Soft-start or Variable Speed Drive? 2

[sry110]

I am a mechanical guy in need of electrical expertise.
I have a motor-driven gear system that has a considerable about of backlash to be taken up before the motor+gear sees the load. When it sees the load, there is a large impact that can damage mechanical components in our system. The impact is due to the motor (3-phase, 1800 RPM) hitting the stationary load at full speed.

I need a way to softly and/or slowly run the motor+gear up against the load to reduce the impact effect. Keep in mind that the motor+gear is completely unloaded (except for the weights and inertias of their own components which is negligible) from the moment it is energized until the driven load appears. Also, once the load is engaged and we are past that point of impact, I need the motor to produce its full locked rotor torque (typically ~300% of Full Load Torque for the motors we use) in order to start driving the load.

Some specific questions:
1) Will a basic soft starter keep the current reduced when there is no load on the motor, such that when our unloaded motor suddenly encounters the load the electrical torque will still be reduced to the set value? For example, let's say I set the soft starter to ramp from 20% to 100% current over 8 seconds (10% per second), and let's say the motor will run unloaded for 1 second until it hits the driven load. Will the motor be at 20% current (and correspondingly reduced torque) when the load suddenly appears, or will the soft starter have already decided the motor is at full speed / no load and therefore bypassed the soft start function altogether?

2) How much torque will a typical VSD allow the motor to produce? I could use a VSD to slowly run the unit up against the driven load, but once the load is engaged I need the motor to produce high starting torque to get it moving. Does the VSD allow the motor to generate 150% torque, 200% torque, higher, lower? I would typically count on ~300% locked rotor torque to get the load moving, but if the VSD only allows %150-200 then I would need to increase the motor size accordingly.

Thanks in advance for any insight.

 

[sry110]

sry110, I apologize for making useless background noise. Good luck with your investigation; you were close to separating the details into proper columns so you can do more than guess in the future. But at least big L gave you the solution.

Wait...I must have missed something here. I guess I'll offer the following:

1) I thought LionelHutz's posts were informative and helpful
2) I thought mikekilroy's posts were informative and helpful
3) If there was a way for me to edit my original post to include all of the relevant details (which I did not think were relevant at the time), I would surely do so.
4) If people are getting frustrated at me because I come off as a person who is ignorant to VFDs and other 'smart' motor controllers....well, that's because I don't know much about them and I'm trying to learn.
5) Just because I haven't responded directly to someone's recommendation does not mean I have discounted it....it simply means I am reviewing/studying/digesting it, as this is the first time I've really encountered any of this subject matter.
6) I have not really noticed any "background noise" here, just people trying to give me an education.

 

[LionelHutz]

You posted a fairly clear question wanting to know how capable a VFD and SS are to take slack out of a drive system. There was already another post with more details when I got here. So, I tried to answer your question of the VFD and SS. My answer would have just been more general if you hadn't posted the extra details. Frankly, you posted you are a mechanical guy so why should I be questioning your mechanical calculations or your ability to properly design the mechanical side of the system or your inability to solve your issue mechanically. Sometime, it can be helpful to hear different ideas but from the start your questions struck me as very direct. I didn't see anything indicating you were lost or needed any help except for answering those questions. However, that is only my opinion.

One thing I should maybe elaborate on a bit more. The VFD being a frequency controller "shifts" the speed torque curve of the motor so the motor always operates above the breakdown torque peak. That is why my comments said the locked rotor current and torque mean nothing when sizing a VFD. Same with not requiring the locked-rotor current from the VFD. You are usually better off finding a motor which has a worse locked rotor torque to get a higher breakdown torque when using a motor and VFD combination on an application which can have high peak torque demands. Motor current and motor torque remain somewhat relative to each other above breakdown toque. 200% torque means you need current in maybe the 200% to 250% range. You could look at the motor curves, but generally they are hard to read in that range. I've also seen many curves where the motor current wasn't anywhere close to 100% at the speed where the motor torque was 100%. When I see this it tells me the curves are useless to use near full speed. No matter what the curves say, in real life the motor won't run at rated speed and rated torque while drawing 50% of it's rated current.

