Author Topic: Aluminium Nitride Thermal Pads  (Read 733 times)

Offline Weston

  • High Voltage Technician
  • ***
  • Posts: 123
  • Karma: +11/-0
    • View Profile
Aluminium Nitride Thermal Pads
« on: August 12, 2022, 04:31:17 AM »
I recently discovered you can order aluminium nitride ceramic thermal pads for TO-247 / TO-220 parts off of Alibaba express / Ebay. They are only about $1 each too. Shipping from China takes a while though.

If these thermal pads are anywhere near the thermal conductivity of bulk aluminium nitride they should be the best thermal pads you can buy by a wide margin. This would be especially useful for hobby power electronics stuff where people are trying to push as much power through a part as possible.

Unlike the silicone conformal pads the aluminium nitride pads require thermal grease on both sides, which will increase thermal resistance a bit. But bulk aluminium nitride with the geometry of a TO-247 thermal pad would have a thermal resistance of 0.02C/W. Based on datasheets for thermal greases I can find online it looks like the thermal grease would add ~0.05C/W per side for a total of < 0.2 C/W.

For comparison the best insulating silicone / phase change thermal pads I can find are > 0.8C/W and those are pretty thin and easy to have arc over as well as having a high capacitance to the heatsink. If you have a good heatsink the thermal resistance of those insulating pad is going to be the dominant thermal resistance in the system, so the aluminium nitride thermal pads represent a big improvement!

As these thermal parts are from alibaba express and you can't get them off digikey or from western suppliers I am a bit suspicious (although western companies regularly use aluminium nitride as the internal insulator for the big power semiconductor modules). I had my TO-247 thermal pads arrive in the mail today and as far as I can tell so far them seem to be true aluminium nitride. I am going to see if I can figure out an easy chemical / density check as it is pretty hard to get good measurements of thermal conductivity without a good test setup.

I set up some power BJTs in TO-247 package on a small heatsink and compared the aluminium nitride thermal pads to some alumina thermal pads. The heatsink is too small so everything starts to equalize pretty quickly, but it does seem look like the aluminium nitride thermal pads (lower left) keep the part cooler than the alumina thermal pad (upper right), so I have a rough upper bound on the thermal resistance.


Offline klugesmith

  • High Voltage Senior
  • *****
  • Posts: 517
  • Karma: +16/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #1 on: August 12, 2022, 12:44:20 PM »
Nice pic.  Applause for quick and to-the-point experiment.  Is the FLIR camera personal or work's?

The visible outline of thermal pad around the lower left transistor, darker than heatsink, creates risk of optical illusion when comparing transistor temperatures by eye.   Can you get spot temperatures or line profiles using FLIR software instead of general image processing software? 

As a novice, I'd be wary of reading much into brightness difference between exposed pad and heatsink; couldn't they have widely different emissivity at thermal IR wavelengths?  Are there occasions where experts cover whole scene with opaque spray paint, to present uniform & well charted emissivity at the expense of disturbing the cooling system under test?

First time I'd heard of AlN for thermal conduction applications.  Google shows its conductivity getting within a factor of two of the notorious BeO, with conductivity of aluminum metal standing between them. 

Offline klugesmith

  • High Voltage Senior
  • *****
  • Posts: 517
  • Karma: +16/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #2 on: August 12, 2022, 06:39:27 PM »
Re. testing the specimens to see if they are AlN.

Would x-ray fluorescence spectrometer, as used by used-metal dealers, register Al and N?

For bulk density:
You can compute volume from mechanical measurements. Can you do better by water displacement, noting apparent weight change of suspended pads or tiny dish of water, upon dunking?

Geologists separate sand grains by density using dense liquids; I did it for work in summer of '74. At specific gravities above 2.8 they (the dense liquids  :) ) tend to be more toxic and harder to get. Standard compounds are mixed to get intermediate densities, and of course the lab has a range of floating hydrometers.
https://www.911metallurgist.com/blog/dense-heavy-medium-separation-hms-dms
Are you still at Stanford, and do they have a geology department?

How about measuring the dielectric constant by making two-plate capacitors?

I guess the sky's the limit when it comes to fancy setups to measure thermal resistance of an insulating pad.
How 'bout attaching a temp sensor to the tab of one of your TO-247 devices, and insulating the non-heatsink side with something fluffy like styrofoam or glass wool or vermiculite?

At work we use soft sticky sheets of TIM to go between a water-cooled cold plate and a board with many hundreds of watts of FPGA's and ASICs. Mechanical compliance is needed to tolerate height variation and non-flatness, but there's no separate thermal grease & it's not a heat cured adhesive. Now you got me wanting to find out the thermal resistance numbers.


