VFD Discussion

[JohnW]

Over on another thread Tom Kitta said:

Quick question about VFD - how does the 5hp VFD work on your lathe - have you tried it yet? I know there is a lot of talk about "some" VFDs having issues with single phase input as they are using 4 out of 6 inputs & there is some talk about de-rating VFDs.... not sure whatever this deals with people simply wanting bigger.

Other thread talks about 3hp VFD being able to start 50hp motor over 20-30 seconds...

With a lathe it seems the startup is critical as the chuck is heavy - not so with a mill.

Yet other people talk about how it is not possible to run multiple motors on a VFD - while other people compute the VFD power needed just to do that.

Great work on the lathe BTW - I see how much I have to do to restore my old K&T 2E - she is older than my dad. So far the mill head and gears were in as new condition.

I thought I'd reply in a separate thread since I can see this discussion going off in its own direction.

I have not yet tried my VFD, although the person I got it from (my cousin) tested it and it apparently works.

To answer some of the above questions, you first need to think a bit about how VFD's work.

The first stage in all of them is to rectify the AC input into DC. That is called the DC bus on the VFD. That is done with sets of diodes. To full-wave rectify 3-phase power you use 6 diodes (two on each phase). To rectify single phase (aka split phase) power you use 4 diodes - again 2 per phase.

Most 3-phase input VFD's do not care if you only feed two of the three input phases. Some VFD's will generate an error code if the third phase is not powered. That is really just a diagnostic feature to warn you that one of the input phases is missing. Those VFD's can generally be fooled when being fed single phase power by connecting one input to phase A, and the other to phase B and C. All the VFD is really doing is checking that there is power at each input, not that it is really valid 3-phase power separated by 120 degrees instead of 180 degrees.

The bottom line is that the AC input will be rectified and feed the DC bus regardless of whether it is 3 phase or single phase.

There are some limitations though. If fed 3-ph power, the load is spread across the 6 input diodes instead of just 4 when it is fed single phase. If you connect one of the single phase inputs to two input phases on a 3ph VFD, you help reduce the stress on the four diodes that are doing half the work, but the remaining 2 are still doing half the work instead of a third.

The input diodes are the first reason you need to de-rate a 3-ph VFD when fed single phase power. The input diodes are not really big power wasters, and since big diodes are relatively easy and cheap to make they are usually oversized anyway, so although you should de-rate a bit based on the input diodes, it is not really that critical.

Next is the issue that the three phases of 3-ph power are staggered by 120 electrical degrees, while with single phase, the two lines are separated by 180 electrical degrees. That means that with single phase power there is an AC peak 120 times a second (based on a 60 hz AC frequency). With 3-ph the peaks come 180 times a second. In other words, the DC bus is "recharged" every 5.6 milliseconds (1/180) with 3-ph, and only every 8.3 ms (1/120) with single phase.

The next thing to understand about the DC bus is that is is backed up by a capacitor bank that charges when there is an AC peak, and discharges between the peaks. With the peaks being further apart with a single phase input, the DC bus voltage will sag more between AC peaks, and need more current to charge up again when the peak does occur. This is the biggest reason that you must de-rate a 3-ph VFD when it is being fed single phase power. The pulsing of the DC bus causes additional AC currents in the DC bus capacitors, which are the real limiting factor.

The DC bus runs at pretty high voltages. With 240V single phase input, the DC bus will run at around 340V DC. That is 1.4 (square root of 2 actually) times 240. With 208V 3-ph, the bus will run at 208 x 1.4 = 291 V DC. The bus us usually designed to handle around 400VDC on a 240/208V VFD.

The engineers have done the math on this and the magic de-rating number is 1.73. That means that the VFD will draw 1.73 times the current when being fed single phase instead of three phase power. So, if your 3-ph motor has a current rating of 14A, the VFD will need to be able to handle about 24A from a single phase source. Or, conversely, if the VFD is rated for 21A, when fed with single phase power, it can only properly drive a motor rated at 12A.

Most VFD's can handle short overloads of 1.5-2.0 times the average current, so there is some headroom for peak currents caused during startup and intermittent heavy loads - usually peaks loads for less than a couple of seconds.

Most VFDs are pretty durable beasts. They are micro-controller controlled and the CPU regularly monitors all sorts of things (temperatures, currents, and voltages) and will almost always shut the VFD down before anything happens to let out the magic smoke.