A good point made by jraef that might have been missed is that some soft-starters have a jog function so that could be used to run the motor slowly until your clutch engages. This would keep the speed down on the higher ratio systems. But, his other comment about VFD's being so cheap at 5HP brings says why bother with the solution requiring extra wiring for timing the switch from jog to run when you can find a VFD with sequencing logic built in and just tell it to go and it takes care of everything after that.

To give a totally off-topic example of sequencing logic. A customer had a drum application that wouldn't start due to material sitting in the bottom of the drum. We sold a VFD and used the sequencing logic to rock the drum to get it started. He could have used external logic to do it with timers and a set of reversing contactors and such, but the VFD gives an all-in-one solution. Similarly, you can use a VFD, connect power, connect motor and connect 2 wires from a run contact and the physical installation is done.

And now I'm adding even more noise to your thread....

 

[iop995]

I think a VFD with torque control is good enough to do this job. Set a torque limit of about 5-10% above no load motor torque and a ramp torque after reaching this value.
Motor start, accelerate, engage clutch and hit load train; when VFD sense that motor reach this torque initiate a ramp (need to find best equation) torque to keep speed (SSS engaged) and accelerate load at desired speed.

 

[sry110]

A good point made by jraef that might have been missed is that some soft-starters have a jog function so that could be used to run the motor slowly until your clutch engages. This would keep the speed down on the higher ratio systems. But, his other comment about VFD's being so cheap at 5HP brings says why bother with the solution requiring extra wiring for timing the switch from jog to run when you can find a VFD with sequencing logic built in and just tell it to go and it takes care of everything after that.

I agree that a single electronic component to do the entire job would be ideal. Let me know if I am on the right track:

* Connect VFD with sequencing logic in front of 5HP, 1800 RPM AC motor
* Set first 'mode' of VFD to run motor at 300 RPM for 1 second
My math is: t (sec.) = 2*Theta (rad.) / omega (rad/sec)
where Theta is the total rotation of the motor in radians from starting to clutch engagement
and omega is the speed of the motor in radians/sec.
Theta = 20 deg. x (pi rad./360 deg.) x 20.5 (gear ratio) = 7.16 radians at the motor shaft
Omega = 300 rev/min x (2*pi rad/rev) x (1 min / 60 sec.) = 31.4 rad/sec at the motor shaft
t = 2*7.16 rad / 31.4 rad/sec = 0.46 seconds

So within the first second, the clutch has engaged at low input speed and the motor is trying to rotate at 300 RPM

* Set second 'mode' of VFD to run motor at 1800 RPM (full speed) for the duration of continuous operation. When this second 'mode' is engaged, the motor can generate 200% of its rated torque which we utilize for breaking away the customer's shaft train.

 

[LionelHutz]

Looks good except you need to be aware that the VFD always ramps from one speed to another. So, after a start it doesn't just jump from 0 speed to 300rpm. There is a ramp time which defines how quickly it accelerates. For a VFD, the ramp time is typically the acceleration from 0 speed to full speed or 0 to 1800rpm. If you use a 10-second ramp time then accelerating from 0 to 300rpm would then take 1.7 seconds (300/1800*10). But, if the ramp was 2 seconds then you're talking 0.34 seconds to 300rpm which basically doesn't affect your plan.

Sizing the VFD can be done a few ways.

You can first pick a motor and then contact the manufacturer to get solid current numbers for the peak torque you need. Then, send that data to the VFD manufacturer and get them to pick a VFD. You spend time going back and forth between manufacturers asking for number guarantees and to make sure everything is nailed down exactly to match what you calculate.

