80m Tx Design

[NZ0I]

Jerry Boyd discovered a serious problem with my 80m transmitter circuit design in schematic Rev B.22 and earlier: it won't work at all. Jerry also proposed a design that should work, and will prevent the final amplifier FET from being left in its conducting state with no signal applied: a state that would result in damaged components, or at least a lot of heat and a power supply shutdown due to excessive current.

According to the Si5351 specification, the signals the Si5351 puts out should have an amplitude of about 1.3V (2.6V p-p). Jerry's proposed revision to the 80m transmitter didn't provide as much output power as desired with that input signal amplitude when I tried modeling it. But with a few tweaks to Jerry's design, a "hybrid" design models very well. Below is a SPICE (TINA) model for the hybrid design showing the results for a 1.3V input signal providing 4W of output power at 3.6 MHz into a 50-ohm load, using a 12V power supply, with a nice looking sine wave. The current-measuring resistor has about 40mV across it, indicating that the final amplifier is consuming about 4.8W of power, making it about 80% efficient.

The hybrid design, like Jerry's, should prevent the FET from being left in a conducting state when no signal is applied.

The hybrid design uses four transistors. Jerry's proposed design uses five transistors, but Jerry's design has been tested and shown to work in actual hardware - though not with the Si5351 oscillator. Others on the list: please feel free to chime in with any suggestions or observations you might have.

The image below shows the output when the oscillator's signal has been removed. The scope shows a flat line for the output signal, and a zero voltage (middle trace) across the current measurement resistor (R2) indicating zero current being drawn by the power FET.

[Patrick R. Sears]

Bummer.  But hey, that's how we learn : )

Charles, the blue labels you added to the sub-circuits in the schematic on the Google drive are really handy for me.

I looked at the screenshots and schematic but ran our of time.

About the drive of the Si5351.  I could probably dig up the experiments I did to get the 80m transmitter to work.   There are screenshots of spice simulations and actual experiments.  But I'm not sure how useful it would be because I was still learning pretty basic stuff; e.g. biasing transistors and input capacitance on FETs.  And I've switched gears to contacting schools to make sure the word is getting out for Apr-8.  So I may not get to it for a couple days.  The cards that get distributed to the schools are being printed and cut tomorrow morning - Kelly has a teacher work day  :  )  so we'll be at the school early in the morning so we're not in anyone's way.

Best,

Patrick

--
You received this message because you are subscribed to the Google Groups "Receiver Development Platform" group.
To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/f0c2fdad-b130-44a4-a31a-02a7f16928f6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Patrick R. Sears]

Hi Charles,

About the Si5351 LT-spice vs experiment:  I just found the attached info.  It's a couple pages from tests I did while building the last transmitter.  The oscilloscope traces shows the output of the Si5351 at 3.5 MHz while it's driving 2n2222 and of the output of the 2n2222 before any of the rest of the downstream circuit is added.  I don't have the schematic with me, but the circuit is basically the current 80m transmitter with all the components up to the 2n2222 soldered on. 

Anyway, I'm not sure this is useful for comparing with LT-spice models of the si5351 drive but I happen to have it with me.  It's amazing how much these small USB sticks now hold : )

Best,

Patrick

On 03/06/2017 05:00 PM, NZ0I wrote:

[Charles Scharlau]

Thanks for the historical information, and oscilloscope trace. Below is a SPICE model with a simulated oscilloscope trace of the Si5351 clock output (top) and the transmitter output (bottom). The Si5351 is being modeled as a square wave signal source with amplitude of 1.3V and an internal resistance of 50 ohms. The result is a distorted square wave, not too dissimilar to your oscilloscope trace. It would be interesting to model your 2N2222 circuit and see if the simulate signal match is even closer. It is likely that the load that your Si5351 is driving differs in impedance from what is being simulated here, and the 50-ohm internal resistance assumption is probably not correct. But I am happy with the results being as similar as they are.

​

To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development-platform+unsub...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/f0c2fdad-b130-44a4-a31a-02a7f16928f6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Receiver Development Platform" group.

To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development-platform+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/a0669acc-cf74-5636-5771-58adbb718d87%40gmail.com.