The Use of Explosion Proofs Motors on Distortion Supplies 6

[ICEMAN]

Can we start a discussion about the use (and safety) of fixed speed explosion proof motors on voltages supplies polluted with harmonics ?

We could also extend it to the use of explosion proof motors fed from VFDs.

If somebody kicks it off I will pick up the baton !

ICEMAN

 

[ozmosis]

ICEMAN
There is no blanket frequency range now on the ATEX certified Siemens motors. Each frame size has the specific range it has been tested. It goes on the origianl certificate.
I think you would be surprised at the level of QA, now that ATEX has taken hold. Looking at the processes of certification and checking (this is PTB mind)then I doubt very much if anyone can really get away with murder. They appear(I'm not a motor man, just a vsd man)to have incorporated QA systems where bodies such as PTB will be auditing the company processes and therefore the certif.conformance will be (much) more than just an addition of any ISO9001 process. So it will be EExe experts auditing.

 

[ICEMAN]

sed2developer...

Tut mir leid ! I was referring to the UK (and elsewhere) re "they were getting away with murder". The Germans were the ONLY one doing it right with regards to certification and technical aspects. That's why I ended up workly closely with Loher.

Siemens did (in the late 1980s) have the blanket certification (I saw it) but since then it obviously been removed as you stated. Remember, I am going back almost 15 years here. I think it is still in place re BASEEFA (or what it is called now) and UK made motors; I may be wrong though.

I am also a drives guy who now is involved with harmonics too (and drives now and again from a consulting angle). I did not set out to be an expert (some people's eyes)in the area of explosion proof motor/VFD packages but ended up having to be one if we wanted to sell the equipment correctly and safely.

In those days there were many, many 'cowboys' in the drives business. The 'ex campaign' really sorted the men from the boys. I was the Managing Director of a drive company and was looking for a niche market for AC drives (and found it in hazardous areas) where honesty and technical expertise stood more chance of ending up in orders. I had then to assess the then situation and was horrified to find noboby knew what they were doing in the UK. I gathered evidence and took it to BASEEFA, the Health & Safety Executive and the Dept of Energy (Petroluem Engineering Division)for offshore installations. The letter of reply from BASEEFA is dated 1st September 1987 and said....."I am having your letter studied and aim to write more fully in the next week or so" - almost 17 years later I still await the reply ! However, I did work with all parties on this serious matter and the initial result was the conference mentioned. The Senior Inspector, Dept of Energy over a nice Italian lunch with me in Aberdeen stated "the UK works on the stable door principle - we may know there is a problem but can't or will not deal with it until it is too late. It'll take a big bang for people to take this matter seriously". Unfortunately, a few months later he had his big bang - the oil platform Piper Alpha blew up killing ~160 guys ! They chaired the conference.

The day following Piper Alpha I was contacted by Scottish TV re the possible cause of the incident (our PR people had sent them a copy of the editorial "Are We Playing With Fire" from the Electrical Review a few weeks earlier. We had long discussion and I was asked to Glasgow to give an interview. I never arrived there !

After the campaign (was included a lengthy campaign in the UK and international technical press), which was designed to warn and educate people, we never lost an order to any other company in this application area. Most stayed out of the ex application area totally - too hot to handle as the customers were more clued up that the suppliers. A nice case of the "7 Ps" principle - proper preparation and planning prevents p*ss poor performance".

However, my real aim was to ensure a level playing field in the application of VFDs to ex motors. This was successful - no cowboys played the game anymore (at least in the UK). However, the whole issue was/is a real can of worms re what people has done in the past. Say no more about that.

I ended up in Iraq, China and other exotic places because of my knowledge in this area, both as a consultant and selling equipment. The main problem was often faceless consultants who would admit they got it wrong and change the spec. Some of things they specified were dangerous. I also done some work others (inc the US which is still way behind Europe)in these issues.

ICEMAN

 

[ozmosis]

I've got a feeling our paths must have crossed when I was based in the UK (now in Melbourne looking after Asia.Pac region), I certainly had dealings with Al Archambault from Mirus whilst I was in Congleton.

