VFD - Minimum Speed

[Jody Boehs]

Hello,

I have a machine with several VFDs on it. The operators of this machine have requested that the minimum speed setting be lowered even more. I have limits set on the touchscreen for 10% - 100%. They are wanting to go slower than the 10%. The motor current increases if the speed decreases. So, how slow can I actually go without harming the VFD or Motor?

My understanding of a VFD is ... The input voltage is converted to a higher DC voltage and then using an IGBT it is inverted back to AC in the desired frequency. I have been told that the IGBT "rebuilds" the sine wave(s) in small increments.

Now my theory is ... as the frequency decreases the sine wave(s) in a sense "stretches out" (maybe this doesn't make sense). Due to the slower frequency, the motor is "started" fewer times per second. And, if the motor is under a load, the amount of current needed to "restart" the motor is greater.

Now, the slower I go with the VFD, the harder it'll be on the motor, correct? Is it equally as hard on the VFD?

Hopefully this makes sense! Any thoughts would be greatly appreciated!

The VFDs I am using are GS2-43PO from AutomationDirect.com .
Here is a link for them: https://www.automationdirect.com/ad...uency_drives_(vfd)/micro/drive_units/gs2-43p0

 

[ptonsparky]

I can’t answer the questions about the VFD. The cooling of a TEFC motor is affected by the speed of its rotation. Newer motors have a speed ratio on them depending on their usage. Constant vs variable torque.

My thoughts would be to change the gear ratios and allow the motors to run at higher speeds. Hindsight is always better.

 

[Besoeker3]

Hello,

I have a machine with several VFDs on it. The operators of this machine have requested that the minimum speed setting be lowered even more. I have limits set on the touchscreen for 10% - 100%. They are wanting to go slower than the 10%. The motor current increases if the speed decreases. So, how slow can I actually go without harming the VFD or Motor?

My understanding of a VFD is ... The input voltage is converted to a higher DC voltage and then using an IGBT it is inverted back to AC in the desired frequency. I have been told that the IGBT "rebuilds" the sine wave(s) in small increments.

Now my theory is ... as the frequency decreases the sine wave(s) in a sense "stretches out" (maybe this doesn't make sense). Due to the slower frequency, the motor is "started" fewer times per second. And, if the motor is under a load, the amount of current needed to "restart" the motor is greater.

Now, the slower I go with the VFD, the harder it'll be on the motor, correct? Is it equally as hard on the VFD?

Hopefully this makes sense! Any thoughts would be greatly appreciated!

The VFDs I am using are GS2-43PO from AutomationDirect.com .
Here is a link for them: https://www.automationdirect.com/ad...uency_drives_(vfd)/micro/drive_units/gs2-43p0

What is the driven load?

 

[blayton1212]

Normally I never set the minimum speed lower then 12-18 hz what is the distance from VFD to the load ?

 

[Jraef]

The concept of “restarting” the motor is flawed, but the rest of your perception is fairly correct. There is no issue with the VFD running at lower frequencies, it’s all about the motor. If the motor cooling is done via fans on the motor shaft (ie TEFC), then the fans move less air at lower frequencies, yet the current in the motor might be the same, depending on the machine function (which is likely why Besoeker was asking).

Given that you said the motor current is going UP as the speed goes down, that’s a sign that this is exactly the type of application where motor cooling is going to be problematic for you.

But we don’t know anything about your motor; someone may have anticipated this and selected a motor capable of this, especially if they selected a good quality “inverter duty” motor. Look up your motor specifications and specifically look for the motor’s “turn down ratio”. So for example if you are running at 10Hz on a motor designed for 60Hz, that is s 6:1 turn down ratio. If you want to run it slower than that, it needs a higher ratio. So a 10:1 turn down ratio means you can run it at 6Hz, a 100:1 ratio means you can run it at 0.6Hz etc. If your motor was NOT selected as inverter duty and does not state a turn down ratio, I typically tell people to assume no better than 4:1, so 15Hz as the slowest speed. You say you are already running at 10Hz, so hopefully this is not the case and someone knew what they were doing.

Post back motor nameplate data data if you want more help with that, including make and model.