« Last Edit: August 12, 2022, 06:53:42 PM by klugesmith »

Online alan sailer

  • High Voltage Technician
  • ***
  • Posts: 166
  • Karma: +10/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #3 on: August 12, 2022, 06:52:58 PM »
One little data point. I used to use AlN substrates for microwave purposes. Since it was high frequency, low loss was important and we always ordered high purity material. Visually it is a light grey color so if your pads are some other color...

Incidentally I tried making some thermal pads from the stuff. It cut into the right size using a diamond blade at work, but the cheap diamond drills I tried to use to make the mounting hole didn't do the job so I gave up.

Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1806
  • Karma: +88/-1
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #4 on: August 12, 2022, 07:26:41 PM »
Presuming thermal conductivity is isotropic, how about a test like this:  Clamp heat source(s) along the top, say a couple TO251 package parts or some SMD resistors up-side-down (resistive element against AlN).  Place bottom few mm in ice water.  View thermal gradient from back side.
David Knierim

Offline klugesmith

  • High Voltage Senior
  • *****
  • Posts: 517
  • Karma: +16/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #5 on: August 12, 2022, 08:28:28 PM »
>> Presuming thermal conductivity is isotropic

Reminds me of a strongly non-isotropic material that popped up in searches related to this thread:
thin sheets of pyrolytic graphite, as seen for example in diamagnetic levitation demonstrations.
Somebody held a sheet between fingertips & used a thin edge to melt a kerf into an ice cube.
Some suppliers state x-y plane conductivity above 1500 W/m-K, far higher than any metal, and within a factor of two of diamond.  (But graphite is not an electrical insulator. ) Even farther up the chart, and also generally anisotropic, we find heatpipeium.
« Last Edit: August 12, 2022, 08:48:02 PM by klugesmith »

Offline Weston

  • High Voltage Technician
  • ***
  • Posts: 123
  • Karma: +11/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #6 on: August 14, 2022, 08:33:16 AM »
Thermal camera is my own. Its a flir E4 I modified the firmware on to unlock full resolution. Pretty useful tool  :)

My initial goal for the BJTs was to use the -2mV/C Vbe voltage coefficient to measure the internal temperature. I quickly realized that power BJTs are very non-ideal at higher currents. I would have to modify the test setup to pulse the power and allow measuring the voltages with a low bias current.

The material is obviously very thermally conductive, I can hold one end of the thermal pad and cut through an ice cube. Even if it's with an order of magnitude of pure AlN the thermal resistance will be dominated by the thermal paste. My initial goal is just to prove to myself its not actually BeO.

The thermal pads I have are 22x28x1mm with a 3.4mm hole. I applied copper tape to both sides of a thermal pad and measured a capacitance between the two sides with my LCR meter. I measured a capacitance of 43pF and a dissipation factor of 0.005. The readings will be impacted a bit by the adhesive of the copper tape, mounting hole, and edge effects, but if I treat is as a plate capacitor I calculate a dielectric constant of 8.7. The published dielectric constant is 9.2 and the loss tangent is 0.002, so I think thats close enough.

Importantly, the dielectric constant of BeO is 6.8, so I can rule out the material being BeO from that measurement.

Measuring the thermal conductivity laterally on the thermal pad is a good idea. I instrumented the TO-247 parts with thermocouples but it was still hard to get measurements because of how fast the heatsink was heating up.

I soldered 2 1206 resistors on copper tape on each side of the thermal pad and clamped the other side against a larger heatsink with some thermal paste and epoxied two thermocouples laterally ~10mm apart. I am just using UV cure epoxy that is not optimized for thermal conductivity but I ran the thermocouples across the entire width to try to get them to equilibrium. They are also pretty thin so not much heat will be conducted away through the wires.

I covered the whole setup in tissue paper to reduce any heat loss due to convection and ran 3W through the resistors and waited a minute until temperature equalized. I measured 46.6C on the top thermocouple and 40.2C on the bottom thermocouple. This is a thermal resistance of 2.1C/W. With a 20x1mm cross section and 10mm distance this equates to a thermal conductivity of 234 W/(m·K). The published value is 321 W/(m·K), so this is close enough where I am not going to worry about it! Might be measurement error, might be variation in the material properties. These ceramics are probably made from a pressed powder so it should not be anisotropic. Not seeing anything in literature about the material being anisotropic except for very thin films.


Happy to say that this material seems legit! Not to promote some random vendor, but here is the listing I bought it from: https://www.aliexpress.com/item/3256801164656742.html . I am also waiting for some TO-220 thermal pads I ordered off ebay. I would expect all the vendors to be selling pretty similar material.