Having now experienced your tour de force in the UK (don't take it the wrong way, I think your insight has been an interesting read) but what about your original question? I'm guessing you have a reason for posting the question in the first place. What are the recommendations on (power quality) supplies when connecting explosion proof motors? Is it a question of isolating transformers to EExe motors if harmonic distortion is high? How many people check the THD when connecting such motors? not many I would say. Every explosion proof motor must have some form of temperature measurement to protect overheating, if this was to get too hot then the system would cut out. I would say that if the motor is fixed speed then investigations would then take place as to why rather than before the system is installed. This, I would say, is the reality rather than ideal scenario. Or are you saying that if the THD was too great then the heating in the motor would create bigger problems (like your big bangs, god forbid)that would not be protected by the thermistors in the motors?

 

[ICEMAN]

Sed2developer....

You aint Steve Barker from Siemens UK are you now gone walkabout ? I heard you escaped from Stalag 17 (that's what an old Siemens Germany pal told me the Germans called the Siemens UK posting. You only got it if you done wrong !)

The problem is that usually the certification of ex motors of whatever protection concept is based soley on sinusoidal supplies with no harmonic voltage distortion (nor the effects of taken into account) - statement direct from BASEEFA. Therefore a 'certified ex motor' more is no longer certified when on distorted supplies and is potentially dangerous depending on the magnitude of the distortio and the rotor design, etc.

The problem is that ALL voltage supplies are distorted to greater or lesser degrees. What I think is required is that certification process has to take into account this and IEC and other international bodies agree on a maximum Vthd to which normal ex motors can be connected while maintaining safety and retaining certification. I really don't see that as a big deal myself if the political will is there.

The normal maximum where harmonic standard apply is 5% Vthd so why don't the certifying bodies and standard organisations get together to ascertain the effect on 5% Vthd will have on ex motors re temperatures (rotor and stator), bearing currents, etc. I would tend to use the standard 6 pulse PWM AC harmonic 'staircase' footprint as the reference harmonic source as they as usually largest polluters.

Depending on the results motors could be modified, outputs reduced, etc. and the motors would be safe and certified for use of supplies containing up to 5% Vthd. There would have to be a provision however such that the 5% Vthd was applicable at the point of connection and not measured miles (and less 2-3% Vthd away)at some remote swichboard or whatever. Obviously, if the site has more than 5% Vthd it has problems anyway which need to be addressed and resolved.

Now you focussed my mind re this I'll maybe take it up with PTB and BASEEFA (or what they are called now) et al. Now that'll really make my popular with the motor manufacturers again.

All joking aside. This needs to be done as it is big hole in the safety aspects of explosion proof motors ALL OVER THE WORLD.

Comments guys ! I have done a lot of talking. Now I'll do the listening.

ICEMAN

P.S. You got a bronzie yet Steve ?

 

[ozmosis]

I'm not Steve. He's still there.

 

[ICEMAN]

Mmmm. If it's not Steve Barker then it must be Patrick Ramsden !

Ja oder nein ?

Wie geht es Ihnen ?

ICEMAN

 

[jbartos]

Comment: The explosion proof enclosures have been around for more than 50 decades. There are fairly mastered technologically. E.g., chemical industry, petrochemical industry, etc. have been using explosion proof motors for long time and worked on their improvements together with industry standards, e.g. NEMA. Electrical design books, e.g. by Bishop dating way back to 50s detail explosion proof installations. With the advent of VFDs, the motor bearings were initially affected by higher harmonics causing EDM on bearings. However, the phenomenon is now well understood and VFD outputs are properly filtered. There should not be much more detrimental effect to the motor bearing from a VFD output than from a normal harmonic contaminated power supply with DOL start.
Another area that has not been addressed yet is the reflected waves that are also posing problems in terms of elevated voltages and their impact on the motor and motor feed insulations. However, this phenomenon is now also well understood and properly treated by the VFD output filters and slower switching devices, e.g. IGCTs.
Consequently, the current VFD - motor technology is approximately on the same safety level as the prior technologies without VFDs or perhaps even better, when the traditional, manual or automatic star-delta starters having been used and accommodated in hazardous environments.

