Thermal fuse for a metal oxide varistor

Thread Starter

xchcui

Joined May 12, 2014
134
Hi.

I saw a circuit that used a thermal fuse/thermal cutoff fuse(TCO)tied
to a metal oxide varistor(MOV)in order to cuttoff the current to the varistor,when the varistor
is reached to a specific temperature.And that is in order to prevent the varistor to burn.
The TCO comes in a different values of clearing temperature(the temperature that it cut the current).
What is the TCO clear temperature that should be chosen for a varistor?
In my case:a 7D391K(230VAC)varistor.In the varistor's datasheet of several brands they mentioned
only the operation temperature of the varistor(max?),which is about 125°C,but i didn't find a mention
to the combustion temperature of the varistor.
I would like to choose a temperature for the TCO to prevent the varistor to burn,but not a temperature that will everytime will open the TCO(permanently),just because the varistor temperature is rised alittle above its limit,while it doesn't burn.

Thanks.
 

MisterBill2

Joined Jan 23, 2018
3,861
Such a thermal fuse might prevent damage to a varistor, but mostly it will allow the protected circuit to return to normal operation. A common failure mode of varistors is to stay in conduction after the transient that caused them to conduct has passed. So at that point the device will continue to flow current until either it burns open or else the circuit's over-current protection switches off. Thus an over-temperature cutoff may provide some protection in some cases. But a better approach will be an over-current limiter device in series just before the varistor. If that device opens quickly enough it may protect the varistor and allow it to reset to a high resistance value.
 

Hymie

Joined Mar 30, 2018
708
My understanding of the failure mode of varistors is that over time they go leaky, resulting in increased power dissipation across the device (under normal voltage conditions), which can result in the component burning.
Some varistors are available with integral thermal fusing to protect against this.


The product safety standards IEC/EN 60950-1 and 62368-1 mandate overcurrent protection for all varistors in a primary circuit to protect against this. You should select an approved varistor that meets the requirements of Annex Q of IEC/EN 60950-1.


If you choose a device without integral thermal fusing, I would recommend you position the varistor such that should it burn, there are no other components in close proximity that could propagate the fire (varistors complying with Annex Q above, meet a minimum specified flame rating for their enclosure).
 

Thread Starter

xchcui

Joined May 12, 2014
134
Such a thermal fuse might prevent damage to a varistor, but mostly it will allow the protected circuit to return to normal operation...
I would like to correct myself.I meant to say that the TCO is tied to the varistor,but the TCO is connected in series to the circuit in a way that it should cutoff(permanently)the current to the circuit,so the circuit won't return to a normal operation.
Refer to the integral thermal fusing,i read about them before,but i would like to understand
my particular situation.Sometimes there isn't a safe space to connect a varistor,so a TCO is a good idea.
Anyway,may you focus on my specific question?since my main issue is what value of TCO temperature should i choose according to my other details?
Thanks.
 

Hymie

Joined Mar 30, 2018
708
Normally the TCO provides protection to the varistor only – it being in series with the varistor (only).


Once the TCO operates, the varistor will not provide any protection against overvoltage spikes from the mains supply – but the operation of the TCO will have prevented the possibility of a fire starting within the equipment.


If the TCO is protecting the complete circuit, then once it has operated it will need replacing to make the equipment operational once more.
 

Thread Starter

xchcui

Joined May 12, 2014
134
If the TCO is protecting the complete circuit, then once it has operated it will need replacing to make the equipment operational once more.
Yes,this is what i refered to.So,may you refer to my main question
about choosing the right temperature of TCO?
 

Thread Starter

xchcui

Joined May 12, 2014
134
I will try to focus my question.
If the maximum temperature of a varistor is 125°C,should i
choose TCO that open at 125°C exactly,should i choose a lower openning
temperature or maybe higher?

