CamBam
News:
 
*
Welcome, Guest. Please login or register.
Did you miss your activation email?
December 10, 2019, 08:28:23 am


Login with username, password and session length


Pages: [1]
  Print  
Author Topic: My VFD destroyed - this is the reason  (Read 255 times)
Dragonfly
CNC Jedi
*****
Offline Offline

Posts: 2224



View Profile
« on: December 03, 2019, 16:41:59 pm »

Yesterday I was sitting listening to the router doing a long job when suddenly there was a popping sound, the router stopped together with the light. Fist thought - a power break. Then I saw the automatic mains fuse down. And when approaching the router - the smell of electric discharge from the spindle.
After some time for careful inspection of the power connector I discovered that a 1/3 circular segment had broken from the side wall of the metal body and obviously caused a short between two phases.
As a result the VFD (popular Chinese Huanyang brand) was fried.
I think I can repair it - one pair of IGBT in one phase are damaged and the optocoupler controlling them too. But it will take time because I ordered the components plus a new connector from China.
Therefore I bought a new VFD today.
On the photo you can see the splintered metal wall which caused a significant damage to my wallet Smiley


* DSC_0118.JPG (1331.63 KB, 2397x1542 - viewed 37 times.)
Logged
lloydsp
CNC Jedi
*****
Offline Offline

Posts: 8077



View Profile
« Reply #1 on: December 03, 2019, 18:52:19 pm »

Now, what the heck would have caused that INSIDE a connector housing?  Plus, if the connector had been fully-seated, it should have 'bottomed-out' in the housing, preventing any chips or debris from getting across the pins!

I don't think I understand how that could have happened.

Lloyd
Logged

"Pyro for Fun and Profit for More Than Fifty Years"
Bubba
CNC Jedi
*****
Offline Offline

Posts: 2706



View Profile
« Reply #2 on: December 03, 2019, 19:00:00 pm »

Sorry to hear that, It does not make sense to me at all. As Lloyd stated the plug wasn't seated all the way for this to happen.
Logged

My 2¢
Dragonfly
CNC Jedi
*****
Offline Offline

Posts: 2224



View Profile
« Reply #3 on: December 03, 2019, 19:08:00 pm »

There is a ceramic cylindrical body with the pins which is inserted and rests on a step inside the housing. The back of the housing is then rolled in to seal the ceramic inside. Perhaps too much force on the step caused a crack in the metal which is as we call it here ZAM cast alloy (Zinc-Aluminum-Magnesium) - very grainy and easy to break.
Not a long time ago I took out the plug for inspection but didn't peek inside the socket. I suppose at that time the broken metal stuck out and when tightening back the plug it was caught between it and the bottom. But the retaining threaded ring tends to loosen in time due to vibrations and at some moment the same vibrations drove the metal splinter closer and closer to the pins.
When I took out the plug yesterday it had some metal traces on the front and the metal piece was lying flat on the bottom.
At first I thought fine aluminum chips got somehow inside but it is practically impossible.
Logged
EddyCurrent
CNC Jedi
*****
Offline Offline

Posts: 4077



View Profile
« Reply #4 on: December 03, 2019, 22:28:20 pm »

When I worked full time in industry we used to use quite a few 120kW inverters, these were used to drive multiple motors. There were others, commonly 7.5kW, driving a single motor but both systems had one thing in common, plugs and sockets between the inverter and motor were often a cause of breakdown.
High frequencies and transients in the supply would cause flashover between pins and the insulator showed signs of discolouration similar to that I've seen in 11kV systems.
In critical processes we tried to make it easy to swap out faulty inverters by making plug and socket interfaces but as I said these often became the cause of a breakdown themselves.
Chokes on the inverter output helped the situation but sometimes these would blow up and cause a fire.
Logged

Made in England
Bob La Londe
CNC Jedi
*****
Offline Offline

Posts: 3807


^ 8.5 pounds on my own hand poured bait.


View Profile WWW
« Reply #5 on: December 04, 2019, 00:37:49 am »

Dang it.  That sucks.  Glad you can afford to get it back up and running quickly atleast. 
Logged

Getting started on CNC?  In or passing through my area?
If I have the time I'll be glad to show you a little in my shop. 

Some Stuff I Make with CamBam
http://www.CNCMOLDS.com
Garyhlucas
CNC Jedi
*****
Offline Offline

Posts: 1311


View Profile
« Reply #6 on: December 04, 2019, 03:42:32 am »

Carbon tracking across phases. Very common problem on 480vac and higher systems. Dust and a little humidity and things go bang. Consider a little silicone dialectric grease on the connector.
Logged

Gary H. Lucas

Have you read my blog?
 http://a-little-business.blogspot.com/
Bob La Londe
CNC Jedi
*****
Offline Offline

Posts: 3807


^ 8.5 pounds on my own hand poured bait.