Or, you assume the current is mostly linear with torque and use best judgements to work through the numbers. You need 210% torque minimum. Add a safety factor to ensure you get enough torque so maybe bump the required torque by 25% to 263% torque. Next take your torque and convert to current, so 263% torque means a minimum of 263% current. Finally, know the current could be a little higher so add another safety factor for this, say 25% again. The result is you're needing 328% current. A heavy duty (or constant torque) VFD is typically rated 150% for 1 minute. So, 328%/150% = 2.2X. This means A 10hp VFD would be marginal and a 15HP VFD should be more then capable of running the load. I know most VFD's can also produce 175% to 200% current for short periods of time so I could use that fact to justify trying the 10HP VFD. Overall though, once you weigh the couple of $100 extra for the 15HP vs the risk of failure do you really want to bother?

Now, if you do 100's to 1000' of these systems a year then sizing the VFD tighter could make a few extra $$ for you. In this case, I would recommend testing a smaller VFD at a site where you can get away with testing to validate it will work before placing ones in the field where failure gives you a "black eye".

 

[mikekilroy]

And if you believe in the KISS principal, you do not even need the two step turn on; as I showed earlier, a 5-10 sec ramp will get you to engagement point around 300 rpm anyway - why add the second step when a single ramp does the job?


You have never said how long you run at 1800rpm; it souinded like only short time. You may not be aware that all these vfd's will run your 5hp motor upto 1800rpm at constant torque, but then happily continue it up to 3600rpm in constant 5hp mode. Maybe 50% (most of the lower cost motors) are speced to limit top speed at 90hz (2500rpm), and most lower cost worm gearboxes are limited to about 2500rpm (better ones like Stober easily go to 3-4000rpm). So Even if you use lower cost components, if it were my design I would reduce my peak torque requirement by another at least 50% by increasing the 20:1 ratio (in this case) to 30:1 so you do not have to oversize that vfd so much and get exactly the same torque results for engagement.

http://www.KilroyWasHere<dot>com

 

[LionelHutz]

Yes Mike, a straight ramp should work fine on the application outlined in this thread. It may not work very well on the next application with a much higher gear box ratio.

 

[mikekilroy]

Food for thought sry110.... Some others here may consider this background noise, but the fact is, you have not separated out how your stored 5hp motor's rotational kenetic energy is assisting at sss engagement. May I suggest that it is not insignificant in the total picture, and if you really want to understand the relationships of torques, speeds, and times on your sss at engagement, you calculate the breakaway vs accel torque requirements.

Here is why I say this:

Typical 5hp motor inertia: .012 #-ft-sec^2
Your reflected 152#-ft^2 load inertia @ motor: .012 #-ft sec^2

So torque required (ignoring breakaway) to accel 0-1800rpm in 100msec DOL 5hp motor is about 50#-ft - close to what you have seen. That is 25#-ft required to accel the motor plus 25#-ft to accel the 152#-ft^2 load.

Then when 'breakaway' happens, you see - about same.

If we assume sss engagement droops DOL motor to 1/2 speed, 900rpm, and it again recovers in another 100msec, which seems reasonable with the values given, it pulls the same 50#-ft during this recovery per your tests. I suggest without scoping speed and current during this DOL start and then engagement of sss, which all basically happens at the same time, you cannot tell that 1/2 the torque load you see is simply torque to accel the load. So I wonkder if you really only need 1/2 the torque you suggest for breakaway. This would bring your peak torque required on a vfd AT SLOW SPEED down to just breakaway torque since you extend the accel out and basically make that 25#-ft portion (accel torque) go away. This would mean your 211% max torque required MAY be closer to 100%.

Looking further, you can calculate that the stored kinetic energy in the motor that is running 1800 rpm before engagement, is .5Iw^2= 260watt-sec. You can further calculate that the energy used during the re-accel portion (after engagement, assuming speed drop from 1800rpm to 900rpm) ALONE is 25#-ft*900rpm/5252= 4.2hp or 5.7kw. Since the recovery was another 100msec, that is 570 watt-sec used just to reaccel the load. But your motor only had 260 watt-sec of energy stored, and we know SOME of that was used for the breakaway, so what does it mean? Seems to me it means most of your present 40 amps measured at 'breakaway' is used to re accel the drooped speed, not for the breakaway itself. If there is any validity in this statement, then a vfd will not require much extra torque for 'breakway' since most of the torque was used in re acceling that load. Means no vfd oversizing required yet again.