As the limiting factor with AlN thermal pads is the thermal compound, does anyone have suggestions? Is pump out / dry out something you need to worry about anymore, or has it been fixed with the more modern synthetic compounds? CPUs still use thermal compound, so it cant be that bad an issue. I would think that the thermal compounds sold to computer enthusiasts would be pretty good but I can't find datasheet's for the most popular brands  ::)

The graphite sheets are pretty interesting, without running complicated thermal models I am not really sure what benefit the increased X-Y plane conductivity is though.

The best solid sheet I can find to replace thermal grease is this graphite sheet, which is only 25um thin (!!).

http://www.tglobaltechnology.com/uploads/files/tds/T69-25.pdf

That has to be a typo? I am not sure how I would handle material that thin and not get creases in it or anything. Even then, its 0.14C/W per cm^2, which is about 3x worse than some thermal compounds I have seen. If my goal for the TO-247 package is 0.2C/W total thermal resistance though that might work in place of the thermal grease.


Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1806
  • Karma: +88/-1
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #7 on: August 14, 2022, 07:31:16 PM »
Great testing and results!

Quote
With a 20x1mm cross section and 10mm distance this equates to a thermal conductivity of 234 W/(m·K). The published value is 321 W/(m·K), so this is close enough where I am not going to worry about it!
Is 321W/mK for single crystals?  The ranges I'm finding for AlN ceramic are 140-240W/mK.  If that is accurate, your material is at the high end for ceramics.

For thermal compound, I've been using an older "Arctic Silver" product rated at 8W/mK.  I think there are well better products now.  I gather that bulk thermal conductivity is not as relevant as resulting actual contact resistance.  Ideally there is some direct contact at surface asperities of heat sink and device.  What matters is how well crevices are filled and final conductivity after the material has been compressed and exposed to multiple thermal cycles.  Don't know if there are any standardized tests for that.
David Knierim

Offline Weston

  • High Voltage Technician
  • ***
  • Posts: 123
  • Karma: +11/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #8 on: August 14, 2022, 10:38:37 PM »
Oops, my bad. I just copied the google blurb which was quoting the portion of the wikipedia article for the maximum possible thermal conductivity, which seems to be for single crystal.

The vendor page lists a value of "Thermal conductivity: 190-260w/m.k". My measurement is close to midway between that. Looking at the picture I took I realized the thermocouple tips are a bit closer together than the lines I drew. They measure at 8.6mm apart, which would lead to a thermal conductivity measurement of 205 W/(m·K). Regardless, it seems to be somewhere in the right range and is low enough where thermal resistance is dominated by the thermal paste.

Thermal resistance with thermal pastes is going to be impacted by the thermal conductivity and the bond line thickness, which is determined by the viscosity, particle size, mounting pressure, and surface flatness. I sadly cant find any figures on the surface flatness of the TO-247 package. The AlN plates seem pretty flat and smooth though.

My goal is to achieve around 0.2C/W for the TO-247 interface. The thermal pad area is ~ 1.4cm^2 so I need a thermal resistance of < 0.28C/W/cm^2 for the whole interface. The 1mm AlN thermal pad has a thermal resistance of ~0.04C/W, which leaves me with a budget of 0.12C/W/cm^2  for each of the two layers of thermal paste / interface.

In terms of thermal paste I found two public reports on testing of commercial thermal pastes  https://aip.scitation.org/doi/am-pdf/10.1063/1.5001835 https://www.nrel.gov/docs/fy08osti/42972.pdf

The NREL report is pretty good and they include artic silver in the tests. They measure artic silver as having so-so performance. However, the artic silver product literature claims that the thermal resistivity decreases after a few initial heating cycles, which these tests do not do



Based on these results it seems that Dow Corning TC-5022 would be good with a thermal resistance of 0.05 C/W/cm^2 at 40PSI

https://www.exdron.co.il/images/Products/files/%D7%92%D7%A8%D7%99%D7%96-%D7%98%D7%A8%D7%9E%D7%99-DOW-CORNING-datasheet.pdf

As would Dow Corning TC-5121 with a thermal resistance of 0.96 C/W/cm^2 at 40PSI: https://datasheet.octopart.com/13310-Greenlee-datasheet-14709059.pdf


I am a bit worried about thermal compound dry-out or pump-out and a phase change pad might be a bit better in that regard. Lard makes some high performance phase change pads: https://www.laird.com/sites/default/files/2018-11/THR-DS-TPCM580%201112.pdf

Looking at the datasheet the different thicknesses are all ~0.1 C/W/cm^2, which implies that the bond line is determined by pressure after the material melts and not the initial material thickness. I am not sure if these pads would not have the same issues with dry-out or pump-out given that the material is semi-liquid at operating temperature.

