Wye-Delta Starter on 6-Lead IEC Motor 1

[sry110]

I need some input on wye start / delta run motor starting. I am a gearhead and V=IR is basically the extent of my electrical capability, so please bear with me.

First some background: I have a turning gear system consisting of an AC motor, driving into a worm gear speed reducer, and an overrunning clutch on the output shaft. When the unit is at rest the clutch is disengaged, and it requires approx. 5 revolutions of the AC motor to engage the clutch (i.e. it has to "twist" into engagement). During this 5 revolutions the motor is essentially unloaded (the only load it sees is its own inertia and the inertia of our worm gear reducer, which is negligible).

The problem: When the clutch engages after the 5 unloaded revolutions of the motor, the motor instantly sees the load of the driven equipment. So the motor goes from unloaded to fully loaded (Locked-rotor condition) in a split second. This rapid deceleration and instant onset of locked rotor torque output manifests as a large impact torque. There are components in the turning gear system that will yield/break due to the large impact torque. Note: we start the AC motor with an across-the-line motor starter, which the customer specs and provides.

We have tried using soft starters, but as soon as the soft starter sees that the motor is unloaded it basically shorts out and gives the motor full line voltage. So when our clutch engages, it's at full speed and full motor torque as though the soft starter wasn't there.

So now I'm considering a wye-delta starter. The motor is a 6-lead IEC motor (made by ABB), 4kW rated for Delta connection. As far as I can tell, if I use a wye-delya starter, the motor starts in Wye with reduced current/torque and stays in that mode until the timer times out, at which point it switches over to full across-the-line Delta mode. Looks like wye-delta starter is a "dummy" device in that it does not receive any feedback from the motor regarding loaded / unloaded condition.

So if I use the wye-delta starter, and keeping in mind that our motor is essentially unloaded and will get up to full speed very quickly, can I be confident that the motor will have reduced current / torque at the point of clutch engagement? Or does the motor react differently once it gets up to full speed (i.e. same problem we have with the soft starter)?

Any help will be appreciated!

 

[sry110]

One more thing: We know that the clutch will engage within ~1 second, so we would set the time delay for switching to Delta for 2 seconds.

 

[waross]

Search this site. This issue has been discussed at great length.
The solution is a VFD drive and a ramped start. I have solved similar problems with excessive backlash and impact damage to machines with the use of VFD drives. Fortunately the machine manufacturer recommended this to the customer as a retrofit and I just had to install the drives without all the customer hand holding, discussions and drama that such an installation often provides.

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

 

[Compositepro]

It sounds like your over-running clutch is similar to what is used on automotive starter motors. With that type of clutch you will always have impact loads because the motor must start before the gears engage. I think you need to fix the clutch, not the motor. Will the clutch engage if the motor starts very slowly? An automotive type will not.

I won't speculate any further with the limited information you have supplied so far.

 

[waross]

That type of automotive starer used to have a large spring called a Bendix spring to absorb the shock, and you are speaking about Ford type starters. Many GM starters engage the drive gear before energizing the starter motor. The overrunning clutch is for after the motor starts so that the high RPMs won't cause centrifugal damage to the starter rotor.

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

 

[jraef]

You are better off with a properly designed solid state soft starter system than the Wye-Delta. The Wye-Delta will just change the timing of when that shock will hit, and in fact will more likely subject your equipment to TWO shocks instead of one.

Most modern soft starters have what is called an "anti-oscillation circuit" that detects an accelerated motor and bypasses any remaining ramp profile, based on the assumption that if the motor is at full speed, you no longer need to worry about it. But as you have observed, that can get in the way. Because of this, some now offer either linear ramp profile options or constant current starting options. In a linear ramp profile, you use a tachometer feedback to the RVSS so that it forgoes the anti-oscillation function and ramps in a linear fashion no matter what. In a constant current operation profile, the soft starter overrides the ramp process in favor of current limitation no matter what, even if the motor s already at full speed. So for example in your case you wouldset it for let's say 300% current and t would maybe ramp to that point (optional) but ce there, hold there until the motor either accelerates (hopefully) or stalls. Then even if it is running full speed, if the load starts to demand more than 300% current, even for a moment, like when your clutch engages, the soft starter artificially cuts it off at 300%. This has the effect of softening the blow so to speak.

The caveat is, you CANNOT do this with one of the types that have an integral bypass contactor, the SCRs have no further control once that is closed. So for your application, get one without that and one that offers constant current operation, then use a separate bypass contactor and close it only after the clutch is engaged.

The VFD option of course works beautifully as well, just a quantum leap in cost if the motor is large. If you are under 15HP however, a RVSS will probably cost about the same and I would go that route.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[ScottyUK]

Is this a turbine application?

