VFD - V/Hz High Breakaway Load

Learn the NEC with Mike Holt now!
Status
Not open for further replies.

W@ttson

Senior Member
Location
USA
Hello,

Here are the facts:

VFD Drive
No Encoder
V/Hz Control mode
Load is Constant Torque, with high starting torque required
Motor = 1800RPM rated with 3% slip, 460V rated

V/Hz typically limited to 150% FLT at 3 Hz. Below 3Hz it get wishy washy and you need all sorts of voltage boost which you don't know if you will have enough torque to start the high inertia load.

Question:

I know the load is is at 0RPM (0Hz) at starting, but why can't I just have the drive in V/Hz send the signal to start the motor at 3Hz (87.5RPM). I am not too concerned with current inrush etc. I just need to make sure the load starts.This will put me in the portion of the Torque/speed curve for the capabilities of V/Hz that would allow 150% FLT. It seems like the drive will try to start at 0.001Hz (something arbitrarily close to 0 Hz) and work its way up with a possibility of the load not starting, but if the drive continues to increase the voltage and current without caring if the motor actually started or not eventually it would get to 100%Voltage (460V)and 60Hz at that point the motor would be able to be started at Full Load Torque and the 150% FLT.

What am I missing? Why is there such concern of using V/Hz at low speeds with high starting torque requirements.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
When a VFD is in Scalar control mode (V/Hz), the VFD puts out a PWM pattern to the motor with the required Voltage and Frequency that it WANTS the motor to turn at, but has no way of knowing if the motor is actually doing that. So if the motor "equivalent circuit" is not exactly what the VFD expects, the motor performance will be unexpected as well. At low speeds, this becomes more evident because the cable has capacitance, the motor stator/rotor relationship has capacitance and the entire setup has other variables like leakage reactance that become a larger part of the overall circuit when the speed is low. So as a general rule, standard motors and VFDs using Scalar control are good for at best, a "turn down ratio" of 4:1, meaning 1/4th of the base motor speed or from a Hz standpoint, 15Hz. You can go lower with the VFD, but that's where you will start finding a lot of instability and the motor problems get compounded. "Torque Boost" is a manual adjustment in the VFD that tweaks the V/Hz pattern coming from the drive that increases the torque, at the cost of starting to saturate the windings. So that feature generally has a point at which it reverts back to the proper V/Hz ratio to avoid overheating the motor.

If your VFD is capable of it, I suggest changing to what's called "Sensorless Vector Control" or SVC instead of V/Hz. That adds the ability for the VFD to "see" what's happening in the motor and automatically adjust the output to get you the desired performance on the fly, down to 1Hz or even lower (depending on the quality of the VFD). Your motor may still not be able to run at 3Hz without overheating, unless it was designed to do so or has a separately powered blower keeping it cool, but that's a separate issue. If your drive is capable of SVC, you will need to do what's called an "autotune" procedure that lets the VFD measure all of those variables in the circuit. Without that, you will not get the desired performance.
 

synchro

Senior Member
Location
Chicago, IL
Occupation
EE
With an induction motor you need slip in order to generate torque, and so if you can set it to start at 3Hz or higher (as I just saw was mentioned) then that should help start the motor...unless you need precise control of the speed as it ramps up at very low RPM. But in that case you should be using a synchronous motor.

The voltage has to be boosted above the constant V/Hz target at low RPM because although the inductive reactanced goes down as 1/f (which requires the voltage to be lowered to maintain the same current), the winding's resistance does not. So you need extra voltage at low frequencies just to maintain the current at a target value. And it is often boosted as you say at startup to provide some extra torque, although the magnetics will saturate if this is done too agressively.
 