Note that all I have talked about is how the DC bus is fed from the input power. That is all that really matters. The back end of the VFD works identically no matter how the DC bus is fed. The back end chops the DC bus at a very high frequency (usually 8-20KHz) to produce an AC current at the desired voltage and frequency. There are three chopping circuits, that produce a signal that offset by 120 electrical degrees so three phases are generated.

Sorry, I know this is all pretty complicated, but hopefully this explanation will help. Ask questions and i will try to answer them as well as I can.

 

[JohnW]

To understand how the VFD and motor interact, you first need to think about a few things about how 3-phase induction motors work.

An induction motor works by creating a rotating magnetic field with the field windings. That is done by alternating the phases that power each winding. So, if the power goes to phase A, then B, then C, and the field coils are ordered in the same way, the magnetic field will move from coil A then B then C, etc. The rotor is the dragged along with the rotating magnetic field.

It is actually a bit more complicated than that. The magnetic field on the outside causes a current to flow in the rotor, which causes its own magnetic field, which is what actually reacts with the rotating field coils and causes the rotor to spin.

Nominally the rotor will spin at a speed based on the AC frequency. So with 60hz (cycles per second) AC, the rotor will spin at 1800 RPM (revs per minute) which is 30 revs per second. 30 RPS is one half of 60 Hz. The motor spins at that speed because there are two sets of field coils - ABCABC, so it takes two complete AC cycles for the motor to rotate once. If there was only one set of field coils, the motor would spin at 3600 RPM or 60 RPS, which some motors do.

Why are motors rated at 1720, 1750 or 1760 RPM instead of 1800? Because if the rotor spins at exactly the same speed as the rotating magnetic field it will not have any current in it, so it will have no magnetic field of its own, and have zero torque, so it can't spin at that speed. If it slips a bit, it generates a rotor current, which generates a magnetic field and hence some torque against the rotating field current. The motor will slip just enough to generate the torque required to overcome the friction in the motor. The more it slips, the more torque it generates and the more power is used. Again, it is more complicated than that, but this model works for most purposes.

The thing to take from that is that an induction motor spins at a speed that is just a little less than the speed of the rotating magnetic field created by the frequency of the AC input current. It will use very little current when no torque is applied to the rotor. As the torque on the rotor is increased, the slip will increase, which will increase the current and magnetic field in the rotor, which will create more torque, and use more power from the AC source.

Now can you run more that one motor from a single VFD? Yes, but there are a lot of considerations - especially if the motors are not identical and not under identical loads. Identical motors under identical loads is doable in practice. With different motors and loads, it is doable in theory, but probably not in practice.

Can you switch one VFD between multiple motors? Yes. But don't do it when the motors are spinning! The motors can become instantanious generators that can generate large voltage spikes, which can fry even the big durable chunks of silicon in the big transistors (IGBT's actually) in teh VFD output stages.

How can you run a very big motor on a small VFD? Remember that the motors will draw a current roughly proportional to the torque they are producing. So even a 20HP motor doing only 1HP of work will only draw about the same current as a 1HP motor doing 1HP of work. (not quite, but it is close). So as long as you are not loading down your big motor, a smaller VFD can drive it.

Now, acceleration is a big issue. Without VFD, you will suddenly apply 60Hz to a stationary motor when starting. The slip between the rotating magnetic field and the stationary rotor is HUGE. Hence the current is HUGE until the motor gets up to speed. The start current in this case can be 10x the fully loaded run current or more. With a VFD you typically use acceleration curves. That is really just telling the VFD to ramp the frequency from zero to 60Hz (or even 90Hz or 120Hz) over several seconds. Since the rotor accelerates along with the frequency, teh slip is never very much, so the current is never very much. The acceleration allows energy to be applied to the spinning chuck on a lathe over several seconds.

For instance a 5HP motor suddenly turned on with 60hz might attempt to generate 20HP initially (along with a huge peak in input current), and do that for 1 second until the lathe is up to speed. With a VFD ramping the frequency from 0Hz to 60Hz over 5 seconds, it may not need to generate more than 2HP for 5 seconds to come up to speed.

Again it is all pretty complex. I hope my explanations work. Ask questions if not.

"Questions are free, answers are expensive, correct answers cost even more."

 

[Tom Kitta]

Thanks for super long and detailed explanation - it puts a lot of things together. For example, no one in other message boards mentioned the capacitors as the limiting factor of the VFD - most people seem to concentrate on the diodes and the 4 vs. 6.