The good thing about trying a vfd is you will then likely have software available with it that allows you to scope these things finally. Then you can take that data and analyze it and fine tune how you react to your sss engagement under your terms.

I would certainlyi go with Lionelhutz' recommendation and buy a 15hp vfd for this first one. I would make sure I bought one with a really good USB/ethernet software communications package that does scope functions - at least for this first one.

Future vfd's could be brought back down in oversize capacity as proved by this first one. Future systems could be designed to AT LEAST go 2500rpm on the motor to reduce torque requirements another 50%.

http://www.KilroyWasHere<dot>com

 

[mikekilroy]

did not get to edit my post; dam cat sat on keyboard and hit send. sorry.

http://www.KilroyWasHere<dot>com

 

[mikekilroy]

One last nonsense background noise comment.... Please tune out if you consider this more noise - there certainly is no requirement you read it.

I recently had a consulting job on a device called a "barrel cam" (you can google it) used a lot in the glass industry. Its effective operation is close to your sss. I DID use kenetic energy calcs as well as accel/decel/breakaway calcs - to solve a multi year problem where the customer was continually increasing the motor HP to try to get control. They went from 5hp to 40hp and still had similar issues to you. Simple KE calcs proved the solution in their case was adding an inertial flywheel at the motor and all of a sudden they were back to a 5hp motor and with perfect control. So this in depth analysis CAN be useful sometimes....

sry110, you are the mfgr of this equipment; THAT is the reason I thought all this in depth analysis was worthwhile; if I mfgred a system using this sss thing, I certainly would make sure I fully understood its total operation in detail. I do not operate my company without proper analysis, and I am assumed you do not either. Although I am a servo engineer at heart, I make no claims that my free calcs are perfect or don't need tweaking by you, the mfgring engineer - I am sure they do: my 'noise' inputs here were just to give you some potential direction if you wanted it. But, on the other hand, for someone else, just using one of these gidgets, sure, the guess to just oversize the vfd 2.5x would be sufficient.

If that guess was all the info you wanted to gain from your OP, I certainly apologize for all my background noise!

http://www.KilroyWasHere<dot>com

 

[mikekilroy]

Yes Mike, a straight ramp should work fine on the application outlined in this thread. It may not work very well on the next application with a much higher gear box ratio.

But wouldn't you consider the adjustable RAMP parameter equivelant to the adjustable extra step JOG parameter for the next slightly different system?

http://www.KilroyWasHere<dot>com

 

[LionelHutz]

It could very easily be that the ramp time required to accelerate up to X speed in the time it takes the clutch to engage is in conflict with the desired acceleration rate after the clutch has engaged.

 

[davidbeach]

First off, I'm not a VFD guy so maybe I'm way off base, but couldn't you program the VFD to run at some very low speed and highly limited torque, then she it stalls change the operational mode to accelerate with a high torque limit (above break away) and the run at nominal speed? Maybe it takes an external mini-PLC to control it, but somehow this seems like a really simple problem to solve. Perhaps, though, that just means I don't understand it.

