A pulse rating is a temperature

A pulse rating is a sentence about heat with the heat left out: a stated current, for a stated time, carrying the junction from one temperature to another. How hot you are willing to arrive is a decision rather than a property of the part, and so is the half nobody writes down, which two nodes the rise you allow is measured between. Two published figures that read like rival budgets turn out to span different pairs of nodes, which is why this page puts no ratio between them.

What it is and how it works

A semiconductor die is a small thing that gets hot quickly. While it conducts it dissipates, and that heat leaves by a stack of materials: the die itself, the solder that attaches it, a copper baseplate, a layer of grease, a heatsink, the air. Every step in the stack resists the flow of heat, and every lump of material in it stores heat.

Max, in a thread on IGBT temperature, gives the analogy that makes the rest of this readable: temperature is voltage, thermal mass is capacitance, thermal resistance is resistance, and the dissipation in the die is a current source charging the first capacitance it finds. Each larger mass hangs off the one before it through a resistance, so the die's temperature is a node voltage part way along a chain, and in his words "the whole system is a low pass filter".

That is not decoration, because it means how long you push decides how much of the chain is in the circuit at all. A pulse shorter than the first time constant charges the die's own heat capacity and gets no further; the heatsink is not part of that answer, and neither is the grease. Push for long enough and every capacitance is charged, the chain settles, and the far end of it, the air, is the only thing that matters. This is why a datasheet carries two different things about heat: one static thermal resistance in kelvin per watt, and a curve of transient thermal impedance against pulse width.

So there are two questions, and they are not the same question. Does the average power get out? That one is answered by the heatsink and the static resistance. Does the peak stay under? That one is answered by the die's own heat capacity and the transient curve. A design can pass either and fail the other, and the instrument you would reach for tells you about the first one only.

And this is where a pulse rating comes from. Read one backwards. Somebody picked a starting condition, picked a temperature they were willing to end at, worked out the power the device dissipates at some current, and read the time off the transient curve. The rating is the answer to that arithmetic. Move either temperature and the answer moves with it, and nothing about the part has changed.

What it does in a coil, and what you decide

A coil is the extreme case of that split, because almost nothing is on. A bang lasts a few hundred microseconds and there are a couple of hundred of them a second, so the bridge conducts for something like six per cent of the time [derived: 300 µs at 200 bangs a second]. The ontime is this site's own typical, "typically no more than about 300 µs a note" on the interrupter is the note; the rate is only an illustration here, and that page says as much, "There is no best rate. It is a note.", while its own worked duty is 5 per cent rather than six. The average power is small, the heatsink is comfortable, and the die still goes somewhere inside those few hundred microseconds that the heatsink never hears about.

Jimmy Hynes put the whole design problem in two sentences on the Tesla list while working out his second DRSSTC (pupman): the problem was allowing the die to absorb more energy during the short bursts, and getting the heat out of the heatsink was not a problem at all. He was proposing to put a piece of copper on top of the die for it, and said in the same breath that it did not have to move any net heat out, it only had to add thermal capacitance. That is the sentence that separates a coil's thermal design from a converter's. You are not sizing a cooler. You are sizing a bucket.

Which two nodes your rise spans

Before you argue about how large a rise you will allow, say what it is a rise of. Two published figures show why that comes first.

Mads Barnkob sets one as a life target on his DRSSTC design guide: "We want T junction to stay below 80ºC and have T case cooled to stay below 50ºC. A advisable rule is to keep the temperature ripple below 30ºC, we often use second hand IGBT bricks and we do not know their usage history." The next sentence names what he is protecting: "To avoid real fast degeneration of die and bonding wires inside the IGBT, we will aim to reduce large temperature swings." His ripple runs from junction to case, inside the package, and his reason is life rather than survival.

davekni is doing something else in the same territory. His sentence is that for most DRSSTC designs the transient rise is larger than the heatsink's rise, and the figures inside it are an illustration of that claim: "If pulse rise is say 100C, and heatsink is at 50C, then peak junction temperature is 150C." That rise runs from junction to heatsink, which is the package and the mounting joint as well as the die, and the word "say" is his.

So the two are not two answers to one question, and this page puts no ratio on them. What they do share is the part worth taking away: the rise is chosen. Barnkob says so himself, in a reply to a commenter on that page: "it all comes down to what temperature rise you find acceptable for your application." And he prices the choice on the same page: driven much harder than his own calculation allows, a CM600 brick's expected lifetime "might be in the hundreds of hours instead of 10s of thousands". The mechanism that eats those hours, the swing rather than the peak, is the bridge's page and not this one.

Whether you trust the rating's own arithmetic

Two builders worked a pulse rating in public on the Tesla list on 1 October 2004, on two different bricks, and a third answered with a model instead of a number.

