(IGBT WELDER) looking at this transistors, can it work like this, and the output current

[hadewaleadelabu9]

 

[rjenkinsgb]

No, for multiple reasons.

The upper and lower FETs must have totally separate gate drive signals, as the gates can only stand somewhere around 20V maximum and you have a DC supply of around 320V.

There are specially designed "high side" driver ICs to control the upper FETs in such H-Bridge circuits; remember that the upper gate drive is relative to the source of each device - which is changing in voltage over a wide range as the various devices switch on and off.

The output from an H-Bridge is taken between the two centre junctions, where you have the heavy link. If either diagonal pair of transistors switched on, they would short the supply.

If you are trying to trace the circuit of a machine, either the transformer is connected between the centre junctions, or both upper devices are driven simultaneously and both lower devices are driven simultaneously.
eg. A half-bridge with parallelled power devices, rather than the crossed gate connections.

 

[hadewaleadelabu9]

No, for multiple reasons.

The upper and lower FETs must have totally separate gate drive signals, as the gates can only stand somewhere around 20V maximum and you have a DC supply of around 320V.

There are specially designed "high side" driver ICs to control the upper FETs in such H-Bridge circuits; remember that the upper gate drive is relative to the source of each device - which is changing in voltage over a wide range as the various devices switch on and off.

The output from an H-Bridge is taken between the two centre junctions, where you have the heavy link. If either diagonal pair of transistors switched on, they would short the supply.

If you are trying to trace the circuit of a machine, either the transformer is connected between the centre junctions, or both upper devices are driven simultaneously and both lower devices are driven simultaneously.
eg. A half-bridge with parallelled power devices, rather than the crossed gate connections.

The device came with two 50t65 which is 100a + 100a =200a, while I couldn't find it, that is why I wanted to use 60N60 which is 60a * 4 =240a.
With the look of things which one is preferable among the diagram

 

[rjenkinsgb]

The 50T65 is rated 50A RMS, 650V maximum.
The 60N60 has a higher current rating at 60A, with a 600V maximum.

They are both rated for 15V gate-emitter, so the 60N60 could be a 1:1 replacement with better ratings, as long as the switching frequency is not too high; it's a slightly slower device than the 50T65.

Paralleling more devices than in the original setup is not a good idea as they have very high gate capacitance and doubling that with two in tandem would further slow the switching and increase device heating and power losses.

 

[ronsimpson]

hadewaleadelabu9 This is a big project.

I wanted to use 60N60 which is 60a * 4 =240a.

The currents do not add up like that. Two devices in parallel will do *2 not *4. Only two devices are on at a time.

 

[hadewaleadelabu9]

hadewaleadelabu9 This is a big project.

The currents do not add up like that. Two devices in parallel will do *2 not *4. Only two devices are on at a time.

But I needed 200a, though the 50t65 maximum power dissipation is 100A, since I couldn't find it I intended to use 60N60 which 60a,
That was why I wanted to pair it

 

[rjenkinsgb]

The 50T65 is rated at 50A continuous.

"Maximum" or "peak" current are for short duration pulses, often how much they can take for eg. 10mS.
They are the threshold-of-destruction figures, not operational values.

See the headline data in the datasheet here: 650V 50a.

FGH50T65UPD datasheet, igbt equivalent, Fairchild Semiconductor

FGH50T65UPD datasheet, FGH50T65UPD igbt equivalent, Fairchild Semiconductor, Features and benefits, Stock and price

datasheet4u.com

It's also part the device name; 50A & 65(0) volts, as with the 60N60: 60A & 60(0) volts.

 

[hadewaleadelabu9]

The 50T65 is rated at 50A continuous.

"Maximum" or "peak" current are for short duration pulses, often how much they can take for eg. 10mS.
They are the threshold-of-destruction figures, not operational values.

See the headline data in the datasheet here: 650V 50a.

FGH50T65UPD datasheet, igbt equivalent, Fairchild Semiconductor

FGH50T65UPD datasheet, FGH50T65UPD igbt equivalent, Fairchild Semiconductor, Features and benefits, Stock and price

datasheet4u.com

It's also part the device name; 50A & 65(0) volts, as with the 60N60: 60A & 60(0) volts.

Check this out

fgh60n60ufd.pdf PDF datasheet. ALL TRANSISTORS DATASHEET. POWER MOSFET, IGBT, IC, TRIACS DATABASE. Electronic Supply. INNOVATION CATALOG. Service

alltransistors.com

MBQ50T65FDSC IGBT. Datasheet pdf - Equivalent

alltransistors.com

the two of 50N60 won't be able to handle 200a load, unlike two of 50t65

 

[rjenkinsgb]

Wrong, sorry.
Look at the actual tables within those two datasheets.

IC values.
50T65 = 100A & 50A, depending on temperature.

60N60 = 120A & 60A for the same temperatures.

 

[hadewaleadelabu9]

Wrong, sorry.
Look at the actual tables within those two datasheets.

IC values.
50T65 = 100A & 50A, depending on temperature.

60N60 = 120A & 60A for the same temperatures.

Yes, you are right
but physically the (50T65) pin and tube is ticker and bigger than the said (60n60)

 

[rjenkinsgb]

They are both TO-247 packages and have the same pinout, so the 60N60 is a direct drop-in replacement.

If you wanted an even higher rated one, you could possible use these; again physically interchangeable, but somewhat higher gate capacitance which may slow the switching.

https://docs.rs-online.com/399b/0900766b812671e9.pdf

 

[hadewaleadelabu9]

They are both TO-247 packages and have the same pinout, so the 60N60 is a direct drop-in replacement.

If you wanted an even higher rated one, you could possible use these; again physically interchangeable, but somewhat higher gate capacitance which may slow the switching.

https://docs.rs-online.com/399b/0900766b812671e9.pdf

Ok thanks

 

[gary350]

WHY do you need 200 amps?

 

[hadewaleadelabu9]

Ok thanks

just changed the transistors now, but unfortunately blown off
What could be the problem?

 

[hadewaleadelabu9]

They are both TO-247 packages and have the same pinout, so the 60N60 is a direct drop-in replacement.

If you wanted an even higher rated one, you could possible use these; again physically interchangeable, but somewhat higher gate capacitance which may slow the switching.

https://docs.rs-online.com/399b/0900766b812671e9.pdf

just changed the transistors now with 60n60 1:1, but unfortunately it blown off
What could be the problem?

 

[hadewaleadelabu9]

hadewaleadelabu9 This is a big project.

The currents do not add up like that. Two devices in parallel will do *2 not *4. Only two devices are on at a time.

just changed the transistors now with 60n60 1:1, but unfortunately it blown off
What could be the problem?

 

[rjenkinsgb]

Was it previously working OK?
Something caused the previous power devices to fail - have you located the cause of that?

 

[ChrisP58]

just changed the transistors now with 60n60 1:1, but unfortunately it blown off
What could be the problem?

Please post a complete, clearly drawn schematic for what you're doing, with details of all components in the system.

 

[hadewaleadelabu9]

Please post a complete, clearly drawn schematic for what you're doing, with details of all components in the system.

ok

 

[hadewaleadelabu9]

Was it previously working OK?
Something caused the previous power devices to fail - have you located the cause of that?

No, the previous 50t65 got burnt, 60n60 got burnt also.
and nothing wrong with the rectifier diode

Was it previously working OK?
Something caused the previous power devices to fail - have you located the cause of that?

I have checked the rectifier diode, is still good
but I don't have oscilloscope to check the driver signals