Down East Microwave - 144-28 Transverter Repair - Helping out someone else.....

I've been very busy at work recently and so not had much time for the 811 or any of my other projects 

BUT

When I was on 40M a few weeks ago I had a really great QSO with someone that had a 144 to 28Mhz  transverter that needed repair, and I talked about some of the repairs that I do, he had a look at this blog and my QRZ.com page and well ........ I ended up offering to take a look (with no guarantee's though !!!)

When it arrived a couple of weeks  it was a top end (Down East Microwave) transverter with a 28Mhz IF, low level input (0dBm). The report was that it didn't transmit. It did a manual with it, although when I opened it up it didn't seem to match the actual layout of the board although the circuit looked pretty close !

First things first, a close visual inspection - nothing to note  here no burnt components (to the naked eye), no overloading by the look of it. The owner did  say that he thought it might have had a near miss with lightening (!!) and so I had expected the worse, but no.

I was very impressed with the build quality, absolutely superb :) here's a picture of it on the bench ;

So i powered it up (via a current limited supply) - checked the Local Oscillator, right frequency - has a wonderful temperature stabilised arrangement - worth looking at the circuit really good for frequency stability, warms the crystal up and keeps it there - very clever, i've seen it in professional kit but never in a transverter before !

Got a suitable IF transceiver (in this case a ten tec as it has a nice transverter output at the right level and does key output for the ptt) connected it up and...................

RX worked fine - injected a -127dBm 2M signal from my test set and this was easily detectable in the IF transceiver so no problems there - great news !

Checked the input level from the transverter correct at 0dBM (224mV into 50R), accroding to the manual this should give more than 10W out on 2M from the transverter.

Key down on CW (as per manual suggested setup) - nothing out into the Bird ThruLine :(
well after putting it onto the test set it was actually giving out ~300mW !!

PA block failure perhaps ?

So worked my way down (or is it up ?) the tx chain, after swtiching the 28Mhz IF signal goes through a MAR 3 MMIC amplifier to take it up from 0dBm and yes 0dBm in but -6dBm out - so the amplifier is an attenuator  :) might be a problem !

Check the voltage supply in the area / biasing and all looked reasonable - so decided to bypass the amplifier effectively making it into a higher input power transverter, injected 1W from an FT817 on 28Mhz and w and behold ~20W out on the Bird.

So a fault MAR3 MMIC, after finding the right manual it said this should be a MAR 6 anyway which was a little confusing, the owner didn't know why and had never changed it so perhaps it was built this way ?

So after chatting with the owner via email, I ordered some MAR 3 and MAR 6 MMIC's (always good to have a small stock and I didn't have any !!). 

Arrived last Friday, quick change (photo of new one installed below) a new MAR 6 was installed as per the circuit diagram (the scratches where its mounted were already there, this is where the track was cut to insert the option MMIC in the TX chain to allow lower drive levels, its detailed in the manual);

For those readers who don't know what a MMIC looks like its the small black dot with 4 legs in a star arrangement above the orange disc ceramic capacitor :)

Fired up the Ten Tec as the IF on 28Mhz and the result is shown below ;

The meter on the left is showing the RF input to the PA Hybrid at 1.6V which is 50mW (what the manual says it should be) and the output on the Bird is showing 35W - this is using CW as per the manual suggests. I set the input control (the lower blue variable resistor on the MAR6 photo)  to give the 50mW input required.

So all fixed and working, gives approx 23W on speech peeks, and the output is nice and clean - see below (apologies for the poor photo!)  each graticule block is 50Mhz and the large spike (!) is of course the 144Mhz signal.

After a few more email conversations with the owner, his IF transceiver gave a little more output than mine +10dBm (100mW) and I wanted to check that the input into the PA block hybrid didn't overload anything and so I constructed a 20dB attenuator and used a 1W input into this from an FT-817 to give 100mW and then adjusted the input resistor pot (as above) to give the right input level of 50mW.

I then gave it a really good test over several hours at that level with simulated long overs etc with breaks etc, the unit get warm but not overly so, and so I think this is fixed - result.

Now Ii've just got to post it back to the owner - hope he likes it - its certainly working better now than before !!!!