difference of MCB TYPE C & MCB TYPE B

[sahu]

what's the difference of MCB TYPE C & MCB TYPE B......

 

[betwixt]

Only the reaction time. 'C' is normally used in industrial applications where short time overloads are preferable to the trip opening.

In fuse terms, 'B' = quick blow and 'C' = anti-surge.

Brian.

 

[sahu]

I want to desing a electrnic MCB . But cont found any stander PROTOCAL for over load and shart circuit.

 

[betwixt]

Most MCBs have only two wires and are in series with the source and load. If you want to make an 'electronic' one you should consider how it is to be powered and how it should react if it's own power source is lost in the event of a fault. I believe electronic over-current trips use independent power sources and sense the load using a current transformer. The voltage on the secondary side of the transfomer is measured and when it reaches the maximum allowed, a relay or contactor is released to disconnect the load. You might be able to measure the current with a series resistor but beware that the voltage at each end of it will be high and constantly reversing polarity so sensing it could be very difficult.

Brian.

 

[sahu]

yes , sensing the load using a current transformer. The voltage on the secondary side of the transformer is rectified & measured and when it reaches the maximum allowed, a relay is released to disconnect the load.

till hear OK....
i cont develop any protocol for 2nd stage for over load & short circuit (as mcb TYPE B & TYPE C)...
pl guide me .....

 

[betwixt]

I'm not sure what you mean by 'protocol'. There is nothng special about MCBs, if you open one, all you will see inside is a mechanical switch and a latch mechanism. The latch is released by a bi-metal spring which bends as the current increases and it warms it up. The only difference between 'B', 'C' and 'D' types is the mass of the spring, the lighter 'B' type has a faster reaction time. Some MCBs also have a magnetic release on the latch so they can operate faster in the event of a massive overload.

If you are trying to make an electronic version, as I mentioned before, it must have it's own isolated and reliable supply and monitor the current through the transformer. To make it an 'B', 'C', or 'D' type all you have to do is decide how responsive your current monitor is to be. Beware that in the event of a short circuit or very heavy overload, the supply may be temporarily unavailable and your electronic circuit must still be able to operate. Whatever you do, make sure it is fail-safe, in other words make sure it stays tripped and it doesn't reset the MCB when power is first applied as this would make it extremely dangerous.

There is some more information on MCB ratings at this web site:
Untitled

Brian.

 

[sahu]

dear friend Mr.betwix very -2 thank u for guidance...
i have design it . it working good for liner load (heater , etc).
protocol mens ..i set it ..form CT o\p
100 % = 0.65 volt
130 % = 0.65*1.3 = 0.845 volt , wait 2 min delay for over load
300 % = 0.65*3.0 = 1.950 volt , wait 12 sec delay for over load
400 % = 0.65*4.0 = 2.600 volt , wait 0 sec delay for short circuit

but it not working properly mix load (all house load )...

 

[betwixt]

Can you show your schematic?
It is possible you need to filter the voltage from the CT to keep harmonics out of the measurement. Many household items do not draw sinusoidal currents so the CT secondary voltage may not be a true indication of real load. In a normal MCB it is heating that operates the switch so they are very tolerant of waveform.

Brian.

 

[sahu]

i think ,my schematic is OK.

 

[betwixt]

I don't think so!

Three problems:

1. the time constant of the filter is far too long. For a fast response you need to remove the capacitors after the bridge rectifier. I would suggest you even remove the bridge completely and use head to tail diodes, a normal diode in one direction and a 4.7V Zener in the other direction. Ideally you would use a zero crossing detector to tell the PIC to start a quarter cycle delay then read the peak of the AC directly into the ADC. You can set the trip response time by counting how many cycles the high readings last for.

2. You don't show where the power for the PIC or the relay comes from. Be very careful with the curent transformer as both ends of it are above VSS potential so it must not be grounded.

3. The 1N4007 is connected in the wrong place. In series with the transistor base all it does is limit the base current which is not a good idea. It would be better to connect it across the relay coil with the cathode (+) end wired to the relay supply. With out a diode across the relay coil you run the risk of the transistor being damaged as it turns off.

Brian.

 

[sahu]

I don't think so!

Three problems:

1. the time constant of the filter is far too long. For a fast response you need to remove the capacitors after the bridge rectifier. I would suggest you even remove the bridge completely and use head to tail diodes, a normal diode in one direction and a 4.7V Zener in the other direction. Ideally you would use a zero crossing detector to tell the PIC to start a quarter cycle delay then read the peak of the AC directly into the ADC. You can set the trip response time by counting how many cycles the high readings last for.

2. You don't show where the power for the PIC or the relay comes from. Be very careful with the curent transformer as both ends of it are above VSS potential so it must not be grounded.

3. The 1N4007 is connected in the wrong place. In series with the transistor base all it does is limit the base current which is not a good idea. It would be better to connect it across the relay coil with the cathode (+) end wired to the relay supply. With out a diode across the relay coil you run the risk of the transistor being damaged as it turns off.

Brian.

like this.

 

[rajudp]

I don't think so!
3. The 1N4007 is connected in the wrong place. In series with the transistor base all it does is limit the base current which is not a good idea. It would be better to connect it across the relay coil with the cathode (+) end wired to the relay supply. With out a diode across the relay coil you run the risk of the transistor being damaged as it turns off.
Brian.

no problem with 1N4007, the only thing you have to make sure is the micro controller will be getting power (volt) even if any short or heavy load, else the controller resets and wont work as you planned

---------- Post added at 19:50 ---------- Previous post was at 19:48 ----------

like this.

you have to remove the 100mf/25v cap for aster response

 

[betwixt]

That's better but still not right. You do not need D1 at all, just remove it and connect R3 to pin 10.

The reaction time is still being limited by the time it takes to charge C1 and C2. Ideally, you want the electrical reaction time to be as fast as possible and use software to create the delay.
I'm assuming you are trying to measure the current by reading the voltage at PIC pin 7. To make it react quickly what you need to do is sample the rectified AC signal at it's peak in the cycle. To achieve this you need to find where the peak is. There are two ways to to that, either you keep sampling and comparing the readings and use the one just before the voltage started to drop, or, the easier way is to use a timer. You know what frequency the AC is, so use a zero crossing detector to find a reference point then wait for one quarter cycle before taking the measurement. The peak will always be one quarter cycle after zero so timing it is simple. For example, if your AC frequency is 50Hz, one complete cycle takes (1/f) 20mS so the peak will be at 5mS from zero. Be careful to use the positive peak and not the negative one!

Brian.