The point I'm measuring is output shaft of planetary g/box that rotates at approx 5rpm, if you notice the highlghted areas in one of my following attachments it says rpm is much faster and waveform contains approx 200 revolutions, I know that its more like 2 revs,is it possible there are impacts or individual teeth meshing that the analyser is pickin up as a complete revolution? could i trouble you guys to see if my set up is wrong?

ap_set_37.doc (66 Kb, 72 downloads)

heres more info

press_G4P.doc (67 Kb, 59 downloads)

havent got the hang of doing multiple attachments

press_drive.rtf (43 Kb, 52 downloads)

Jonesy,

I can think of two possibilities:

Either the machine set up is incorrect (maybe a wrong reference speed), or if you have any of the "Locate RPM of the Spectra if" buttons ticked (under plot data globals)

Is this a variable speed machine?

If you can post the details of the machine set-up, it would probably be helpful.

Ian

You said its 5 RPM. also i see Press section. are yu sure its not 5 HZ which wouild be 300 rpm
I have a pres section with planetery gearboxes and the output is around 5 HZ. As far as planetery gearboxes go I hate them lol...
Very dificult to analyze...

Mike

If this is on a paper machine (Press Section) and it is variable speed, what have you set the referenece RPM as, and what type.

What does the meter ask for when you first press "Enter" on the first point of the route or machine, rpm or fpm? Or does it even ask you anything? What is the fpm to rpm factor?

What does the other points on the gearbox read as far as rpm and number of revs goes,correct or incorrect?

My guess is the machine is setup with the referenece speed as 60 rpm and then does not ask for this speed on the first point, so all points may be defaulting to 60 rpm.

If the machine is setup as "RPM", how do you determine the speed to enter, if the meter asks for the speed?

Here is an example of what I am asking.

Ralph's suggestion is a good one.

I would add that you should check for conflicting information between dbase and the configuration in Wizard (if any).

A special twf to 80 orders might not be a good set-up either since you don't need it with PeakVue and if it is variable speed, it could change your fmax and hp filter.

A few answers to your Q's
Ian S
previously unlocated is ticked
Mike
Its definitely 5 rpm not HZ
Ralph
I enter rpm determined by taking a reading at motor then plot data to find 1 X and use the cpm figure in the bottom left corner (2130 using RBMware)
Danny
what do you suggest in regards to your comment about 80 orders not being a good setup with STW

G4P_info.doc (64 Kb, 17 downloads)

Jonesy,

ALL of the boxes in the "Locate RPM of the Spectra if" option should be un-ticked.

You have a machine reference speed of 2250 RPM, and an output shaft point speed of 5 RPM (implies a gearbox ratio of 450:1)

It's a variable speed machine, and your speed type code is RPM.

The speed that the output shaft will register as will be the input speed you enter, divided by 450. But I'm unsure of the effect of having any of the "Locate RPM..." boxes ticked.

My opinion is that you should NOT collect a special time waveform with peakvue unless there is a particular reason to.

If you still can't get it to work, you can email me at ian dot seward at paperlinx dot com dot au, and we can go through your set up in more detail over the phone, if you like.


Ian

quote:

Ralph
I enter rpm determined by taking a reading at motor then plot data to find 1 X and use the cpm figure in the bottom left corner (2130 using RBMware)

You do not enter the speed at the 1st point before taking data? You wait until you get back to the computer and correct the speed? This is ok to a certain degree but using the "locate speed" option in globals is not the correct way to go, IMO. It looks at something "close" to what it thinks is running speed and it may be WAYYYYY off. I never use this option.

With the "machine" setup as "rpm" I am thinking the factor should be "1" rather than "3.808". This odd or fractional factor is for FPM or MPM not RPM. The meter and software uses this factor to calculate rpm from fpm or mpm.

Only my opinion and may not fit your analysis technique.

Hi Jonesy,

I'm pretty much in agreement with most of what has been said above. "Locate RPM of the Spectra if" is definitely a big trap. The RBMWare default software installation sets this so beware, whenever upgrading your version, need to untick everything again!!