 

[ICEMAN]

Jbartos

I note your comments but have to wonder as to whether you have read and appreciated all that has been said.

I agree that explosion proof ‘enclosures’ have been around a long time since emanating from the mining industries. However, we are not discussing the effect of harmonics on a piece of metal but on windings and other components parts contained therein.

Tell me what ‘filtering’ is applied to VFD now as standard other than the use of higher switching frequencies (2-~10khz) which produce higher frequency harmonics ? If that is the case then why are R-L-C sinus filters often required by VFD both to minimise both output harmonics and the standing waves problems (I will say more above this below but as is not deemed as much of a potential danger I did not touch on it) ? In any case we are not just considering the effect of harmonics on bearings; that is pretty minor in comparison with other issue if one ignoring the catastrophic effects due to bearing collapse, etc.

If you think that the harmonic spectrum on the mains supply is the same as that on the VFD output you need to get your text books out again, hopefully not the 1950s ones ! In any case we have the same problem; the ex motor is operating outside the conditions envisaged when it was certified, no ?

OK, you mentioned standing waves. (I did not really intend to address this as is relates to all motors, not just explosion proof types). This may be an issue on some applications, mainly on voltages above 440V. It was standard practise some years ago to fit output inductors on VFDs where the cable run from drive to motor was in excess of 50m (~55 yards). On some applications these can result in accelerated motor insulation failures (but not overheating of the motors). There was considerable concern some 8-10 years ago, mainly in the US, regarding failures on applications where IGBTs were used (i.e. higher output frequencies – up to ~8khz) and the mains voltages was 460/480V and above. These failures were not phase to earth (ground) insulation failures nor overtemperature related but short circuits, conductor to conductor within a given stator slot.

Oscillatory voltages (aka ‘standing waves’) considerably in excess of the motor insulation inception voltage can be applied to the motors when fed from the VFD, irrespective of type. The level and duration of these overvoltages were/are dependent upon several factors, namely :-

Cable length between VFD and motor
Presence of a VFD output reactor (inductor)
VFD PWM pulse pattern
VFD PWM repetition rate

The actual dv/dt of the individual VFD PWM waveforms at the drive terminals is only one aspect of the problem; once cable length between the VFD and motor reached a critical length, resonant wavefront reflections occurred which caused oscillatory overshoots on the voltage applied to the motor at each PWM edge leading to a magnitude of around double the dc bus voltage level (up to ~ 1300V on 480V supplies).

If the PWM strategy was such that a line to line voltage step of –dc and +dc level occurs at one instant, or the dwell between –dc bus switch and =dc bus switch is small (relative to the cable resonance characteristics) then the resulting line to line overvoltage could actually be FOUR TIMES the normal DC bus level. The frequency of the overvoltage transient oscillation is usually very high and decays rapidly, and as such the motor may (rpt may) survive.

However, in applications with long cable lengths, output inductors were often fitted (to minimise capacitive coupling to earth (ground) and subsequent drive tripping). The inclusion of the output inductor had the effect of significantly reducing the resonant frequency of the motor-cable circuit. The overvoltages produced at each PWM edge remained of the same magnitude but the frequency of the oscillation associated with the overvoltage is considerably reduced thus extending the time delay. This had the effect of causing motor insulation discharge inception voltage to be overcome for a longer period during each PWM pulse, accelerating insulation failure.

Studies of insulation characteristics at the time did show the insulation can recover from oscillatory transient voltage in excess of its inception voltage providing these was adequate time between transients. Therefore, as VFD PWM repetition rates increased (i.e. the switching frequency) so insulation failure was accelerated.

Regardless of the type of devices used in VFDs the dv/dt associated with them is usually sufficient to excite motor cable resonance producing an overvoltage to at least twice the dc bus level once a critical cable length in reached. The actual voltage rise times at the motor terminals became determined by resonance characteristics. The resulting rapidly rising wavefront was not evenly distributed across the motor winding due to parasitic capacitance leaving the first few turns of the motor winding exposed to full overvoltage stress, hence the failures.