Steve Conner took the datasheet line at face value and pushed it as far as it would go (pupman). For a Semikron SKM300GB123D the maximum pulse rating is 600 A for 1 ms, and the datasheet says the pulse width is limited by junction temperature. From that he proposed the temperatures behind the rating, and labelled the proposal as one: "That suggests that 600A for 1ms will raise the junction temperature from 25'C to 125'C", with the rating falling to 440 A at an 80 °C case being what "reinforces this hypothesis". Then he treated average current times time as the invariant, so "1800A average for 333us is the same as 600A for 1ms", and carried that through a full bridge to an energy per bang.

He flagged his own assumptions in the same message, which is why it is worth reading rather than quoting: the arithmetic assumes the forward voltage does not change with current, and in practice "the temperature may be higher with shorter pulses as the heat has less time to escape to the baseplate."

Jim Lux answered with the model rather than the rating (pupman): does the manufacturer publish a two or three node model, die mass, theta jc, case mass, theta cs, because with those masses and resistances "it's just like a cascaded RC network, and you can calculate the time constants." That is the chain at the top of this page, offered as a replacement for the front page. He produced no number of his own.

Jimmy Hynes had already taken that route, on a different part (pupman): "I didn't use that number for my calculations, I looked at the transient thermal impedance charts on the datasheet. Those make it look even better than your method." His device was a Mitsubishi CM300DY-24H and Conner's was the Semikron, so the charts coming out kinder is one part read against another part's rule, not one datasheet read two ways. Nobody closed the argument and nobody in that exchange could have. What closes it is the transient impedance curve for the part in your own bridge, at the width you actually run, which is what the top of this page said the datasheet carries.

Our own reading of why a rule like Conner's is a rule and not a law, and it follows from a sentence already on IGBT or MOSFET: current times time is the adiabatic answer for a device whose loss is V_ce·I, because then the energy is proportional to I·t. A MOSFET's loss is I²·R, so its adiabatic invariant is I²·t instead. Halve the pulse and an IGBT takes twice the current for the same rise, a MOSFET only √2 times [derived from those two loss laws]. Neither rule survives past the first time constant in the chain, where the die starts genuinely sinking into the copper, which is Conner's own caveat restated.

What building the model actually costs

Hynes built his in SPICE and said what he had to find (pupman): all the values for the IGBT came from the datasheet except the thermal capacitance of the baseplate, which he calculated, and the heatsink's, which he got by weighing it. The output stage of the simulation was a gain block converting joules to degrees, which is the thesis of this page stated as a schematic.

He also planned to instrument it live, and the plan is the useful part, though that message is in the future tense throughout and nothing in it was built yet. A thermistor on the die was rejected on paper, its thermal time constant too long and the differentiator needed to compensate amplifying the noise too much. He proposed an infrared photodiode instead, which measures emitted infrared with "only 1uS of delay", so a bang could be cut short from inside itself, and he noted the mismatch honestly: the wavelength of those diodes is "near 900nm, which is a little short for the temperature I'm looking for (~100C or so)".

Whether the measurement you took means anything

davekni splits it in the same thread, and the split is the one this page opened with. For average power, the top of a TO-247 case centred on the die is a fair approximation, and he gives one reading of his own for it. For transient thermals during an enable pulse, "most people rely on calculations", and the two measurement options he offers are not equivalent. The first characterises a temperature dependent parameter of your own part and uses it as a transfer function, and that one alone carries the warning that has a box to itself further down this page: what it returns is the die's average temperature, while the specification is written for its hottest point. The second is half measurement and half calculation, and it ends back at the datasheet, scaling the measured average power by one over the duty cycle and reading the rise off the transient thermal resistance graph.

He also puts a condition on the calculated route that is easy to skip, and it is the grease at the top of this page: the estimate presumes the case to heatsink resistance is low. If that joint is poor, the rise you calculated is landing on a baseline you have not measured.

