Notice: This technical discussion forum has been established by the SRE to assist all reliability engineers, not just SRE members. To post your technical problem, solution, question, or answer here, send an e-mail to SRE webmaster. Your e-mail should follow the format of those already posted. Postings are accepted from anyone, as long as they relate to the reliability field. It works best if your posting contains contact information for possible follow-up. But, if necessary, postings can be anonymous - just so state in your email! Also, let us know when you do get workable solution or answer. We want to post it so everybody will benefit.

Problem/Question(Sep 30, 2004):

Dear sir/madam,
I have some question related to vibration equipment. Our steam turbine serviced by Bentley Nevada 3300, due to our needs to perform a vibration analysis We want to up grade our vibration system Bentley Nevada 3300.

Right now we have 2 options for this issue, doing up grade by using Bentley Nevada 3500 with system one Or doing up grade by using other brand such as movisys and divaline (from 01 db stell france).

My question is:

Is it possible to up grade Bentley Nevada 3300 to be movisys system (still use bently probe and proximity) Just changing Bentley 3300 (multi channel vibration monitoring system) to be movisys.And then install divaline to provide on - line vibration monitoring.

Where is this up-grade have implemented, using Bentley Nevada installation (3300) and use movisys as multi-channel vibration monitoring system and use divaline as a on-line software.

Your help will be appreciated.
I am looking forward to hearing from you soon.

Sincerely yours,
rafael a. rajagukguk
condition monitoring engineer
Rafael.Rajagukguk@kpc.co.id

Reply, April 13, 2005

Sir,
There is a possibility to use the Benley sensors and a VM600 system (vibrometer) system for vibration monitoring and analysis.This system is full programmable.


Dé Verschuren

D.Verschuren@istec.nl

Reply,July 18, 2005

Sir,
The proximity probe and proximitor used in the 3300 system is pretty much a standard eddy current displacement sensor. Be advised that BN has many stages of obsolescence, so do not change this equipment out just because you are told it is 'obsolete'. If it works- it works.

If you wish to develop the condition monitoring and analysis side of your system you may not need to remove the protection system at all.

Regards,
Scott Robertson IEng MEng AMSOE AMIPlantE
srobertson@commtest.com
Principle Diagnostics Engineer
Commtest Instruments
Christchurch New Zealand
Phone: +64 (3) 374-2337
Cell: +64 (2)121-59637
Fax: +64 (3) 374-2339
www.commtest.com

Reply,July 18, 2005

Sir,
Stick with the Bentley 3500. I have commissioned several in power plants and there is none better in my opinion. This will save you money in the long run. And also think about operator training.

RDM

Markle Control Engineering LLC

Solution/Answer: Can you help?

Problem/Question (Sep 20, 2004):
Hi, I am a testing engineer working on some accelerated life test projects. I would like to know if there are any methods or procedures to conduct accelerated life test on incandescent bulbs and fluorescent tube? I am working on a few different method suggested by factories and 3rd party lab. It seemd that they are not using the same method. Pls help.

Michael Chow
michaelc@techpas.com

Solution/Answer: Can you help?

Problem/Question (Sep 1, 2004):
I am trying to evaluate software programs for computing subassembly and system level MTBF on a MS Windows platform. From this end, the task looks to be somewhat daunting. Can anyone point me to a site that features reviews of commercially available MTBF prediction calculation software, or provide insights from personal experience? I have some limited experience with RelCalc and ReliSoft products but that is it.

I would appreciate direct e-mail responses as well as postings.

Best Regards,
Mike
michael.englehart@sbcglobal.net

Solution/Answer: Can you help?

Problem/Question (Jun 25, 2004):
Hi, I am facing a problem in converting Management Expectations of durability into reliability target. Suppose the management wants a useful life (durability) of 10,000 hours for a system, which is expected to follow an exponential distribution for failures in field.
What should be its target MTTF (hrs)? Is any method available?

Pl advice

Thanks,
Shashank S Kotwal
kotwal.shashank@mahindra.com

Solution/Answer: Can you help?

Reply, October 25, 2004

The cumulative distribution function (CDF) for the exponential distribution is F(t) = 1 - exp(-xt), where x (aka lamda) is the failure rate and MTTF = 1/x. F(t) is the probability of failure by time t. Management will have to decide what a satisfactory definition of "useful life" is: 95%, 99%, 99.9%, etc, chance of success at time t = 10,000 hours. Call this probability P.
Then you need to solve the equation:
F(t) = 1 - P = 1 - exp(-t/MTTF)
P = exp(-t/MTTF)
ln(P) = -t/MTTF
MTTF = -10000/ln(P)

A few sample values:
P = 95% --> MTTF = 195,000 hrs
P = 99% --> MTTF = 995,000 hrs
P = 99.9% --> MTTF = 9,995,000 hrs

The exponential distribution is the simplest distribution to work with and will work fine in the "useful life" region of the product's life--that is, the flat bottom of the bathtub curve. If the product might wear out at a time of the same order of magnitude as your MTTF (mechanically or by corrosion), then the exponential failure rate will not be appropriate.