To overcome these problems ‘sinus’ filter (R-L-C) were developed which also have the effect of attenuating the VFD output harmonics.

Re your comments regarding VFDs being as safe as DOL or star-delta starters….If you mean as seen from the mains side. OK. I have not problem with that – that is not what we are discussing however. However, if you mean from the explosion proof motor side, whether on fixed speed operation on distorted supplies or on variable frequency supplies you are very, very much mistaken. Does anybody else concur here or I am just a lone voice in the wilderness ?

I think this discussion is starting to drift and the plot is being lost. Can we bring it back ?

ICEMAN

 

[jbartos]

Comment: This is essentially eng-tips Forum. Lengthy discussions do also appear; however, any practical usage often needs to be either proven, derived or referenced. This is the biggest strength of engineering and eng-tips.

 

[ICEMAN]

jbartos...

OK. I note your comments. Lets leave it at that them.

Over and out.

ICEMAN

 

[ICEMAN]

OK. It's been while since I've been on this site. JBartos unwarranted comments last time kinda peed me off; at least I have contributed something meaningful and important.

I have spoken at length with the Europeans and in the US regarding fixed speed explosion proof motors on distorted supplies. As far Europe is concerning standard explosion proof motors (flameproof {EExd}, increased safety {EExe} and pressuried {EExp}) are permitted only 2% Vthd (HVF - harmonic distortion factor) before they are 'operating outwith the conditions envisaged when they were certified" (i.e now uncertified). With EExN (non sparking) motors for Zone 2 only; 3% HVF is permitted. There are no IEC standard applicable to higher levels of voltage distortion therefore for operation above those levels special testing and certification is required.

In North America we have a problem. To cut it short....UL who is responsible for these matters does not know. It was refered to NEMA who referred it back to UL. You have a problem in the US obviously which needs urgently to be resolved. NEMA and UL need to get together pronto before a big bang happens !

This reply will help those outside the US but in the US it's seems like a case of 'you pays your money, you takes your chance !"

ICEMAN

 

[ICEMAN]

OK.

HVF as mentioned above is course "harmonic voltage factor" not harmonic distortion factor. Sorry about that - fingers move faster than brian at times !

Also, NEMA does restrict the use of NEMA rotor designs C and D (i.e. double cage/deep bar type rotors)to NON VFD fed and NON distorted supplies, even for general purpose (i.e. non hazardous location/area) applications (MG-1).

Europe on the other hand has no restrictions on double cage or deep bar rotors (as per NEMA C & D) for explosion proof motors or any protection concept. Kinda crazy that !

A wee bit confusing to a lot of people.

ICEMAN

 

[ozmosis]

ICEMAN
there's quite an interesting article (not aimed at flameproof/explosionproof specifically) on

http://www.pdma.com/PowerQualityFaultZone.pdf

but concentrating on NEMA motors.

 

[pareshcvora]

Iceman - Thanks for the insightful information on the applciation of VFDs to Ex motors.

Have been involved in recent times with a lot of requirements of retrofit of old (>10 years) Ex motors in oil refineries. A few questions to which am yet to find an answer or develop an opinion. May be many of you can guide

a) Since dv/dt at the motor terminals is one of the issues with application of a synthesised waveform to a motor, are there any useful guidelines to dv/dt limits.

b) Which of the technologies if VVVF generation i) the current source invertor (CSI) or ii) the voltage source invertor (VSI) more suitable for Ex environment motor? Can you share any of your experiances in both fields?

c) For a VSI the solutions proposed to remove maximum harmonics for retrofits is to provide a sine wave filter? Is there a better / econimical alternative to this?

d) Are there any issues with CSI technologies wrt effects like high di/dt or others? Do CSI also require sine wave filters?

Thanks to the eng-tips team too. This is a great site.

Paresh Vora

 

[ICEMAN]

Paresh

I have also been involved in similar projects in the Middle East and have some real horror stories as to what end users, consultants and drives companies DO NOT know and try to do thinking they know better (a real case of a little knowledge is dangerous !!). Price, not safety is usually the main criteria. I still do consultancy on drives and harmonics in that area now, both in hazardous areas and general applications.

