Lathe VFD Addendum: How to Tune Acceleration and Braking 

Thank you! I appreciate the feedback.

The L510 230V model doesn't support a braking resistor.

You're welcome! I'm glad you found it helpful.

Good points to consider. Let me see if I can address them. The lathe has six belt positions, for 150, 300 and 540? RPM on the back gear (belt), and 720, 1200 and 2400 RPM direct. I have the belt in the 1200 RPM position. This gives me quick access to everything from 120 RPM (6Hz) up to 2400 RPM (120Hz) with the flick of the knob. I've done some turning since finishing this video, and it's been working very well. Being able to tweak the speed while it's in the cut to get the chips to break the way I want is really game-changing. It does get cold here in the winter, but the lathe castings never get below about 40F (5C) in the shop. At that temperature, the lathe came up to speed easily at 1200 RPM, but struggled a bit on the first start of the morning at 2400. However, since I'm always in the 1200 RPM belt position now, I should have roughly double the torque most of the way up to 2400. It certainly gets there faster now than it ever did with the stock motor and the 2400 RPM belt position. Low RPMs are a different matter. For anything under 1200 RPM, while the vector drive maintains significant torque all the way down to 6Hz, it can never compare to the added torque of the belt reduction. So far, I've dropped it to 6Hz (120 RPM) to gently scrape out large chamfers with high-speed steel tools and to tap up to 1/4-28, both in mild steel. Neither of these requires much torque, and being able to flick the knob to access the slow speed has been really useful. The worst-case scenario for torque will be, as you suggest, a large-diameter steel workpiece running slowly. For carbide in steel, this would be around 8" diameter at around 400 RPM. I've never turned anything like that, but if I do, I can always swap out the belts.

Thanks for the additional information. My current lathe is a bit different with only a high and low speed belt position. It's variable speed single ph with no VFD (yet) Low speed belt position and about as low as I can go on the variable gives roughly 50-900 rpm and 900-2400 in the high position. But there's not all that much torque at the bottom speed. Certainly not as much as I'd like at it's maximum swing of 11". Even with a 2 hp vector drive I very much suspect there still won't be enough. As you say for good bottom end torque then some type of mechanical torque multiplier is needed. So it's either build some type of jack shaft arrangement or try and somehow shoehorn in the back gear set up the machine should have been built with anyway. The step pulley, back gear plus VFD on my mill works very well and gives me roughly 30 - 4500 rpm with ample torque within normal cutting conditions.

You're welcome.

I have spent a lifetime reading and learning from others. RU-vid is a great resource, and I have access to machinists and engineers at work. I'm constantly taking on projects that are just beyond my grasp and learning what I need to know to complete them.

If you kill power, the load coasts. This VFD model does not support a braking resistor, unfortunately.

I run half a second stop with the same VFD, no issues. 4-5 seconds isnt fun when threading. I actually cut stop time to .2 seconds when threading and use the jog buttons (added to my remote mount control panel) to power the lathe. I never touch the half-nut and simply spin the lathe in reverse to reset the carriage. Very handy and saves a ton of time.

I hope people who do this only have D type chucks.

The G0602 has a threaded spindle nose, but also has locks to prevent the chuck from unscrewing.

i think that's 15% above its rated power (so say it's a 1 kW motor; you'd be demanding 1.15 kW)

You can overcome that by fitting a 5C collet chuck that will give minimal inertia.....no worry about spinning a scroll and loosening the jaws.

This VFD model has no connection for an external braking resistor. The braking transistor just dissipates the energy into it's own heat sink, so it is pretty limited. Higher voltage models have the external resistor connection.

I've thought about that quite a bit. The one thing that CNC would really add to the lathe would be the ability to turn long tapers without a taper attachment or fiddly setups. I did the CNC conversion on my mill, and I'm very happy with it, but I really miss being able to turn the dials. With ballscrews on the mill, pressure on the axes will easily back-drive the non-energized motors, so I think operating it manually would be pretty tricky.

The VFD comes set with safety limits that do what you suggest. If you try to ramp it up faster than it can handle, the magnetic field in the motor will slip or the drive will detect an over-current situation and shut down. When decelerating, the VFD monitors the voltage on the DC bus and shuts down automatically when it reaches a preset threshold. When this happens, it essentially disconnects from the load and lets it coast to a stop on its own, just due to friction. For a manual lathe, this is fine. For other applications, like a heavily-loaded downhill stretch of conveyor belt, cutting it loose might get pretty ugly, but then you wouldn't be using a $150 VFD for that, either. The trigger thresholds are configurable (I played with some of them in the video) but I don't think you can turn them off completely.

The motor is running at double the rated speed, but I'm not worried because I believe (without evidence) that they use the same bearings in the 1800RPM and 3600RPM motors. The spindle of the lathe isn't actually running any faster than stock. Stock, in the fastest belt configuration, the lathe runs at 2400RPM. I have the belt in the 1200RPM position, so at 120HZ, it's turning at the same 2400. Honestly, I'd be nervous to run it much faster. Maybe with a collet chuck and small workpiece something like 3600RPM might be reasonable, but with a bit import chuck spinning, that seems awfully fast.

Welding on a lathe is generally a bad idea. Electronics like a VFD can be damaged, but if the weld current goes through the spindle bearings it will destroy them.

I have mine set up with a contactor. When I hit the E-STOP, it disconnects input power.

@@Clough42 Thank you

You're welcome.

That's news to me. I think a static phase converter has to be derated, but I'm not aware of limitations like that with VFDs. VFDs generate electrical noise and can damage motor windings if they're not rated for inverter duty, but they can also soft start the motor and vary its speed.

