VFD DC-Link Quick Disconnect? 2

[WWWJD]

Hello all,
We have a little project where we're tying ten 5HP 480VAC drives together on a common buss setup. These drives are on portable tables, and I'm wanting to find a way to do a quick disconnect on the DC connections. 1kV, 5-6A, 2 or 3 pole, finger safe on both the male and female side. My other option would be to some up with some kind of safety interlocking via contactors, but I haven't really thought that through thoroughly. The risk is that operators will potentially be making/breaking connections to a hot DC link. The easy answer is "don't do that" but reality is it will probably happen at some point. Just trying to make sure the electrons stay on the right road.

Any thoughts or recommendations? I'd like to find a good fingersafe connector and be done with it. Not sure where to look first.

Thanks,

 

[ScottyUK]

Bill,

Slightly off-thread -

I've been reading the official report from the Public Inquiry into the Piper Alpha disaster by Lord Cullen. It's a long and tiring read, but hugely educational. 30 years later similar mistakes are being made today: lessons haven't been learned, or maybe the lessons learned at such a cost are being forgotten as those who learned them retire.

 

[WWWJD]

I'll provide a little clarity into the setup that I'm trying to glue together. What this is in a nutshell, is a dyno setup, using a set of regen drives (ABB 880 and ABB ACS 800). The setup is as follows:

One 40HP ACS 800 regen drive fed from 480VAC, mounted to the wall. The DC Buss connection is brought out from this drive.

Ten 480V VFDs will be running ten motors, the devices under test.

Those motors are going to be coupled to ten additional motors as loads.

The load motors, are going to be connected to ten ACS 880 drives, running in torque mode. The ACS 880 drives will be sharing the DC Buss with the big 40HP regen drive on the wall.

The idea here, is to find another way to dyno test motors without the classic powder brake or H-brake setup (very expensive at our torque levels; 2000+ lb.in. at low RPM).

The caveat is that these motor combos are going to be spread on 5 somewhat portable tables, so they can be rolled in and out of an environmental chamber, where all this magic sauce is going to be stirred.

I've gone back and forth with another engineer, and I think we've decided that the only practical way to bring a table in/out of service is to bring the whole system offline. There will still be some quick disconnect going on for AC mains and drive motor interlocks. We're assuming that to keep the DC side of the system happy, we'll have to sequentially start the drive motors. So the original post was centered around a way to quickly and safely disconnect those DC buss connections. At this point though, I'm thinking there won't be a practical way to do it. We're also gambling on if we can even get this complex setup to work. We'll save thousands of dollars in hardware cost. We'll probably spend 3x as much in labor trying to make it work.

I will add that yesterday, we at least confirmed that the pre-charge circuit in the 40HP unit has enough capacity to bring itself and the 10 slaved drives up simultaneously without losing it's lunch.

 

[waross]

Thank you for the very good description of the background to the problem. lps

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

 

[stevenal]

Regarding the KIRK spare key story, all locks can be defeated. If regular locks are meant to keep honest people honest, then interlock keys are meant to keep safe people safe. I've seen them used on gang operated air break switch handles where the whole locking mechanism could be removed with a wrench much faster than a spare key might be found. Idiots need not apply, since it was never intended to be idiot proof.

 

[itsmoked]

I'm still trying to get my head around this.. I'm a visual person so it's hard for me to visualize descriptions.


40HP regen on wall.

Bunch of 10HP non-regens fed only from the 40's DC-Link.

Matching 10HP regens matched to each of the 10's above. They regen directly to the mains.

If that sums things up...

1) The only regen would come out of the 10HP regens, so why is the 40 a regen?

2) Why the 40HP at all?



Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

Keith, The motors under test will be driving the load motors (not braking them). Power then goes from the load motor to the load VFD to the 40hp VFD on the wall back to line power.

The decision to power everything off when changing the connections is by far the best decision that could be made.

 

[waross]

As I visualize the setup, the 40 HP is the input power to the dyno. The DC buses will all be tied together so that the 10 HP drives may regen into the 40 HP DC bus.
I don't believe that the 40 HP needs to be a regen.
I don't believe that the 10 HP drives need to be AC regens.
If the 10 HP drives are regen, then you don't need to agonize over connecting/disconnecting.
You could let the drives regen through the AC mains and disconnect the AC supply to the drives.
When the Dc links are tied together you may regen through the DC link more efficiently and do not need full AC regen drives.
If I am wrong, then I need more help with the explanation also.

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

 

[waross]

Does the 40 HP drive drive a mechanical load or is it used only to regen back to the AC mains?

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

 

[LionelHutz]

Also, the OP posted high torque at low rpm. So the 10hp load motors likely aren't running anywhere close to their rated rpm which means they are not running anywhere close to 10hp. Hence the ability of a 40hp regen drive to handle the regen energy from 10 x 10hp motor+VFD equipped loads. Based on the component sizing, the load motors will be running at <40% of rated speed.

 

[WWWJD]

"Does the 40 HP drive drive a mechanical load or is it used only to regen back to the AC mains?"

It's only there for regen back to the mains; more volume in the accumulator so to speak as I understand it. I'm kinda in the middle here; another guy spec'd all of this stuff out. So I can't honestly comment if the 40HP is really necessary or if it's just there as a just-in-case maneuver.

 

[jraef]

Another danger to connecting the DC busses of the downstream loads after the system is already energized is that you will be circumventing the pre-charge circuits. One mishap could take down the entire setup.

You must, for several important safety and reliability reasons, design it to disallow any reconfiguration while anything in the system is powered (plus the wait time). Force the power down, open a line-side contactor that controlls a delay timer that engages an electrical door interlock controlling access to any DC circuits. Having a line contactor is not ideal, but necessary if you proceed like this.

A better idea I have used for test skids, albeit not with the DC bus ties, would be to leave all of the drives tied together all of the time, whether used or not, and simply put pin and sleeve connectors on the outputs of the drives going to the motors. So your “test tables” have the motors on them, but not the drives themselves. Then use the receptacles that are tied to a disconnect switch with an aux contact on it and run that six contact into the enable circuit for its respective VFD so that they can’t plug in or unplug the motor with the drive running.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden