Here is the teaching vedio of class E by keysight using CGH60030D hemt model on youtube.

Here is the datasheet of CGH60030D hemt. from cree, https://cms.wolfspeed.com/app/uploads/2020/12/CGH60030D.pdf

Recently, i started my university research on RF power amplifier, while I was trying to find some data for the GaN hemt applied on the class E power amplifier. I found that most of the GaN hemt used on the power amplifier is depletion mode with negative Vgs. I wonder what might cause this and the advantages of this design comparing to using enhancement mode GaN hemt, since firstly i need to do simulations on ads. So i am just very comfused about this. What will the results be if i use enhancement mode GaN hemt.

 

I'm going to hazard a guess and say the reason is related to the geometry and the process of creating the device. It was the case for MOSFETs where the n-channel enhancement devices had clear advantages over it's p-channel cousin. Similarly there don't seem to be any p-channel JFETs.

 

My understanding is that the primary reason depletion-mode GaN HEMTs are used is because they're much easier to fabricate. Enhancement-mode GaN is apparently really hard to properly make and be comparable to the depletion-mode versions in operation. One of the easiest way we can make enhancement-mode GaN HEMTs is to take a depletion-mode one and combine it with an enhancement silicon FET. This runs into its own problems, such as its inability to handle high frequency or low voltage.
The PDF I've attached goes into this in much greater detail.

Basically, an enhancement-type GaN would be much more convenient for almost all applications, but depletion-mode GaN (in its current state) has the advantage of being easy to make, handle a wide range of power levels, and has an enormous bandwidth.

 

My understanding is that the primary reason depletion-mode GaN HEMTs are used is because they're much easier to fabricate. Enhancement-mode GaN is apparently really hard to properly make and be comparable to the depletion-mode versions in operation. One of the easiest way we can make enhancement-mode GaN HEMTs is to take a depletion-mode one and combine it with an enhancement silicon FET. This runs into its own problems, such as its inability to handle high frequency or low voltage.
The PDF I've attached goes into this in much greater detail.

Basically, an enhancement-type GaN would be much more convenient for almost all applications, but depletion-mode GaN (in its current state) has the advantage of being easy to make, handle a wide range of power levels, and has an enormous bandwidth.

WoW. Thank you for your patient guidance. Does it mean that for n-channel GaN hemt, the reason we always prefer to use depletion mode is that it is more easy to fabricate and has such advantages like low voltage and inability. Yes, the paper is really helpful (many thanks!!). But when it comes to power amplifier. Can the specific reasons be like larger electron mobility, which can improve the gain (maybe)?

 

Does it mean that for n-channel GaN hemt, the reason we always prefer to use depletion mode is that it is more easy to fabricate and has such advantages like low voltage and inability.

I believe that is correct, yes.

But when it comes to power amplifier. Can the specific reasons be like larger electron mobility, which can improve the gain (maybe)?

More or less. High mobility helps with the frequency range primarily, which is desirable for a lot of modern RF power amplifiers. It might be helpful for the gain as well, but mobility is primarily for frequency. The power levels of the device are due to the wide bandgap of GaN; it takes more energy to cause the device to conduct, but that also means it can handle more power. GaN transistors themselves don't have particularly high gain by themselves; high gain comes from cascading multiple stages together.