Proximity vibration sensor

hussam295,

There are NOT any additional proximity sensors on the unit--in addition to the -X and -Y proximity sensors on each turbine bearing, and the ones on the generator (methinks there is something amiss with the generator proximitors....).

To gather the required information for vibration analysis (one of the really excellent features of proximity vibration sensors and systems), one needs to have two (2) proximity vibration sensors on each bearing, located 90 degrees from each other. To get this 90 degrees of separation it is easiest to place the sensors through the bearing housing at the 10:30 o'clock and 1:30 o'clock positions (which are separated by 90 degrees!). Placing them at 12:00 o'clock and 3 o'clock or 9:00 o'clock and 12:00 o'clock would place one of the proximity sensors at the bearing split/joint--which would be difficult (though not impossible)--and would make physical access to at least one of the "probes" (as proximity vibrations sensors are often called) for adjustment and replacement more difficult. So, since the only real requirement is 90 degrees of separation, they can be placed at any positions--as long as they are separated by 90 degrees.

I am NO vibration expert (which is something of a "black art"--which means there is a LOT of hands-on experience, some training and a lot of "gut" feel (instincts) involved in vibration analysis, though with the data and analysis tools available these days and with improvements in data and software the "gut" feel aspect is diminishing more and more, though it can still be a VERY important aspect of troubleshooting a difficult vibration problem. Anyway, The other "sensors" you are referring to are signals which are related to multiples of the vibration (1X means at the actual speed of the shaft; 2X means at double the speed of the shaft)--which are useful when analyzing higher than normal vibrations, as well as determining the location of the cause of the vibration (if it's a balance or out-of-balance) condition, for example.

The phase angle is another part of the detailed information that can be obtained from proximity vibration sensors, and helps with identifying the source of the vibration and with correcting it (for example, with a "balance shot"--placing a weight on the shaft at a proper location and of a proper size to nullify the effects of some cause of vibration (usually an out-of-balance condition).

The keyphasor, as was aptly explained in the URL provided in the previous thread you started about this topic, is for measuring speed for the proximity vibration sensors. It's just another speed sensor, but it's one that is required for the proximity vibration sensors to work correctly (see the article).

As an operator, most of this information is not useful at all--except for capturing screen prints when vibration increases to share with the Instrumentation & Controls and/or the Mechanical Departments. As an I&C technician or mechanic trained or experienced in vibration analysis, the information on the display is critical to analyzing the problem. But, for operators--it's "clutter." UNLESS the company provides training about how to interpret the information and what to do when the information indicates an impending problem.

Now that GE has assimilated Bently-Nevada, they are able to offer the information you see on the HMI display as an option to Customers. (Prior to purchasing Bently-Nevada, GE would have had to pay a fee to Bently-Nevada to offer the data on the HMI display.) Having the information on the HMI display gives a very quick way to gather and send information to an analyst, without having to have additional monitoring equipment.

There are LOTS of YouTube videos on vibration analysis using proximity sensors (probes; pick-ups), and there are LOTS of documents available on the World Wide Web. In the many years I have been contributing to Control.com, I haven't seen a lot of people with detailed knowledge of and experience with vibration analysis willing to share their information and knowledge and experience. (The information and experience and knowledge is very valuable, and some people make their living for themselves and their families performing this analysis, and so are probably not so willing or anxious to share detailed information--not that you're looking for detailed information.)

In the Operations & Service Manuals provided with the turbine-generator and auxiliaries, you should find some Bently-Nevada manuals which would provide some good information about the equipment, though sometimes it's very technical information and not very useful for someone just getting started without a mentor or experienced person to discuss the information with.

Lastly, ToolboxST lists all of the inputs (sensors and probes and pick-ups and transmitters and switches and such) connected to the Mark VIe. You can look in the GT configuration file for detailed information about the proximity (and seismic) inputs (number; scaling; options; etc.).
 
Hussam 295,

I guess some aspect/parameters of the vibration monitoring shown on this screenview are not clear .

For example:

Max vib is shown on right top case "Turbine data" equal to 13.2 mm/s , whereas it is displayed 18 on 32Y .

Also Key phasor 77RP11, is showing different value (42rpm) from Turbine data ( 2989.6rpm....).

Do you know how much is radial vib sensor average gap for 31X & 32Y ? as it is not displayed on this screen view, maybe there a reason for that.


I would suggest you to check these datas, by this way you can get a better approach and understanding on vibration monitoring process .
Also they are interesting document on web describing vibration monitoring process , you can get lot of crucial informations.

Is the(se) unit(s) has been commissionned for long time??

I am bit confused after seeing some displayed datas on this screen!

Please give us some clarifications so we can support you!

Controlsguy25
 
Controlsguy25,

When I worked on the Mark IV (which had a black-and-white, text-based CRT and "soft-switches for an operator interface) I would occasionally have to change a few rungs and Control Constants (well, maybe more than just occasionally, but it wasn't ever necessary to make a lot of changes).

