What method of detecting bearing/machinery failure?

Hello Scott. It would be helpful to know exactly what kind of plant you are in, and in what country. Sounds like a power plant I think...

The big three for portables in the US are Emerson CSI, Entek IRD, and SKF.. Bently Nevada is the biggest for online monitoring of turbines. To be quite honest, SPM is not a very big player, and their results in the market have been mixed.

My personal experience with SPM was not positive, but some people like their gear. My opinion is any device that does some sort of high frequency enveloping (peak vue in the CSI) is fine.

My personal preference is CSI for portable and Bently for online, but the plants I deal with run both Entek and SKF as well and they have been fine.

Hello Steve,

Thanks for your comments. The plant is wind turbines and the country is England. The advice about peak vue is really useful. At the moment CIS does seem to be the best portable system.

Best Regards

Scott.

I will offer this as a suggestion. First, speak to a large bearing manufacturer such as SKF. They have made a great inroad into rotary vibration analysis concerning rotating precision elements. We use vibration sensors and a main recording/analysis system in our plant, and we use SKF as the source for much of the data as far as what is expected/acceptable and what is not.

Best Regards,

Ing. Robert Forbus

Hi Scot.r111,

I can tell you if you like,

I did a lot of research on ball-bearing related problems since my professional beginning in 1974 with a gyro-system manufacturer.

We did investigate into error mechanisms and measurement of these in ball-bearings (fluid- and air-bearings too).

Five major areas of problems exist:

A. Bearing quality as delivered and matching surfaces (these often good, those almost always bad).

B. Mounting procedures and cleanliness.

C. Proper lubrication and elastohydrodynamic lubricant-film and measurement of these.

D. Imperfections (out of roundness, waviness of balls and raceways and matching surfaces) and vibrations generated by these: may be benign may be start of failure.

E. Dents from improper mounting and "big" dirt particles (fibers being worst) , initiation of spalling, prevention of electrical erosion and measurement of these.

All these have to be discussed - and results implemented - to get a really good and reliable production and product.

The devices you are thinking about assume that you start with a good bearing assembly - most often not really true.

With fine raceway deteriorations you will get high frequency noise, with overload and denting you will get low frequency impulses at any ball rolling over the dent. So this is related to bearing geometry and speed.

With these devices you will have to watch several units from beginning to failure in order to build up your knowledge.

With lubricant film measurements we experienced around 1 million more revolutions after first lubricant-film failure to reach total breakdown. This was measured in small high frequency bearings.

With noise measurements it is much more difficult to define a threshold good/bad. There are noisy bearings not failing and quiet bearings that are failing early.

I heard a rumor that the future of the wind-turbine-bearings will be hydrodynamic,do you agree? Or is this application for a hydrodynamic bearing - then detection of failure is easier.

You are heartily invited to a in depth discussion.

RHABE

Hello RHABE,

Thank you for your response! I have been away for a day or two so apologies for the delay in getting back to you.

I really like questions like the Hydrodynamic one as it prompts me to do some homework. I have to say that I am not an expert on these bearings and have not heard of their use on wind turbines....yet. That said, after reading about them, it sounds very interesting and I would suggest they will be found sooner than later.

My only question is their ability to deal with extreme loads during frequent stops and starts. They appear to come under some critisism under these conditions although the use of a secondary bearing apparently helps.

What about hydrostatic bearings? Where an external pump is used to maintain the fluid layer. This apparently is not affected by stops and starts.

The perferct wind turbine would never stop, technology and weather permitting. This is not the case and on low wind days numerous stops and starts will be seen, would this cause problems on the hydrostatic over time as these rely on bearing motion to suck in the fluid.

I hope this complete novice is making some sense and not rattling on about the wrong thing!

All the best

Scott

ps I have heard that ceramic bearings are finding their way onto wind turbines...

If you are doing wind turbine monitoring in England, I would have a look at B&K vibro, and SKF. They both have good wind turbine experience and are more or less local to you. CSI has made inroads into the European market, but they don't have the support there that they have in the US...

Thanks Steve S

I have a meeting with SKF this week, so will let you know how this goes. I did meet with Emerson last week and they do have a good presence here in the UK, although probably nothing like in the USA.

They have not re branded there equipment with wind, like some of the others have done and they informed me this was a company decision and not that they had not thought about it. I have to say I was really impressed with their portable equipment, so thanks for the pointer.

I should be able to make a decision after meeting SKF, who do have a dedicated 'wind' system (online). I guess this will probably be a coventional system, rebranded to make purchasers feel more confident.

Either way, it looks like either Emerson or SKF, so thanks for your thoughts on both.

All the best

Scott

I had at one time an older SKF, a CSI 2120 and a Commtest VB3000. I sold the 2120 and stayed with the VB3000. Commtest did research with wind turbines and they have files on their website. I believe it is available from Commtest files free of charge. I personally do not use the Online monitoring program. The software and instrument I started up without reading the manuals. [ I did have to eventually] It is a Windows based program that does not need special training. [I believe it is still available however] there is no fee for upgrades to software for 5 years. there is no yearly "maintenance fee". I used to be a Commtest rep for a short period of time.