 

[mikekilroy]

sry110, ur up. misc questions/comments given tofor u. u shud reply?

http://www.KilroyWasHere<dot>com

 

[waross]

Torque limit of the breakaway clutch vs the torque required to start the load.
If the setting of the anti-back-drive device is too low, you just won't be able to start the load.
Motor torque: This is more a reflection of the load on the motor than on the electrical supply to the motor.
Why do you need a torque limit?? The torque will be limited by the load. Absent other factors the motor torque will increase until the load breaks away. An artificial torque limit on the motor will, if set higher than the required starting torque, probably never be active. If the artificial torque limit is set lower than the required starting torque then the load may never be started.
Motor torque; This is related to slip frequency. If you look at any torque curve you will see a speed base. At the point of maximum torque the speed may have dropped 40%. The slip frequency has now increased from the normal 2% or 3% to about 40%. The current is well into the overload range. The motor is heating up. VFDs are used to allow a motor to operate at this speed while avoiding the factors that put the motor into overload.
Maybe you should back-drive the motor to the limit of the backlash of s aloppy gear train and give the motor a running start so as to take advantage of the rotor inertia to break away the load. This is not a serious suggestion but more an attempt to step back and look at all possible effects on this application.
I see two possibilities for a low torque start.
1: Enough torque to take up the slack in the gear train.
2: Enough torque to take up the slack in the gear train and engage the SSS clutch.
Consider this:
A VFD to provide enough torque to engage the SSS. Then bypass the VFD and go DOL to get the maximum torque to start the load. Remember that torque limit may not be needed here as the motor will not develop more torque than the load needs and if it does not develop enough torque the load will not start.
Consider a design D motor which has maximum torque at locked rotor.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[sry110]

Sorry again for the radio silence. I'm putting out some other fires at work and I am trying to get back to this ASAP. I have been reading the replies but have not had the time to write my own. Hope to get back to this tonight, thanks for your patience..

 

[sry110]

Annnnnnd I'm back.
Many good questions were asked, and I will try to address all points:

Acceleration time to full speed:
We do not have a limit, but typically for this type of turning gear the time elapsing from energizing the motor until the driven shaft train gets to full continuous turning speed is less than 5 seconds.

Inertia/flywheel effect of motor:
I acknowledge and agree that the torque overload problem here is not due to the electrical torque of the motor, but instead due to the inertia/flywheel effect of the motor rotor having to rapidly decelerate when the SSS clutch engages. The inertia/flywheel effect isn't so easy to calculate, nor do most of our Customers want to believe it's a "real" contribution that we can count on to achieve breakaway, versus the motor electrical torque and gearbox efficiency which are easily calculable and therefore reliable from a design standpoint. So the goal here is to remove the inertia/flywheel effect altogether (by slowly engaging the SSS clutch) and select my motor size based on locked rotor torque x gear ratio x efficiency.

Consider this:
A VFD to provide enough torque to engage the SSS. Then bypass the VFD and go DOL to get the maximum torque to start the load. Remember that torque limit may not be needed here as the motor will not develop more torque than the load needs and if it does not develop enough torque the load will not start.
Consider a design D motor which has maximum torque at locked rotor.

YES! You're onto it. From a mechanical standpoint, this is exactly what I have in mind - creep the clutch into engagement, and then once it's engaged let the motor go DOL so it can develop its full output torque and speed.
Now getting into the fine details: Can a VFD be had with a built-in bypass and a timer so that the VFD and DOL starter could be wired in series? I'm picturing the VFD is set to run the motor at 300 RPM for 1 second. When 1 second passes the VFD times out and the bypass contactor closes allowing full line voltage to go to the DOL starter. Then the DOL start contact closes and sends full line voltage to the motor. Right/wrong?

 

[waross]

Back to another suggestion I made. Not as sophisticated but possibly cheaper and simpler. Use three resistors to limit the current and torque of the motor. Energize the motor with the resistors in series for a second to let the motor take up the slack at low torque and then stall. Then hit it DOL.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

A small soft start operating in current limit would achieve the same thing. It wouldn't need to be rated for the full motor current, just enough to develop sufficient torque to take up the 'slack'. Obviously the motor would stall, but if it is in current limit then that wouldn't matter. Once stalled, go for a standard DOL start.

 

[iop995]

By series resistor, as soon as clutch engage, motor speed down, lose full engaging and motor DOL will lead to a high strike in SSS mechanism, maybe not so good. That, I think by VFD may reach slowly and keep clutch engaged and then increase torque to speed-up load.