Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Synchronous MOSFET rectifier at 200 MHz

Status
Not open for further replies.
H

htg

Full Member level 5
Joined
May 1, 2010
Messages
248
Helped
5
Reputation
10
Reaction score
3
Trophy points
1,298
Location
Poland
Activity points
2,862
Can I expect good rectification efficiency (say 70% or more) if I use synchronous MOSFETs at 200MHz? I consider using MRF6V2300 RF MOSFETs.
 

Thanks for the info. But I wonder if I can use MOSFETs for rectification at 200MHz because of the inductance of the connections, which cannot be compensated.
 

Hello,

What voltage and current would you like to rectify? Mostly synchronized rectifiers are used in low voltage systems where forward voltage drop reduces efficiency. For 200 MHz, I don't think synchronized rectification will help you.

I assume your question does not relate to small signal instrumentation. Do a search on "rectenna". They achieve reasonable efficiencies at 2.45 GHz, so for 200 MHz >70% must be possible with suitable rectifiers. You need real schottky diodes with high reverse voltage so that you can accept relative high output voltage (10V or more).
 
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
I am thinking about 10A and 100V or more. Based on the specifications of the best small-signal diode I found - MBD301, 30V, 100mA, 1.5pF, 300ps, I expect 70% efficiency at 200 MHz. I will try to do a search on "rectenna".
 

I was expecting something like that because of the power transistor you mentioned. This is a difficult one, GaAs, GaN, SiC, etc comes in my mind.

The "problem" with GaAs diodes like IXYS GS150TC25110 is their high effective capacitance (capacitance increases with reduced voltage), so you get large reactive current through your diode limiting the actual rectified current. It will also be a narrow band detector. If the drive level varies, you may expect mismatch problems.

My feeling is that there must be a diode design somewhere in a research lab that will enable such rectification (10A/100V). If you can't find them and this is a real (nice) challenge, you probably need to divide the power. I assume you don't like vacuum devices...
 
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
The "problem" with GaAs diodes like IXYS GS150TC25110 is their high effective capacitance
Click to expand...
Compared to conventional Si-Schottky diodes, the IXYS RF rectifiers have unique low capacitances, which together with the low package inductances gives hope for a usage at 200 MHz, being effectively impossible with standard 10 A schottky diodes. The targetted frequency range of the new devices is however the several 10 MHz domain and it's challenging to operate them at the tenfold.
 
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
Hello FvM,

You are right, for the intended application their capacitance is extremely low and they will outperform every Si-device, thanks for the addition.

I was looking from his 200 MHz perspective. I don't have experience with these diodes so I don't know whether there will be some diffusion capacitance effect also. Maybe others can confirm this. With a careful design, my feeling is that it should be possible to get some hundreds Watts DC from these rectifiers.

I considered such GaAs devices for an RF > 10 kW application, but the analysis turned out that I needed higher supply voltage, so I had to divert to 600V SiC (with significantly more capacitance, so snubbers were required). The rectifiers were used to convert back to DC the stored energy from an LC circuit. It saves me from slow decay of the RF envelope and regained the unused RF power.
 
Last edited:
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
I looked at IXYS GS150TC25110 data sheet. They show reverse recivery at I = 5A, dI/dt=150A/us, so the trr is near 30 ns.
I considered vacuum devices, in particular a field emission diode. It would be promising if someone made it, but it would require nanotechnology.
 

Trr is large with respect to your repetition time (5ns), but as long as this reverse current is caused by a capacitive effect, you can compensate for this in your matching network. The disadvantage is that you get high capacitive current through the junction that reduces DC output power and efficiency.

I don't know your final goals and resources, but there is also a cyclotron wave converter that increases the kinetic energy of an electron beam with the RF input power. After harvesting the electrons the output voltage is higher than the initial acceleration voltage to create the beam.
 
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
Trr is large with respect to your repetition time (5ns), but as long as this reverse current is caused by a capacitive effect, you can compensate for this in your matching network.
Click to expand...

There has been a partly confusing discussion about trr of schottky diodes in a predecessor thread. By their design principle, schottky diodes don't utilize minority carriers and have no respectively only ps order of magnitude reverse recovery time. There is however a reverse current peak caused by the diode capacitance, and some manufacturers specify a reverse recovery time based on it. IXYS also does in when comparing waveforms in the GaAs diode application note. But the important difference to the real reverse recovery of bipolar diodes is, that it's not necessarily involving switching losses.
 
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
I thought about using a series LC curcuit resonating at twice the rectified frequency, placed after the full-wave bridge. Do you think it would enable me to use a diode, whose trr seems to be about 30ns if I want to rectify 200MHz?
 

Hello,

I don't know whether the second harmonic circuit behind the full-wave bridge will be the best alternative, but the Trr may not be an obstacle (my datasheet says 15ns).

As what FvM says, in real schottky rectifiers (so not the hybrid/compound devices), the reverse current transient is because of the diode capacitance (like in a varactor). The transit time (the "tt" parameter in the diode spice model) will be in the ps range.

IXYS does not provide a model, so you have to do some extraction yourself (based on the datasheet). If the simulation looks promising, it is time to build it.
 
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
Whatever the reason, if it looks like trr=15ns, I am sure it will not let me rectify 200 MHz. So IXYS is right to consider the apparent reverse recovery time to be trr. It lets you know what to expect.
 
Last edited:

Hello,

Best is to study standard PN junctions (diffusion capacitance, carrier life time, transition time, etc) and schottky junctions as they behave different with respect to the forward-reverse transition. You may also search for the working principles of varactor and step recovery diodes. In many application they do the same, but the physical background is different, hence efficiency, etc.

I would advice you, to do the modeling and see what happens. You may put aside components that may help you to reach your goal.

If your application may be of interest for IXYS (for example a thesis), you may contact IXYS for support during modeling
 
  • Like
Reactions: htg

    H

    htg

    Points: 2
    Helpful Answer Positive Rating
Like the rest of modern physics, solid state physics is full of lies. They talk about diffusion of holes to the n-side of a p-n junction and other things that cannot really happen. But my current interest in VHF rectification is motivated by a practical application. If anyone wants to know what happens in semiconductor devices, he should start by noticing that quantum mechanics is completely false and it is time to discover the truth about the structure of matter.
Anyway, I need a very fast rectifier. If I have to, I will go to lower frequencies, and then IXYS GaAs Schottky diodes may be of interest to me. So thank you for telling me about them.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top