We scrapped a VFD / induction motor scheme in favour of a DC drive and resistor timestarter fed from the turbine battery. The main driver for this was the need to be able to establish the machine onto barring gear from stationary without AC power, a requirement of black start at our site. The original VFD was fed from a static inverter converting 110V DC to 415V AC 3-phase. I think this arrangement caught a machine coasting down once in this history of the site, and we had a fire in the inverter after trying to start from a stationary shaft. The DC system never failed: simple and robust.

A few things from bitter experience:

When thinking of a VFD or soft start, consider that the break-away torque at standstill is very high and may approach or exceed the stall torque of the barring motor.

Simple usually equates to reliable. Big and simple tolerates abuse too.

What happens when the mains goes off? Do you care if the machine develops a thermal bend and can't run for a prolonged period?

If you have a battery then use it direct - multiple stages of power conversion work better in theory than in practice.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[edison123]

I have seen only DOL start shaft turning gear ac motors. Lube oil pump motors are normally DC.

Muthu

http://www.edison.co.in

 

[ScottyUK]

I think MHI and possibly Westinghouse went back to DC on their combustion turbines, or at least offered it as an option, after using AC motors on machines of late 80's / early 90's vintage. Of course, the cost difference between the two systems makes DC a more expensive option, so the default option is likely to be AC unless spec'd otherwise. There are some really bad CCGT stations - usually independent generators owned by investment groups - which appear to have been designed by the OEM's engineers and the client's accountants.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

A suggestion;
I understand that you have a soft starter installed but that it is not solving the problem. If you identify the soft starter someone here may be able to suggest a configuration or setting hat will solve your problem.
Tweaking settings may be cheaper and quicker than replacing equipment.
Some would say more professional also. (Grin)


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

 

[sry110]

Thanks all for your replies!

@ waross: Agreed, I found a bunch of wye-delta related threads but most were from at least a few years ago and consequently had been closed so I couldn't revive them.

The subject application is a new turning gear unit (albeit a retrofit application), so the starter is not yet determined. (The existing turning gear has an across-the-line motor starter, but we are looking to change it with some form of reduced current / torque starter). We have another unit in the field where the Customer supplied an Allen Bradlet SMC-3 soft starter, but I think he is having the problem jraef described where it sees no load and closes the contactors to give the motor full line voltage. Here is the link to AB's website for the SMC-3:

http://www.ab.com/en/epub/catalogs/12768/229240/229262/3073021/1179231/

@ jraef: Using your suggestion of a soft starter with current-limiting capability and external bypass contactors, is it accurate to say that the current will remain limited to a pre-set value regardless of motor speed / load, until the external contactor closes? And is it feasible to control the external contactor with a time delay circuit so that we can control when, chronologically, the motor goes across-the-line?

@ ScottyUK: Yes, this is a steam turbine / compressor drive application. We are concerned with loss of AC power and thermal distortion of the turbine and compressor rotors, so we have a manual turning provision on our gearbox to allow the operator to manually bar the rotor incrementally and hopefully prevent a severe bow from setting in. Unfortunately 110VDC is not available so we cannot use a DC motor - we are stuck using 415VAC/3-ph/50Hz. I agree, a DC motor with a starting resistor(s) would be the simplest approach. That's essentially what we are trying to mimmick using a wye-delta / soft starter / VFD.

Our system is designed primarily to achieve breakaway torque, i.e. the motor and all mechanical components are sized to achieve "infinite" cycles at the breakaway load. So the motor will be sized to achieve breakaway with the turning gear in the "locked" condition, which is where we expect it to be after creeping the overrunning clutch softly into engagement using the reduced torque / current starting means.

Typically on a brand new application (not a retrofit such as this), when the required breakaway torque is "x" we design the system to generate "1.5x" torque, and the system components are sized so their yield point is approx. 3x torque. So the theory is that if we stay below 50% of the component yield point (endurance limit for steel), then we have "infinite" cycles. HOWEVER, on this retrofit job we need to generate "x" torque for breakaway but we are limited to "2x" torque as a max limit due to small diameter of the existing turbine shaft that we are driving. So, we need to find a way to generate right around the "x" torque value so that we are able to break away, but we don't eat too far into the margin between torque required and torque max limit. By using a reduced torque / reduced current motor starting method, we want to rotate the unit through its unloaded condition to get the clutch engaged, i.e. remove the backlash from the system. We will size the motor so that its Locked Rotor Torque is sufficient to achieve breakaway from this "clutch engaged" position.

 

[sry110]

And a follow-up question:

On a 3-phase AC motor, is it possible to use a resistor to reduce starting voltage the same way it is done on a DC motor?

