[amb]

The second run of the boards is being done as a group buy at Headwize. If you're interested in the board and have not signed up yet, you can do it now.

 

[amb]

FYI, official websites are now up:
- The ε24 Power Switch Driver Circuit
- The σ24 Power Transformer Board

 

[luvdunhill]

amb:

I have a quick question about choosing a thermostat. Say I have a device that has a junction temperature rated to 150C and I want trip the thermal protection circuit when junction temperature hits 75C. Would I choose a 75C thermostat, or something closer to 50C, which is what my heat sink / device case temperature will be when the junction temperature is 75C?

Thanks!

 

[amb]

The trigger temperature should be quite a bit less than the target junction temperature, because what you measure at the heatsink will be less than the device's junction due to the thermal resistances of the device case, isolation pad and heatsink.

 

[luvdunhill]

cool, that's what I figured. Here's a thermal plot of the max heat sink temperature. Available trigger temperatures are 75, 70, 65, 60, and 50 for the Airpax devices. These results are pretty close to what I'm measuring. In this simulation, the max junction temperature of all 6 devices ranges from 75.3C-78C, which is pretty close to where I want them. They are each dissipating 14.8W and are TO-3P packages.

So, maybe try the 50C thermostat located to the left of the left-most-center device? Or maybe go with a 60C device? What do you think?

 

[amb]

Which trigger temp to choose depends on how much leeway you want to give for operating variations. Too close to the actual "normal" temps may lead to frequent false cutouts, too far and you get reduced protection. You choose a comfortable midpoint. This of course also depends on where you mount the thermal breaker. Ideally you want it to be within the hotspots, but that isn't always possible, so you go with a lower trigger temp to compensate.

 

[luvdunhill]

amb:

What are the factors in determining if a 1.2VA transformer falls in the ~1.5VA-2V range?

How much current is needed in addition to the coil current in determining this number?

Also, do you recommend using grease under the thermostats?

 

[amb]

The ε24 circuit itself draws very little current. Most of the current is pulled by the relay coil, so your relay choice is what determines whether a particular transformer is good enough. The 1.5VA spec is good for the Omron G8P relay.

Yes, thermal compound should be applied to the thermostat's heatsink mating surface.

 

[luvdunhill]

Quote:

Originally Posted by amb /img/forum/go_quote.gif The ε24 circuit itself draws very little current. Most of the current is pulled by the relay coil, so your relay choice is what determines whether a particular transformer is good enough. The 1.5VA spec is good for the Omron G8P relay. Yes, thermal compound should be applied to the thermostat's heatsink mating surface.

ok. The G8P seems to have more than double the current draw of many of the SSRs I've seen (including the D1225), so I think I'll be okay. This way, I don't have to wait for the other transformers to get back into stock, and can instead go ahead and use Digikey part 567-1028-5-ND.

Thanks!

 

[amb]

luvdunhill, see Note 4 under the "Parts list" section of the ε24 website. Many SSRs' control inputs draw so little current that the thermal sensing feature may not work without adding a resistor as specified.

 

[luvdunhill]

Quote:

Originally Posted by amb /img/forum/go_quote.gif luvdunhill, see Note 4 under the "Parts list" section of the ε24 website. Many SSRs' control inputs draw so little current that the thermal sensing feature may not work without adding a resistor as specified.

I figured that in. I should still be well under 80mA or so total current draw.