As for multiple motors on the same VFD the explanation for sizing was:
"take the 1st motor max running amps and add to it the locked rotor amps of the second motor. The locked rotor amps of a new motor are high, like 17x running amps while for older motors they are like 13x."

From above I take it that for multiple large motors on the same VFD starting in sequence the VFD would have to be huge.

However, for small loads compared to main motor this seems like not a big deal - for example, when the milling machine main motor ramps up and is running at normal speed & I turn on coolant pump which is just around 0.3 - 0.4 amps the VFD should be faced only with extra 5amps - which should be fine given that main motor is not under stress.

Now power feed is also around 0.3 - 0.4 amps and turned on while mill is not stressed - so it should be fine as well.

My new mill is a 5hp - it has in its main board a little multi-transformer for the power feed (120v) and lamp (24v). The coolant pump appears to be 240v. I know I can connect both little motors to 120v (if coolant is 240v it's not a big deal to get a new one). The lamp should not be a big deal as it is a constant amp load. I don't really want to re-wire everything if I don't have to. I assume the ramping up from 0hz to 60hz by the VFD will not harm the small transformer over 5 sec.

My in the mail VFD is a cheap Chinese 5hp. Opinions vary - some people say it works fine for years - others claim that it would not last over 1y in constant use - I plan maybe max 10h per week so it should last a long time unless capacitors age (someone says they die of old age - they liquid insulation in the dries out).

Given that you picked 5hp VFD for a 5hp lathe - which I consider harder load than a mill (I don't think over sized bridgeport can actually use 5hp without shaking to death) I think I did well with my VFD pick. The next level up was 7.5hp and 10hp - both similar price. Opinion is that they build 3hp and 5hp on the same platform with bigger cooler and 7.5hp and 10hp also on the same base.

I also got a second mill - it has 3hp 1950 vintage motor the size of a small house. I assume I can switch between both mills with the same VFD if I keep the program for the 5hp motor.

Do I make sense above or was I again misled by the net?

 

[JohnW]

I'm getting teh impression that you want to just put the VFD in "front" of your lathe or mill without doing any re-wiring on the machine. I do not think that is a good idea.

A VFD is not really just a simple phase converter. VFD's are designed to control a single motor and provide all sorts of parameters specifically to control a motor. If at all possible you should indeed re-wire your lathe such that the VFD only controls the main motor, and the other accessories run independently.

One of the huge benefits of a VFD is the ability to have a variable frequency drive and hence variable speed. A VFD can generally deliver constant torque from a motor over a wide speed range. Assuming you are dealing with a nominal 1800 rpm motor, you you should be able to run it from as low as 6 hz (180 RPM) up to 60 Hz with a constant torque. You can run it faster too, usually up to 90 or 120 Hz, although the torque starts to drop off as you raise the frequency higher. You also need to worry a little bit about the motor balance and the conditions of the bearings as you go faster.

In terms of the VFD I will be using, it is more of a case of getting a really good deal on that 5HP capable VFD and I will make do with it without difficulty. It is really just barely capable of controlling the motor I have if I run it flat out. I can make do with it for a couple of reasons. I can easily live with a start up acceleration curve of a couple of seconds. I'll need to experiment with just how quickly I can accelerate the motor. If I get too aggressive, the VFD will detect an overload and shut down on a fault. Then I just need to back off on the acceleration. I have successfully programmed a 5HP 3-ph VFD to run a 10HP motor on a large lathe. You can't actually use 10HP or it will overload the VFD with that set up. In my case I will be quite happy to only have 3-4 HP available for steady state cutting, although I think I'll still get most of 5HP. I doubt I'll ever use more than 2 HP anyway.

The sizing scenario you quote makes sense. If the VFD is running one motor under load, it could need to be supplying the FLA (Full Load Amps) to that motor at 60Hz. If yo connect another significantly sized motor that is currently not running the new motor will indeed try to use lots of current (I believe I quoted that situation as 10+ times the usual full load value in my previous posting. 13x is in the same range.) The situation is not as bad as that though. The motor that is already rotating will turn into a generator once the surge comes from the stopped motor, so some of the power will come from the rotating motor which will slow down as some of its energy is transferred to the accelerating motor.