 

[Jody Boehs]

What is the driven load?

It is a feed roller that squeezes dog treat dough through a forming die to make ropes of product which is then sliced after it is baked. There is another feed roller that is next to it but it has its own motor and VFD.

 

[Jody Boehs]

Normally I never set the minimum speed lower then 12-18 hz what is the distance from VFD to the load ?

No more than 10 FT.

 

[Jody Boehs]

The concept of “restarting” the motor is flawed, but the rest of your perception is fairly correct. There is no issue with the VFD running at lower frequencies, it’s all about the motor. If the motor cooling is done via fans on the motor shaft (ie TEFC), then the fans move less air at lower frequencies, yet the current in the motor might be the same, depending on the machine function (which is likely why Besoeker was asking).

Given that you said the motor current is going UP as the speed goes down, that’s a sign that this is exactly the type of application where motor cooling is going to be problematic for you.

But we don’t know anything about your motor; someone may have anticipated this and selected a motor capable of this, especially if they selected a good quality “inverter duty” motor. Look up your motor specifications and specifically look for the motor’s “turn down ratio”. So for example if you are running at 10Hz on a motor designed for 60Hz, that is s 6:1 turn down ratio. If you want to run it slower than that, it needs a higher ratio. So a 10:1 turn down ratio means you can run it at 6Hz, a 100:1 ratio means you can run it at 0.6Hz etc. If your motor was NOT selected as inverter duty and does not state a turn down ratio, I typically tell people to assume no better than 4:1, so 15Hz as the slowest speed. You say you are already running at 10Hz, so hopefully this is not the case and someone knew what they were doing.

Post back motor nameplate data data if you want more help with that, including make and model.

I will post a picture of the nameplate on Monday. The machine is running currently and I am unable to open guards with shutting the machine down.

However, this machine was preowned and I have no idea who built it or what kind of motors are used. I doubt they are inverter-duty motors. All I know right now is that they are ancient!

 

[Besoeker3]

It is a feed roller that squeezes dog treat dough through a forming die to make ropes of product which is then sliced after it is baked. There is another feed roller that is next to it but it has its own motor and VFD.

Sounds like a constant torque load. Running the motor at such a low speed may cause overheating unless there is a separate cooling system. Gearbox change has been suggested and that may be the way to go.

 

[RumRunner]

Sounds like a constant torque load. Running the motor at such a low speed may cause overheating unless there is a separate cooling system. Gearbox change has been suggested and that may be the way to go.

The technical term for this type machine is: EXTRUDER.

It a widely used machine in metal fabrication down to food manufacturing. Years before VFD drives, aluminum extruders were popular using regular induction motors with appropriate gearboxes..

While the molten metal was still in its high malleability (easily forged) it is squeezed through an opening that resembles the cross section of the finish product, and then cut to desired length as it cools or hardens.

When applied to food products or household items, like bar soaps. . . the procedure is the same as mentioned in regard to metal forming.

The bar soaps are cut and then stamped with a press to have the impression of the company logo..

I agree, the gearbox design has always been the mainstay in most manufacturing plants. . . and operators from "old school" still love them. The motor is a constant torque and were even DC motors.

The extruder head in food manufacturing is usually steam-heated for uniform consistency. Some of these factories are no longer in existence. . but the products are still available using VFDs.

 

[Besoeker3]

The technical term for this type machine is: EXTRUDER.

Yes. I've done VSDs for quite a few from plastics to solder.

 

[StarCat]

On a note of practicality I will say that the " Flux Optimization " program found on some ABB drives saved the day for me on a similar application which was slowed down, but not as slow as is being described.
If the current is reasonable can an AUX cooling fan arrangement be made to assist would be my question for the experts?

 

[paulengr]

My employer is actually called Electric Motor Shop. So I might know a little more.

First off it’s not “higher voltage DC”. It’s the peak of the AC voltage. Remember AC is RMS.

Second hard to explain but basically if we know torque of the load and speed we know kilowatts. We can use motor efficiency to calculate input HP and at 460 V amp draw. At half that speed but the same torque current is still the same but we need half as much voltage. This is Volts per Hertz mode. But as we approach zero Hz we notice that torque falls off because actual losses affect operation. So we boost the voltage to compensate...torque boosting in V/Hz mode.