When the Mark V first came on the market and started shipping with new GE-design heavy duty gas turbines, the operator interface was a PC running a proprietary disk-based, multi-tasking operating system, IDOS, which ran MS-DOS as a scheduled task and used text fields to display turbine operating data. It was crude, but adequate, and I still had to change a few rungs and Control Constants on most start-ups, but the displays on the operator interface (called an <I>) were pretty good.

Then, the operator interfaces (still running disk-based OS's) started having color graphics on the displays (screens) which were configured using text-based files executed by an animation program. It was crude, but adequate--and people (more or less) liked it because it was graphical and had colors (16 different colors, I think!). But, that's when the problems started.... There was NEVER a standard for displays and that meant that every requisition engineer (the factory person who configured a Mark V for a particular application) could make the displays be whatever he/she deemed appropriate. There was no oversight or review. And, there was never any check of any display to make sure they were correctly configured, displayed the proper values, and all the "targets" buttons worked.

Then I, and other commissioning personnel (we were called start-up engineers then) began spending more time--a LOT more time--fixing problems with displays. AND, Customers started demanding customized displays (which was a whole 'nother friggin' story in itself). The problems with rungs and Control Constants were still there, and over time the problems with Control Constants actually worsened, the problems with rungs actually got better (in my personal opinion)--but the problems with displays continued to get worse and worse and worse.

Start-up engineers, who usually had a lot of spare time on their hands which they never properly used or understood, started spending a lot of that time changing and modifying and "improving" displays and graphics. To the point that sometimes when we went to some sites the displays were completely unlike anything we had ever seen before, and quite often we were sent to site to fix problems with the displays.

THEN, along came CIMPLICITY. And the problems with displays got worse and worse. And the number of start-up personnel who modified displays to their own whims and for Customers increased. The displays which came from the factory were often almost unusable and had to be extensively modified (to add Peak Load, for example) or enable a second fuel, or make the buttons for Off- or On-Line Water Washing work, and so on and so on and so on and .....

SO, it doesn't surprise me that the display hassam295 photographed and attached has imperfect and incomplete data along with incorrect display values.

Anyway, that's kind of a little bit of the history of how the graphical operator interfaces for GE-design heavy duty gas turbine Mark* turbine control systems evolved. The tip of the iceberg, actually. But, the point is--it should be no surprise that Mark* CIMPLICITY displays have problems. Worse, the operators at the plants where the displays are used to operate the units--and their supervisors--allowed the commissioning personnel to leave site with all the display problems. And, the commissioning personnel didn't identify the problems for the factory personnel to fix before they left site.

I have learned (as should you) not to make too much of the displays we get to see on Control.com. Try to stick to the question at hand. You can mention that this or that doesn't appear to be working, but, generally the poster doesn't know there's a huge problem and there's no one at site who can fix the problem and the commissioning personnel are long gone and the warranty has expired.

I believe most of the questions we receive here about Mark* turbine control systems and GE-design heavy duty gas turbine operation and control are from people who didn't get the training which was provided when the units were new. They were hired recently, or promoted to a new position working with the control system, and they are trying to learn what they can--and aren't getting a lot of help from the more experienced people at site. OR from their supervisors or management. We just try to do what we can to help.

The most worrisome thing about your remarks is the keyphasor value being so different from the TNH value (converted to RPM). If that is, indeed, the value the Mark VIe is using for the proximity values, well, then the proximity values are WAY off. It would be necessary to be able to know what the value in the Mark VIe for the keyphasor input is--because I venture the scaling for the CIMPLICITY (or PROFICY) display is probably wrong....
 
Controlsguy25,

When I worked on the Mark IV (which had a black-and-white, text-based CRT and "soft-switches) for an operator interface I would occasionally have to change a few rungs and Control Constants (well, maybe more than just occasionally, but it wasn't ever necessary to make a lot of changes).

When the Mark V first came on the market and started shipping with new GE-design heavy duty gas turbines, the operator interface was a PC running a proprietary disk-based, multi-tasking operating system, IDOS, which ran MS-DOS as a scheduled task and used text fields to display turbine operating data. It was crude, but adequate, and I still had to change a few rungs and Control Constants on most start-ups, but the displays on the operator interface (called an <I>) were pretty good.

Then, the operator interfaces (still running disk-based OS's) started having color graphics on the displays (screens) which were configured using text-based files executed by an animation program. It was crude, but adequate--and people (more or less) liked it because it was graphical and had colors (16 different colors, I think!). But, that's when the problems started.... There was NEVER a standard for displays and that meant that every requisition engineer (the factory person who configured a Mark V for a particular application) could make the displays be whatever he/she deemed appropriate. There was no oversight or review. And, there was never any check of any display to make sure they were correctly configured, displayed the proper values, and all the "targets" buttons worked.

Then I, and other commissioning personnel (we were called start-up engineers then) began spending more time--a LOT more time--fixing problems with displays. AND, Customers started demanding customized displays (which was a whole 'nother friggin' story in itself). The problems with rungs and Control Constants were still there, and over time the problems with Control Constants actually worsened, the problems with rungs actually got better (in my personal opinion)--but the problems with displays continued to get worse and worse and worse.

