Testing the thermal performance of my transistor adapters. (TO-3P to MT-200 and 2SD746/2SB706 packages for Pioneer receivers.)

UPDATE: 2024.03.03 They’re now available in the Hoppe’s Brain Shop!

The MT-200 package has terrific thermal performance, due to their large copper heat-spreader. Copper conducts heat nearly twice as quickly as aluminum, and the large contact area with the heatsink makes for more efficient thermal transfer than a TO-3P. As you can see from this image, many MT-200 devices are actually constructed with a TO-3P substrate soldered to the larger MT-200 plate. They are often rated for an additional 50W dissipation versus a TO-3P version of the same transistor, and this is entirely due to the action of the heat spreader.

So the question is: Does the Hoppe’s Brain adapter with a TO-3P transistor soldered to it work as well as a real MT-200?

The experiment:

Very simple: We’re going to mount the transistors to a large heatsink, and plot the ∆ temperature rise with a constant 50W dissipated.

I don’t have the computer interface for my Fluke 287, so I’ll just take pictures. It’s really hard to take a picture of an LCD screen, but I find it works OK with the flash on, taken from a high angle.

Test setup:

A large heatsink is tapped for MT-200 mouting holes. A hole is drilled in the center for the thermocouple, and a thermal pad is cut out of Parker CHO-THERM T441.

A 5K resistor is tied from V+ to base, ground goes to emitter. Power supply is set to deliver constant power at 50W. This develops approximately 13V from C-E, at 3.8A.

Each run is 1 minute long, starting around 23C, and I let the heatsink cool completely between runs.

Results:

MT-200 2SA1494 – Start 22.7C, End 49.0C = ∆ 26.3C
Hoppe’s Brain MT-200 to TO-3P adapter with NJW1302 – Start 23.4C, End 45.8C = ∆ 22.4C
NJW1302 on it’s own – Start 22.0C, End 49.8C = ∆ 27.8C

W00t W00t! The Hoppe’s Brain adapters did better than MT-200 by 3.9C! You can also see that at the 12-second mark, the MT-200 is at 40C, and the adapter is at 39C! This could be due to a few factors; The NJW1302 has a slightly larger die, 6mm versus the 5mm die in the 2SA1494, and the heat-spreader plate on the MT-200 is actually slightly smaller than the adapter’s plate, due to the plastic all the way around the edges. And then there’s the tolerance of the test, which could be a few percent.

Interesting how they all came up to a steady-state around 49C, and this was true with the smaller back-plate of the TO-3P. The difference is, it takes longer to come up to temperature with the MT-200, or with the adapters. This is important in audio applications where there are lots of short-duration, high-power events. Notice how much more sharply the temperature rises on the TO-3P.

Conclusion:

Hoppe’s Brain MT-200 adapters perform at least as well as original MT-200 devices.