Hi,

I can get both the transistors to switch on and off alternately. I'm not sure what sort of gate driver i need. I'm using dual transformers

 

Class D is generally meant for lower frequencies, such as audio.

What is the range of RF you are going for?

Are you sure you didn't mean Class C? This is generally a high efficiency RF amp.

 

Class D involves chopping the waveform up at a much higher frequency then amplifying it digitally and putting it back together. I can just see trying to make a 100 MHz amp by chopping the signal into 5 GHz sections.

My guess is he's talking Class C unless this is some sort of exercise in theory which, if it is, serves a questionable teaching purpose.

http://en.wikipedia.org/wiki/Electronic_amplifier

 

hi,

I'm using the class d for RF purposes, with frequencies ranging in the ISM band of 6.7MHz.

Class D RF amps are used for inductive coupling power and data transfer. I can get the switching function to work but when i used the tuned circuits to filter the harmonics, it does not work.

 

There is some difficulty in intepreting your statement -

I can get both the transistors to switch on and off alternately. I'm not sure what sort of gate driver i need. I'm using dual transformers

If you have the transistors switching, doesn't that imply a gate driver in use?

As to the whole amplifier, we really need to see schematics before anyone can help.

 

Hi,

Sorry, I've got the transistors switching using a dual transformer which works well producing a square wave. however, i cannot get a sine wave after the bandpass filter.

i've attached the schematic

 

Hi,

I can get both the transistors to switch on and off alternately. I'm not sure what sort of gate driver i need. I'm using dual transformers

Transistors do not have gates. BJT (Bipolar Junction Transistors) have an emitter, a base and a collector. MOSFETs (Metal Oxide Semiconductor Field Effect Transistor) have a gate, a drain and a source.

Words mean what they mean and not what you might think they mean.

 

Transistors do not have gates. BJT (Bipolar Junction Transistors) have an emitter, a base and a collector. MOSFETs (Metal Oxide Semiconductor Field Effect Transistor) have a gate, a drain and a source.

Words mean what they mean and not what you might think they mean.

IMHO, BJTs and MOSFETs are both transistors. After all, they both have the word "transistor" in their full names.

 

In a literal sense this is true, but in my experience the unqualified term "transistor" pretty much always refers to a BJT. FETs are seldom, if ever referred to as "transistors", but rather JFET, MOSFET etc. Maybe you dwell in an environment where the rules are different: nothing I can do about that.

 

I'll have to go with Ron on this one, transistor is a generic term, and is used as such by most people on this site.

I'll have to admit it makes me pause when I see (or say) BJT transistor or MOSFET transistor.

I'll say it again, I think the OP does not understand Class D correctly, it is not used for RF, and is not really practical for RF. A schematic would clear the issue up. Getting Class D to work for audio using a 100Khz or 1Mhz base frequency already pushes the envelop of frequency response for the vast majority of components.

Pulse Width Modulation

 

OK, found where the OP had displayed his schematic. Translating it into something this site can display...

Nope, no RF there. A 60 Hz square wave, with a series 159Khz resonant circuit following. If the series resonant circuit was even close to the 555 oscillator those transistors would burn out, since series resonant is a short at its resonant frequency, as is this circuit really doesn't make much sense.

The OP is continuing some earlier trends.

RF Class D Power Amplifier Design

This is not a Class D amplifier, nor is there any RF other than the harmonics from a 60Hz square wave. Any harmonics over 10X are going to be very small indeed (600Hz), above 100X (6Khz) it will be practically nonexistent.

The term RF is not even close to this circuit and how it works. If uceesdp would go into how he wants to accomplish what he is after I suspect we could help.

 

From the very sketchy document provided, there is a lot of design yet to go on. The use of a 555 timer put's an upper limit on the frequency (now 60 Hz). The dual secondary transformer is probably not suited to run above 60 Hz. The 2N5018's are 10 ma JFET's. There are no resistors in the gate circuits, so it's no wonder that the output is a square wave - indicating saturation on the JFET's.

To the OP - what are your goals in this design? What power output level do you hope to reach?

 

I'll have to go with Ron on this one, transistor is a generic term, and is used as such by most people on this site.

I'll have to admit it makes me pause when I see (or say) BJT transistor or MOSFET transistor.

I'll say it again, I think the OP does not understand Class D correctly, it is not used for RF, and is not really practical for RF. A schematic would clear the issue up. Getting Class D to work for audio using a 100Khz or 1Mhz base frequency already pushes the envelop of frequency response for the vast majority of components.

Pulse Width Modulation

Maybe it's because I graduated during the Johnson Administration

 

Maybe it's because I graduated during the Johnson Administration

So did I.
As you said, environmental factors are involved. I designed many an analog circuit before I retired, and until 1996, transistors were, in my mind, BJTs. Then I went to work for Micron, designing memory chips. As you can imagine, transistors rapidly (suddenly, even) became MOSFETs.

 

OK now I understand. You see I never had the pleasure of working on a slab o' sand. All of my experience is firmly rooted on FR4 So I'll just have to plead a dearth of possibly relevant experience and a distinct cultural deficit - LMAO

 

I still don't quite understand the application involved here, I'd just use a commercially available IC power amp that covers that frequency.

I also don't think the inductor is large enough unless you're absolutely insistent on the filter being tuned, in that case simple tolerance variations are going to upset the cart too. Even at resonance I'd have more poles in that filter and weigh more heavily on the inductor than the capacitor.

 

Actually the LC circuit is backwards, if the intent was to generate RF. The coil will block RF, and the cap will short it.

If they were reversed it would be different, the cap would pass the RF, and the coil would block it, making it available for whatever. The test point would have no DC what so ever.

Either way, the transistors don't care. It all looks the same to them.