Another way to look at it is that we have two currents, a torque current and a flux current. If we measure speed or estimate it by measuring the actual current we can adjust the output voltage to control speed accurately. This is vector mode. There are some other variations but these are the big two. Vector mode doesn’t overflux the motor so it runs lower amps (cooler). This helps. The version that just measures current to estimate it is called sensorless mode. It works down to around 2 Hz. At slower speeds the algorithm can no longer tell where the motor is at. To go lower you need an encoder on the motor. Then you can go clear to dead stall.

So an extender is a constant torque load. Current doesn’t go down much as you slow. But the motor integral fan does. So below about 15-30 Hz you run a real risk of burning up the motor. If you use a 1.15 service factor motor and expect name plate torque out then 30 Hz is it. If the load is not 100% of name plate there is more wiggle room. Centrifugal fans and pumps run torque about even with the motor. So as speed (and cooling) drops, so does the load torque. So until the torque boost problem hits you, you can run almost arbitrarily slow...down to around 6 Hz is practical minimum but the book number is 15.

If you put an external fan on your motor and use vector mode you can get to about 2 Hz. The 2 Hz limit is because sensorless mode does work at slower speeds. With an encoder you can go any speed even stall (braking). There are motors modified to do this. Inverter duty just means it is more tolerant of long distances to the drive. Actual VFD rated means either extra cooling or an outright separate fan and blower motor. Extra cooling will be listed on the name plate as a turn down ratio like 10:1 which means it is rated to run at 10% speed or 6 Hz even under full load. True blower cooling has no limit since the fan is separate from the motor. With both of these the consequence is the motor will be larger.

Other options are going from say a 4 to 6 or 8 or more pole motor which decreases the name plate RPM at full speed...essentially magnetically gearing it down, or changing mechanical gearing or belts to do it mechanically, or using say a BLDC or PMDC motor which are DC motors that can run on some VFDs but are made to run at slow speeds. I’m throwing lots of options out there because one is usually much cheaper than others.

 

[RumRunner]

(Bold texts are mine)

My employer is actually called Electric Motor Shop. So I might know a little more.

First off it’s not “higher voltage DC”. It’s the peak of the AC voltage. Remember AC is RMS.

By higher voltage I assume OP meant voltage difference between AC input and the resultant DC output after the rectifier. In this case the SCRs that feed the DC bus.

DC output does increase after rectification. This really doesn’t need a sesquipedalian narrative to prove the point.

You can easily find out by hooking up of say a 9 vac transformer. . . hook it up to a full wave bridge rectifier and install a filter capacitor. Whip out your DMM and take measurement . . . voila. . . you will read between 12 vdc to 15 vdc.

So, the voltage is increased.


So below about 15-30 Hz you run a real risk of burning up the motor. If you use a 1.15 service factor motor and expect name plate torque out then 30 Hz is it. If the load is not 100% of name plate there is more wiggle room. Centrifugal fans and pumps run torque about even with the motor. So as speed (and cooling) drops, so does the load torque. So until the torque boost problem hits you, you can run almost arbitrarily slow...down to around 6 Hz is practical minimum but the book number is 15.


VFDs’ prime purpose is to control speed which previously was untenable with a generic induction motor. Controlling the three components--ie current, voltage and frequency is the most efficient approach in achieving this. Unlike DC motors that we can control by varying the voltage or simply deploying a rheostat.
During the AC motor’s operating range, the wide spectrum of speed, current and frequency-- it is necessary to control all facets in order to make a controllable induction motor. . . and that is, to have these three components (voltage, current and speed) operate in concert to achieve what we want.
We can run an AC motor to a point where it will enter the stall threshold without suffering heat damage compared to a DC motor. In the stall event (no rotation) the VFD will continue to provide a rotating magnetic field that is if the VFD is still on.

A motor is actually like a transformer whose winding create magnetic field and convert electric energy to mechanical energy. Unlike a transformer, the motor has rotating magnetic field. The transformer converts a higher or lower voltage from a single stable voltage source—converting it to different voltage hence still a form of energy.