Why is this only a problem with VFDs? Its not. In a big building there is generally a very large electrical service connected to a very large utility company. That tends to be able to supply surges to start up motors without a problem. Still, in my days doing computer and network support I did trouble shoot problems that came down to short power drop outs and surges caused by large air conditioning compressors and elevator motors in the same building. A VFD is like a very small electric utility without a large reserve capacity.

I am definitely re-wiring my whole lathe to support my VFD. The VFD will only run the main motor. It will use low-voltage signalling from the fwd/off/rev switches, and the mechanical brake switches to control the motor. I also hope to implement dynamic braking in conjunction with a staged switch on the mechanical brake. A light touch will just let the motor coast to a stop, while a firmer press will engage both the mechanical brake and use dynamic braking in the VFD to quickly stop the motor. My VFD supports a jog button that can run a slower acel cure and slower max speed, so I will be wiring the jog button to do that.

I will also be installing a separate cooling fan for the main motor. If you use a VFD to run a motor very slow, it can overheat. The motor ends up making almost as much heat when run at say, 10Hz as it does at 60Hz, but the problem is that the cooling fan is usually on the motor shaft, so the slowly spinning motor does not blow a lot of air over itself. The VFD can send a signal to an auxiliary fan to run whenever the motor is spinning.

I will still keep the big input fuses and one big relay so that a simple small 120V switch on the front panel can turn off everything in the machine. Using low voltage wiring makes everything a lot safer as well. When I got my "new" lathe there were a couple of micro-switches that had 120V on them that were covered in oil and chips. That was just waiting to short and possibly start the oil on fire. i will publish details on my wiring once I get to that stage.

As for a cheap Chinese VFD, it may be good or not. It is certainly not true that all cheap Chinese stuff is crap. It certainly is true that some of it is real crap. Most stuff seems to be OK. My Allen Bradley unit is made in China, maybe from the same plant, maybe even using the same design and components as yours, or maybe not. It might use the same quality of components or not. You just never know. A lot of big former engineering companies have turned into marketing companies that just resell stuff, so you never know. The bottom line is if it works for a while, it will probably be OK for a long time. Electronics do not wear out if they are designed correctly. You can over stress components such that the repeated stress will make them fail, but that actually happens quite quickly. If they survive for a while, they will likely survive for a long time. If your VFD does not have valid CSA and or UL certifications, make sure it is properly fused and grounded just in case. That 400V DC bus can pack a pretty big punch.

I think that you are correct in that it is not a big deal to put a .5A coolant pump in parallel with a 10A motor. The same thing is true with the power feed and transformer. The transformer will not work very well with other than 60hz feeding it, but if won't hurt it to ramp up with the motor. The transformer will only be on one phase of the input power, which will cause a bit of an imbalance for the VFD, which should not be a big deal since it is a small load. The issue with the transformer is that its output frequency and voltage will ramp up as the VFD ramps up the main motor. Will devices like the power feed be happy to have 60V at 30Hz feeding it for a second or more when it is expecting 120V 60Hz?.

You said the coolant pump is 240V. If it single or 3-phase? usually 3-phase stuff is 208V instead of 240V. If it is single phase it will also only be on a single leg of the 3-phase power.

Without a VFD, the motor is normally controlled via a set of contactors (aka big relays). Forward and reverse is usually accomplished by switching two of the phase wires using two relays. A VFD will not like having the main motor connected and disconnected by relays. Disconnecting is sort-of OK, but reconnecting to a spinning motor can generate serious voltage spikes that can destroy stuff. The forward and reverse function is performed by the VFD, usually in response to front panel controls or low-voltage signals. What is the primary voltage of the relays? If you have the relays in the circuit, they may not close until the VFD is well up on its acceleration curve because the actuation voltage may not be high enough.

As I said in my previous note, it is in theory possible to run more than one thing off of the VFD, but in practice it gets very complicated and may not work for many reasons. It would take a careful look at the schematic of everything you want to connect to the VFD to try to figure out where the gotchas and gremlins may be hiding. VFD's are really designed to run and control a single motor. Doing anything else may or may not work.

Glitches in the loads or stuff that might be connected to only one phase would unbalance the load and may cause a fault to be detected by the VFD. A 24V pilot light not be a big issue, but the transformer that creates the 24V might be. Even starting a .5 A coolant pump on a single phase could cause a significant phase imbalance when the main motor is idling and only drawing 0.5 amp per phase. The imbalance might not hurt anything, but it might cause the VFD to signal a fault.

