#### Re: QCX+ mini 2W o/p

Hans Summers

Hi all

There's one more thing, not related to inductance.

The MPS751 transistor (Q6) hFE varies wildly from one batch to another. The ON Semi datasheet specifies a minimum of 75, with no typical or maximum values. I have never measured one as low as 75, they usually measure several hundred.

BUT - if you did get one that had a rather low hFE, then what would happen is that the base-emitter current, through the 1.2K resistor R42, would not be enough to saturate the transistor in the "ON" state. Collector-Emitter current is hFE multiplied by Base-Emitter current. So you can see that if hFE is low, and the base-emitter current is limited by the 1.2K resistor, R42, then the Collector-Emitter current maximum amount may be not as much as the PA wants to draw. In that case the voltage at the collector will be lower than it should be, and power output will be too low.

How to check this:

Measure the voltage at the emitter of Q6, and measure the collector voltage, during key-down. The difference between them should be about 0.7V. If it is more than that (for example collector 12V, emitter 10.5V, so a 1.5V difference), that is an indication that the transistor is not saturated, the collector-emitter current limit is being too low because the hFE is on the low side (but still within spec for the MPS751 according to its datasheet).

The resolution to that is to decrease the value of R42. The easiest way to do this is to put another resistor in parallel.
10K in parallel with 1.2K gives 1071 ohms
4.7K paralleled with 1.2K gives 956 ohms
3.3K paralleled with 1.2K gives 880 ohms
2.7K paralleled with 1.2K gives 830 ohms

You should find that the difference between the voltage at the emitter and the collector drops to around 0.7V. On the rare occasions we have seen this issue here, caused by a lower-than-average hFE of the MPS751 Q6 transistor, 10K or 4.7K in parallel fixes it always.

It's perhaps not an ideal situation as the parameters of the rise/fall shape of the RF envelope are also affected; but this is an analog radio, you don't have the luxury of a real raised cosine key-shaping etc., so the answer is to live with it; and if you really want to investigate and attain perfection then be prepared to measure a bit and adjust a few component values.

On Tue, Jun 21, 2022 at 1:20 AM Curt wb8yyy via groups.io <wb8yyy=yahoo.com@groups.io> wrote:
Paul

Sorry you had previous issues, but with 2.5 watt consistent output rig may be healthy.

My 20m mini started out with similar output. I had to remove turns on the inductors. Try removing one or two turns on the PA side of the LPF to see if power goes up some. I reduced turns on both outside inductors to get its power up above 4.5 watts.

Curt

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