The best solid pads I can find are graphite based. A company called Jones makes some films that are 0.2C/W/cm^2 but you can only seem to buy them on digikey as a marketplace product https://media.digikey.com/pdf/Data%20Sheets/Jones%20Tech%20PDFs/21-655-0200.pdf

Laird is a bigger company but the best graphite films they sell are 0.42 C/W/cm^2 https://www.laird.com/sites/default/files/2018-11/THR-DS-TGON-800%201109.pdf



tl:dr:

Total thermal resistance of < 0.2C/W for TO-247 should be achievable with AlN thermal pad and Dow Corning TC-5022 at a bond line of < 150um, which is pretty large as far as thermal interface gaps go. Lets treat it as 0.3C/W to account for improper application or pump-out / dry-out.

Its a bit of a pain to sort though thermal pads on digikey, but it seems that the best insulating silicone /fiberglass thermal pads seem to be something in the range of ~ 0.6C/W to 1C/W.

 

Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1806
  • Karma: +88/-1
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #9 on: August 15, 2022, 12:58:37 AM »
This thread reminds me of paid work ~20 years ago measuring thermal resistance of TO220 FETs.  We first characterized gate threshold voltage vs. temperature in a roughly isothermal environment (externally heated).  Found very little part-to-part variation, which made testing easy.  Then ran thermal tests with a fixed gate-drain voltage (zener or battery) and external pulsed drain current source.  Current source pulsed from some high value to dissipate power to a low trickle current for threshold voltage measurement.  We were after transient thermal information.  The hardest part was accurate rapid Vgs measurement immediately after a higher-voltage state (off-scale on scope to precision measurement).  For looking at case-to-sink resistance differences, slower pulses and slower measurement would work fine.

The biggest puzzle was that we kept measuring internal thermal drop around 0.5C/W for parts specifying 1.0 to 1.1C/W.  Infineon finally explained the discrepancy.  Indeed the average drop (what we were measuring) was around 0.5C/W.  The published specification was much higher to account for the occasional void in die attach solder, making local hot-spot temperature rise about twice as much worst-case.  Many parts survived almost twice rated power.

BTW, our first version of this test used body diode forward voltage drop to measure temperature.  Until we learned that the body diode was far enough below the active power dissipation structures that diode temperature was not an accurate measure of FET gate/channel temperature.
« Last Edit: August 15, 2022, 01:01:39 AM by davekni »
David Knierim

Offline Weston

  • High Voltage Technician
  • ***
  • Posts: 123
  • Karma: +11/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #10 on: August 16, 2022, 05:56:49 AM »
That sounds like a cool test setup. Interesting that the P-N junction is too deep in the silicon to be accurate. I wonder what the depth of the diffusions are for modern power semiconductors.

What you are describing sounds pretty similar to the test setup that I was reading about in this app note https://fscdn.rohm.com/en/products/databook/applinote/discrete/common/rthjc_measurement_and_usage_an-e.pdf

It's pretty cool that by taking transient measurements of the actual junction temperature of the device they can calculate the thermal impedance of every interface in the package.

I came across that app note because I was curious as to how accurate the datasheet thermal resistance from junction to case was. The naive approach would include at least the thermal resistance of the thermal paste, or would require it be calculated and backed out. But with the method in that app note they can give the true value, so there is no margin in that.

Your experience is really not assuaging my fears on the accuracy thermal resistance values. But I guess its still a worst case value, so I won't be blowing anything up, just missing out on additional performance. I wonder if modern devices still have the same margin in the junction to case thermal resistance. Or is it account for delamination over the life of the device? I only recently realized that thermal cycles are really damaging to the die attach on power semiconductors. In tesla coils that really gets pushed...

I am currently trying to resist the urge to create a test setup similar to what you described / what that app note details. Another project that seems like it would be a great leaning experience but not actually useful in my current projects, or in graduating with my PhD in a timely fashion....


I ordered some thermal pastes and thermal pads from some vendors in China today. It's basically impossible to order any of the high performance thermal pastes / pads in small batches from the manufacture or from digikey. I found a company that sells some thermal pastes in re-packaged tubes, they seem legitimate based on online reviews, so hopefully I will get what I ordered.

