What's new
What's new

VFD on a lathe: Fine at 20Hz, no torque at 40Hz+?

MauledByBears

Plastic
Joined
Apr 27, 2017
Hi,

I recently installed a VFD to convert 110V 1ph to 220V 3ph for a 10" lathe.

I gave it 10% torque compensation at the low end so I can get it started and run at low speeds (20Hz is the minimum), and it starts great and runs with plenty of torque. As I increase the output frequency, I get pretty normal behavior (speed and torque increase) until about 40-45Hz where the motor gets a little louder, the spindle decreases in speed, and I get so little torque I can stall the spindle with my hand. This effect is exacerbated up though 70Hz which is the max frequency I set.

This freaked me out, so I immediately cut power. I didn't want to let it run long enough to trip an error on the VFD and risk burning out the motor. After sleeping on it, I'm guessing that I have a voltage drop at the 110V outlet and that the increased current draw at 40Hz+ is causing a drop in voltage at the input, and a proportional drop at the output, resulting in the poor torque and speed outputs. (The electrician who wired my house was, I'm convinced, actually two children stacked inside a trench-coat with a pair of lineman's pliers).

When I get home I'll be able to test my theory with a DMM across the VFD terminals, but does anyone have any ideas in the meantime for things to check? I'd rather not let the VFD fault to get a diagnosis, I just bought it and the risk of letting out the magic blue smoke is making me wary.

Thanks!

Sam
 
There are many VFDs available that have a "voltage doubling" input setup.

But without any more information than is given in the post, not much to say. Need the VFD make/model, and motor data. A photo of the motor data plate would be best for the motor data.

Also, there is a sub-forum for this sort of question.
 
Thanks for the replies! Sorry I didn't post in the more specific sub-forum, I couldn't find it. :D

The drive is a TECO FM50-101-C. It steps up the voltage from 110V-120V 1ph to 220V-240V 3ph, the manual can be found here: http://manuals.chudov.com/TECO-FM50-Manual.pdf

The motor is a Rockwell 3/4HP motor, 220V, 3ph, 1725RPM and 3A at full load. I believe it has a service factor of 1.25. I can take a picture and post one soon.

That's a good idea about the carrier freqency. I believe this comes from the factory at 8kHz, so I'll drop it to 6 or 4kHz and see what happens. For this VFD, anything above 8kHz causes the motor to lose torque at higher speeds.

As for the DC bus I think it's something like 10V? That's just for the potentiometer speed control if I remember correctly. Not sure exactly what you mean.
 
Last edited:
Sam, your theory makes sense to me. DMM test and, as others say, specifications on the motor and the VFD will help to prove, disprove, or suggest alternatives ...
 
The 3/4 HP is not normally too much for a 120V circuit, but VFDs tend to draw a higher rms current that the actual power would indicate. Still, if it is not popping the breaker, you would not suppose it is seriously overloading the circuit enough to drop major voltage. And the specs specify a 20A breaker. (they also specify 14 ga wire, which is wrong, use 12 ga)

If the VFD is not new, it may have an issue.

The ratings suggest it is being operated within its appropriate capability.
 
DC bus : soryy im used to dealing with large cnc machines where one power supply will feed several servo drives . I mean the dc volts that are switched into the motor , on a 220vac in put this would be something above 300vdc (can usually befound near the break resister terminal). I was wondering if the voltage doubler is working correctly or , as suggested above there is some fault with the unit (the voltage doubler is usually based around a couple of capacitors which can fail under load while apperently ok at other times).

Bill
 
Billtodd: I don't think I have a way to test that. Using the DMM across the output leads doesn't give me a consistent reading. So by that you mean the DC line off the recifier/smoothing caps that is fed into the IGBT? I'm not sure how I would test for that without opening up the device and I'm not keen on doing that at this point.

I put my DMM across the input leads, had about 125V when the drive was off (open circuit), and it drops to about 120V as soon as I flip the switch. When I increase the voltage, the current drops to about 115-117V at the problem frequency of around 40Hz and stays there after I keep increasing the frequency.

I've fiddled with the low-end torque settings and keep getting nothing. Changing the carrier freq all the way down to 4kHz also did nothing that I could notice.

I couldn't resist the urge to turn a little so I ran the machine a little at 25Hz and made some really light cuts on some aluminum. I noticed the motor got pretty hot - too hot to keep my hand on for more than 5 seconds. Maybe that's a symptom? I have the overtemp setting on the VFD to 71% of its rated 4.2A capacity, for 3A which is what the motor is rated for. Never got an over current alarm.
 
Your motor is probably running too slow to effectively cool itself.

Does this lathe not have a belt or gear range you can use? Ideally, you should be running it between 60 and 120 Hz, with maybe the brief occasional foray down to 30 for some really light work. At 20 hz, you've effectively reduced your motor power to 1/4 hp, barely enough to pull the hat off your head :D
 
Ah, see I thought it was the other way around, that higher frequency reduces torque and pushes excess current into the motor, making it hot. Thanks!

...Still doesn't help me get up to 60Hz without this weird behavior though :D
 
Ah, see I thought it was the other way around, that higher frequency reduces torque and pushes excess current into the motor, making it hot. Thanks!

...Still doesn't help me get up to 60Hz without this weird behavior though :D

Are you even certain your motor windings are healthy on all three phases, VFD or not?

Is it OLD enough to be at-risk of arcing/corona discharge?

Are the VFD's output side wires run directly to the motor, or are there still ANY legacy devices in between them?
 
Has the vfd got connections for a braking resistor? if so, one will be the dc bus

On many vfds the dc voltage can be read on the display using a display parameter.

That said , I'd expect a low dc bus to flag up as a fault… Is the motor voltage parameter correct? to run faster it needs more volts , hence my questions above.

Are you sure of your motor's connections? low power and hot motors suggests somethings wrong (phase loss etc.)





Billtodd: I don't think I have a way to test that. Using the DMM across the output leads doesn't give me a consistent reading. So by that you mean the DC line off the recifier/smoothing caps that is fed into the IGBT? I'm not sure how I would test for that without opening up the device and I'm not keen on doing that at this point.

I put my DMM across the input leads, had about 125V when the drive was off (open circuit), and it drops to about 120V as soon as I flip the switch. When I increase the voltage, the current drops to about 115-117V at the problem frequency of around 40Hz and stays there after I keep increasing the frequency.

I've fiddled with the low-end torque settings and keep getting nothing. Changing the carrier freq all the way down to 4kHz also did nothing that I could notice.

I couldn't resist the urge to turn a little so I ran the machine a little at 25Hz and made some really light cuts on some aluminum. I noticed the motor got pretty hot - too hot to keep my hand on for more than 5 seconds. Maybe that's a symptom? I have the overtemp setting on the VFD to 71% of its rated 4.2A capacity, for 3A which is what the motor is rated for. Never got an over current alarm.
 
Nope, VFD is new, straight from a Chinese warehouse. I checked the windings by putting my DMM across the windings on an ohm setting. Each one was in the neighborhood of 1-3 ohms which is fine.

I'm going to run a big thick extension cord from an outlet where I know I have 20A. I might be using a 15A breaker. Like I said, the electrician who did this house didn't know shit from Shinola.

Thank you all for your input, given me a lot to think about!

Billtodd I just saw your post as I posted mine. My VFD doesn't have a breaking resistor, that only comes on the beefier 240/460V versions of this drive. I checked my motor and it all looks good. I redid the wire nuts just in case I had a weird contact. Getting continuity from the VFD busbar to the windings.
 
Last edited:
Ah, see I thought it was the other way around, that higher frequency reduces torque and pushes excess current into the motor, making it hot. Thanks!

...Still doesn't help me get up to 60Hz without this weird behavior though :D

The torque is constant up to 60 hz, but that means the horsepower (the amount of metal you can remove) is proportional to the rpm up to 60hz. After you get to 60 hz, then indeed, the current is cut back to protect the motor. It will only ever produce 3/4hp maximum, not matter how much you overspeed it.

Unfortunately with lathe operation, when you want to slow down the spindle to cut larger diameter work, the torque of your motor is constant from the VFD, the very exact opposite of what you want. You need more torque to cut at a larger diameter. So gear range is still very important, or an appropriate step in a step pulley should be chosen to try to keep the motor operating nearest to 60hz.
 
The torque is constant up to 60 hz, but that means the horsepower (the amount of metal you can remove) is proportional to the rpm up to 60hz. After you get to 60 hz, then indeed, the current is cut back to protect the motor. It will only ever produce 3/4hp maximum, not matter how much you overspeed it.

Unfortunately with lathe operation, when you want to slow down the spindle to cut larger diameter work, the torque of your motor is constant from the VFD, the very exact opposite of what you want. You need more torque to cut at a larger diameter. So gear range is still very important, or an appropriate step in a step pulley should be chosen to try to keep the motor operating nearest to 60hz.


That makes a ton of sense. I'd gotten bits and pieces of the whole picture from what I've read, but it's hard to get a full explanation. Thanks!

I only have a four step pulley right now, I'll probably make myself an oversized pulley special for turning large diameter stuff to get more torque.

In other news, changing the outlet solved the problem. Not sure why I was having this problem, maybe the outlet I was using was cursed. Thank you guys for all your help and knowledge. It's appreciated!
 
In other news, changing the outlet solved the problem. Not sure why I was having this problem, maybe the outlet I was using was cursed.

Going by what I found in this house, vintage early 1970's, there are MILLIONS of outlets out there all over America that are 'cursed'.

You will find push-in wire termination on the back of most of those. The 'spring' type, worse than those tightened by screw. That can be dangerous, as they carry common household low-current loads, but overheat on higher-current due to the tiny contact area of the worst among that class of connector. The odd small electric space-heater, in another room, and several outlets down the line is possibly one of the more common culprits.

I found my 'cursed' ones as I upgraded to GFCI, thankfully done one by-one, not 'daisy chained' except in a scant few quad-outlet boxes.

On several, closest to the breaker of a chain of several outlets, the dual-terminations of the outlet had been used as the pass-through to the downline chain. Bad practice, that. The string of loads should be carried wire-to-wire, wire-nutted, with each outlet attached with 'pigtails' so it only ever has to carry its OWN local load. Byproduct is that it is easier to push back into the box, which means reduced risk of strained body or loosening of terminations.

On the worst of mine, the white wire had turned brown from long years of slow-cooking and the heat had charred the body of the outlet. That worsens the problem as springs lose tension and/or the supporting phenolic chars away from supporting them.

Separate 'project', but you might want to look inside more than just that one bad-actor of an outlet and take corrective action.
 
By my experience however GFI circuits and VFDs don't play well together. So avoid a GFI retrofit if a drive
is going to be run off that circuit.
 
By my experience however GFI circuits and VFDs don't play well together. So avoid a GFI retrofit if a drive
is going to be run off that circuit.

This is very often the case, yes.

DC Drives as well if not (at least) placed the other side of a drive isolation transformer. As-in NOT 'autotransformer' or buck/boost wired as such.

The ones I referred to were for the residence side of the wall. Only about half the shop outlets here are GFCI - the 120 VAC 1-P ones.
 








 
Back
Top