Question On Aligning The Lathe Spindle And Bed Twist

David,
I'm no expert in this area, I think we are all learning as we follow along here. I have several VFD's in my shop but never encountered this before. Definitely have had belt issues over the years. Those Chinese motors they put on these machines may be slightly out of balance at that RPM causing issues, just WAG. Never know!
Ken
 
Matt at PM is sending me 2 new pulleys to try. I also have a link belt on the way, should be here tomorrow. Mat suggested I try running the belt in the "A" or high speed position at 35-40hz which would have the upper pulley running at the same RPM as it does when the belt is in the low speed position and 62hz. I did see the same harmonic vibration at about 44 hz so this seems to point to the upper pulley. It also points to the VFD/motor interaction as not the issue as the vibration occurs at different hz. Not conclusive but maybe narrowing the search. I really hope the new belt/pulleys will get this solved, I'll post an update tomorrow.
 
Yeah my guess is the upper pulley, there is a lot more rotating mass there and it did it with that HZ Test we talked about, so that just about rules out the bottom one.

I did have a long conversation with people from the factory last night. Through a translator, but still kind of got some good information. They actually suggested to change both pulleys to a different type, but some of the exact reasoning was lost in translation there. If I get any more info on that I will post it, they are supposed to email me.
 
Sounds like a mechanical issue. Have you tried to check the run-out/wobble of the pulleys? Replacement of both pulleys sounds like the best solution at the moment.

You can get some electrical noise feedback in the cabling if the speed pot cable shielding is not correct or the cable is routed near the motor cable. You can see rapid Hz flutter, the WJ200 speed noise filter eliminated it on my unit as it has a longer sampling time restricts speed changes to 0.1Hz increments.
 
Last edited:
mksj said:
Sounds like a mechanical issue. Have you tried to check the run-out/wobble of the pulleys? Replacement of both pulleys sounds like the best solution at the moment.

You can get some electrical noise feedback in the cabling if the speed pot cable shielding is not correct or the cable is routed near the motor cable. You can see rapid Hz flutter, the WJ200 speed noise filter eliminated it on my unit as it has a longer sampling time restricts speed changes to 0.1Hz increments.
Click to expand...

"speed noise filter" Is this one of the settings?
 
A016 (analog input filter) is set to 31. Prevents speed fluctuations from the analog speed input due to transient electrical noise impulses. As mentioned you should be able to set the Hz/speed, the display reading should not change. Hz will be shown to 0.1 increment. It is also possible to have the display show RPM which is extrapolate (29xHz). With a RPM meter on the spindle, for the most part the spindle speed is rock steady. You can see the spindle RPM flicker between two values, which is a sampling error of the RPM system. Interestingly, with the sensorless vector setting, I see no change in the RPM with increased cutting load. A credit to what a little VFD achieves.

"A016=31 is a special value. It configures the inverter to use a movable deadband feature. Initially the inverter uses the 500ms of filter time constant. Then, the deadband is employed for each subsequent average of 16 samples. The deadband works by ignoring small fluctuations in each new average: less than ±0.1Hz change. When a 30-sample average exceeds this deadband, then the inverter applies that average to the output frequency reference, and it also becomes the new deadband comparison point for subsequent sample averages."
A0 16 setting.jpg
 
mksj said:
You can get some electrical noise feedback in the cabling if the speed pot cable shielding is not correct or the cable is routed near the motor cable.
Click to expand...
Mark, when I first looked at Dan's lathe vibration, it would kind of come and go - almost like a stutter, so I thought initially it was some kind of electrical noise issue with the VFD making the motor stumble. Then he took off the motor pulley, put it back on, retightened the belt and the behavior changed just a bit - the onset of the vibration moved up in VFD frequency from about 62 to 64 Hz and smoothed out to a more consistent vibration at that frequency rather than coming and going as it had before. Of course, in the process of taking off the pulley, putting it back on and adjusting belt tensions, the wiring harness that runs through that area was repositioned. I also observed the outside of the pulleys were wobbling, but the belt tracking looked good, and Dan's run-out test on the pulley inside surface where the belt runs was showing only 0.003" runout. It's possible he has both a pulley balance issue going on as well as some noise being induced into the wiring harness that's causing the VFD to shimmy the output causing the motor to stumble on an intermittent basis.

David
 
Tried the link belt this morning. Pretty disappointed at the results. The resonance peak moved down to about 54hz and was worse if anything. There were some paint drips and build up on the inner v groove of the upper pulley so I ran it with a drill and sanded the paint off. Didn't make a difference. I removed the pulley and checked the shaft, zero runout. I chucked the pulley in my other lathe's 3 jaw and checked the v groove surfaces. The inner/larger v groove had a series of high and low areas (4 I think) that were about 2" long and .003" to .004" of indicator movement. Nothing else really jumped out at me.
 
Did you ever try without the spindle turning, gearbox in neutral? I'm still thinking input shaft.
 
You must log in or register to reply here.
Back
Top