The bottom line is that using a VFD really means re-wiring the machine. Doing it any other way is asking for trouble. Rewiring the machine with the VFD controlling the main motor only also provides many new features to your tool.

I have not mentioned the concept of dynamic braking. If the VFD is only running the main motor, it is possible to have the VFD quickly bring the motor to a stop using dynamic braking, which is a nice feature. It depends on whether your VFD supports that,and it may require the addition of braking resistors.

 

[PeterT]

JohnW, you mentioned the relatively constant torque of VFD. So is it correct to assume you could select a low mechanical engagement gear, dial the VFD down to a low rpm & have the best of both worlds; ie tickover rpm on the spindle much low(er) than the regular motor would turn the spindle, but not sacrificing much torque at that rpm? That would really increase the useable range & make things like power tapping nicer.

When you have a mechanical brake, how is that then configured for VFD-d motor? I assume on a regular motor the sequence is power off (motor de-energized & freewheeling) then friction brake to come to a quick halt. But don't VFD's have some sort of ramp down over time, its not like an on/off switch?

 

[Louis Dusablon]

I purchased a Huanyang 4 KW Chinese VFD, running a 5 hp. 3ph lathe, I gutted all old components and kept existing wiring, modified a couple of micro switches and added a 3 wires for the variable speed dial (potensionmeter) which I mounted on the console. The VFD is mounted inside the electrical panel on the he machine, I manage to keep the transformer for 110 and 24 volt. Everything works great had it for about a year now. I don't use any braking resistors at this point , when I touch the brake pedal it shut off the input to the motor and I can stop it mechanically which is ok. I pick a range for whatever I'm turning and increase or decrease as required. I am using all original levers for/rev/jog/emergency/off/on/coolant. also the adjustable ramp time for start and stops works great,

 

[JohnW]

To understand the constant torque the VFD can provide at low frequencies, you need to thing about how the motor generates the torque. The torque generated by the motor is proportional to the strength of the magnetic fields that are generated. The strength of the magnetic field is proportional to the current that flows though the windings. If the VFD causes the same current to flow though the windings, the motor will have the same torque.

Now it gets messy, what limits the current in an electromagnet? The rate of change of current is limited by the inductance of the coil. Basically, the inductance of the coils limits how fast the current can increase. The motor is designed to run at 60Hz, that gives the coils 1/60 of a second to go through a full sine wave cycle from current flow in one direction to current flow in the other direction and back to the initial direction, so it takes 1/120 of a second to go from current flow in one direction to the other. 1/120 sec is about 8 milliseconds. So the motor is designed such that when 240V is applied for 8 ms the current through the coils will reverse.

At a lower frequency the coils will have longer for the current to build and will build to a higher current = stronger magnetic field = more torque. EXCEPT, there is a limit to how strong the magnetic field can be. It reaches what is called saturation. Saturation is BAD. When the coil reaches saturation the motor will effectively be a dead short and the current will become very high very quickly. That uses lots of power and turns it into heat, not torque! Normally the motor has a bit of headroom designed into it. At 1750 RPM is will not be saturating the coils. It might not do that until it slows down to 1500 RPM, But if it is loaded down below 1500 RPM (in this example) some time in each cycle will begin to be spent with the coils in saturation, which then just heats up the motor, and soon enough it melts the insulation on the coils and the magic smoke comes out - sometimes very dramatically.

So how do you avoid saturation when using a lower frequency? You use a lower voltage. If the voltage is reduced to half, it will take twice as long to saturate the coils, so the motor that saturated at less than 1500 RPM at 240V will not saturate until it is going less than 750 RPM at 120V.

The VFD has a voltage curve that can be programmed into it that reduces the voltage as the frequency goes down. If you set it correctly, the motor will have just enough voltage for the magnetic fields to reach just shy of saturation as the RPM changes.

The key here is that the coils still reach a current that is just shy of saturation, at all speeds up to the rated speed of the motor. That current will generate the same strength magnetic field, so the torque of the motor will remain constant.

The power generated by the motor will go down as the RPM goes down since power is RPM x Torque.

There are limits. Below 5-10Hz or so the rotor segments are not going past the field coils very quickly, so the torque will not be absolutely constant since each time the rotor segment passes a field coil is like a power stroke on a gas motor. If the RPM is too slow the motor looses some of its smoothness. If yo look way back, there also most be a differential between the actual rotor speed and the rotation of the magnetic field to generate torque. If the magnetic field is only rotating at 5 hz, and it needs 4hz of slip to generate the magnetic fields in the rotor to create the torque, there is only 1hz (60 RPM left).