 

[jraef]

@ jraef: Using your suggestion of a soft starter with current-limiting capability and external bypass contactors, is it accurate to say that the current will remain limited to a pre-set value regardless of motor speed / load, until the external contactor closes? And is it feasible to control the external contactor with a time delay circuit so that we can control when, chronologically, the motor goes across-the-line?

The continuous current limitation capability is not universal, but many will do this. When you have it, it will indeed maintain the current setting until the motor accelerates (or stalls). But once the motor gets to full speed and the current drops below FLA, there is another difference in design that takes place. Some designs turn off the firing board, some do not. those that do not will continue to provide current limitation on a running motor as long as a bypass contactor is not closed. But this issue is somewhat irrelevant for you, all you need is for the current limit to remain ion effect long enough for you to engage the clutch and get past the shock. Most will do this for at least 30 seconds, some for longer and that should be plenty.

And yes, in those cases engaging the bypass with a separate timer is the way to do it. The bypass really is just to reduce the heat rejection from the SCRs into a sealed enclosure. A few extra seconds added delay is relatively unimportant. But just to be clear, the motor should be across-the-line before the bypass contactor closes!


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[sry110]

@ jraef: So we want a soft starter with current limitation capability that does not turn off the firing board when the motor is at full speed and current is less than full-load amps. With our motor being unloaded and requiring only about 5 revolutions until the clutch engages, it should see the load within a fraction of a second. I basically want the motor to sit there and hum (stall) for a second once the clutch is engaged (so I can be sure it is engaged), and then give the motor full line voltage so it can generate full locked rotor torque and achieve breakaway. My understanding was that it would not go across-the-line until the external bypass contactor closes - but then you said that the motor must be across-the-line before the contactor closes. Can you explain this further?

P.S. - off topic, but how do you quote a post on this forum? I can't seem to find a button to do it.

 

[jraef]

A soft starter only limits voltage / current but not frequency, which governs speed. Depending on load, your motor may (should) get up to full speed even with the voltage / current being limited. Understand that when I am referring to limiting current, I am NOT suggesting limiting it to a value BELOW FLA, but rather 3-4 TIMES FLA. So with a constant frequency and enough torque (as represented by current) the motor will accelerate, just more slowly and gently. But once it does, the reduction in voltage / current becomes less important and is no longer necessary. That's what the anti-oscillation circuit did; it detected the precipitous drop in current when the motor got to full speed and determined that since the motor had finished accelerating, the remaining ramp time (assuming a voltage ramp) was circumvented and the phase angle was set to full on; making the motor running across-the-line. Only then would the bypass contactor circuit close.

If you use a current limit start method that overrides the anti-oscillation circuit, you must STILL ensure that the amount of current you provide to the motor is sufficient to eventually accelerate the load, otherwise you stall and trip on overload. So a proper setup would be that you immediately limit current at start-up, the clutch engages, the current remains limited but is still enough to accelerate the load. The load accelerates, the current drops back down to normal indicating that the motor is getting full conduction (across-the-line) and THEN you close the bypass contactor.

If you close the bypass contactor before the motor is fully accelerated, the slip is still high, the flux penetration is still too shallow and the current will spike just as high as it would have had you started across the line (but more briefly) and there will still be a corresponding torque spike. That by the way is the same thing that happens with Wye-Delta starting too.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[waross]

jraef,
Apparently this is in the design stage. Would you care to contrast soft starts with a VFD? The OP may choose to spend more money for a less demanding setup.

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

 

[jraef]

Well if there is some additional benefit to using variable speed, then a VFD wins hands-down. But if not, a VFD makes for an expensive soft starter and the run time losses can significantly increase the cost of ownership, especially on larger sizes. Remember that bypassing a VFD when at full speed is a much more tricky operation that it is for a soft starter, so most people cannot do it and run on the VFD continuously. Even at 97% efficiency, that's 3% losses all the time.

But as far as functionality for this application, I think your idea is sound but I'm sort of on the fence. For years I always had 3 "Don't Bee" rules of VFD applications (only people old enough to remember Romper Room will know what that means): Don't Bee applying them on motors for Hoists, Cams and Clutches! But that was in the days before Vector drives, so I'm not so sure any more. Lots of people have used them successfully on all three of those applications, including myself on cams and hoists, so I may have to modify my stance on clutches as well. I have just never done it, mostly for lack of opportunity; clutches are a lot more rare than they used to be.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[waross]

You the man Jeff. Thanks for the comparison. LPS.
Before anyone may be misled by one of my previous posts, this was fractional HP motors with single phase power. We used VFDs to drive three phase motors. A great solution but not applicable in this instance.

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