The heat generated by the motor’s rotating magnetic field is dissipated by integral fan. And when the motor is slowed to a point where minimal movement of cooling takes center stage--nature comes to the rescue.

This is the where physics law of thermodynamics become your friend.

The surrounding metal will absorb the heat.

Remember that the VFD controls the voltage current and frequency. With these components in an almost diminished functionality, (during slow speed) everything is at a minimum to a point where the motor will simply bog down and eventually stall.

The operator-- noticing that there is no motion would be alarmed and hit emergency stop.

So, damage is averted.


Other options are going from say a 4 to 6 or 8 or more pole motor which decreases the name plate RPM at full speed...essentially magnetically gearing it down, or changing mechanical gearing or belts to do it mechanically, or using say a BLDC or PMDC motor which are DC motors that can run on some VFDs but are made to run at slow speeds. I’m throwing lots of options out there because one is usually much cheaper than others.


Motors designed to run on either AC or DC are usually supplied with wound rotor. It would be unwieldy to a deploy a wire wound rotor that is controlled by VFD when you can readily control it without spending money on complicated VFD-- and besides that option doesn’t bode well with the notion that it would be cheaper.

Motors with wound rotors are usually universal motors and come in very limited sizes.

They are not designed to run on extended period. . .much like your drill motors or grandma’s dough--kneading machine in the kitchen.

Happy baking!

 

[Russs57]

Hmm, some interesting comments. In my experience wound rotor motors are seldom less than 5 HP and may be as high as 15,000 HP. Perhaps confusion with a (commutated) series wound motor? I would also say few modern AC to AC VFD's us SCR's. Of course it is dangerous to make such generalizations. Not trying to nit pick here as much as understand others experience as we all see different things depending on our jobs.

Jody, in a nutshell, I would say the majority of modern 3 phase AC to AC VFD's are like this; You bring the input AC to six diodes where it is turned into DC with AC ripple voltage. The resulting DC voltage will be almost 1.41 times the input AC. The diodes will only conduct near the peak of the sine wave of the AC input voltage. You pass that voltage off to a capacitor bank (which may also have an inductor) to get rid of the AC ripple and store energy so you can power the load for the part of the sine wave in which the input diodes aren't conducting.

Assuming you have adequate capacitance this means the DC bus voltage is nearly a constant, typically around 650-670 for 460/480 VAC input.

Next this DC voltage is "pulsed" to the motor by IGBT's. These pulses take place at the carrier frequency of the drive. This could be a few kilohertz or as high as 15khz.

When you want a lower motor speed you send narrower pulses to the motor and when you want more RPM the pulses are "on" for a greater length of time. These pulses are square waves but because the motor windings are basically an inductor the resulting current draw is close to sinusoidal.

Now this is rather simplified and there are different types of drives with different control methods.

 

[Jody Boehs]

The concept of “restarting” the motor is flawed, but the rest of your perception is fairly correct. There is no issue with the VFD running at lower frequencies, it’s all about the motor. If the motor cooling is done via fans on the motor shaft (ie TEFC), then the fans move less air at lower frequencies, yet the current in the motor might be the same, depending on the machine function (which is likely why Besoeker was asking).

Given that you said the motor current is going UP as the speed goes down, that’s a sign that this is exactly the type of application where motor cooling is going to be problematic for you.

But we don’t know anything about your motor; someone may have anticipated this and selected a motor capable of this, especially if they selected a good quality “inverter duty” motor. Look up your motor specifications and specifically look for the motor’s “turn down ratio”. So for example if you are running at 10Hz on a motor designed for 60Hz, that is s 6:1 turn down ratio. If you want to run it slower than that, it needs a higher ratio. So a 10:1 turn down ratio means you can run it at 6Hz, a 100:1 ratio means you can run it at 0.6Hz etc. If your motor was NOT selected as inverter duty and does not state a turn down ratio, I typically tell people to assume no better than 4:1, so 15Hz as the slowest speed. You say you are already running at 10Hz, so hopefully this is not the case and someone knew what they were doing.

Post back motor nameplate data data if you want more help with that, including make and model.

