Metal Oxide Varistors burns out

Go To Last Post
25 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi

The below Power Supply Circuit is a part of Control Panel. the panel is manufactured in USA and installed here India in various locations.

The Metal Oxide Varistor  MOV1 and MOV2 burns out on couple of panels and on one panel even LM2576HV blown out.

The MOV is SMD TYPE Blue Color chip after searching i found look similar is Liitle Fuse Make CH Series Varistor.

i think the panel manufacturer is using 40Vrms Metal Oxide Varistors since the Transformer secondary is 36V.

Input 230V is connected through surge suppressor in some case with out surge suppressor
 

i suspect that if the Input 230V more than 250V the secondary voltage go close to 40V of MOV then the smoke starts

but how the MOV1 and MOV2 burns out Not MOV3

any other reason for MOV1 and MOV2 to Fail??

 

 

P.Ashok Kumar

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What happens if the 0V rail is taken to earth?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

In Working panel it works ok the panel indicate that earth Fault.

the panel has earth fault detection circuit.

 

 

P.Ashok kumar 

Last Edited: Thu. Feb 19, 2015 - 07:46 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Nothing like having a 110V US transformer being fed with 230V by any chance?

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No the transformer is 230V only. the panel works ok for many months

 

 

Last Edited: Thu. Feb 19, 2015 - 08:51 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:

What happens if the 0V rail is taken to earth?

 

Agreed. Unless I'm missing something the only way a voltage can appear across the MOVs is for 0v, or I suppose the 24v rail, to be connected to mains earth.

 

Also, to the OP, I think you really need to find out what those MOVs are, there's too much guesswork based on them being blue going on.

#1 This forum helps those that help themselves

#2 All grounds are not created equal

#3 How have you proved that your chip is running at xxMHz?

#4 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand." - Heater's ex-boss

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have used many an MOV, never had an issue

 

You can physically hear them working if you ever do any EMC testing, transients and surges can be heard being shunted!

 

But I think the issue here is a 36V AC output will have a peak of 50.9V

 

Surely this is the issue here, to me if an MOV is activating then its voltage has been exceeded

 

Best tip is to also fit a TVS diode this makes a much better sytems

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes the 0V or 28V rail always chance to go to mains earth or  any metal body on the field.

the panel suppose should to be work for single earth fault and the panel has earth fault detection circuit.

 

at present we don't have the MOV details we requested the details from manufacturer.

when the panel comes for service we have to remove the MOV and test for operating / breaking voltage.

 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bignoob wrote:

 

You can physically hear them working if you ever do any EMC testing, transients and surges can be heard being shunted!

 

But I think the issue here is a 36V AC output will have a peak of 50.9V

 

 

the panel is UL approved panel so all EMC testing should have passed

The parent company is in USA manufacture supply the panel with out any surge suppressor.

the local office add the surge suppressor (which don't have any name) for reducing the cost parent company

might have removed the surge suppressor or due to the supply voltage is different from USA.

 

Last Edited: Thu. Feb 19, 2015 - 10:05 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The USA is actually behind the rest of the world when it comes to EMC and energy efficiency stuff

 

They just don't care, in EU EMC is a massive thing its something I have had to learn and I am quite experienced with EMC filters and problem solving, its so satisfying to see your product produce very low levels but I digress

 

 

You need the details of the MOV but I can assure you that you will be exceeding its ratings so I would just source a proper rated part and also fit a TVS diode, its not optionl IMO

 

 

Good luck

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ok we will try to get  MOV Details.

 

what is normal voltage deviation on mains power supply in USA

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I live in the UK but I am sure it will be of a similar quality

 

As an ex elec trician I can tell you its 230V +6%, -10%

 

they changed these numbers so we were harmonised with Europe, nothing physical ever changed except the colours

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For equipment designed to European standards, if the rated voltage is 230v single phase or 400v three-phase, then the tolerance is +/-10%.

#1 This forum helps those that help themselves

#2 All grounds are not created equal

#3 How have you proved that your chip is running at xxMHz?

