On Tue, 30 Nov 2004 08:47:38 GMT, Stewart Baker wrote: > On Tue, 30 Nov 2004 07:59:31 +0000, Andy Pevy wrote: >> Hi Stewart >> Try Rather than directly connecting the decoupling cap to the 5v supply, try >> feeding it via a 100 ohm (or so) resistor. at 1ma you will only drop .1 v >> but you have created a low pass filter for the noise. >> Rgds >> Andy Pevy >> G4XYW >> A future K2 owner (when the company bonus arrives). >> On Tuesday 30 November 2004 06:24, ext Stewart Baker wrote: >>> I think that I have found the problem, but have yet to work out a >>> solution. >>> To combat the lack of TX audio system gain I have a 1 transistor pre-amp >>> mounted on the microphone header plug on the front panel board. This gives >>> about 9dB of gain which provides sufficient audio for the compressor IC >>> and >>> the VOX. The amplifier is supplied from the 5V pin on the header. This >>> rail >>> is the 5A supply which is used by the MCU logic, plus other functions. >>> The amplifier only draws about 1mA, however because of noise on the 5A >>> line, mainly from the LED bar graph I have had to heavily decouple the >>> pre-amp supply to stop the noise from being superimposed on my speech. >>> What I think is happening is that this decoupling is affecting the rise >>> time of the 5A line which is used for the MCU reset amongst other things. >>> I have tried reducing the value of decoupling. This seems to stop the >>> "strange" problem, but I am back with the noise. It is a shame that there >>> is no way to get at the +8V rail on the front panel board as that would be >>> an ideal mic supply. >>> The way things are looking I will have to put the pre-amp outside my K2 >>> and >>> power it separately. The only alternative seems to go back to shouting ! >>> 73 >>> Stewart G3RXQ > Hi Andy, > Yes, that's what I have been trying. The noise is really a pulse as the LEDs > in > the display go on and off. Got up to 1.5k with a 47uF cap, however I now get > the > VOX switching on/off by it's self very occasionally, so I think that the 5A > line > does not appreciate any time constant being hung off it. > Latest thoughts are to use an op-amp instead of a transistor as I believe that > they have a greater immunity to supply noise than a simple common emitter > transistor amplifier. That should mean even less RC. > 73 > Stewart G3RXQ
_______________________________________________ Elecraft mailing list Post to: [email protected] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com