Thanks.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
3,861
Time is what makes determining the answer to your question a challenge. The rate of temperature rise in a varistor that has turned conductive is quite high, while the transfer of enough heat to raise the temperature of the thermal fuse will take some time. And then there is the other side, which is that if the varistor becomes leaky it will heat up fairly slowly.
So my suggestion is to forget the thermal fuse idea and use either a fuse or a fairly fast acting circuit breaker. I do not recall what the thing that you are protecting with the varistor is, but it must have some sort of turn-on characteristic current profile. If it does not have any larger inrush then just a close rated fuse or breaker will work. So please clarify about the turn-on inrush current, if there is any.
 

Hymie

Joined Mar 30, 2018
708
Millions of products are on the market having varistors placed between live & neutral without any problems. There should be no need for thermal protection where the varistor meets Annex Q.
However I would recommend that the varistor is rated at 300Vac (as required by IEC/EN 60950-1), a 230Vac rating could result in problems where the supply voltage may be as high as 250Vac.
 

MisterBill2

Joined Jan 23, 2018
3,861
Millions of products are on the market having varistors placed between live & neutral without any problems. There should be no need for thermal protection where the varistor meets Annex Q.
However I would recommend that the varistor is rated at 300Vac (as required by IEC/EN 60950-1), a 230Vac rating could result in problems where the supply voltage may be as high as 250Vac.
The product of mine that included a varistor also had a regular fuse protecting everything, including the varistor. When the varistor failed the fuse popped and the varistor stayed in the "ON" mode, short circuiting the power. So I removed it and the TV lasted many more years. It was a physically small device, not intended to do more than protect against very small line spikes. And with only a part number it was not possible to find a replacement. And the local TV shop had totally no clues about it. It seems that they were mostly a bunch of tube changers.
 

Thread Starter

xchcui

Joined May 12, 2014
134
The varistor is in a circuit that control 220VAC 250w 1.2A
(normal operation current)One phase electric motor.
I agree with you that the varistor+a fast acting glass fuse is enough,but i
thought to add a TCO to the varistor in order to prevent from the varistor
to burn and damage other component in the circuit.Even with the right act fasting
glass fuse value,the varistor manages to cause damage(to melt)other components
around before the fast acting fuse blow and no matter how you try to arrange it in a good and safe position.
In regard to inrush current that you mentioned,Do TCO fuses(those with the fusible alloy
wrapped with epoxy sealant)suitable to operate under inrush currents?
I mean if 4A fast acting glass fuse is okay for the motor,will TCO fuse rated
for 4A will also be suitable or TCO fuses have issue with inrush currents of the motor?
 
Last edited:

MisterBill2

Joined Jan 23, 2018
3,861
The varistor is in a circuit that control 220VAC 250w 1.2A
(normal operation current)One phase electric motor.
I agree with you that the varistor+a fast acting glass fuse is enough,but i
thought to add a TCO to the varistor in order to prevent from the varistor
to burn and damage other component in the circuit.Even with the right act fasting
glass fuse value,the varistor manages to cause damage(to melt)other components
around before the fast acting fuse blow and no matter how you try to arrange it in a good and safe position.
In regard to inrush current that you mentioned,Do TCO fuses(those with the fusible alloy
wrapped with epoxy sealant)suitable to operate under inrush currents?
I mean if 4A fast acting glass fuse is okay for the motor,will TCO fuse rated
for 4A will also be suitable or TCO fuses have issue with inrush currents of the motor?
A thermal cutoff device is triggered by a heat rise external to the device. The intention is for them to not be current sensitive, that is the job for regular fuses. I have seen varistors encased in a silicone tube so that they will not damage other parts. so for it to work the TCO device must be sensing the temperature rise of the varistor. That will take several seconds at least for the heat transfer. That is why a TCO is not a good choice here.
 

Thread Starter

xchcui

Joined May 12, 2014
134
A thermal cutoff device is triggered by a heat rise external to the device. The intention is for them to not be current sensitive, that is the job for regular fuses. I have seen varistors encased in a silicone tube so that they will not damage other parts. so for it to work the TCO device must be sensing the temperature rise of the varistor. That will take several seconds at least for the heat transfer. That is why a TCO is not a good choice here.
Yes,of course.My intention,as i mentioned before,was to tie(the body of the TCO to the body of the varistor)while the leads of the TCO will be in series with the line.
But i was convinced,by your replies,that i better use regular fuse.
Refer to the silicone tube that you have mentioned,can you attach a link to
that kind of encasing?since i search for photos in the web and i saw only
large encasing boxes that are too large for my application.
 