Start-up engineers, who usually had a lot of spare time on their hands which they never properly used or understood, started spending a lot of that time changing and modifying and "improving" displays and graphics. To the point that sometimes when we went to some sites the displays were completely unlike anything we had ever seen before, and quite often we were sent to site to fix problems with the displays.

THEN, along came CIMPLICITY. And the problems with displays got worse and worse. And the number of start-up personnel who modified displays to their own whims and for Customers increased. The displays which came from the factory were often almost unusable and had to be extensively modified (to add Peak Load, for example) or enable a second fuel, or make the buttons for Off- or On-Line Water Washing work, and so on and so on and so on and .....

SO, it doesn't surprise me that the display hassam295 photographed and attached has imperfect and incomplete data along with incorrect display values.

Anyway, that's kind of a little bit of the history of how the graphical operator interfaces for GE-design heavy duty gas turbine Mark* turbine control systems evolved. The tip of the iceberg, actually. But, the point is--it should be no surprise that Mark* CIMPLICITY displays have problems. Worse, the operators at the plants where the displays are used to operate the units--and their supervisors--allowed the commissioning personnel to leave site with all the display problems. And, the commissioning personnel didn't identify the problems for the factory personnel to fix before they left site.

I have learned (as should you) not to make too much of the displays we get to see on Control.com. Try to stick to the question at hand. You can mention that this or that doesn't appear to be working, but, generally the poster doesn't know there's a huge problem and there's no one at site who can fix the problem and the commissioning personnel are long gone and the warranty has expired.

I believe most of the questions we receive here about Mark* turbine control systems and GE-design heavy duty gas turbine operation and control are from people who didn't get the training which was provided when the units were new. They were hired recently, or promoted to a new position working with the control system, and they are trying to learn what they can--and aren't getting a lot of help from the more experienced people at site. OR from their supervisors or management. We just try to do what we can to help.

The most worrisome thing about your remarks is the keyphasor value being so different from the TNH value (converted to RPM). If that is, indeed, the value the Mark VIe is using for the proximity values, well, then the proximity values are WAY off. It would be necessary to be able to know what the value in the Mark VIe for the keyphasor input is--because I venture the scaling for the CIMPLICITY (or PROFICY) display is probably wrong....
CSA,


I agree with your statements on how graphical operator interface for GE HDGT Mark* evolved.
I Also have been faced as lot of field engineers to various issues as you explained.

You saying that i should do "not too much on the displays we get to see on Control.com and Try to stick to the question at hand."
Ok you and i understood that , but as a Technical advisor I can't see such thing and leave it like it is without mentioning it to the original poster, I mean I come to this great forum to share various experiences/expertises from Engineers or other staff personnel.

You abloslutely right by saying that "I believe most of the questions we receive here about Mark* turbine control systems and GE-design heavy duty gas turbine operation and control are from people who didn't get the training which was provided when the units were new. They were hired recently, or promoted to a new position working with the control system, and they are trying to learn what they can--and aren't getting a lot of help from the more experienced people at site. OR from their supervisors or management. We just try to do what we can to help.

I clearly understood your message and thank you for the advises.

We all here have a good opportunity to extend our circles of competencies/skills /know how , and that is a wonderful thing !

After saying that , It is now the turn to Hussam295 to write us some answers/clarifications, on the questions I asked him, so we can try to support him as much as we can from this forum .

With my kind regards,
Controlsguy25
 
Dear CSA/ Control Guy/ Hussam295

Thanks for good explanation on the thread. I just want to add and may be repeat some points to complete my comment.

Radial vibration is measured for HD Gas Turbine for bearing# 1, 2 & 3.For generator @ DE and NDE. The screen shared is. Only sharing as generator one radial and no mention of DE or NDE.The Thread initiator is required to go through the P&I D to better understanding the vibration sensors physical location including the key phasor sensor. The sensor gap can be measured at Bently Panel and can be converted to distance by 7.87V= 1mm. As long as the gap voltage is in between-8V to -12V the sensor can work fine.

I believe that the threat initiator is not much Familiar with the vibration or key phasor. The key phasor senses the speed based on the pulse. At the sensing point there is a notch on the shaft. During the rotation of the shaft, the notch gives pulse to key phasor & key phasor calculate the speed. Usually keep phasor is connected with generator.

The turbine TNH speed is measured and calculated through Magnetic speed pick up sensors.

This is confusing the key phasor is Only showing 44 RPM and speed pick up sensor is showing around 2990 RPM. This can be sorted out from Bently Monitor software. Toolbox st is receiving values from Bently most Cases through modbus.
Please look at P&ID for location of key phasor, look at instrument list for the range. It will give idea what speed is sensing by this key phasor. Then check the Bently panel if you have access. Then check the tag of the key phasor at Toolbox what raw value receiving.

Other factor used in the graphics for the vibration analysis and more handy to the analysers incase of any vibration issue arise.
Operator the simple is follow the normal and alarm ranges.

The graphics of the complicity is depending on who supplied. GE oil & gas (currently Baker Hughes) or GE power & water or third party supplies.Different group uses wayof building the graphics in cimplicity which some time a problem to have consistency.

it is time for Hussam95 to give feedback.
 
Top