I ordered:

Dow Corning TC 5022, which is one of the thermal compounds I previously discussed and has a thermal resistance of 0.06C/W/cm^2

Dow Corning TC 5026, which has a thermal resistance of 0.03C/W/cm^2 and a much lower dielectric constant / dissipation factor than Dow Corning TC 5022. The datasheet for Dow Corning TC 5022 says a dielectric factor of 18 and a dissipation factor of 0.56 at 1KHz. That seems to be so high I wonder if its correct. Or is the resistivity interfering with the measurement?

Honeywell PTM7950 phase change thermal pad, which has a thermal resistance of 0.04C/W/cm^2. Its non-conductive but I cant find a published dielectric constant.


I think that the phase change thermal pads are the way to go, as you don't need to deal with the hassle of the grease. However, thermal resistance is still limited by surface flatness / bond line thickness and the clamping has to have enough compliance to deal with the ~0.1mm change in thickness when the material melts at first heat and gets pushed out.


If you don't want to deal with random Chinese vendors (but then again, thats the only way to get the AlN thermal pads....) Laird TPCM 580 phase change pads are avaliable on digikey and are 0.1C/W/cm^2, but they are a bit expensive https://www.digikey.com/en/products/detail/laird-technologies-thermal-materials/A15405-01/2634225






Offline klugesmith

  • High Voltage Senior
  • *****
  • Posts: 517
  • Karma: +16/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #11 on: August 16, 2022, 08:30:17 AM »
Yup, it's hard to measure or model those very-close-to-junction thermal dynamics.
IN some cases it eventually gets simpler.  For example, tolerance of very brief single pulses of current, when adiabatic model is not too conservative.  No time for any heat flow, just thermal inertia of the material where the Joule heat appears (e.g. a bond wire). This leads to specs in the form of a maximum allowable I-squared-t product or integral.   Product can be larger when pulse is long enough for heat to be already diffusing away.

Here's a question about resistance of connections made with non-hardening thermal paste.  Doesn't squishing proceed until actual contact between high spots of the solid surfaces ( component, hard insulating pad, or heat sink) ?   That would make the paste layer thickness go to zero in some fraction of the interface area.  Over whole area, the thickness distribution would depend on roughness and non-flatness of the solid surfaces.  Do pastemakers, or electronic cooling literature, offer charts or rules of thumb to get "effective thickness" of layer where paste replaces air?

How do those thin graphite sheets help, unless they are soft enough to compress at high spots & move laterally to displace air at low spots?   
« Last Edit: August 16, 2022, 08:34:51 AM by klugesmith »

Offline Weston

  • High Voltage Technician
  • ***
  • Posts: 123
  • Karma: +11/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #12 on: August 16, 2022, 09:24:00 PM »
I think the graphite sheets are intended to be soft enough to deform and fill the gaps, same as the silicon pads. For a material that does not flow like a grease / phase change material I assume the pad has to be at least as thick as the deepest concavity to be effective. So perhaps the super thin graphite sheets are not too good.

Perhaps due to this, for rubber/graphite pads there is a strong dependence of the thermal resistance with pressure as it needs to make thermal contact. You can see that in this random example datasheet: https://www.ohmite.com/assets/docs/sink_pad_tim.pdf



For thermal greases / phase change material the material is already in full contact with both surfaces at very little pressure. The thermal resistance is determined by the bond line thickness between the transistor back and the heat sink / ceramic thermal pad. Bond line thickness is determined by the surface flatness, the viscosity of the material, any solid fillers in the material, and the pressure. It seems that for thermal greases there is little improvement as pressure is increased as the bond line just gets a bit thinner. I am not sure what the physics are, but it seems the bond line thickness asymptote in the 10's - 20's of um range as pressure increases. So if the surfaces are more planar than that you can just assume its all thermal paste with no actually contact between the surfaces?



The NREL paper used glass microbeads (same as a large particle size filler) to get controlled bond line thickness and they still get good thermal resistances using thermal pastes, even at a 100um bond line, which would be pretty extreme in terms of surface roughness / flatness. It seems like phase change materials / phase change pads should give good performance for surfaces with pad planarity and bad surface roughness. They should be able to get into even very deep groves in the heatsink while solid pads would not make full contact.