W@ttson

Senior Member
Location
USA
When a VFD is in Scalar control mode (V/Hz), the VFD puts out a PWM pattern to the motor with the required Voltage and Frequency that it WANTS the motor to turn at, but has no way of knowing if the motor is actually doing that. So if the motor "equivalent circuit" is not exactly what the VFD expects, the motor performance will be unexpected as well. At low speeds, this becomes more evident because the cable has capacitance, the motor stator/rotor relationship has capacitance and the entire setup has other variables like leakage reactance that become a larger part of the overall circuit when the speed is low. So as a general rule, standard motors and VFDs using Scalar control are good for at best, a "turn down ratio" of 4:1, meaning 1/4th of the base motor speed or from a Hz standpoint, 15Hz. You can go lower with the VFD, but that's where you will start finding a lot of instability and the motor problems get compounded. "Torque Boost" is a manual adjustment in the VFD that tweaks the V/Hz pattern coming from the drive that increases the torque, at the cost of starting to saturate the windings. So that feature generally has a point at which it reverts back to the proper V/Hz ratio to avoid overheating the motor.

If your VFD is capable of it, I suggest changing to what's called "Sensorless Vector Control" or SVC instead of V/Hz. That adds the ability for the VFD to "see" what's happening in the motor and automatically adjust the output to get you the desired performance on the fly, down to 1Hz or even lower (depending on the quality of the VFD). Your motor may still not be able to run at 3Hz without overheating, unless it was designed to do so or has a separately powered blower keeping it cool, but that's a separate issue. If your drive is capable of SVC, you will need to do what's called an "autotune" procedure that lets the VFD measure all of those variables in the circuit. Without that, you will not get the desired performance.

yes, I was going to suggest open loop control but I just wanted to get the theory if I could have the drive in V/Hz mode start outside of the “muddy” low range. Have it start at 3Hz, 10Hz, 15Hz what ever it takes to still be able to use V/Hz but still get the overload capacity needed to get the load going. Is that a possibility? If the drive doesn’t have feedback and just tells the motor what to do without checking in on it, what is stopping it from sending it a 15Hz signal with the appropriate voltage to be able to start the load? I mean if I didn’t have the drive and across the line started the motor it would be at 60Hz, 460V, and be able to develop ~ 160%FLT for the starting torque .
 

W@ttson

Senior Member
Location
USA
With an induction motor you need slip in order to generate torque, and so if you can set it to start at 3Hz or higher (as I just saw was mentioned) then that should help start the motor...unless you need precise control of the speed as it ramps up at very low RPM. But in that case you should be using a synchronous motor.

The voltage has to be boosted above the constant V/Hz target at low RPM because although the inductive reactanced goes down as 1/f (which requires the voltage to be lowered to maintain the same current), the winding's resistance does not. So you need extra voltage at low frequencies just to maintain the current at a target value. And it is often boosted as you say at startup to provide some extra torque, although the magnetics will saturate if this is done too agressively.

yeah that is the question. What’s stopping the drive from starting at 3hz or whatever larger frequency so that high starting loads can be started with V/Hz mode.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
yeah that is the question. What’s stopping the drive from starting at 3hz or whatever larger frequency so that high starting loads can be started with V/Hz mode.

A drive always starts from zero speed and ramps. All you can do is set a minimum speed, but from a dead stop, it still ramps to that point.

The drive’s ability to create torque in the motor is better than across the line, the drive is capable of a short burst (2-3seconds) of Break Down Torque, which is 200-220% of FLT. The problem is that in V/Hz mode, that ability is limited by the inherent inaccuracy of V/Hz control. With SVC you can have all the torque the motor is capable of at almost zero speed.
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
What comes out of a drive in 'sensorless vector mode' is simply PWM synthesizing a given frequency and voltage for the desired torque and the rotor speed. Of course these values are constantly changing as the motor accelerates. So if you know the correct frequency and voltage to use to start the motor, you could just directly send those values, but without actually knowing the motor speed it is pretty hard to adjust the values as the motor accelerates.

As you note, you _could_ simply start the motor across the line, and would see a torque at zero speed that was set by the torque/speed curve of the motor, which is probably greater than 150% of FLT, but it really depends on the particular motor. Of course, when you do this the motor will want to draw high current depending on the motor characteristics, often on the order of 5x nominal full load current. If you wanted to use a VFD in this fashion it would need to be hugely oversized to supply the excess current at 0 speed.

