U might laugh at this
Spectrum 187 Plasma Cutter
Collapse
X
-
You can use the same tests for MOSFETs and IGBTs. On the diagrams, below shows the equivalent leads MOSFET. IGBT
G. G
S. E
D C
I wouldn't worry too much about the coefficient of the Miller Effect. There is a certain amount of energy (gate charge) and voltage you need to provide to the IGBT/MOSFET gate to make sure it fully conducts (acts more like a closed switch and less like a resistor) so it doesn't overheat.
The reason why you need to check the gate drive circuit when replacing the IGBT is you'll blow the IGBT if there's more than 20V powered to the gate, or the gate drive has shorted to + 15V or so, and the transistor conducts all the time, which saturates whatever inductor and the current rises until the IGBT overheats.
Comment
-
-
-
Ok most everything seems c85 v arnt tho down on half bridges sr1pm1 getting 240 got the juice there could you tell me the 28 volt 12 for pilot arc stepdown transformers are??just lik5a coming out of rc11everything seems Lewis that a damaged igbt ? The over heat on main transformer. Could this a cause. How about boost inductors ummmmmm
Comment
-
-
Hey yall waiting on parts as my moduals where like from the dinosaur age got ahold of vishey dude he says run vsgt140da60up if overcharge hit again would ent this being a larger amp #do damage on putin kinndaa scale both my boys 101 airborne hope they both don't go oh well the igb they the gate collector drain an source what with the kelvin????Attached Files
Comment
-
-
Hoping your boys stay safe, great unit!
The newer IGBTs are built better (lower gate capacitances), so it may not be a big impact to the gate driver, and as long as it has the same packaging configuration, I bet you are good.
I see you pulled OC4. If you find the datasheet for OC4, track down the input to that optocoupler and measure resistance across the Input (check resistance across outputs too). Also check from the anode side of the input coupler to the + voltage side of whatever is signalling it - you want a high resistance there. You just want to be sure none of the diodes, capacitors, transistors, or microcontrollers driving it are shorted or otherwise bad. Once you confirm it's good all the way back to the power supply (there are two power supplies since you're using an optocoupler) you can be confident it won't pop when you turn on the machine after installing.
Comment
-
-
It might work, though it will switch and conduct differently.
Being 1200V instead of 600V of IGBT, It likely has a higher on voltage (Vsat in datasheets), which means it will make more heat when powered on. If it's the same size package, it's probably built to handle the extra heat generated in conduction (really just for the short time the pilot is on). You can compare Rth(j-s) or whatever thermal resistance it has in the datasheets to see the temp drop. I think those pilot circuits are at most 20A average, so you can get an idea from Vsat how much heat is generated (from the chart I'm the datasheet, go to 20A, assuming 15V for Vge, take Vsat and multiply by 20 to get power dissipated). Divide that by Rth(j-s) or whatever is listed to get an idea on temperature drop across the chip. Hopefully it's about the same for both the 600V IGBT and 1200V IGBT. Your heatsink might be a few degrees higher for the 1200V IGBT, what's the difference in Vsat?
Compare the Qc from the datasheets to see how it will affect the gate driver (OC4? - please confirm based on tracing the board leads) and the Cies, Coes to see how it would impact the snubbers on the board. If Qc is much larger, the IGBT will draw more power from OC4. Whether or not it's a problem depends on the switching frequency. If Cies and Coes are much larger, you will have higher switching losses - probably not too big of a deal for the IGBT, but it might raise your heatsink temp a bit, which could make it an issue. What size (ohms) resistor is between OC4 and the IGBT?Last edited by jjohn76; 02-26-2022, 07:55 AM.
Comment
-
-
Comment