Hi guys and seasons Greetings to all and I hope 2007 is the best year ever. Are there any guys out there who are taking vibration readings on a hoist drum on P&H shovels.Or is there anybody who can make any suggestions on taking readings on this machine.It is a very low speed,you have limited time,the speed is not constant either.I am using Odyssey software and a Data Pac.Thank you for any information anyone can supply.

Keith,

I used to do draglines and shovels. The crowd is the hardest to get. Time Wave Form is about the only thing you have time to do. You can get a spike energy reading, but you have to only take the spectrum, or set them up as two different points (overall one time, spectrum the next). I used to use two different collectors.
The easiest thing to do is punch the store button BEFORE it reaches speed. Let it range while the stick is coming up.
Also have the operator load the dipper while your running the collection.
For 300 RPM, 80,000 with 1600 lines will get you about 6 revs of the drum. Will take about 1.18 Sec's for collection time. If you've already got the box in the mode (punch button early)you can get the data. Spike is harder to get on the shovels than on the drags.

Dave

A fellow by the name of Rich Pratt used to visit the Reliability-Magazine website and he worked for P and H.

I havn't heard anything from him for quite a while, but I think he may be with CSI now.

Maybe someone will get word to him that you could use his help.

I helped set up a program at the recently closed Trans Alta mine in Centraila, WA on BE draglines and shovels. The gentleman running the program had a fantastic track record of identifying faults, even getting to the point of being able to accurately determine to the shift how long the machines would operate with some types of failures. It's not an easy task, and some of the traditional ideas go out the window to make it work.

I can talk specifically about the draglines, but the same ideas apply to shovels as well. Data acquistion time is limited on the hoists, and the speed is constantly changing (as the cable winds around the drum, the load decreases because less cable is being lifted). The solution ro both of these problems is LESS resolution. High take a longer time to acquire, and, since the speed is constantly changing, no peaks show up. Taking 200 line data speeds up acquisition and since the bin-width is wider, the data isn't smeared as much. The results are still "humps" instead of peaks but with a little effort, problems can be seen.

You need to look carefully at all the data. There have been some interesting findings. A vurious one is a peak around 45 Hz, corresponding to nothing in the hoist and drag gearboxes. When this peak appears, there is a crack in one of the gear teeth. This might be a resonance of the gear or support structure but combing through the data after the first broken gear tooth found this identifying characteristic.

Jon
Spintelligent Labs