#4 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand." - Heater's ex-boss

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For equipment designed to European standards

But he asked abourt the tolerance of the mains supply, this is in the hands of the power companies and beyond anyones design, the tolerance is currently +6% -10% for voltage magnitude it was changed when the harmonisation came into effect so that the EU electricity could be harmonised without the UK having to change all the transformer tap settings

 

The frequency is much more tightly controlled because its used to measure time in many cases

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Having blown up many protection devices, i consider myself an expert! MOVs tend to degrade, so if you have lots of small transients it is death by 1000 cuts. If you sketch out the possible current paths, you'll see it is a silly placement for movs1,2. What are they trying to protect against? Anyway, ehen they degrade, they get resistive then they get hot, degrade some more, get hotter until you get smoke. If it were a gross overload, they would vapourise. Rule of thumb - if you have protection devices, ensure they are protected.
TVS diodes like going short circuit, so unless you want a burnt circuit board, place fusing inline. What's the use of placing a mov or tvs to protect something if it fails and the unit no longer works? I dare say for UL, you need fusing for MOVs as they are well known for going up spectacularly.

For this little problem, i would look at the whole system. Especially where there could be transients or voltages generated or current paths from mains earth to 0V. I'd scope the 0V line in relation to earth and see what is happening. I think there's something goung on that wasn't anticipated. I also think mov1,2 are poorly placed and mov3 should be fused. I know this from bitter experience.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman, good post

 

Seriously dude I can't keep up with you!, I need to learn that MODBUS stuff (its high on the list) and get my IoT data capture stuff going but lately I am fighting battles on many fronts and I just do not have time but its coming!

 

I am interested in your take on MOV's TVS diodes, we have to gfit them to our products for EMC, transients and surge protection, withouit them the products fail

 

Now I never seen a problem so far (doesn't say a lot) but I know some of our customers use MOV and TVS diodes in systems with a 20 year lifetime and these products have worked flawlessly for years and years in horrendous environments like railways etc

 

maybe a proper sized device is half the battle, like I said I have physically heard them doing their job which has given me confidence in them but your comments have me wondering if my confidence is justified?

 

Looking at the schematic I would have the MOV's on the mains supply where the transients and surges are coming in, I also do not see why the secondary side has its ground refrenced to earth, I know its sometimes required but looking at the schematic its hard to see why!

 

Edit

 

What are you like with analogue Kartman?, Guru?, rusty?

 

I have a big interest in  this and I have worked on some nice simple current limitting circuits lately, I am looking to improve my knowledge a bit

 

 

Last Edited: Thu. Feb 19, 2015 - 12:08 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bignoob wrote:
... Best tip is to also fit a TVS diode this makes a much better system....

Yeah, I would fit a TVS diode which works to a narrower margin. It appears the varistor voltages are both too low (not uncommon line noise is making them clamp and fail) and not low enough (they are clamping to a voltage that is too high for the regulator).

- John

Last Edited: Thu. Feb 19, 2015 - 12:51 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:

Having blown up many protection devices, i consider myself an expert! MOVs tend to degrade, so if you have lots of small transients it is death by 1000 cuts. If you sketch out the possible current paths, you'll see it is a silly placement for movs1,2. What are they trying to protect against? Anyway, ehen they degrade, they get resistive then they get hot, degrade some more, get hotter until you get smoke. If it were a gross overload, they would vapourise. Rule of thumb - if you have protection devices, ensure they are protected.
TVS diodes like going short circuit, so unless you want a burnt circuit board, place fusing inline. What's the use of placing a mov or tvs to protect something if it fails and the unit no longer works? I dare say for UL, you need fusing for MOVs as they are well known for going up spectacularly.

For this little problem, i would look at the whole system. Especially where there could be transients or voltages generated or current paths from mains earth to 0V. I'd scope the 0V line in relation to earth and see what is happening. I think there's something goung on that wasn't anticipated. I also think mov1,2 are poorly placed and mov3 should be fused. I know this from bitter experience.

 

Normally we see the burned out MOVs on Mains Supply side and the reason can be understand that the MOVs fails are due long operation degraded by surge or dead by high voltage but in this circuit we are not able understand the logic how the MOV1 and MOV2 only fails we may have to gather details from users it may take long time and some user may not be able to give the details like voltage fluctuation, surge  and cable grounding/Earth.

the manufacturer says the hundreds of panel which sold in USA are not having this issues. so we suspects the  supply voltage

here in India the input Voltage can between 210 to 255.