MisterBill2

Joined Jan 23, 2018
3,861
Yes,of course.My intention,as i mentioned before,was to tie(the body of the TCO to the body of the varistor)while the leads of the TCO will be in series with the line.
But i was convinced,by your replies,that i better use regular fuse.
Refer to the silicone tube that you have mentioned,can you attach a link to
that kind of encasing?since i search for photos in the web and i saw only
large encasing boxes that are too large for my application.
There are quite a few vendors of suitable tubing materials, including the wire manufacturers. And I believe that most of the electronics suppliers, such as Newark and DigiKey, sell ready cut sections of tubing for exactly this application. The problem usually is that it is sold in lots of 1000 pieces. But there may be exceptions now, my experience with such tubing dates back a few years.
 

Thread Starter

xchcui

Joined May 12, 2014
134
Are you refering to the heat shrinkable tube?
Will it handle the high temperature of the varistor?
I don't know,but it looks like that the varistor reaches to an higher temperature
than the tube can handle,while the tube,also,prevents from the varistor to conduct the heat
out from its body.
It is just my logic thought.Is it really efficient?
 

MisterBill2

Joined Jan 23, 2018
3,861
Are you refering to the heat shrinkable tube?
Will it handle the high temperature of the varistor?
I don't know,but it looks like that the varistor reaches to an higher temperature
than the tube can handle,while the tube,also,prevents from the varistor to conduct the heat
out from its body.
It is just my logic thought.Is it really efficient?
There is heat-shrink and then there is fire-proof, they look similar but the fireproof is often a bit thicker. AND the supplier makes a big point that it is fireproof safety tube. They are not the same at all.
 

Thread Starter

xchcui

Joined May 12, 2014
134
There is heat-shrink and then there is fire-proof, they look similar but the fireproof is often a bit thicker. AND the supplier makes a big point that it is fireproof safety tube. They are not the same at all.
Okay,the issue is much more clear now.
Thank you very much for your help,you save me alot of time
that i was going to waste on the wrong method.
The fireproof tube seems to be the right solution.
Only for knowledge(since i didn't find the answer on google),what is the temperature
(more or less)that the MOV start to catch in fire?
 

MisterBill2

Joined Jan 23, 2018
3,861
It must be less than the temperature that melts the attaching solder, since on those that I have seen failed the solder was still OK. Of course the heat would need to be conducted a distance down the leads to the solder joint. So there is an area for caution, which is that in the circuit board assembly process, have the MOV spaced away from the circuit board by a minimum of 2mm, with 3mm being a better spacing yet. That way the failure will not damage the circuit board and the assembly should be repairable. (I am presuming that it is intended to be repairable.)
 

Thread Starter

xchcui

Joined May 12, 2014
134
It must be less than the temperature that melts the attaching solder, since on those that I have seen failed the solder was still OK. Of course the heat would need to be conducted a distance down the leads to the solder joint. So there is an area for caution, which is that in the circuit board assembly process, have the MOV spaced away from the circuit board by a minimum of 2mm, with 3mm being a better spacing yet. That way the failure will not damage the circuit board and the assembly should be repairable. (I am presuming that it is intended to be repairable.)
I didn't understand,you say that the MOV can catch in fire at less than 188°C,which is the melting temperature of the solder(60/40)?
 
Last edited:

MisterBill2

Joined Jan 23, 2018
3,861
I didn't understand,you say that the MOV can catch in fire at less than 188°C,which is the melting temperature of the solder(60/40)?
No, at least that is not what I intended to imply. The connecting leads do not get hot enough to melt solder, I have no idea how hot the part that burns up gets. But the tubing used for that purpose does not catch fire, so it must be adequate for the temperature. And as long as the protective fuse opens and cuts off the power, the length of time that the varistor is getting hot will be limited, and so the heat will not spread a lot. That is why a fuse is used. It limits the total amount of heat energy when the fails.
 
Top