the capacitor lengthens the time to crest and rate of rise of voltage at the machine terminals.
.....

6.7.3.9.2 Rotating machine surge protection practice
Much documentation exists relating to the surge protection of rotating machines. An ideally protected installation requires the following:
a) A strictly effectively shielded environment
b) Arresters at terminals of machine
c) Surge capacitors at terminals of machine
d) Strict adherence to good grounding practices

.......................

6.7.3.9.3 Special care required for proper installation of surge capacitors
Exploratory observations confirm the presence within shielded environments of voltage transients that approach arrester sparkover magnitudes and have exceedingly steep fronts (0.1 µs front-time). Although lightning does not usually entail such steep fronts, certain switching events do; for example, insulation breakdown, capacitor switching problems, or discharge of high lightning current-to-ground. As established previously, separation distance between protective equipment and apparatus to be protected invokes (sometimes serious) depreciation of protection. This is particularly true when steep wavefronts are involved. Surge capacitors, and preferably arresters also, should be connected directly to the machine terminals so that added inductance of the power cable circuit and of the surge capacitor lead will not interfere with their action. This limits the arrester and capacitor total lead lengths to one or two feet, thus requiring extreme care in the motor terminal box equipment arrangement.

Each application should be reviewed on its own. If several machines are fed from a common bus, for example, it may be suffcient to connect arresters on the line side of the feeder circuit breaker, placing only the capacitors at the machine terminals. Such practice generally requires that the insulated conductors of each motor feeder circuit are continuously enclosed in a grounded metallic raceway and that more than one feeder will be closed at the same time, along with a careful analysis of the arrester-protective level and the capacitor wave-shaping action as a function of the feeder length involved. A direct, low-impedance path between machine winding and surge-protective devices must exist on both line and ground sides of the circuit. A good ground connection to the machine frame is essential.

=================================
The only thing I did find that mentioned or recommended putting protection at the breaker terminals was Siemens literature about a device called a Surge limiter. I have attached some info. Since I need two attachments and this is a new topic, I will continue in the next post.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

The only information I found about placing any motor surge protective device at the switchgear was some Siemens literature about a device called a "surge limiter".

First, a quote from the attachment to my PREVIOUS post 29 Aug 08 16:30:

Siemens Tech Topics #3
Limiters and arresters differ in several fundamental respects. The type 3EF surge limiter can absorb the trapped energy associated with a vacuum interruption, whereas a surge arrester has a greater energy absorption capacity to deal with system phenomena, including lightning strikes and switching surges from all sources. The 3EF surge limiter has a lower (i.e., better) protective voltage level than an equivalent surge arrester....


In capsule form, our recommendations for vacuum circuit breaker application (vacuum contactor application recommendations differ) are as follows: ....
2. For transformers of reduced BIL rating, add some form of protection. (Either surge limiters at the switchgear, or surge capacitors at the transformer, or surge arresters at the transformer).
3. For motors with locked rotor current under 600A, add surge protection (surge limiters at the switchgear, or surge capacitors at the motor, or surge arresters at the motor).

Notice they don't treat a surge arrester and a surge limiter as interchangeable parts. If we use limiter we put it at the switchgear... if we use arrester we put it at the motor. And they give a hint of the differences between these devices – limiter has a lower (better) protective level..

Now let's explore the differences a little more by looking at the attachment to THIS post.

Type 3EF surge limiters

* Lower protection level than in the case of common arresters
* The protection characteristic is relatively insensitive to steepedge impulse waves
* Especially suitable for limiting switching impulse overvoltages
...
The housing is made of plastic. Inside the housing there are gaps and non-linear resistors connected in series.

The resistors are Siemens SIOV metal-oxide varistors...
As a result of these characteristics, the surge limiters respond very quickly to switching surges and restrict them to low values.

So in contrast to the typical surge arrester (gapless Zn O), the surge limiter is a series gapped arrester with Si O element.

These are completely different devices. The limiter is optimized to provide better protection for switching surges than a standard arrester (clamps to a lower value at the arrester terminals) and to provide a smaller footprint (for installation in switchgear), at the espense of ability to handle other types of surges. The only device they allow putting at the breaker is the limiter which offers a lower (better) level of protection. I would assume that this is needed to compensate for the distance effects which make arresters less effective as we remove them away from the protected device.

So we have found only one group of references that allows us to put the protection at the switchgear (the Siemens literature), but the protection is not an arrester, it is a limiter. And the Siemens recommendations are careful to identify they are not interchangeable – the arrester is to be used at the motor and the limiter is used at the switchgear (one or the other, not both).

Gunnar – is it possible that the cases you are citing involve installation of limiters at the switchgear, rather than arresters? From looking at the attachment, we see that limiters have a smooth plastic case, while arresters typically have the rippled porcelain external surface.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

pete

Just got the Nailen book today. First glance - very informative. Thanks for your tip.