KitsAndParts.Com Forum

Brent,
My first inclination is to find out whether the buffer is working properly. Take it to work and hook it up to the same equipment you did before to test the amplifier. Put a small signal to the input of the buffer and connect the output to your spectrum analyzer and see what you see. Vary the pot and see the level change directly on the analyzer.  The buffer input impedance might not be a great match with the output of the signal generator but you should still see if you're in the ball park. You could also skip the signal generator and use the DDS to drive the buffer. *Personally, I like to know how each part works on its own. Once you know that each part is doing its job correctly, it's easier to troubleshoot them collectively.
On the N3ZI DDS, how do you get your TX frequency? In other words, if you set up the DDS with an IF offset of 11MHz for the receiver, how do you get the 20M signal for the transmitter? Are you using the RIT setting?

Tom, ak2b

*One of the problems of early homebrewing is confidence. You get more of this with each circuit you build. This is one of the reasons I was attracted to Diz's little modules in the first place. Getting each board to work and knowing how they behave indvidually, makes it easier to know what is supposed to happen when they are all connected together to make something, like a receiver.

Tried the buffer and amp again today at work. Didn't get time to try the Dds, that's Mondays job, but the amp looks great driven at 0 dbm from the Generator, with the second harmonic >30db down. No low pass filter. With the buffer, things are still reasonably good, as long as I don't overdrive things. -10 dbm seems good for the buffer input. It took me a while to realize that one of our newest signal generators (agilent) has relatively poor harmonic suppression of -25 db, and what I thought where amp problems where actually being caused by the generator. More testing required, but I really wonder if I might have to re think my layout...trying to cram too much in a small box. I think I will try mounting the amp and buffer in their own metal box. My main antenna is a 60ft ladder line doublet fed using a z match. The antenna works great on my yaesu, but I'm wondering if I am getting some form of rf in the shack feedback when everytgin is just laid out on the bench. As I mentioned before, I can't get a decent swr to the amp regardless of what I do with the match, and the amp will go into spurious oscillation. Once in this state, I can completely unplug the buffer output th amp keeps oscillating on it's own until I kill th power. No such problems using the dummy load. I read an article somewhere about some of th simple ten tec kit qrp rigs having similar problems when trying to match an end fed wire.

Bseres wrote:

For example, in T1 does it matter which of the red wires connects to which of the green wires?

No.
So far it seems:
1. The transmitter into a dummy load works fine.
2. The buffer/amp driven with -10dBm at its input and 0dBm on its output into the transmitter with a dummy load works fine.
3. Connecting to an antenna causes a tilt condition.
What happens when the entire food chain is driving a dummy load? i.e. The DDS, buffer, transmitter, and LPF into a dummy load.
Can you get this to work and can you adjust power up and down from the buffer or the transmitter (R6) while transmitting a clean signal?

Tom, ak2b

Seems the RF from the antenna setup is getting back into the breadboard circuitry...my best guess.
If so, need to rebuild the project with RF enclosures and RF filtering etc.
...but what do I know...-Diz

Ok it seems I'm finally back on track. I was using a small dip relay on a piece of per board to switch the dds output from the receiver to the buffer in transmit. When I bypass the relay, the problem goes away. I guess the one inch piece of unshielded wire going from the connector to the relay is the problem. Although I shielded the amp and buffer, the dds itself and the relay board were not shielded yet. Good thing I bought a pin diode switch from Diz to replace the relay! I guess I will also have to re organize and build some sort of small shielded cover from pc board for the dds, receiver and diode switch. I think with all of the gain in the buffer, I'm getting the rf equivalent of 'hum'..like when you touch a screwdriver to the input of an audio amp. The good news is I know the project will work..it just needs some more tweaking

Bseres wrote:

It works! Had my first qso using the new rig tonight on 40. I'm beginning to appreciate the challenges of building a multi band rig though. Only problem now is that when I set the drive up for 20 meters, the rig is overdriven and distorts on 80. When set up properly for 80, very low output on 20. Any suggestions? Diz, could I remote the amplifier drive pot to the back panel to allow easier adjustment?

That is a common problem when using a multi-band amp without AGC. The easiest and simplest fix is to remote mount the gain pot (R6). The R6 circuitry was specifically designed for remote operation to handle this problem.
Better approach would be to design an AGC circuit; use a detector diode at the output through a resistor, amplify the signal and then use a MOSFET to control the gain via negative feedback from the diode.
The ELECRAFT K2 has such a circuit to control the gain of the VCO.
73, Diz, W8DIZ

Has anyone been able to get 5 watts out on 30 meters?  I can get 6 watts out on 40 meters, but only 3 watts on 30 meters.  I finally purchased the kitsandparts 30 meter low pass filter.  That got me up to the 3 watts out.  I was at 2 watts before that with my homebrew filter.  This is with the 5 watt cw amp.

I made my first contact on 80 meters and it was 5 watts out to a G5RV.  I got a reply instantly when I responded to a CQ.  He was about 200 miles away and gave me a 579.  He said I had a good, clear signal with no odd sounds or anything.  I wish 80 meters was open for more of the day. I'm extremely pleased with the kitsandparts 5 watt cw amp.  I have made plenty of qso's on it. It's set up for 4 bands now with great results. I'm using alligator clips to change low pass filters for the different bands.  I have a picture of my rig on QRZ.com.    Brock AE4GM

Thank goodness I had no part in building my wife. Ha. Brock.

ae4gm2003@gmail.com wrote:

I have had some SA612 mixers go bad in my TenTec T1056 receiver running them off the N3ZI DDS VFO buffer.  I think it is too much voltage.  I tried putting a dB pad out of the buffer with a comedy of errors.  Finally I put a 20k ohm pot out of the buffer, set it to max, and hooked it to pin 6 after bending it out from the socket, and it seems pretty happy there.  My Accuprobe measures 43 millivolts with it set at 20 k ohms and it pegs my DVM in the millivolt range with it turned down to 0 ohms.

The Oscillator injection to the 602/612 should be between 250 and 300 mV peak-to-peak
73, Diz, W8DIIZ

The Gain Control is not at RF potential so can be quite a distance...1 to 2 feet no problem - Diz

Hello Brock,
The gain control wires have nothing to do with output...just the pot that connects to them
It sets a current trought the PIN Diode that controls gain of Q1.
As for 5 watts on 20 meter, not sure why your not getting full power.
I always use 14 volts to get full power.

You might experiment and optimize the value of C7 for 20 meters...
increase to 82 or 100 pF or decrease to 47 or 56 pF

If you have the caps available, remove the 68 pF cap (C7)
then power up and touch different vaues to C7 location while observing output.

-Diz