Offline klugesmith

  • High Voltage Senior
  • *****
  • Posts: 517
  • Karma: +16/-0
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #13 on: August 16, 2022, 10:54:19 PM »
>> "large particle size filler"

reminiscent of proppant granules, one component of the secret sauce used by frackers in oil/gas wells.  Can't let the gap get too thin after fracturing pressure is turned off.
« Last Edit: August 16, 2022, 10:56:43 PM by klugesmith »

Offline Uspring

  • High Voltage Engineer
  • ****
  • Posts: 306
  • Karma: +29/-1
  • Physicist
    • View Profile
Re: Aluminium Nitride Thermal Pads
« Reply #14 on: August 18, 2022, 01:01:41 PM »
Weston wrote:
Quote
I came across that app note because I was curious as to how accurate the datasheet thermal resistance from junction to case was. The naive approach would include at least the thermal resistance of the thermal paste, or would require it be calculated and backed out. But with the method in that app note they can give the true value, so there is no margin in that.

This app note is an interesting read. Thank you for the link. At the end of the note the authors write, that the equation

𝑇𝐽 = 𝑅𝑡ℎ𝐽𝐶 × 𝑃 + 𝑇𝐶 [°C]

must not be used to calculate the junction temperature. But it is just this equation that one wants to use for that purpose. So while the Transient Dual Interface method yields reproducible values, there is some doubt about the usefulness of this value to estimate junction temperature.
I believe the problem results from the definition of the criteria for the branching point of the 2 temperature curves, i.e. the ones with and without grease and the effect of the grease. The branching point could be defined by e.g. by a 10% difference in temperature, but the criteria should also depend on the effect of the grease. It is an advantage of the method, that the dependence on the criteria and on the effect of the grease is only logarithmical, but an uncertainty remains. The criteria will probably also have to be adjusted for differing distributions of heat conductivity and capacity inside the case.

Design engineers could be tempted to extend the TDI method to include the heat sink in order to get a thermal resistance value from the junction to the heat sink. This could be done e.g. by measuring 2 heat curves, one with forced air cooling on and one with it turned off. Sadly this will be inaccurate due to an additional element having heat conductivity and capacity properties, namely the heat sink. That will make the heat curve branching criteria change. There are criteria, which will get the correct value of RthJC, but they are unknown.

High Voltage Forum

Re: Aluminium Nitride Thermal Pads
« Reply #14 on: August 18, 2022, 01:01:41 PM »

 


