2SC2879 vs. 2SC3240

[Shockwave]

When the original source for the Toshiba 2SC2879 dried up, I noticed several of the newest HF mobile rigs started using the Mitsubishi 2SC3240. Just today I looked up the datasheet for this part and it was very impressive! I was hoping it would be close to the 2SC2879 and had no idea it could be better. It has 20 more watts of collector dissipation at 270 watts and 2 more volts of breakdown capacity at 20 volts. It equals the continuous current rating at 25 amps too! The only area it appears to be a little lacking in is gain at 11.5 db. So it takes 7 watts of drive. Has anyone used this transistor to see how it handles abuse and if it lives up to having specifications greater than the old 2879?

 

[Highway Man]

where's the popcorn emoji? I'm all in

 

[kopcicle]

20 more watts of collector dissipation at 270 watts and 2 more volts of breakdown capacity at 20 volts. It equals the continuous current rating at 25 amps too! The only area it appears to be a little lacking in is gain at 11.5 db. So it takes 7 watts of drive

$44.95

MRF300AN
27mhz 50V , 340w cw, 80%+ efficiency,27.3db gain, TO247-3 package, tab is source (ground), complimentary part available (gate and drain swapped) . The rest is in the pdf
$25
50v you say? Too bad. Get a boost converter. Life moves on.

 

[Shockwave]

$44.95

MRF300AN
27mhz 50V , 340w cw, 80%+ efficiency,27.3db gain . The rest is in the pdf
$25
50v you say? Too bad. Life moves on.

That's great for new designs where 50 volts is an option but this 2SC3240 seems like a much better replacement for existing 12 volt equipment than the HG transistors mislabeled as 2SC2879. They can be purchased for less than $20 too. The $44.95 is the most expensive price that can be found at RF Parts. The fact that Yaesu used them to replace the Toshiba rather than the HG part, is somewhat telling also.

Notice the 20:1 VSWR test hasn't been conducted at the industry standard of 12.6 volts for these devices. To demonstrate ruggedness, Mitsubishi performed this test at a VCC of 15.2 volts. That indicates much more power can be produced at a lower VSWR without exceeding breakdown voltage.

 

[kopcicle]

meh

 

[nomadradio]

2SC3240 is a discontinued part. Wasn't re-tooled to sell as a ROHS toxin-free design like the red-dot Toshiba parts.

Specs look good, but when the pipeline goes dry, they'll be gone.

RF Parts will sell you single parts, but matched pairs are out of stock. Probably forever.

73

 

[Shockwave]

I will admit when you posted this part number I was impressed with its specifications. Seeing they now sell PC boards designed to demonstrate the parts abilities makes it much easier to work with too. I can tell you from experience that working with a high power RF device in a TO-247 package with over 27 db of gain, is very problematic. Success with any stability are highly dependent on the construction and layout of the PC board.

You're forced to address everything from the high inductance leads to lead creepage during thermal cycling. Those 3 stiff narrow pins can apply enough force through expansion and contraction to rip the leads from the chip within the package when not properly bent. That means even more inductance must be added. The layout of the die inside this package does not dissipate heat well and as you see, requires the use of a copper spreader that is normally not seen on devices below 1kw.

The single hole mounting screw provides uneven pressure across the back of the device to the heatsink. You must compromise between enough torque to transfer heat while not using too much so that the bottom edge of the package lifts from the heatsink. For this reason the TO-247 is being abandoned in many high power devices in favor of the TO-247 PLUS. This package is the same shape but has no hole. It uses a clamp over the top to apply even pressure and allows a larger die space for the chip inside.

You're still dealing with those high inductance leads. The source lead must be low inductance or the circuit will oscillate at RF. One thin wire out of the base cannot accomplish this goal. Therefore, the back metal surface of the transistor is also the source terminal. Make sure that surface and the heatsink are as smooth as possible and use as little heatsink compound as possible. Otherwise, you can easily compromise the RF connection between these two surfaces.

 

[Shockwave]

2SC3240 is a discontinued part. Wasn't re-tooled to sell as a ROHS toxin-free design like the red-dot Toshiba parts.

Specs look good, but when the pipeline goes dry, they'll be gone.

RF Parts will sell you single parts, but matched pairs are out of stock. Probably forever.

73

Agreed but singles pulled from the same lot will usually behave well together and they are still much more abundant than any Toshiba today. Ebay has loads of them under $20. They are also not known well enough to have created a counterfeit market. The unique package shape alone makes that difficult to do. I'm not promoting the use of this part as much as I was seeing if anyone has tested its true capabilities yet. Someone will now.

 

[Shockwave]

I also have to say that whole ROHS thing ruined a lot of good parts that were once available. The beo ceramic is the best material for electrical insulation and thermal conduction we have. To ban its use rather than encouraging proper recycling was counterproductive. They could have doubled the original price to add a "core charge" to the part so that defective ones would be properly recycled or disposed of. New equipment manufacturers could have received wavers. Millions could have been kept of of landfills if there was a $20 core charge.

 

[blasphemy000]

The two original ones in my IC-751 (non-A) have held up great. I wouldn't say that I've abused them, but I have put them in some less than ideal situations by accident and they could easily exceed the radio's rated 100W output when the ALC op-amp decided to give up. The previous owner daily drive this rig since it was knew until he passed 8 months ago, and it's been good to me. They seem like tough little buggers, but I don't know how they would withstand being in a typical CB type amp, because I've never seen them used that way. Don't go buying them all up though, I might need spares some day. Lol.

73s

 

[Highway Man]

i was just watching a YouTube video of a japaneese guy with a 12 pill with 2sc3240s
and was wondering about those transistors myself. he also had a box of 100 Toshiba 2879s.

 

[Shockwave]

I'm bumping this thread because this transistor still seems to be the best by far to replace transistors being used in 12 volt applications. Real BeO ceramic, 20 more watts of dissipation and 2 more volts before breakdown failure! Why go backwards when you can still go forwards? Choosing an amplifier with switch mode FET's while these are still available for less than $20 each does not seem logical. I now have a pair in a modified HF rig that modulates the RF stage on AM to over 170 watts PEP before I decided to stop do to an inadequate heatsink and questionable power handling of the low pass filters.

 

[Shockwave]

OK, it's taking too long for someone into 12 volt amps to finally try this part so I'll give some info in order to start getting some good parts back into old 2SC2879 amps...

Understand the Mitsubishi 2SC3240 was designed years after the 2879 and is a step forward in many areas from voltage handling to power dissipation. In stock today at RF parts for $44 and on eBay for less. It replaced the 2879 in HF rigs as the last newly designed high power bipolar transistor before the switch to FET's

They probably would have been in Toshiba amplifiers years ago if they made any power as a drop in replacement. They don't because their input drive impedance is nearly 4 times lower than the 2879! This problem could not be any easier to overcome. Simply drop a 4:1 RF transformer in front of your first RF splitter and you're 95% of the way there. Adjustment of input and output padding caps should dial you in from here.

 

[443]

I'd be interested in playing around with them. However, looks like all the 2SC3240's on eBay are from China, so they will all be fakes, if lucky maybe some are just used pulls that still have a little life left in them. But I'd be leery of ANY shipped from China.

Question, did MA/COM stop making the MRF454? Just curious, thought you may know.

 

[nomadradio]

According to their web site, they still have the MRF454 in stock, among others.

https://www.macom.com/products/rf-power/rf-power-transistors-silicon-bipolar

73