Ryan Turner
Risk Analyst
ARES Corporation

Problem/Question (Jun 25, 2004):
Hi, My name is Quynh Nguyen, Component Engineer from DragonWave Inc. I'm performing the MTBF of the system and need to calculate the first year multiplier rate based on Telcordia SR-332. One requirement is the burn in of either device/unit/or system. My question is that the "burn-in" mentioned here is from one representative lot or from every lots ship out to customers???

Please help.

Thanks,
Quynh Nguyen
qnguyen@dragonwaveinc.com

Solution/Answer: Can you help?

Reply, July 30, 2004
Hi Quynh,

I believe you need the whole thing BURN-IN in hour in order to complete your task. Another way you can do it is by each device and then add the Failure rate together.

Roc Bui.
HNS QA Engineer.

Reply to answer, November 23 2004

Thanks Roc for your reply. Now, I have another question. I've been performing the MTBF calculation based on Telcordia on electronic components for the whole communication systems. How do I relate this MTBF number to the time for warranty??
Thanks in advance.

Quynh Nguyen
qnguyen@dragonwaveinc.com

Problem/Question (Jun 25, 2004):
I am looking for a general equation for a bath-tub curve so I can model the life of eletrical equipment. Can anyone help?

Thanks,

Chris Nye
nyeguy21@hotmail.com

Solution/Answer: Can you help?

Reply, July 16, 2004
Design and Analysis of Accelerated Reliability Tests, Part 1,” ASQ Reliability Review, Vol. 24, No. 2, pp. 11-20, June 2004 and “Design and Analysis of Accelerated Reliability Tests, Part 2,” ASQ Reliability Review, Vol. 24, No. 3, pp., Sept. 2004 propose piecewise linear failure rate functions. This is proposed because:
1. they're simple
2. they represent the left-hand part of the bathtub curve
2. they capture infant mortality
3. testing usually doesn't test enough items, long enough, and with enough failures to estimate more complex models.
Contact me if you would like more info, pstlarry@yahoo.com , or send data for free analysis.

Problem/Question (Jun 7, 2004):
While Calculating Reliability (217F std) for diodes, we need to find out Junction Temperature( Tj ). In this calculation we will take Junction to Case temperature into consideration. When we calculate Temperature factor, do we need to consider Profile temperature again along with Junction Temperature (Tj ) or only Junction temperature is enough. The same case with IC's also.

Need your valuable guidelines in this regard

Kirana.N.A
Sr. Software Engineer
Satyam Computers
Hyderabad
India
Phone : 91-40-55233816

Kirana_Anantharao@satyam.com

Solution/Answer: Can you help?

Problem/Question (Apr 08, 2004):
I am seeking to procure some reliability software for our test facility here at NVESD. I need some help from people who have already purchased or used software packages (e.g. Relex, PRISM, Reliasoft). We are interested in knowing if there is a package that is yielding good results. We would also appreciate knowing about any shortcomings (e.g. no training/poor training, too expensive, poor/no support). Thank you for your help.

Ronald L. Edwards
Electronics Engineer
ronald.edwards@nvl.army.mil

Solution/Answer: Can you help?

Reply, September 16, 2005
RISK SPECTRUM, a swedish software for reliability analysis is very good and simple to use. Not too expensive. There are some training courses too but i dont have experience. but i have used the software extensively for reliability analysis of Nuclear power plant safety systems.

-c.senthil kumar
cskumar@igcar.ernet.in

Problem/Question (Jan 23, 2004):
Hello

I've a problem in using a component defect rate in %ppm and "translating" into FIT failure rate, for predicting reliability( MIL-HDBK-217) ! What do I have to request to component provider to use this datum? Sample size seems to me not sufficient? Perhaps something about "life test " on the component could help me?? Can anyone please help me on this??

thanx a lot

Luigi

Luigi Di Gioia
ldigioia@cgspace.it

Solution/Answer: Can you help?

Reply, March 29, 2004
Hi,

ppm has nothing to do with FITS/fpmh. ppm is parts failed per million parts, and is a value how good the parts will be manufactured against a specification. FITS / fpmh is the dimension for parts failed in field.

ppm - process related <-----> FITS - field related

Regards
Wolfgang Baumann

Reply, Jan 10, 2005

The component defect rate do not include environmental stresses and other electrical or mechanical stresses which lead to component failures during life of the component. Component defect rate is taken at an instant of time not for a life span. A separate life testing which may be accelerated life testing should be conducted on the components received from component manufacturer to estimate FIT failure rate of the components. The two terms defect rate and failure rate can not by used interchangeably. The failure rate must be estimated by modeling the stresses faced by the final product. For better explanation on carrying out life testing a book "Reliability Engineering" by Elsayed A. Elsayed, published by Prentice Hall PTR may be referred.

Neeraj Goyal
Research Scholar,
Reliability Engg. Centre,
IIT Kharagpur,
West Bengal - 721302
INDIA
Tel. - ++91-3222-28-1681(office)
++91-3222-28-1116(hostel)

Problem/Question (Jan 12, 2004):
Hello Please can anyone help me to understand the term: Secondary cause of plant asset failure and how can this be identified?

Thanks
Dhevan

Dhevan Naidoo
Dhevan.Naidoo@engenoil.com

Solution/Answer: Can you help?