Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment   Page 2 of 2      Prev   1   2
fburgos

Sr. Member
Registered:
Posts: 670
Reply with quote  #16 
Quote:
Originally Posted by Danny Harvey
Ferdy,

I'm not sure what you are asking. I delete the old AP and AL sets and replace them with my own. I always start mine at 100 and just delete anything below that after reassigning the sets. I have a database I call Source and I use Dbutly to copy my AP and AL sets into a new database when I build one. It's important to leave the old ones in place while you reassign them or you will get error messages for every point.

The wizard will add new ones as you configure machines but the same process works then too.  It's just a little housekeeping.

i use a simple designation of something like 120000/75/5 for a Velocity reading with a 120,000 cpm fmax, 75 cpm bin width and 5 averages and 120000/75/1/2000 for the corresponding PV with a 2000 hz hp filter. The velocity sets are always even and the PV's are always odd. That helps me keep them straight when I can't see the description, just the number. 

Mine are based on the discrete fmax settings of the 2120. The 2130 had lots more but still used discrete fmax settings.  I'm not sure about the 2140 but I suspect that they have joined the rest of the world and use what ever number you tell it. I never saw much reason to expand it even when we went to 2130's.  I usually start low with something like a 1200 fmax and make two different ones for velocity and PeakVue with two different bin widths.  Then I almost always use the smaller bin width except when it gets below about 12000 fmax and it starts taking up real time.

This plan keeps it to about 50-60 sets and by starting with number 100 it is easy to tell which are mine and which are the Wizards.

Using this method, it is really fast to use the Modify MP utility that is built into Dbase to make the changes on the machines after they have been configured in the Wizard. After practice, about 30 seconds per machine or so. Some take a little more thought, though.

Then after making the changes, just delete any AP and AL sets under 100 although there's nothing wrong with leaving them there. I just like to know for sure that they are gone and no MP's are using them. Plus if you missed any, it will tell you when you delete a set that is in use.

Please remember that this is based on my use of RBMWare up to Version 5.42 and up to the 2130. Things have changed and there may be a better way or maybe computer code writers have become better vibration analysts and have written good AP and AL sets of their own.[wink]


im planing to replace all the "default AP set" with "customs AP set", for example each set with will have 10 PA (10 trends/parameters)

pk-pk wave
crest
sub-harmonics
1x
2x
3x
3-10x
10x-45x
45x-90xFmax
1k-20Kz

I have been taking masurements with "default AP set" and every AP has its own set of trend/parameter, then if I replace "Default AP set" with a "custom AP Set", what happen each trend?
Walt Strong

Sr. Member
Registered:
Posts: 889
Reply with quote  #17 
I think you will get trend plots and alarms, but they won't have much meaning. I think the trend is using Parameter-1, Parameter-2, etc. place-holders with the name and definition you provide. Similarly the Alarm Set is linked by Parameter place-holders. I could be wrong, so my best advice is to backup the database and try your scheme on one machine and possibly only one measurement point. I like your new parameters, but you might have to live with the consequences of the change!

Walt


Danny Harvey

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,402
Reply with quote  #18 
Walt is right.  Trends will change.

But if they were less than desirable, the change is for the better.

For the short term, I usually take a look at every spectrum and try to rationalize the changes based on whatever changes were made in the AP set.

As far as AL sets, I make them to correspond to the AP sets.  That way AP set 100 will use AL set 100, etc. Set Alerts and Alarms for each parameter and after a few times through, you should have a good grip on things.
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #19 
The problem comes if you have collected lots of data on a machine and the change or modify the AP set. Suppose you have a set with 6 parameters. Every time you collect a route point, those 6 parameters are calculated and stored (along with the overall value). Suppose you decide to change parameter 6 in your AP set. The next time you collect data, parameter 6 will be calculated differently, but the data will still be stored in the same location. When you do a trend plot of parameter 6, you may have a sudden change. There will be nothing in the database indicating when the parameter was changed, or how it was previously calculated. This could create chaos.

What I do for new machines where I am unsure of the parameters I will need, is create a basic parameter set with those I *know* I will want (sub-synchronous, 1X, 2X, 3X-10X, max peak acceleration, crest factor). Then I can add additional parameters as I need to, without affecting the trend data I’ve collected before I changed the AP set.

If you really want to change the AP sets, you could always delete just the “trend data” and keep any spectra and waveforms you’ve collected. Not sure if this deletes the overall values - you might end up with no trends at all.

__________________
"The trend is your friend"
Danny Harvey

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,402
Reply with quote  #20 
If you are changing the AP set, I would presume that you are changing it because the set in use isn't giving you the information you need.

I'd rather have a bump in my trend than a trend based on something that doesn't give you what you need.
fburgos

Sr. Member
Registered:
Posts: 670
Reply with quote  #21 
today did some some test

1. Data manager: Deleting trend, also delete overall.
2. data base utility: exporting without trends also skips overall trend.
3.  Manual data entry: deleting trend deletes overall, but you can edit overall or other parameters but only by point


cant find a way to delete only some parameters trends I belive ill start with no trends at all
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #22 
"I'd rather have a bump in my trend than a trend based on something that doesn't give you what you need."  Danny, I agree, but if the guy who changes the setup suddenly disappears for some reason, there could be a lot of confusion for those left behind (unless you document database changes much, much better that I do).

Ferdy, fwiw I agree with your test results.

If you have a parameter set with say 10 parameters, and you like the first 8 (for example), you can just change the number of parameters in the parameter set from 10 to 8, and your data will no longer "show" parameters 9 & 10 in the plots.  But the old data will still be "attached" to the point. There is no way that I can find to *not* copy that data if that point or machine is copied to a new machine.  I assume that going forward, parameters 9 & 10 would not be calculated, but if you were to ever change the parameter set from 8 back to 10 parameters, then the old data will still be there when you plot it, even if you change parameters 9 & 10 to something other than the original setup.


__________________
"The trend is your friend"
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #23 
If you're a CSI user, and "standard" AP sets appeal to you, you might look at the discussion of "Building a Database from a Master" in another thread (scroll toward the bottom - but read the rest of the thread as well).
__________________
"The trend is your friend"
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.