|
Paul
In addition to Ward's comments ...
To be more consistent with the original design (to avoid the "I built it just like ... except" syndrome) move the driver board close to the FETs to shorten the gate twisted pairs. Resistors / diodes should be right at the gates.
HP3325A ... set frequency x2, square wave, amplitude 5 volts, dc offset 2.5 volts. Have used the HP3325A here for countless numbers of years.
51 ohm resistor is there to properly terminate the end of the coaxial cable connecting the HP3325A (or other exciter) to the amplifier.
On the gate waveforms, what's going on above the FET turn on / off voltage (several volts) is of little concern unless there's significant 'anomalous action'. Some rounding above the turn on voltage is normal.
On the drain waveforms, what's going on above zero volts on the waveform is of less concern than what the waveform is doing when at zero volts (FET conducting). During conduction (zero volts) the waveform should go to zero and stay there for the duration. During the non conduction period (voltage higher than zero) the peak of the ringing should be well below the FET breakdown voltage. Chris's waveforms look pretty normal.
The prototypes of the doubler and amplifier were all run with no shielding for years. In fact, even after building the packaged amplifier I continued to run the 'out in the open' prototypes rather than reconfigure the station.
Always start out on the dummy load. Test in stages. Test the driver into the FETs with the FET drain supply off. Verify the waveforms with a dual trace scope to see that there's a small 'dead zone' between the two phases. This prevents the possibility of cross conduction. Once the gate waveforms are confirmed to be correct, turn on the FET drain supply. Once the amplifier is verified to be working properly add the doubler.
Gotta run ... more later if necessary.
Jay W1VD
----- Original Message -----
Sent: 4/8/2018 4:57:26 AM
Subject: Re: LF: W1VD amp help
Paul,
I may have missed some of the previous posts, but here's my $0.02 worth. When I built my first version of the amp, I set the X2 option on the U3S and drove the amp directly from the CLK output on the U3S synth. Everything worked fine. On the next one, I had all sorts of weirdness until I removed the 50 ohm resistor on the input. I assumed the U3S at 3.3v couldn't provide enough drive across the resistor. Since then, I have left the resistor out of every amp driven by the U3S, with no issues. A level translator would probably be better, but I have thousands of hours on a couple of these amps with no problems.
Ward K7PO
-----Original Message----- From: JD < [email protected]> To: rsgb_lf_group < [email protected]> Sent: Sat, Apr 7, 2018 10:37 pm Subject: Re: LF: W1VD amp help The 74F74 is an edge-triggered flip-flop. so the correct input for the main board is a square wave at TTL voltage levels, with fast rise and fall times, but with a current requirement greater than TTL because of the 51 ohm load resistor. As for the oscillation when the doubler is used, the issue may be whether there's enough shielding. The phase detector input of the 74HCT4046 that provides signal conditioning is very sensitive...maybe not enough to break into oscillation spontaneously, but perhaps enough to sustain it once a feedback path has been established with actual RF through the power stage. 73 John
| 