Choosing a MOSFET for class-A linear operation

[vu2tve]

Hi,

I've posted before on this forum concerning a class-A R.F. PA that I've been building for a multi-carrier FM broadcast project. The current status of the design is that I'm using a NXP MRF300AN for my design, as it has a lower case-to-junction thermal resistance than my previous choice of MRF101AN (0.55 C/W as opposed to 1.1C/W).

Now, the design is performing reasonably well at this point, but I'm exploring options for pushing the power envelope.

Because this is class-A, thermal management is one of the prime concerns. I was exploring some NXP transistors and found this device.
MRFE6VP5150N - 150W CW, 1.8-600MHz.

https://www.nxp.com/products/radio-frequency/rf-industrial-scientific-and-medical/1-600-mhz-broadcast-and-ism/1-8-600-mhz-150-w-cw-50-v-wideband-rf-power-ldmos-transistors:MRFE6VP5150N?tab=Design_Support_Tab

While it's max rated power of 150W is lower than the MRF300AN's 300W it has much better thermal specs. Case to junction thermal resistance is 0.21 C/W (as opposed to 0.55 C/W) and max operating junction temperature is 225C as opposed to 175C.

However, since there are two MOSFETS on one die, this chip was designed with class-AB operation in mind, it seems.

I was wondering what if I just parallel up the two transistors and run them both in the linear region. Thoughts? Has anyone have experience with using these transistors in the linear region?

Thank you!

[jonpaul]

Typical power FETs are specifically for SMPSand poor performance linearity in linear application especially subject matter UHF  parasitic oscillation.BJT may be a better choice


Jon

[vu2tve]

If you refer the link in the original post, you'll see these are specifically RF LVDMOS technology MOSFETS.

[RoV]

However, since there are two MOSFETS on one die, this chip was designed with class-AB operation in mind, it seems.

I was wondering what if I just parallel up the two transistors and run them both in the linear region. Thoughts? Has anyone have experience with using these transistors in the linear region?

You may have difficulty at stabilizing the operating point (two devices, one temperature...). I also suspect you'll have to apply some derating, because the individual cells in each MOSFET may have some degree of umbalance (my opinion only).
But why don't you consider a push-pull in class AB? Distortion would be more or less the same, but with a much lower power consumption for the same PEP power if you have several FM carriers.

[vu2tve]

I do want to consider class-AB push pull as well, but because it's a multicarrier setup, having class-A linear operation reduces the in-band imd. Atleast that's the theory. The current MRF300AN is working quite well in linear region. Efficiency is not really a concern at this point, but thermal management definitely is a challenge that would be much easier to manage in class-AB.

Perhaps I'm too afraid but class-A feels like a safer bet for this application, and once I have one design nailed down, then go and explore class-AB.

I do also have a MRF101AN and MRF101BN lying around, so could in theory try a design. Need to search for reference designs for class-AB, and learn about matching baluns