You could do the same thing at _any_ frequency and supply voltage; for any giving electrical input parameters the motor will have its torque/speed curve, and will draw current and accelerate the load depending on that curve. If you pick a combination of voltage and frequency which, given the motor characteristics, will produce a torque/speed curve and a current/speed curve which are both acceptable from 0 speed on up, then you could use this programmed value to accelerate the load.

Doing this would almost certainly operate the motor less efficiently and require higher current than adjusting the drive frequency and voltage to match the rotor speed and required torque, but it clearly has to work in the same way that across the line starting works.

-Jon
 

W@ttson

Senior Member
Location
USA
A drive always starts from zero speed and ramps. All you can do is set a minimum speed, but from a dead stop, it still ramps to that point.

So that is the question right there. Suppose you set the min speed. It is at a dead still. The motor cannot move at 0Hz, 1Hz, or 2Hz. The drive still keeps ramping to the 3Hz mark (it doesn't care that the motor didn't start spinning, it is simple V/Hz, no feedback). It now reaches 3Hz, has enough torque to move/start the load. Why is this scenario not possible? Is it that the drive would overload/fault out before it reaches the 3Hz if the motor didn't start turning.
 

W@ttson

Senior Member
Location
USA
What comes out of a drive in 'sensorless vector mode' is simply PWM synthesizing a given frequency and voltage for the desired torque and the rotor speed. Of course these values are constantly changing as the motor accelerates. So if you know the correct frequency and voltage to use to start the motor, you could just directly send those values, but without actually knowing the motor speed it is pretty hard to adjust the values as the motor accelerates.

As you note, you _could_ simply start the motor across the line, and would see a torque at zero speed that was set by the torque/speed curve of the motor, which is probably greater than 150% of FLT, but it really depends on the particular motor. Of course, when you do this the motor will want to draw high current depending on the motor characteristics, often on the order of 5x nominal full load current. If you wanted to use a VFD in this fashion it would need to be hugely oversized to supply the excess current at 0 speed.

You could do the same thing at _any_ frequency and supply voltage; for any giving electrical input parameters the motor will have its torque/speed curve, and will draw current and accelerate the load depending on that curve. If you pick a combination of voltage and frequency which, given the motor characteristics, will produce a torque/speed curve and a current/speed curve which are both acceptable from 0 speed on up, then you could use this programmed value to accelerate the load.

Doing this would almost certainly operate the motor less efficiently and require higher current than adjusting the drive frequency and voltage to match the rotor speed and required torque, but it clearly has to work in the same way that across the line starting works.

-Jon

From Jraefs post above the drive always seems to have to start at 0 and ramp up. I would ultimately would like to send 50% voltage and 50% speed command initially to the motor. This will guarantee that the motor will be able to start. But from Jarefs post. It looks like the drive in V/Hz will need to ramp up to get to that point, and I am assuming the issue is that it will fault out by the time it gets to a point that can move the load at starting. The thing about across the line starting it is that you hit it with the 100% voltage and 100% frequency right off the bat. And the motor doesn't have time to overload/fault out. It rips through its locked rotor torque position and keeps moving along its curve.
 

ElectricMatt

Senior Member
Location
Waco, tx
From Jraefs post above the drive always seems to have to start at 0 and ramp up. I would ultimately would like to send 50% voltage and 50% speed command initially to the motor. This will guarantee that the motor will be able to start. But from Jarefs post. It looks like the drive in V/Hz will need to ramp up to get to that point, and I am assuming the issue is that it will fault out by the time it gets to a point that can move the load at starting. The thing about across the line starting it is that you hit it with the 100% voltage and 100% frequency right off the bat. And the motor doesn't have time to overload/fault out. It rips through its locked rotor torque position and keeps moving along its curve.

I would setting the ramp time to .01 sec and disable faults during ramp. That should get you to your ideal speed relatively quickly and prevent nuisance trips.