The numbers

• The rise is the free variable, and it is not worth the same to everybody. The thermal impedance you are allowed to spend is that rise divided by the power, so it moves with the rise. The current it buys does not: at V_ce·I the current moves nearly with the rise, and at I²·R it moves as the square root [derived from the two loss laws above]. Whatever you settle on is worth more to an IGBT builder than to a MOSFET builder.
• What the site's own calculator does with the field. Switch thermals defaults to a full bridge of eight devices, two per position, 160 A peak, a hot R_DS(on) of 0.19 Ω, a 2 ms burst at 14 per cent duty. That is 40 A RMS in each device, 304 W during the burst and 42.6 W averaged [derived from those inputs: 160/2/2, then I²R, then times 0.14]. At the 100 K the ΔT field arrives set to, the transient budget is 0.329 K/W and the page calls the requirement moderate. Put 30 K in, change nothing else, and it is 0.099 K/W and the same page tells you to go and read the datasheet curve because the static R_th(j-c) will not do [both derived from the page's own arithmetic]. Same bridge, same current, verdict changed by one field. Two things that machine is not, before you carry the demonstration anywhere: its loss model is I²R, so the device it is costing is a MOSFET, while both of the rises argued over above came off IGBT threads; and a 2 ms burst at 14 per cent duty is a ramped coil, not the few hundred microseconds this section opened with.
• What pulsed operation is worth, on Barnkob's worked example. He reads 0.0081 K/W at a relative duty of 0.02 off a transient impedance graph, against the 0.17 K/W a 50 per cent duty gives in his datasheet example, and calls it "a factor 21 lower". That one recomputes [derived: 0.17 over 0.0081 is 21.0], and it is the size of the discount a bang buys over a converter. Two things travel with it. The page says of that graph that "it is not specific to a CM300 or CM600, it is only showing how to read out a value from a duty cycle", so what is being demonstrated is the method and not a figure for a brick. And the second ratio printed beside it, "a factor 11.33" at D of 0.064 and 0.012 K/W, does not recompute against its own inputs [derived: 0.17 over 0.012 is 14.2]. Asked about the worked example below it, Barnkob answers that his spreadsheets used more conservative values and that the article "does not describe it in enough details". Take the method, and do the read on your own part's chart.
• Steve Conner's chain, and where he stopped. 600 A for 1 ms, 440 A at an 80 °C case, a proposed 25 to 125 °C, "1800A average for 333us", and from a 650 V bridge a peak of 5902 A, an RMS of 4173 A and 1149 J per bang, which he labelled an absolute theoretical maximum in capitals. He then declined to use it: the short circuit rating is 2000 A and not guaranteed past a thousand pulses, so "I would keep my peak current down to 1000-1500A and that limits me to 50-70 joules." The derating he states on current is a factor of 3.9 to 5.9 [derived: 5902 over 1500 and over 1000]. His joule figure is not that same bang: his own energy expression at 333 µs and those peaks gives 195 to 292 J [derived: 650·(4/π)·(I_pk·0.707)·333e-6, which at his 5902 A returns 1150 J against the 1149 J he printed], so the shorter bang behind 50 to 70 J is not stated anywhere in the message. Take the current limit from him, not the energy.
• The case is not the die, and one reading is not a calibration. davekni measured "~140C on the top of the case when die temperature was 150C", on a TO-247 of his own, at average power, which is the only case where he says the method works at all. That is 10 °C of offset on one part [derived: 150 minus 140]: the difference of two approximate numbers, growing with dissipation, so what transfers to your bench is the sign and not the size. A case reading is a floor under the die.
• How far past its rated peak a builder will actually run a part. Steve Ward, on the Tesla list (pupman): "If only the bigger IGBTs could withstand such abuse. I only run my CM300s at 2X their peak rating." The abuse he means sits in the message he was answering (pupman), where Terry Fritz reports an IRG4PH50UD, a "24 amp rated TO247AC IGBT with diode", "happily eating 700+ amps" against the "180 Amp peak rating" it gets because "it uses a great big die in a small package". Those two multiples are nothing like each other, and the reason Fritz gives for his is the die rather than the package, which is where the top of this page puts a short bang's heat as well. Note also where Ward's routine sits. If his peak rating is the datasheet's pulsed figure, then twice it is where switch thermals puts davekni's rule of thumb for where devices fail.

What will get you

What goes wrong

• Devices die weeks in, well below Tjmax, with nothing wrong on the scope. Not the peak temperature, the size of the swing, stressing the attachment between die and heat spreader and the wire bonding with it. Weston and Hydron both say so in the same thread, and Hydron's version is the useful one: the swing becomes problematically large long before you reach Tjmax, and that drove a lot of his QCW design. The mechanism is covered with the bridge.
• The heatsink is cold and the bridge fails anyway. You sized the static path. At a few hundred microseconds the heatsink is not in the circuit, and what killed the device was the first rung of the chain.
• A device that was fine on the bench dies on the arc. The pulse rise did not change. What changed is the baseline it sits on, because the arc run has a higher duty and the heatsink came up under it.
• Same current, longer ontime, sudden failures. Transient impedance rises with pulse width, so the same power for longer is a larger rise, and the ontime is the term you moved. Rate reaches the same place by the other road, through the average, and the interrupter page has the duty arithmetic for both.
• You measured 140 °C on the case and called it safe. On the one reading davekni reports, his die was at 150 °C while his case read that. The offset is his part's rather than yours, but the direction is everybody's.
• The budget passes on paper and the device still dies. Check whether the switching loss field was left at zero. With soft switching working it is close to nothing, and when it stops working there is no predicting it, so a conduction-only figure is a floor rather than an answer.
• A paralleled pair with one device much hotter than the other. Not thermals at all: gate drive mismatch, and IGBT or MOSFET has it.

Where next

• Switch thermals does the arithmetic above. Change the ΔT field first and watch the verdict move, because that field is the argument this page is about.
• The bridge and what kills it, for the swing as a failure mechanism rather than as a budget, and read the note above before you take the ratio on it.
• IGBT or MOSFET, for the loss law that decides how a pulse rating scales.
• The interrupter is the note, because rate and ontime are the two knobs that set the average.
• Paralleling bridges, for when the answer is that the pulse does not fit in one die.

---

Every figure above is quoted from a fetched source or marked as derived with its inputs. Nothing here was measured on this bench, and the transient impedance curve for your own part is the one step none of it replaces.