Re: Possible PA low output problem.

Jerry Gaffke
 

As I have said several times already, I'd like to see these boards going out with some basic tools.  

1)  1khz square wave out of a Raduino digital pin to inject into the microphone, with suitable attenuator. 

2)  Wide ranging RF from a spare Si5351 channel, with an attenuator that can give a range of output levels.

3)  An RF voltage probe, such as the one N5ESE described.  That design has been around for most of a century.

4)  A low pass filter to look at DC levels in RF circuits

HFSignals could make money selling those parts in a bag for a dollar as an option with the Bitx40, to be dead bugged into the ground plane somewhere.  It's a dozen resistors and caps and a schottky signal diode.  Ideally, the board someday gets laid out to have those parts always in place, so we can just tell folks having trouble what gets shorted to what and where to measure with the rf probe.  Some minor adjustments to the Raduino code would be needed to get the two signal generators up, they would normally be shut down to avoid crosstalk.

Perhaps if I ever get a Bitx40, I'll get some of that going when I have some time.  And write up a debug procedure.

Could be that R136 between Q13 and Q14 in the final got left set too low (or has a bit of dirt on the wiper), that determines the drive level into the IRF510.  And twiddling that pot is a lot safer than playing with 20v into the IRF510.  But remember where it was set, it was more than likely set there for a reason.

This post has a simple oscillator that VE7BEE successully used for transmitting CW.  Perhaps build that up and wire it to the mic input.  Should give 6 or 7 watts out the final.    https://groups.io/g/BITX20/message/20638 

From my simulation, it looks like you should get full power out to the dummy load if you inject around 0.1 volts pk-to-pk at 7.2mhz into the base of Q13 through a series 0.1uF cap, and with the drive level at R136 set to full throttle.  That and the numbers in my LTSpice simulation post should allow somebody with a scope and a 7.2mhz signal generator some idea what an amp is to figure out if your final is working properly.  

But it would take an hour to figure this out if we had a few extra R's and C's and a bit of Raduino code up.  Not a week and $1000 worth of lab gear.  And so long as I'm on this rant, will reiterate that I'd like to see pads for a prescaler and an AD8307 on the main board (and parts available in an optional bag), so we could include an freq counter and a Sweeperino and the ability to look at low level RF in that tool kit.  Oh, and that capacitance meter of post 19453, required no additional hardware.  Measuring voltage on a Raduino ADC pin would also be very useful in the field.  This is aimed at budget minded experimenters, not appliance operators.  Nothing is quite so discouraging to some green squirt as to be told "Oh, it's easy, read this book, you'll figure it out.". Sometimes it isn't easy, especially if all you have is a voltmeter, and your dad works sacking grain.  Some of us can still remember having been there. 

Jerry, KE7ER 






On Sat, Feb 4, 2017 at 10:11 am, John Smith wrote:

Aside from trying a different electret mic, how could I test the audio input circuitry?

 

Join BITX20@groups.io to automatically receive all group messages.