To discuss the converse situation with a higher frequency (just for completeness) - if you set the VFD to 120hz, the time to build the magnetic fields drops in half (compared to 60Hz). Since the voltage applied to the fields can only be as high as the DC bus voltage, the coils will not have time to reach near saturation, so the torque of the motor will drop as you exceed the motor nameplate frequency. If you had a higher DC bus voltage, you could maintain constant torque at higher frequencies. The limitation here is the quality of the insulation on the wires inside the motor. If the insulation fails, the motor will usually fail quite dramatically.

The constant torque over a wide RPM range is a major benefit of the VFD. It makes it possible for instance to dynamically increase the RPM of a facing cut to maintain a reasonable cutting speed near the centre where the radius is small, or just to be able to change the cutting speed without have to change gears (within limits). Especially on a machine where you have to change belts on pulleys, changing the motor speed with a VFD is way easier.

The second question is on braking:

The VFD can be told to just "let go" of the motor to stop. In that case the VFD turns off all its output transistors, the field coils have no magnetic field, so the rotor will have no magnetic field and the motor will coast to a stop based on friction.

The VDF can also be told to use a deceleration curve. In that case it will ramp the frequency down to zero over whatever period you specified. Sine the field coils are energized, the rotor generates torque and brings itself to a stop. If it is trying to go faster than the nominal frequency because of inertia, that is causing reverse slip on the rotor, which generates a reverse torque which is just as strong as the motor's forward torque, so it can be rapidly brought to a stop.

There is a big BUT here though. The motor will now act as a generator feeding power back into the VFD. The VFD output transistors, then end up feeding that back into the DC bus. That can be VERY BAD. There is nothing draining the DC bus (the input diodes cannot feed it back into the grid) so the voltage will increase by charging the capacitors. If the DC bus voltage goes too high, the capacitors (and possible the diodes) will eventually fail and the VFD will fail very spectacularly!

The VFD controller will monitor the DC bus voltage and if it gets too high for any reason with generate a fault and disconnect from the motor, which will stop the DC bus voltage from getting higher, but will also let the motor free-wheel.

So, many (somewhere between most and some), VFDs have dynamic braking transistors built in. They are actuated if the DC bus voltage goes higher than the line voltage, but before components will explode. They direct the power to braking resistors (separate from the VFD) that bring down the DC bus voltage by turning the extra energy into heat. In my lathe build, I have ordered up a pair of 225 watt, 100 ohm braking resistors that I will run in parallel. That will create a 50 ohm 450 W resistor. The resistors I'm using are rated for a 10x overload for up to 5 seconds, so it will be able to dissipate up to 4500 watts for up to 5 seconds. A full stop on my lathe will probably only be 10-20% of that. As long as I do not do a full dynamic stop more than a couple of times a minute the braking resistors will have a chance to cool between stops.

The third thing a VFD can do for braking is called dc injection. By putting a constant DC current into the field coils it can force the motor coils to saturate, which will make the motor coils themselves into braking resistors. It works, but put heat into the motor.

I don't quite remember, but IIRC, dynamic braking works best at higher frequencies, while DC injection works best as the motor is nearing a stop, so some VFD's will dynamically switch over.

My planned implementation (I'm not absolutely sure I can make this work), is to install staged switches on the mechanical brake foot pedal (it is already done). A have a light tap will tell the VFD to just disconnect and let the motor coast. Pressing a bit further and harder tells the VFD to use dynamic braking, while at the same time applying the physical brake to bring the machine to a stop very quickly.

One more gotcha (nothing to do with VFD's specifically) is that doing hard braking can be hard on the gears, since it snaps them from one direction to the other across the existing lash.

 

[JohnW]

Louis, I agree that that is the best way to do it. All braking resistors add is the ability to ramp down the speed quickly.

Make sure the electrical box is well vented since the VFD goes generate some heat.

 

[Tom Kitta]

I am in very similar situation to Louis - I got Huanyang 4 KW Chinese VFD that should arrive tomorrow or day after. It is to power a 5hp milling machine. I also have a 24V and 110 transformer - we must have similar machines. The 24V feeds the lamp, the 110V feeds X feed and coolant pump.

