13284	2016-11-09 07:21:51	Ravi Miranda	Automatic bias for a MOSFET power amp and power up
Message	Date	From	Subject
Hi All First of all, thanks to everyone who responded with good advice on the relay power question I raised a few days ago. I now am considering building two sub-circuits for the RF Power Amp which will use 2 x RD16HHF 1. Two additional MOSFETs that will trigger a power-up of the PA board based on a uC voltage (IMHO, this has been used in popular commercial rigs) 2. Automatic bias drive based on a voltage from the UC. Explanation: usually the bias is set either for CW or SSB (not sure which class for which) so I could only use the amp (at least theoretically for a particular mode and NOT both). What if I was to use a device such as digital pot (MAXIM? or TI? other?) that would set the bias voltage based on the signal from the uC. Again this has been used in commercial rigs. Could some learned members please help with the following? 1. Suggest an appropriate circuit for pt. 1 and part numbers? Preferably popular devices (easy to source) 2. Suggest a device that would replace the conventional biasing circuit for MOSFET amplifiers, again easy to source devices please? Best 73 Ravi/M0RVI -- I'm here to add more value to the world than I'm using up.
13285	2016-11-09 08:12:00	Thomas S. Knutsen	Re: Automatic bias for a MOSFET power amp and power up
To start, there is no significant difference between a SSB signal, and a CW signal from the amplifier point of view. A CW signal can be made by injecting a keyed single tone into a SSB transmitter. As such, a class AB amplifier is quite suitable to amplify both. Changing bias point of a amplifier would most of the time change the input impedance somewhat. Thus the benefit of using a single bias for all operating classes is simplified driver design. As for the proposed RD16 amplifer, I don't think a 20-30W amplifier warrants a large design excerice, exept perhaps for the value of the experience, so I would have considered a bias regulator with a P-MOS switch to turn bias off. The whole thing could probably be implemented with 5 components, and most any cheap MOS device (or even bipolar transistor) could be used. It could be interesting taking the whole a bit furter, and control the output with a digital pot. Or perhaps adding in a bit of pre-distortion to improve the linearity of the amplifier. Grebennikov outlines this in may 09 issue of High Frequency Electronics, available free online. As for determining the bias point, I use 2 signal generators with power amplifiers to do 2-tone measurements, and measure the intermodulation products with a spectrum analyzer while watching the current. Usualy there is a point where one find a good trade-off between supression of intermodulation products and bias current. I'd like to reccomend William Sabins book, "Single Sideband Systems and Circuits". This covers a lot of the needed information to design a SSB transceiver. The amplifier chapters are worth a read. 73 de Thomas LA3PNA. 2016-11-09 16:13 GMT+01:00 Ravi Miranda ravimiranda@gmail.com [emrfd] <emrfd@yahoogroups.com>: Hi All First of all, thanks to everyone who responded with good advice on the relay power question I raised a few days ago. I now am considering building two sub-circuits for the RF Power Amp which will use 2 x RD16HHF 1. Two additional MOSFETs that will trigger a power-up of the PA board based on a uC voltage (IMHO, this has been used in popular commercial rigs) 2. Automatic bias drive based on a voltage from the UC. Explanation: usually the bias is set either for CW or SSB (not sure which class for which) so I could only use the amp (at least theoretically for a particular mode and NOT both). What if I was to use a device such as digital pot (MAXIM? or TI? other?) that would set the bias voltage based on the signal from the uC. Again this has been used in commercial rigs. Could some learned members please help with the following? 1. Suggest an appropriate circuit for pt. 1 and part numbers? Preferably popular devices (easy to source) 2. Suggest a device that would replace the conventional biasing circuit for MOSFET amplifiers, again easy to source devices please? Best 73 Ravi/M0RVI -- I'm here to add more value to the world than I'm using up.
13287	2016-11-09 16:13:59	Chris Howard w0ep	Re: Automatic bias for a MOSFET power amp and power up
Is the drive level also a variable in this determination?
13288	2016-11-10 02:48:57	Ravi Miranda	Re: Automatic bias for a MOSFET power amp and power up
Hi Chris Nope, I haven't thought of that at the moment. Just a newbie designer trying to build on things and see what happens, drive level would be fixed with and attenuator set at the input. Is a variable drive something that is easily achievable, I could give that a go? Best 73 Ravi/M0RVI
13289	2016-11-10 15:31:48	Sandeep Lohia	Re: Automatic bias for a MOSFET power amp and power up
Raviji, Class A or push-pull AB can be used to amplify both CW & SSB. Will amplify both positive upper shine & negative lower shine... But class C we don't use with SSB. However Class C more efficient, will provide double the power for FM, CW, & some digital modes... there was a nice article in SM0VPO - Harry's Homebrew Homepage or the same can be find in BL Theraja , ECE first year. What I have learnt practically, change in input RF drive lavel also changes proper biasing threshold value, same goes with VDD on drain. some short of diodes can be used on heatsink as automated biasing control, as to decrease in case if too hot. for example if Class C = O volt DC bias in DVM Class A = approx 2.5 +/- for Fets. AB push-pull even lower biasing say 1.5 will work on same drive lavel... Power BJT R amp controlled on base where as FET voltage controlled...
13290	2016-11-10 15:39:33	Sandeep Lohia	Re: Automatic bias for a MOSFET power amp and power up
Sometimes it starts ringing if things go wrong ( destructive self- oscillation ) Best of luck, & don't forgot to use dummy load across... ;) On Nov 11, 2016 5:00 AM, "Sandeep Lohia" <sandeeplohia12@gmail.com> wrote: Raviji, Class A or push-pull AB can be used to amplify both CW & SSB. Will amplify both positive upper shine & negative lower shine... But class C we don't use with SSB. However Class C more efficient, will provide double the power for FM, CW, & some digital modes... there was a nice article in SM0VPO - Harry's Homebrew Homepage or the same can be find in BL Theraja , ECE first year. What I have learnt practically, change in input RF drive lavel also changes proper biasing threshold value, same goes with VDD on drain. some short of diodes can be used on heatsink as automated biasing control, as to decrease in case if too hot. for example if Class C = O volt DC bias in DVM Class A = approx 2.5 +/- for Fets. AB push-pull even lower biasing say 1.5 will work on same drive lavel... Power BJT R amp controlled on base where as FET voltage controlled...