 

thank you all for valuable inputs

 

 

 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bignoob wrote:
I also do not see why the secondary side has its ground refrenced to earth, I know its sometimes required but looking at the schematic its hard to see why!
I second that thought.

Usually there's a high reliability Y capacitor (0V to earth).

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bignoob wrote:
I am interested in your take on MOV's ...

...

in horrendous environments like railways etc

Another challenging environment is by the transformation of rural into light industrial by petroleum operations; lightning can be a contributing factor to the lifetime of surge suppression.

DELTALA Why a Delta Silicon Arrestor

http://www.deltala.com/whydelta.htm

...

MOV devices are much like Delta silicon arrestors EXCEPT that the metal electrodes are separated by zinc oxide rather than silicon. The functional differences include: ...

P.S.

An example of how lightning degrades a retail device :

Bench Talk for Design Engineers

THAT Doesn’t Need Circuit Protection!

On February 11, 2014 in Circuit Protection by Kelly Casey

http://www.mouser.com/blog/that-doesnt-need-circuit-protection

"Dare to be naïve." - Buckminster Fuller

Last Edited: Thu. Feb 19, 2015 - 08:54 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ashokok wrote:
we may have to gather details from users it may take long time and some user may not be able to give the details like voltage fluctuation, surge  and cable grounding/Earth.
Would the custormers and operators state the exposure to lightning?

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Noob, i'm probably technician grade when it comes to analog. I have a knowlege of most building blocks, but don't expect me to design them!
Normally you'd use movs and tvs for transient and lightning protection. In these applications the total amount of energy involved can be handled by the device. In some cases you might expect an overvoltage to happen and the energy involved can destroy the device. If you've ever plugged a pc psu set for 110V into 240V and looked at the results. The poor old m205 fuse could not break the fault current so the movs exploded and the switching transistors died. With a lot of gear i designed we has rs485 comms in a multicore along with 24VDC. This is generally not seen as good practice - for good reason! Invariably the site wiring was done wrong so you got 24VDC on the comms. The average rs485 transceiver doesn't like this and promptly dies. So you add TVS diodes for protection. Sounds reasonable? But the tvs diodes are thermally limited so applying 24V to a 5V1 tvs ends up burning the pcb. So you add polyfuses in series to limit the current. Net result you get a design that has transient protected comms as well as overvoltage without dying in the process. Except if the customer applies 240VAC! That was a lesson in 'poke yoke' (Google that). At some point you need to draw the line as to what would be expected and what the design can tolerate.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ashokok, if you can find out the day the unit died, search the weather reports to see if there were storms that day. Can you set up the system as you would expect the customers to be wired as? Or at least sketch the schematic noting cabling etc. it could be sonething somple like running the dc parallel to some high current cables. You could also collect some history info - how long does it take to fail and how many.
My suspicion is that it is to do with earth vs 0V. In which case the placement of the movs is not ideal.
With these situations you need to collect as much data as you can. The answer is there - you just need to find it.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Kartman,

 

My use of MOV's TVS diodes has been to do with the EMC directive, as you know some standards have very harsh levels

 

I know a fair amount of analogue electronics and its an area I put a lot of effort into getting better at, I always used diodes for polarity protection but these days I like using MOSFETS which can give polarity, inrush current limit and overcurrent protection its all really neat but I do wish I was much better with transistors, its actually amazing the amount of things you can do with transistors

 

The OP needs to seriously consider this system, its a WTF system

 

Its like the MOV's should be a couple of Y caps from an EMC filter, the Y caps and the common mode choke has been removed from the input as has the X cap

 

And there isnt actually any surge protection to protect against power surges at all, because power surges not surprisingly come on the power supply, maybe thats why they called them that!

 

And then for extra WTF they reference the 0V line to the main earth!

 

WTF?

Last Edited: Fri. Feb 20, 2015 - 12:17 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Noob, I think my solution would be a IEC input filter & fuse unit for the mains input and a X rated cap from 0V to earth. The root cause is external to what we've been shown and MOV1,2 are causing currents to flow in the wrong paths when there is a transient.