* Recent Topics and Posts

post Re: Driverless Single Mosfet Relay interrupted SSTC? - [All kinds of problems]
[Solid State Tesla Coils (SSTC)]
TiagoBS
Today at 05:32:02 PM
post Re: Is this a pulse cap
[Spark Gap Tesla Coils (SGTC)]
55#sofcopper
Today at 04:39:00 PM
post Re: Audio modulation questions, chords
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Max
Today at 10:28:42 AM
post Re: Audio modulation questions, chords
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
Today at 04:49:55 AM
post Re: Audio modulation questions, chords
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Max
Today at 12:32:58 AM
post Re: RF mosfets and laser generation
[Light, Lasers and Optics]
dru
October 03, 2022, 11:29:26 PM
post Re: are these small guys flybacks?
[Transformer (Ferrite Core)]
Alberto
October 03, 2022, 10:43:21 PM
post Re: Is this a pulse cap
[Spark Gap Tesla Coils (SGTC)]
MRMILSTAR
October 03, 2022, 10:22:59 PM
post Re: Is this a pulse cap
[Spark Gap Tesla Coils (SGTC)]
klugesmith
October 03, 2022, 08:18:40 PM
post Is this a pulse cap
[Spark Gap Tesla Coils (SGTC)]
55#sofcopper
October 03, 2022, 07:36:49 PM
post Re: Getting last joules out of batteries
[Laboratories, Equipment and Tools]
rikkitikkitavi
October 03, 2022, 05:20:17 PM
post Re: RF mosfets and laser generation
[Light, Lasers and Optics]
Da_Stier
October 03, 2022, 10:26:07 AM
post RF mosfets and laser generation
[Light, Lasers and Optics]
dru
October 03, 2022, 03:31:12 AM
post Re: Getting last joules out of batteries
[Laboratories, Equipment and Tools]
klugesmith
October 03, 2022, 01:57:50 AM
post Re: Getting last joules out of batteries
[Laboratories, Equipment and Tools]
davekni
October 03, 2022, 01:45:48 AM
post Re: Audio modulation questions, chords
[Dual Resonant Solid State Tesla coils (DRSSTC)]
dru
October 03, 2022, 12:17:07 AM
post Audio modulation questions, chords
[Dual Resonant Solid State Tesla coils (DRSSTC)]
dru
October 03, 2022, 12:14:34 AM
post Re: are these small guys flybacks?
[Transformer (Ferrite Core)]
Mads Barnkob
October 02, 2022, 08:10:51 PM
post Re: are these small guys flybacks?
[Transformer (Ferrite Core)]
Alberto
October 02, 2022, 05:47:51 PM
post Re: Getting last joules out of batteries
[Laboratories, Equipment and Tools]
rikkitikkitavi
October 02, 2022, 05:31:05 PM
post Re: Radiometer Copenhagen OSG 42b Oscilloscope Repair, From The 1950'ish
[Laboratories, Equipment and Tools]
rikkitikkitavi
October 02, 2022, 05:27:23 PM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
mthome4
October 02, 2022, 02:24:19 PM
post Re: are these small guys flybacks?
[Transformer (Ferrite Core)]
Mads Barnkob
October 02, 2022, 01:11:56 PM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
October 02, 2022, 04:48:00 AM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
mthome4
October 02, 2022, 12:46:19 AM
post are these small guys flybacks?
[Transformer (Ferrite Core)]
Alberto
October 01, 2022, 10:13:29 PM
post My audio modulated sstc circuit, an easy one for beginners
[Solid State Tesla Coils (SSTC)]
dru
September 30, 2022, 08:05:55 PM
post Re: [WTS] large selection of IGBT bricks, large electrolytic capacitors
[Sell / Buy / Trade]
dbach
September 30, 2022, 04:17:35 PM
post Re: Driverless Single Mosfet Relay interrupted SSTC? - [All kinds of problems]
[Solid State Tesla Coils (SSTC)]
davekni
September 30, 2022, 05:01:14 AM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 30, 2022, 04:46:59 AM
post Re: Driverless Single Mosfet Relay interrupted SSTC? - [All kinds of problems]
[Solid State Tesla Coils (SSTC)]
TiagoBS
September 29, 2022, 10:23:27 PM
post [WTS] (EU) electrolytic caps 4700uF + film caps 30uF
[Sell / Buy / Trade]
costas_p
September 29, 2022, 10:13:52 PM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
JCF
September 29, 2022, 05:36:58 PM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Anders Mikkelsen
September 29, 2022, 12:05:51 PM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 29, 2022, 05:18:23 AM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Mads Barnkob
September 28, 2022, 08:28:16 PM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
JCF
September 28, 2022, 06:05:46 PM
post Re: controlling the negative gate bias voltage of the Gate drive transformer
[Dual Resonant Solid State Tesla coils (DRSSTC)]
prabhatkumar
September 28, 2022, 05:16:13 PM
post Re: DC charging solution
[Spark Gap Tesla Coils (SGTC)]
davekni
September 28, 2022, 05:43:34 AM
post Re: Recommended development boards for QCWDRSSTC
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
davekni
September 28, 2022, 05:20:55 AM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 28, 2022, 05:15:19 AM
post Re: Recommended development boards for QCWDRSSTC
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
dru
September 27, 2022, 07:11:42 PM
post Re: Recommended development boards for QCWDRSSTC
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
dru
September 27, 2022, 07:07:00 PM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
JCF
September 27, 2022, 05:08:45 PM
post DC charging solution
[Spark Gap Tesla Coils (SGTC)]
Conrad73
September 27, 2022, 04:57:38 PM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 27, 2022, 03:12:49 AM
post Re: Ud failure
[Dual Resonant Solid State Tesla coils (DRSSTC)]
klugesmith
September 27, 2022, 02:23:44 AM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 26, 2022, 07:55:31 PM
post Re: Help with PLL induction heater driver
[Electronic Circuits]
dbach
September 26, 2022, 04:04:11 PM
post Re: What am I missing?!?!?
[Spark Gap Tesla Coils (SGTC)]
davekni