View Profile WWW
« Reply #7 on: December 04, 2019, 18:32:04 pm »

Carbon tracking across phases. Very common problem on 480vac and higher systems. Dust and a little humidity and things go bang. Consider a little silicone dialectric grease on the connector.

Interesting. 

I would note that my new more expensive high speed ATC spindles do not have a power connector.  They have a wired pigtail coming out of the spindle case.  I connected them with wire nuts and filled the nuts with silicone dielectric grease. 

My older spindles did in fact have a familiar 4 pin electrical connector, but they were not all wired the same, at least one did not have a connected ground pin, and they were not all the same size connector.  I used a meter on every single one to identify windings, and ground.  I do not worry about phase rotation, although I have a meter that is supposed to have a way to determine that (useful for some types of pumps which can be damaged if run backwards).  I just power it up, and if it runs backwards I swap any two phase leads.   
Logged

Getting started on CNC?  In or passing through my area?
If I have the time I'll be glad to show you a little in my shop. 

Some Stuff I Make with CamBam
http://www.CNCMOLDS.com
Dragonfly
CNC Jedi
*****
Offline Offline

Posts: 2224



View Profile
« Reply #8 on: December 04, 2019, 20:25:51 pm »

Carbon tracking across phases. Very common problem on 480vac and higher systems. Dust and a little humidity and things go bang. Consider a little silicone dialectric grease on the connector.
I don't have much knowledge or experience with HV. So those are valuable things to know. Have to see whether such a grease is available here.
Although rated as 3 phase 220vac it's I suppose RMS. As the mains 220vac is rectified and about 300V DC is being fed to the IGBT bridges.
The connector itself is called "Aviation type" as far as could learn. I ordered a new one (its type is H17) together with the semiconductor components that got fried from Aliexpress and intend to repair the VFD.

The internal overcurrent protection is helpless in case of a short on an open IGBT as it is turned immediately into a true conductor and a shut off gate signal has no effect on it.

I was thinking, are there fast acting fuses which could eventually prevent the damage?
« Last Edit: December 04, 2019, 20:31:53 pm by Dragonfly » Logged
Garyhlucas
CNC Jedi
*****
Offline Offline

Posts: 1311


View Profile
« Reply #9 on: December 05, 2019, 00:01:56 am »

It is tough to protect from an arcing fault phase to phase or phase to ground for motor loads. This is because an arc has resistance and the current may not even be as high as normal running current.  If the fault current is flowing to ground then a ground fault protector would work. Simple concept, put a current sensing coil around all three phases. If you read any current you have a fault to ground.

There are now arc fault sensors that sense the characteristic of arcs but I don’t know of any yet for motors.
Logged

Gary H. Lucas

Have you read my blog?
 http://a-little-business.blogspot.com/
Dragonfly
CNC Jedi
*****
Offline Offline

Posts: 2224



View Profile
« Reply #10 on: December 05, 2019, 09:09:37 am »

Gary,
I don't think there was sparking but rather an instant complete short between two phases. The VFD has two internal current transformers on two phases and it can detect relatively fast rise in current but not an instant jump when current is times higher than the maximum rated collector-emitter current of the IGBT. In my case 80A DC. This is worsened by the presence of a large high voltage capacitor which discharges in an instant.
When the spindle is not running there is no voltage (resp. current to detect) by means of a sensing coil. Better to measure coil resistance and terminals leakage to chassis or ground.
IMHO only fast blow fuses could eventually save the VFD.
Logged
Garyhlucas
CNC Jedi
*****
Offline Offline

Posts: 1311


View Profile
« Reply #11 on: December 05, 2019, 15:50:48 pm »

Dragon,
Lets say a chip got between two phases.  The current needed to vaporize that chip could easily be well less than the current needed to run the motor and still do a lot of damage like carbon tracking on the insulator.  Fuses for motors also have to be well oversized to handle the higher current during acceleration and momentary overloads like when the tool goes into a tight corner.  So a static device like a fuse has no hope of protecting your equipment.
Logged

Gary H. Lucas

Have you read my blog?
 http://a-little-business.blogspot.com/
onekk
Wookie
****
Offline Offline

Posts: 497


View Profile
« Reply #12 on: December 06, 2019, 08:10:35 am »

In critical applications, such those in automotive sensors, one old method was to put the connector not on the componed but with a little piece of wire away from the component, the piece of wire acts as a "shock absorber" for the vibrations, but the cable has to be built with vibrations in mind, plus the connector could be some industrial or even military grade ones, I've found many interesting connectors in "old" '50 military equipments sold in the "surplus" that are virtualy bullet proof, and under 10 dollars for a 7 or 8 pin connector, the downside are generally dimensions and weight, and somewhat the olive green color.

Regards

Carlo D.
Logged

Carlo D. (onekk)

eShapeoko #343 750x1000 mm + GRBL + bCNC + CamBam
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Valid XHTML 1.0! Valid CSS! Dilber MC Theme by HarzeM
Page created in 0.132 seconds with 20 queries.

Copyright © 2018 HexRay Ltd. | Sitemap