@@Clough42 Wow, thanks for writing back so quickly. On a video this old, I was not expecting that. Anyway, I had read that a VFD is not that different from a static phase converter in that you needed to de-rate the motor. I don't know the facts, which is why I am asking since you are more up on this than I. While I am not electronically or electrically inclined, I recall that when the discussion came up at work about wanting to retrofit VFD's to our large air handling units (100 - 200 hp motors), the motors were not VFD rated. After some looking into by our Engineering department, it was determined that motors made from the '90's on are fine with VFD's, regardless of whether they were rated for VFD usage or not. Come to think of it, I think it might have applied to high efficiency motors produced from the '90's on (?), which is what we had by then, and not the "normal" efficiency motors, but this is going back to when I started at my current job about 16 years ago, so I may be mis-remembering things. I know you're thinking that motors from the '90's should probably all be replaced by now, not exactly, motors in that size range get re-built (new bearings, not sure what else may be replaced) and re-wound (when needed) and not fully replaced until they can no longer be repaired, economically.

@@paulprobusjr.7597 Those are big motors. The largest VFD I've ever used is 3HP. I've met a lot of people with vast experience who believe all sorts of things--and probably for good reasons--but I would look to the VFD and motor manuals or consult with the supplier. Most industrial supply houses can answer technical questions about their products.

* mode to read rpm

Most VFDs have a way to configure a ratio so it can display RPM. I didn't bother because I was planning to include a tachometer in the ELS.

I don't think so. The real issue on this unit is the lack of support for an external braking resistor.

Yes. An external braking resistor does the job.

Praveen Kumar depending of what vfd you are using.

Paul Morley you can put a thermal switch, normally open, on the IGBT heatsink and wire it inline with the fan positive. Get a unit that closes at the desired temperature, which will turn the fan on. The right temp might be something like20 degrees over room temperature in your shop.

It shouldn't. I've done it several times with no ill effects. Worst-case, the back-EMF from the motor would power the drive until it stops or shuts down due to over-voltage.

If I put something big in the chuck and run into trouble, I'll adjust it. I'd like to have a brake someday, but I didn't realize this VFD didn't support it when I bought it.

@@Clough42 alright. I had a mechanical brake on my storebro gk195 and that felt nice. If something bad happens. But how come it doesnt work with a vfd?

The VFD I'm using doesn't use a braking resistor. I think the 460V models can, but I don't have one.

I have used the lathe on several projects since the conversion and I am very happy. I've done some parting, tapping, drilling and wide chamfers at 6hz with no issues. Yes, the torque is low and it's possible to stop the spindle, but in practice, it hasn't been a problem. I still need to try single-point threading. If torque becomes an issue for a particular job, I can change the belts. Being able to go from 120RPM to 2400RPM with the flick of the knob has been wonderful. At the high end, 2400RPM is already pretty fast. I could see 3500RPM someday with a collet chuck, but with a three- or four-jaw, that's a lot of energy, regardless of whether the bearings could handle it. Parting is all about rigidity, speed and lubrication. I have had pretty good results parting in aluminum and steel, but you have to be careful. Keeping a brush with cutting fluid (A9) in the groove, keeping the speed up and keeping the parting tool engaged and making a chip works pretty well in aluminum. The same goes for steel, but at lower speeds. Going too slowly can make the tool grab just as easily as going too fast can make it ring. In general, I prefer parting with the bandsaw if I can. I've even taken the chuck off the lathe and clamped it in the bandsaw with the workpiece still in it to cut it off. Parting in an eighth of an inch, bandsawing the rest and then facing often works well. Then again, parting is tricky anyway. I've blown up parting blades and even stopped the motor cold on a Hardinge HLV-H before, so there's that.

Clough42 great minds think alike - my more recent go to parting method is a Milwaukee portable band saw -haha - good point on Chuck speed-energy too.

The microcontroller uses the spindle encoder to calculate the speed in real-time. I'm making no assumptions about the spindle speed. In fact, I expect it to vary during the cut.

I just realized this was a comment on one of the VFD videos. I was assuming at the time that I could calculate the speed from the lathe speed chart, the motor plate, and the VFD frequency. In reality, it's running a little faster than that. I suspect the belt ratios are only approximate. The electronic leadscrew I'm building now has a tachometer generated from a spindle encoder, so the speed will be measured precisely.

Thanks for the clarification, I'm working through the backlog now and I kinda forgot about your new ELS... :)

The chucks that come with the G0602 lathe have locking clamps that should prevent this. I probably still wouldn't run it in reverse under load.

@@Clough42 I am curious do you know a way that you could safely run it in reverse? As that makes it basically non usable then..

No. That's a separate issue. Motor bearing pitting is probably just normal bearing damage, or common mode electrical current that should be blocked with a big toroidal choke.

@@Clough42 why not just use a grounding ring to ground the motor shaft to the frame of the motor ?

That would be a good solution, probably best used in combination with a filter. Filter toroids are available for about $20. The grounding rings I've seen cost ten times that.

This particular VFD has no resistor and has no terminals to connect one. The braking energy is very limited because it's dumped through the transistor into the heat sink in this model.

@@Clough42 Actually it seems yours doesn’t even have a braking transistor and they left the setting in the firmware anyway. Here’s a quote from the quick start manual: “Notes: *1: Models 460V 1 ~ 3 HP ratings have a built-in braking transistor. To use this braking transistor a braking resistor can be connected between P and BR.”

@@mman454 looking at it again, I think you're right. I hadn't noticed the diode in the diagram. This model appears to have only DC injection braking.

@@Clough42 I mean, it will still do 20% braking through the IGBTs that it uses to run the motor. So you aren’t wrong in saying that it is doing some regenerative braking and dumping the energy into the heat sink.

Yup. I also have it tuned to stop as fast as possible, but this VFD doesn't have a braking resistor terminal, so braking capacity is very limited.