[Amps] Some thoughts on a transistorized legal limit

[kg7hf at comcast.net]

Hi Marv,

When I saw your post, I also thought of using class E as well, since you have to restore anyway.  More efficiency = smaller heatsinks, smaller powersupply, less expensive parts.

But my guess is that engineerig costs to produce a bullet proof militarized amp such as this in a commercial venture would still exceed the $1/watt goal.

Paul

> From: Radio WC6W 
> Subject: [Amps] Some thoughts on a transistorized legal limit 
> amplifier... 
> To: amps at contesting.com 
> Message-ID: <924973.94241.qm at web63908.mail.re1.yahoo.com> 
> Content-Type: text/plain; charset=iso-8859-1 
> 
> In order to meet the price, weight and broadband 
> requirements discussed herein the past few days, I 
> offer the following approach: 
> 
> Start with the 1500 watt peak output amplifier 
> described in the old APT application note, APT9701, 
> which posits 25?/watt for a Class C amplifier section 
> with 70% efficiency at 27 MHz (ISM band, not CB!). Add 
> some ferrite to the transformers for wideband 
> operation. 
> 
> Connect a double pi-net (3C's & 2L's) to the output 
> of the above amplifier for both impedance matching and 
> harmonic suppression -- No filter bank required. 
> Switching the pi-net into 7 or 8 band segments (the 
> L's and one of the C's) plus padders for 160 should be 
> about right. These components, including the 
> switch(es), will have to handle some high currents 
> but, only the center C will need to be rated for more 
> than 1KV so, reasonably sized air variables will 
> suffice for the two variable C's. 
> 
> Employ a power factor correcting (PFC) front end and 
> a single ended pulse width modulated (PWM) regulator 
> for the power supply. This will be a direct off line 
> unit with AC isolation provided by the RF input & 
> output transformers. A line input of 12 amps at 
> 120VAC should be sufficient for 1500W PEP (750W 
> average) out. 
> 
> The heaviest part of the unit will be the heatsink & 
> fans which will be required to remove about 600 watts 
> of heat, figuring an overall wall socket to antenna 
> efficiency circa 55%. 
> 
> Now, recall this started with a class C amplifier, 
> so a bit of inexpensive Digital Signal Processing 
> (DSP) will necessarily be inserted between the mike 
> and the exciter to provide for envelope restoration 
> (linearization) of the amplifier via the PWM 
> modulator. The exciter will only be required to 
> provide around 30 watts drive for this scheme to 
> function. Note that the exciter must be operated 
> without compression, limiting, or ALC in this system. 
> Those functions may be included in the DSP section. 
> 
> The above numbers are figured for SSB operation. AM 
> could be implemented by setting a CW carrier at 375 
> watts output (which will coincidently be the SSB tune 
> point) and directly modulating the final with the PWM 
> regulator. For CW/FM/FSK, the capacity of the 
> heatsinking and the PFC section would have to be 
> doubled. 
> 
> 73 & Good morning, 
> Marv WC6W 
> 
> http://wc6w.50webs.com/ 
> 
>