September 26, 2022, 05:13:23 AM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 26, 2022, 05:10:13 AM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 26, 2022, 05:03:46 AM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 26, 2022, 04:33:12 AM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 26, 2022, 03:01:21 AM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 26, 2022, 01:30:04 AM
post Re: Help with PLL induction heater driver
[Electronic Circuits]
Anders Mikkelsen
September 25, 2022, 10:57:23 PM
post Re: Ud failure
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 25, 2022, 09:59:53 PM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
mthome4
September 25, 2022, 09:43:24 PM
post Re: Recommended development boards for QCWDRSSTC
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
davekni
September 25, 2022, 07:33:57 PM
post Re: Ud failure
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 25, 2022, 07:06:33 PM
post What am I missing?!?!?
[Spark Gap Tesla Coils (SGTC)]
Conrad73
September 25, 2022, 06:35:43 PM
post Radiometer Copenhagen OSG 42b Oscilloscope Repair, From The 1950'ish
[Laboratories, Equipment and Tools]
Mads Barnkob
September 25, 2022, 03:18:22 PM
post Re: Recommended development boards for QCWDRSSTC
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
Rafft
September 25, 2022, 02:53:30 PM
post Recommended development boards for QCWDRSSTC
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
dru
September 25, 2022, 09:05:37 AM
post Re: Ud failure
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Mads Barnkob
September 25, 2022, 08:29:09 AM
post Re: Ud failure
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 25, 2022, 05:27:41 AM
post Re: Gauss meter for strong magnetic fields
[Laboratories, Equipment and Tools]
klugesmith
September 25, 2022, 02:12:21 AM
post Ud failure
[Dual Resonant Solid State Tesla coils (DRSSTC)]
flyingperson23
September 25, 2022, 01:50:45 AM
post Re: controlling the negative gate bias voltage of the Gate drive transformer
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 24, 2022, 09:36:58 PM
post Re: Gate drive transformer with negative gate bias voltage
[Dual Resonant Solid State Tesla coils (DRSSTC)]
prabhatkumar
September 24, 2022, 06:56:35 AM
post Re: Gate drive transformer with negative gate bias voltage
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 24, 2022, 06:09:03 AM
post Re: Gauss meter for strong magnetic fields
[Laboratories, Equipment and Tools]
davekni
September 24, 2022, 01:32:08 AM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
mthome4
September 23, 2022, 10:55:21 PM
post Re: Gauss meter for strong magnetic fields
[Laboratories, Equipment and Tools]
AstRii
September 23, 2022, 10:39:13 PM
post controlling the negative gate bias voltage of the Gate drive transformer
[Dual Resonant Solid State Tesla coils (DRSSTC)]
prabhatkumar
September 23, 2022, 05:48:17 PM
post Gauss meter for strong magnetic fields
[Laboratories, Equipment and Tools]
Peregrine
September 23, 2022, 02:01:10 PM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 23, 2022, 03:04:31 AM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 22, 2022, 04:15:36 PM
post Re: SSTC - PCB design
[Beginners]
Przemekk00
September 22, 2022, 12:39:51 PM
post Re: Possible Leads on Single Photon Camera
[General Chat]
alan sailer
September 22, 2022, 12:05:55 AM
post Re: Possible Leads on Single Photon Camera
[General Chat]
Mads Barnkob
September 21, 2022, 10:32:29 PM
post Possible Leads on Single Photon Camera
[General Chat]
alan sailer
September 21, 2022, 03:57:38 PM
post Re: Driverless Single Mosfet Relay interrupted SSTC? - [All kinds of problems]
[Solid State Tesla Coils (SSTC)]
davekni
September 21, 2022, 05:50:15 AM
post Re: Musical interrupter problems
[Solid State Tesla Coils (SSTC)]
RoamingD
September 20, 2022, 10:29:41 PM
post Re: Musical interrupter problems
[Solid State Tesla Coils (SSTC)]
Mads Barnkob
September 20, 2022, 08:24:30 PM
post Re: Driverless Single Mosfet Relay interrupted SSTC? - [All kinds of problems]
[Solid State Tesla Coils (SSTC)]
TiagoBS
September 20, 2022, 06:23:54 PM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 20, 2022, 05:20:07 AM
post Re: TO247 mounting
[Beginners]
davekni
September 20, 2022, 05:15:35 AM
post Re: MidiStick V2.0: Next gen tesla coil interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
TMaxElectronics
September 19, 2022, 11:43:47 PM
post Re: TO247 mounting
[Beginners]
Weston
September 19, 2022, 09:58:56 PM
post Re: Some questions about my first drsstc.
[Dual Resonant Solid State Tesla coils (DRSSTC)]
mthome4
September 19, 2022, 09:05:35 PM
post Re: Musical interrupter problems
[Solid State Tesla Coils (SSTC)]
RoamingD
September 19, 2022, 12:58:35 PM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 19, 2022, 12:13:46 AM
post Re: MMC capacitor crash
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 19, 2022, 12:09:22 AM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 18, 2022, 11:38:51 PM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 18, 2022, 10:41:54 PM
post Re: Musical interrupter problems
[Solid State Tesla Coils (SSTC)]
davekni
September 18, 2022, 10:31:32 PM
post Re: Musical interrupter problems
[Solid State Tesla Coils (SSTC)]
RoamingD
September 18, 2022, 10:09:33 PM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 18, 2022, 09:58:34 PM
post Re: UD2.7 dual Purpose Use
[Dual Resonant Solid State Tesla coils (DRSSTC)]
FonziDaytona
September 18, 2022, 08:45:21 PM

Sitemap 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 
SimplePortal 2.3.6 © 2008-2014, SimplePortal