IGBT with a Fourth Leg

Many might call it a "four-legged IGBT".

So is there such a device? What would you think from what I describe?

See #5 below.

You cannot call it; it will not come to you no matter how many legs it has.

The typical casing has 3 "legs" out the " bottom " and a mounting tab. the tab is usually directly connected to the centre " leg ".

For clarity of future answers please specify the IGBT part number from the spec sheets you have.

Agree, a "Kelvin emitter" corresponds to the "auxiliary cathode" in high power thyristors.

A Kelvin bridge uses 4 terminals to measure a resistance: two "force" terminals to apply a current and 2 "sense" tyerminals to measure voltage. The resistance is measured excluding the contact errors you would have when measuring with 2 terminals only.

In a similar way, the drive circuit does not "see" the voltage drop between emitter or cathode to heatsink.

More information can be found in the devices datasheet and application notes, which generally explain what all the legs do.

No it is usually a dedicated terminal for the return path for the gate. It will generally be of a much smaller conductor size than the emitter but they are electrically connected.

Pl look up link http://www.irf.com/product-info/datasheets/data/auir3317.pdf.

These were originally developed with MOSFETs and are called current sense leads. Imagine a MOSFET rated for say 10 or 20 or 40 amps. tO measure MOSFET or IGBT load current one will to use mili or even micro ohm resistors and it can present many practical problems. Hence MOSFET / IGBT is internally designed with 2 leads for the source or emitter side. The area for current carrying is for these two paths is divided in the ratio of 1:100 say. Then by ohms law current in the two paths is also shared in the ratio of 1:100. Now it is easy to sense current which is say 400 mA in 4th lead and 40 A in main path going to the load. Hence these are called current sense devices. Such MOSFETs have been around for more than a decade.

This scheme has been used for power thyristors for more than 30 years.

With the "auxiliary cathode" terminal the drive circuit will not see the voltage drop from cathode to heat sink. Power SCRs need a short, steep and high-current trigger current to switch on reliably.

Again there is some confusion. My statement is to sense emitter current without using milio hm or micro ohm resistors and NOTHING to do with gate drive circuit feedback mechanism.

What you are referring to is an additional cathode lead in SCR - which can be used along with gate drive circuit - where cathode to heat sink voltage feedback does not affect gate drive waveform.

What I am stating is ENTIRELY different from what you are stating. Anyway - kindly let me have a link to the thyristor you are referring to- I am interested to see a data sheet.

Sorry xyz,

my confusion! As you mention, " The area for current carrying for these two paths is divided in the ratio of 1:100 say." Power MOSFETs and IGBTs are made up of many small cells on one chip, so it is possible to have a separate contact for a smaller group of cells (not so for SCRs) with a high precision ratio.

I would rather call this a "current sampling" scheme, not a Kelvin one.

This is an example of SCR modules with an insulated base (no electrical contact to heat sink) but in configuration TT, page 3, terminal 2 is main (current carrying) cathode, 4 is the aux. cathode.

http://www.infineon.com/dgdl/tt500n.pdf?folderId=db3a304412b407950112b42f6f3e4b72&fileId=db3a304412b407950112b42f95574c4a

Here you can see the case for hockey puck packages:

https://www.google.com.br/search?q=hockey+puck+scr&espv=210&es_sm=122&tbm=isch&imgil=xpWi4bGxCJvcXM%253A%253Bhttps%253A%252F%252Fencrypted-tbn2.gstatic.com%

Red cable for the gate and white for aux. cathode.

After my post graduation in 1975 I joined a reputed company in India which obtained technology for manufacture of power semiconductors from Siemens, Germany. Basically these are Power SCRs and diodes. I was incharge of testing, QC , maintenance and process/ test equipment fabrication. I was trained at Siemens in Frankfurter ring (then) in Munich.

The auxiliary cathode connection you are referring to is just electrically shorted additional wire to hockey puck packages. Basically gate drive is between cathode and gate terminals. Any thyristor screw type or disc type carrying say 20 amps upwards have solid (meaning not easily flexible) cathode terminals. As any application engineer finds it difficult to attach gate drive return path - he is provided a soft, flexible wire attached to the cathode. This has nothing to do with any modification on the silicon inside. I am aware of internal, physical construction f such high power SCR and diodes. Hence I am making statements about MOSFET & may IGBT and SCR with hands on experience. For fourth wire on MOSFETs search in Google for current sense FETs, for short it may be called sensefets

sure, the HK has nothing to do with the silicon inside. I must be gettig older... and I haven't dealt with IGBT chips for some 15 years now.

Never been to BHEL, but I also had my time at Frankfurter Ring and Pretzfeld...

mfG

Snel