I have a lot of experience on large single / double stage epicyclics. If you and the others can't figure it out, I would be happy to have a look as well. Send me the *.rbm file - no promises! Am in Melbourne. info at commotionsystems dot com dot au. Rgds.

quote:

With the "machine" setup as "rpm" I am thinking the factor should be "1" rather than "3.808

If the speed type code is RPM, then the value in the FPM to RPM factor is ignored (I think).

Thanks Ian & Vibe-rater
I'm going to plug away and change some settings one at a time according to all the suggestions in this thread, and see what happens, if no improvements I'll take you up on your generous offers,
thanks to all.

Jonesy.

quote:

IanS quote
If the speed type code is RPM, then the value in the FPM to RPM factor is ignored (I think).

You may be correct. Have never entered anything but "1" here for rpm. The "3.808" might be from copying the machine of one which was set at this number. Will have to experiment and see what happens, ignore or not. Sounds logical for it to be ignored.

Jonesy,

What does this thing drive to go from 2000+ down to 5 rpm? Some sort of oscillating doctor blade?

Ralph,
Its a DC motor driving 452:1 planetary box direct coupled to 2.4mtr drum on a seimplekamp medium density fibre board press.
at this stage I've unticked the "locate rpm if" box and the STW box, in ap set, but final rpm is still way out.

Jonesy,

I setup 2 points, motor with a speed of 2250 rpm and the gearbox with a speed of 5 rpm and collected data by entering 2250 when the meter asked for the speed and the results were as shown here, ok.

Can't understand what is happening on yours but the setup I used worked perfectly. If you are waiting until you collect data to set the motor speed this is not going to affect the output speed. You will have to ratio the true input speed to get the output speed. So the best way to collect the data, IMO, is to enter the correct speed in the meter when collecting the data. Ratioing after collecting is ok if done correctly.

Hope this helps.

Motor Spectrum

Ralph,
when I collect data I have a "dummy" point that i use to get the speed and then enter rpm when asked for on motor, I have all the g,box shaft speeds referenced so the 2130 should calculate this, I admit my method for getting speed isnt entirely accurate but within 10%, when reduced by 450:1 this wouldnt be a 1 rpm difference at the output shaft, so I'm still scratching my head.

thanks
Jonesy.

quote:

Ralph,
when I collect data I have a "dummy" point that i use to get the speed and then enter rpm when asked for on motor, I have all the g,box shaft speeds referenced so the 2130 should calculate this, I admit my method for getting speed isnt entirely accurate but within 10%, when reduced by 450:1 this wouldnt be a 1 rpm difference at the output shaft a 10% error could very well be a large error on a slow turning shaft. I have seen where 5 to 10 feet per minute error on a 600 fpm speed would chane the running speed on a 30 rpm shaft enough to make the gearmesh fault and the bearings and running speed be "WAY" off, so I'm still scratching my head.

thanks
Jonesy.

Hmmm! A "dummy" point? 10% is good for higher speeds but bad for slow speeds.
What is the speed when the dummy point is used? Is it close to the 2250 machine speed setup? It really doesn't have to be close, but just wondering. If you correct the speed on the motor after downloading the data, are you correcting it on the gearbox also at the correct ratio? I am sure this goes without saying but I am sure you know that if the rpm you put in the motor is not 2250 that the gearbox is not going to be 5 rpm when calculated by the software and meter. (Dumb remark Ralph)

Why not extract out the machine into a small database and email it to me and I will run it through my meter and see what I get and maybe both of us together can solve this strange problem.

Are you familiar with using the DButly program for extracting? If not, I have an MSWORD DOC help page I have typed up for some clients and friends to use when sending me data. If you like, I can email you a copy of it so you can email the machine in its own small database.

ralph_stewart@alertanalytical.com

or
info@alertanalytical.com

put attention Ralph Stewart in the email if you use the second address

Jonesy,

Did you resolve this issue?

Ian