The mill already has variable speeds - it has a small wheel to the right of the head - when you turn it speed indicators change - it is identical head as on http://www.grizzly.com/products/Vertical-Mill-Variable-Speed/G0748

The mill has speeds in 50Hz and 60hz - I assume it was for dual market. Top speed at 60Hz is 4200 - I assume with VFD I can probably push it to say 70Hz and 5000 rpm as I suspect the bearings cannot take more than that.

John, I am correct understanding from your DC bus timings with a single phase vs 3 phase input that a VFD fed with 60Hz single phase has more capacity running at say 30Hz output than 90Hz output?

Louis, in practice on your lathe does the VFD has any issues - have you managed to overload it under heavy cut? I assume you use the 24V lamp and 110V coolant pump through VFD? Or are you feeding the transformer with single phase 240V?

I also have mechanical break on my mill so no urgent need for VFD breaking. Besides mill head is much easier to stop than lathe with heavy chuck with some heavy part in it.

 

[Louis Dusablon]

I did manage the overload the VFD as I was simulating a hard braking sequence at 660 rpm, the VFD shut down and had to be manually reset, I have tried heavy cuts with no issues or loss of torque.
I'm not sure about running the frequency lower than spec as the motor may heat up, I did however use my sample 3 ph motor and change the frequency to 100 hz it change the rpm fron 1725 to 4000 I didn't have the nerve to go any higher. I certainly would not try it on a lathe because of the the gearing. Not sure on a mill, the mill I have is Bridgeport clone belt drive, as you turn the variable speed handle in opens up the clutch similar a snowmobile , no sure how you could use the variable speed.
The transformer that I am using is 220 step down to 110 to 27 and 24 volts.
As far as breaking resistors I don't use the brake as a rule only an emergency, their is no danger of feed back to the VFD as the lathe would stop in a millisecond. For regular Joe like myself a 3 second ramp down works fairly well.

 

[Tom Kitta]

Well, I got my new Chinese VFD today. Instruction manual is an actual manual not some kind of web print out. English used feels British with most of the grammar at OK level - so maybe it will be of use. I try it out first with my 1950s K&T 2E machine - it only has one motor and a simple on/off switch so not much work needed to get this to test.

 

[Tom Kitta]

After about 12h of use now on the K&T the VFD worked great. No complaints at all. Basic setup was super easy and took just 5 min - simple programming. VFD runs very cool - no hot air is blown out - probably b/c I am pushing the machine maybe to at most 10% - but still there is some mechanical energy wasted. I ordered same identical VFD for my other machine and I just got it yesterday.

 

[Louis Dusablon]

was your K&T variable speed before the VFD installation?

 

[Tom Kitta]

Depends what you mean by variable speed control.

K&T has to be stopped and speed has to be changed with a "spin" type of gear box. This is not usually an issue through and the 8 speed selection is fine (could go faster a bit with a VFD but worried about old bearings). The reason is it is rare to need to change a speed for a particular end mill \ cutter - thus if you are changing speeds you are changing cutters => changing cutter means stopping the machine.

Model H - (newer) has a great feature - it has a disconnect on the spindle - so motor is running yet spindle is still - can position work with power feeds, change spindle speed, cutting tools and re-engage.

Variable speed on a mill still has the fine tune advantage - the 560 to 1000 speed range has 440 difference with which VFD helps.

On a lathe I can see variable speed of great use - especially when facing large pieces - hard to do with constant speed of engine lathe.

I guess K&T can be used as a short bed and super large capacity lathe so VFD when facing would be of great help.

 

[Louis Dusablon]

wasn't sure on what type of mill you had, I though you had a knee mill with variable speed adjustment,

 

[Tom Kitta]

I also have a second mill with a knee / turret - it has variable speed head with a little turning knob.

 

[Louis Dusablon]

will you be installing a VFD on that mill also ?

 

[Tom Kitta]

Yes, already tested the other mill with a VFD - works great. Just using the VFD as a phase converter - at some point I start hooking up all the controls to it.

 

[Payner]

Johnw. Thank you for your in depth explanation, now quick question .
I ordered a 3 phase input by mistake ( senior moment ) now if I try hooking it up to 1 phase input am I better to just hook up the 2 legs ( 220 V ) to see if it shows an error code or go straight to connect " A " to one leg and " B,C " to the other leg ?
I have 3 other VFD's up and running at the moment all with great results.
Thanks Bill