Sent from my iPhone using Tapatalk
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
From Jraefs post above the drive always seems to have to start at 0 and ramp up. I would ultimately would like to send 50% voltage and 50% speed command initially to the motor. This will guarantee that the motor will be able to start. But from Jarefs post. It looks like the drive in V/Hz will need to ramp up to get to that point, and I am assuming the issue is that it will fault out by the time it gets to a point that can move the load at starting. The thing about across the line starting it is that you hit it with the 100% voltage and 100% frequency right off the bat. And the motor doesn't have time to overload/fault out. It rips through its locked rotor torque position and keeps moving along its curve.

That is more of a software/firmware question, and Jraef certainly has more experience than I with actual drives that you can purchase. Start at 0Hz and ramp up might be what all available drives actually do.

In principal there is no reason that a drive could not simply start at a particular frequency and do the equivalent of an 'across the line' start at that frequency. Or start at 0V and XHz and quickly ramp to operating voltage at that frequency. The latter is what we do with the experimental drives we do in our lab, but we are writing the actual PWM software ourselves.

-Jon
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Won’t make a difference. The drive is only capable of, at BEST, 150% current for 60seconds and will current limit itself. The only way to get a VFD to start a motor across-the-line (in the sense that is equivalent to a standard starter) is to start with a VFD that is 4 times larger than the motor FLA.

But there is no point to all of that. Because the VFD is controlling BOTH the voltage and frequency together, it gets equal or better performance out of a motor that you can get with across the line starting. The frequency you start at makes zero difference.

The REASON your motor doesn’t perform well at low speeds is simply because you are using an old technology (V/Hz mode).l that is inherently sloppy at low speeds as I explained. If you need full starting torque capability from the motor at the very outset of a run command, use vector control of any sort, ie SCV. That’s what that is for.
 

Besoeker3

Senior Member
Location
UK
Occupation
Retired Electrical Engineer
Question:

It seems like the drive will try to start at 0.001Hz (something arbitrarily close to 0 Hz) and work its way up with a possibility of the load not starting, but if the drive continues to increase the voltage and current without caring if the motor actually started or not eventually it would get to 100%Voltage (460V)and 60Hz at that point the motor would be able to be started at Full Load Torque and the 150% FLT.
The VFD will limit the current and the voltage. If it could get up to full rated voltage before it started you would six to eight times full load current. Ain't gonna happen.
 

W@ttson

Senior Member
Location
USA
Won’t make a difference. The drive is only capable of, at BEST, 150% current for 60seconds and will current limit itself. The only way to get a VFD to start a motor across-the-line (in the sense that is equivalent to a standard starter) is to start with a VFD that is 4 times larger than the motor FLA.

But there is no point to all of that. Because the VFD is controlling BOTH the voltage and frequency together, it gets equal or better performance out of a motor that you can get with across the line starting. The frequency you start at makes zero difference.

The REASON your motor doesn’t perform well at low speeds is simply because you are using an old technology (V/Hz mode).l that is inherently sloppy at low speeds as I explained. If you need full starting torque capability from the motor at the very outset of a run command, use vector control of any sort, ie SCV. That’s what that is for.


I spoke with the drive manufacturer. Their model drive in question comes standard in Open Loop control, however, the model drive has such a rudimentary torque control model even in Open loop vector control that it still might work. They recommended a model up where in Open Loop vector there would be no problem at all.
 

W@ttson

Senior Member
Location
USA
I spoke with the drive manufacturer. Their model drive in question comes standard in Open Loop control, however, the model drive has such a rudimentary torque control model even in Open loop vector control that it still might NOT work. They recommended a model up where in Open Loop vector there would be no problem at all.

.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I spoke with the drive manufacturer. Their model drive in question comes standard in Open Loop control, however, the model drive has such a rudimentary torque control model even in Open loop vector control that it still might work. They recommended a model up where in Open Loop vector there would be no problem at all.
Who's drive is it?
 
Status
Not open for further replies.
Top