Here are a couple photos of the motor nameplate!

 

[Jody Boehs]

Hmm, some interesting comments. In my experience wound rotor motors are seldom less than 5 HP and may be as high as 15,000 HP. Perhaps confusion with a (commutated) series wound motor? I would also say few modern AC to AC VFD's us SCR's. Of course it is dangerous to make such generalizations. Not trying to nit pick here as much as understand others experience as we all see different things depending on our jobs.

Jody, in a nutshell, I would say the majority of modern 3 phase AC to AC VFD's are like this; You bring the input AC to six diodes where it is turned into DC with AC ripple voltage. The resulting DC voltage will be almost 1.41 times the input AC. The diodes will only conduct near the peak of the sine wave of the AC input voltage. You pass that voltage off to a capacitor bank (which may also have an inductor) to get rid of the AC ripple and store energy so you can power the load for the part of the sine wave in which the input diodes aren't conducting.

Assuming you have adequate capacitance this means the DC bus voltage is nearly a constant, typically around 650-670 for 460/480 VAC input.

Next this DC voltage is "pulsed" to the motor by IGBT's. These pulses take place at the carrier frequency of the drive. This could be a few kilohertz or as high as 15khz.

When you want a lower motor speed you send narrower pulses to the motor and when you want more RPM the pulses are "on" for a greater length of time. These pulses are square waves but because the motor windings are basically an inductor the resulting current draw is close to sinusoidal.

Now this is rather simplified and there are different types of drives with different control methods.

This is down the line i was thinking for the VFD aspect of this subject. Then the motor cooling factor was brought in which is equally as important. These motors have exterior fans on them and as the the nameplate states the gear ratio is 214:1. I know this isn't the turn down ratio, but I guess the nameplate gives the information fort the gearbox too.

 

[Jraef]

This is down the line i was thinking for the VFD aspect of this subject. Then the motor cooling factor was brought in which is equally as important. These motors have exterior fans on them and as the the nameplate states the gear ratio is 214:1. I know this isn't the turn down ratio, but I guess the nameplate gives the information fort the gearbox too.

Yes, that's common nameplate data when it is sold as a "gearmotor" unit.

So as I interpret that nameplate, the final shaft speed is 3 RPM, with a 214:1 gear ratio, which means the motor shaft speed would be just 642RPM, so it must be a 10 pole motor (720RPM sync speed, 11% slip). So when you say they want to turn it down to under 10%, you mean they want the work shaft speed to be under 0.3 RPM? Must be an extruder then...

If the motor has a separate blower then, they must have understood the application. I'd watch this really carefully though, that 2HP 10 pole motor is NOT going to be on the shelf anywhere! If it burns up, your machine will be down for weeks.

 

[Jody Boehs]

Yes, that's common nameplate data when it is sold as a "gearmotor" unit.

So as I interpret that nameplate, the final shaft speed is 3 RPM, with a 214:1 gear ratio, which means the motor shaft speed would be just 642RPM, so it must be a 10 pole motor (720RPM sync speed, 11% slip). So when you say they want to turn it down to under 10%, you mean they want the work shaft speed to be under 0.3 RPM? Must be an extruder then...

If the motor has a separate blower then, they must have understood the application. I'd watch this really carefully though, that 2HP 10 pole motor is NOT going to be on the shelf anywhere! If it burns up, your machine will be down for weeks.

It is 8 RPM. So that would be 1,712 Motor RPM. The external fans I was referring to are mounted to the motor shaft.

There is also a possibility that 100% would be running at 90Hz. I'll have to verify that when the machine is between runs though. The individual that programmed those drives has done that before.

 

[Russs57]

That company is still in business. Might be worth giving them a call. Motor looks old, old enough that I'm inclined to doubt it was intended for use with a VFD (it wasn't clear to me if the machine always had a VFD).

I remember a well known cooling tower manufacturer that would quote tower efficiency numbers based on running the fan motors at a hertz that was too low for the splash lubrication to work in the gear reducers. Sometimes the problem can be other than your area of focus.

FWIW at 90 hertz you would only have (at best) 67% of the motor's rated torque at your disposal.