I will try and answer you questions as follows :-

1. No there are no real guidelines (as far as I know) for dv/dt other than minimise it. Older VFDs did tend to be bad but IGBT ones which switch up to 8-16khz have steep dv/dt too. 600 V/uS and upward can cause real problems on the winding insulation, especially on older motors or those with Class B and less insulation (E).

2. Current source inverters are kinder to the motor as they do not cause as high losses. The problem is the CSI are not 'sexy' and cannot be mass produced at low prices. CSI require the motor to have high levels of magnetising current as it is this which commutates the inverter output bridge; so motors and converters have to be matched closely. Such close matching is not a usually a requirement of PWM drive which introduce additional stator and rotor heating to the motors. This can be attenuated by high quality sinus filters but these are expensive. Also with CSI the displacement power factor varies with speed and load (displacement pf is the pf without the harmonics). PWM drives have a dpf CSI harmonics are not as 'stair cased' as PWM).

3. My experience is that CSI is better with ex motors, but more expensive and not feasible at all at smaller powers (<300kW). I am talking here about new installations where the ex motor/VFD was supplied at a certified and fully tested package. Suggest you look a

http://www.loher.de

for info on this as they do PWM to 500kW and CSI to 8000kW for gas, oil and petrochemical industries. They are one, if not, the best in world in this area.

One crucial fact here re CSI or PWM is that whenever you retrofit an VFD to an ex motor your motor certification is no longer valid as the motor will "operating outwith the conditions envisaged when it was certified". If this is an explosion due to the motor the industrial insurer may not pay out if it is proved the use of uncertified equipment was the cause. Do not take this lightly please ! Fitting sinus filters will make no different legally!

4. CSI drives do not require sinus filters but you have to consider the voltage commutation notches on the motor (every 60 deg). If a new application this can be optimised via carefull motor and inveryer output bridge capacitor design. It is usually well within the capability of the motor insulation.

I would not recommended retrofitting CSIs to ANY application (never mind hazardous area - as per the comments above)without careful consideration of all the motor and application (starting torque, etc) issues. I would approach Loher, Siemens and others; high quality CSI manufacturers for advice.

Paresh, I have a couple of papers which WILL be of interest to you. If you want to email me seperately on mailevans@tiscali.co.uk I will send them to me; they are not for open distribution. If you have any other specific questions use the same address.

I hope I have helped. The advice about NOT retrofitting VFDs to ex motors in the most important here !

ICEMAN

 

[nunion]

I am a process engineer involved in commissioning a pharmaceutical facility and we seem to have become embroiled in issues surrounding the drives, inverters and ATEX. I would appreciate any help and guidance that anyone can offer.

In essence, the problem is that we have 6 drives supplied by one manufactuerer, 2 drives supplied by another, and 1 drive supplied by yet another manufacturer. Eack drive is connected to ABB inverters on a control net network. One of the drive suppliers has told us that their drive needs to be replaced as it has not been certified as ATEX compliant in combination with the specific inverter. This has naturally made us concerned about all of the other drives as none of these have been ATEX certified with the specific inverter although all have been ATEx certified for use with inverters.

My question is:

a) Do drive and inverter combinations have to be ATEX certified together
b) Do the thermistors in motor windings have to be matched with the thermal relays in the MCC or VSD.

I would appreciate any guidance that anyone can give me

 

[aolalde]

nunion (Chemical):

Although I am in America were the standards on this subject are not deeply involved, I think your questions are answered into the brilliant deep exposure by ICEMAN.

 

[DickDV]

I've found this discussion thoroughly enlightening and useful. Being in the US, I have always insisted on not only an XP label on the motor of a Drive/Motor system in a rated atmosphere, but also an XP Inverter duty nameplate.

This nameplate specifies PWM power, speed range in Hz, and Current Limit setting in the drive. Most of these motors have been Reliance and I have been quite comfortable using them. However, the above comments especially about the field weakening point raise some concerns that I find completely valid. Having always set the field weakening point at 460/60, I don't think there is a problem but variations from that would be a real issue.

Thanks for an especially thoughtful discussion of this important subject.