High speed solid state motor contactors?

[jraef]

Anyone know of a source for high speed Solid State contactors that can switch off an AC motor of around 90A 460V in 11ms? I am using an SCR based contactor now and the disconnect time is approx. 80ms from command, but that is apparently too slow. The client was using an IGBT based contactor for this up until the supplier went under and he is looking for a replacement. Because it did not wait for a zero cross, it cut off power in 7 - 11ms from command (anecdotal, I never saw it work). The application is a 30HP 460V 37FLA high speed cutter motor with a worst case duty cycle of about 6 starts per minute.

The SS contactor is only used as an emergency disconnect when the blades of this machine jam, but I mention the duty cycle because this contactor, upstream from the normal reversing motor controller, will still see all of the starting current from a thermal standpoint. I picked 90A as a rating to derate for the extra current it will see.

So far the 90A SCR contactor is holding up thermally, but they want something faster because they have a 20ms delay on jam detection, plus this disconnect delay, plus the motor field collapse delay of about 28ms before the motor torque is gone and in that time, the high speed blades are at risk for serious damage and or destruction. At the 11ms disconnect with the IGBT contactor they had before, it was apparently fast enough to prevent catastrophic failures. I can't seem to find anyone selling IGBT based contactors for anything other than DC applications.

 

[Skogsgurra]

Hi Jeff,

If it is used as an emergency disconnect only, the life of a normal AC coil contactor would be almost infinite. And they are quite fast. Maybe not 11 ms, but I would say 20 ms. And it doesn't have to wait for zero either.

Disconnecting the motor is one thing. But how do these guys stop it? Will it not coast to stop when disconnected?

Next thing I would try (yes, probably way too expensive) is to use a vector drive. They have torque loop step responses in the sub-ms region. Guaranteed to stop the motor in a 11 milliseconds if power rating is correctly chosen.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Why wouldn't they use a magnetic clutch? That would shed the whole inertial train-wreck resulting from a jam.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Hi Smoked,

Most magnetic clutches, at least the ones I have seen, are DC with time constants in the 1+ second region. I think that would be too slow.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[jraef]

Thanks for the thoughts guys.

Keith,
Gunnar is right, way too slow on the uptake. They don't need to stop the blade, just stop applying forward torque to it. It's actually already pretty much stopped by the time they detect the jam with the current sensor. I had said high speed but that is for normal operation. The fast disconnect is for when they can't do their job.

Gunnar,
Although you can disconnect a contactor coil at any time, the mechanical opening time of the main power contacts is on the order of 80ms to full arc interruption for the fastest ones I could find (Siemens naturally). The worst were 180ms, with most of them bunched up at around 110 - 140ms.

OEM is too scared (and cheap) to put in a VFD and there is no need for variable speed for any other reason. Still, I brought it up from a component count perspective alone because it eliminates the high speed disconnect contactor plus the reversing contactors plus (I believe) the current sensor they are using. It also would help with the duty cycle issue in terms of wear and tear on the motor. But they are extremely reluctant to change something that works for them. They apparently changed some things years ago and it almost bankrupted them when it didn't work right, so now the owner is gun shy. As an example, they use a 10 year old Mitsubishi PLC, to which they had to add a micro box PC just in order to attain communications! The owner will not consider changing PLCs until Mitsubishi officially obsoletes the one he is using now (which is imminent).

 

[Skogsgurra]

Don't laugh now, Jeff.

You can use DC SSRs. And, to make it more attractive, you can use it in two phases only.

This is how: Put rectifier bridges in two motor phases. AC in/AC out. Then put a DC SSR across each DC out. Activate the SSR to short output and let current through to the motor. Open SSR to stop motor current. Snubbers or MOVs may be needed.

There may be some issues with motor voltage unbalance. If so, then use three rectifier bridges.

Now, you can laugh...

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

To continue my angle why not a pneumatic mechanical clutch?
They could be faster than a contactor. Then you just stay with a simple run-of-the-mill contactor, turning off the motor when you get around to it.

Or any one of the dozens of torque limiting clutches that work instantly. If the tool doesn't break at the motor's normal torque it certainly won't matter if it is stationary with the same torque.

I just think this is one case where electrical is a whole lot bigger pain in the rear if VFDs are out of the running.

Oh, I had no idea eddy current clutches were that slow.. Thanks for the info.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[ozmosis]

Jeff
The latest Sinamics G120 drive including the Failsafe functions (CU240S DP-F) will give you a Safety Torque Off in under 1ms. I'll email you some of the training foils on the subject as I guess they take their time to make it over to California....

 

[ozmosis]

actually, I gave you duff information. It is <4ms, not 1ms. Still pretty quick though.