An Awesome RF Speech Clipper for the FT817

 

I operate QRP phone with a Yaesu FT817 and a 9.6 volt homebrew battery pack.  I’ve been wanting to increase its on-air effectiveness and remain QRP portable with a system I can back country camp with.  My antenna for these forays is a simple 80m dipole made with 26 awg wire and I tote a Z-11 auto tuner for HF all-band use.  I was looking for a speech processor to raise my effectiveness without undue size, weight, or appetite for my all precious current.

 

After searching out information and the experience of others, I decided that the RF Clipping method of speech processor was the best way to go to get the most bang for the buck.  See my article Speech Processing in Ham Radio for some detail as to why I made this choice.  It should be in the same folder where you found this review

 

I was prepared to put one together myself, but was delighted to find a source in Europe that had already done so.  I’ve been down the design and build road many a time and this, I can tell you, saved me a great deal of time and effort.

 

The Holy Grail!

I learned, in fact, that a new model had just been developed and (get this) it fits inside the designer’s own FT817’s stock microphone, drawing only 11ma from the mic’s 5v line!

No increase in size.  No more weight.  No extra piece of equipment.  No additional cables or connectors to forget.  No battery burden to speak of.  What more could I want?  I ordered one immediately!! 

 

It took between 3 and 4 weeks for my bank draft to arrive and then receive the little blessing in the mail.  It seemed to take forever!

 

The unit comes built and tested, is about 1” x 1 ½” x ½” size, uses a 0.1” matrix board with a combination of surface mount and conventional components, The 3 ICs are DIPs and comprise two mixer stages and a dual OP amp. No RF problems have been experienced.

 

A 453Khz ceramic resonator that can be trimmed to match the operator’s voice spectrum controls a local oscillator.  An electret mic element is fed through a low noise preamp and mixed up to dsb at 455khz.  A ceramic filter extracts only the upper sideband which then gets amplified. Two back-to-back diodes do the clipping at RF.  A second ceramic filter suppresses the myriad of unwanted frequencies and produces a “nice” ssb signal.  The second mixer demodulates this down to the audio base band again.  This compressed audio, bereft of harmonics is then low pass filtered and a potentiometer restores it to mic levels ready for the rig to inhale

 

While originally designed for the stock MH-31 mic (the dynamic mic module is removed to create lots of space for the electret and clipper), I wanted to use my MH-36 DTMF mic.  There’s a lot less room in the DTMF mic due to the DTMF board in there.  That plus the fact that it uses a tiny electret element that wouldn’t make room for a housefly.  After some experimentation, I was able to find the correct connection points and just managed to shoehorn the clipper into the housing. I even installed a tiny switch to make A-B tests easy to do.

 

The Bottom Line!

Initial checks done, I plugged it into the FT817 fitted with a 50ohm dummy load.  It seemed to behave as expected.  In PTT mode, I could plainly see the power meter flicking up to full scale but remaining mostly at the low end during my transmissions.  Once I cut in the clipper, the modulation was nearly always varying between 75% and full scale!  I took the whole works out to the shack and hooked it up to my multi-band GAP vertical.  I checked into the 40m Trans Provincial Net and asked for signal and audio reports with the stock setup.  I had to repeat my call to get on the list and was told I was “barely readable” above the noise level by just about everyone.  When my turn came around I announced that I was QRP and I got a 3x3 report.  I flipped the clipper switch.  “Whoa, that’s much better” came back. I confessed to using the RF Clipper and asked for an audio quality report as well.  “Excellent audio! Q5 copy” was the response.  One fellow in Nova Scotia agreed to go up to 20m with me and conduct more tests.  To quote Yogi Berra, “It was déjà vu all over again”.  I was 3x3, barely copyable in the stock mode and this guy gave me a 5x7 in clipper mode!  This has been typical of the on-air performance with the RF clipper.

 

I did have a problem in that the VOX wouldn’t transmit in that mode.  I suspected a slight change in circuit topology would correct this, so I contacted the designer.  He concurred, but said he’d had some trouble with his FT817 staying keyed all the time in VOX and had been in a hurry to try the clipper out on the air.  He’d disabled VOX from working and hadn’t gone on to delve into the cause.  A friend of his had removed the mod and reported that his FT817 then worked fine on VOX??  I did my little mod and set the VOX gain lower on the FT817 and VOX works like a charm now.

 

35 Watts out of an FT817. You must be mad!

Now, 5 watts is 5 watts and no more!  The difference is that during voice transmissions, I’m now putting out 5 watts MOST of the time the key’s down rather than for brief moments.

 

The dynamic range of speech (the range between lowest and highest volumes as we speak normally) for most of us is in the order of 40db.  Even a skilled operator aware of this phenomenon can rarely achieve a dynamic range of less than 6db.  In ssb, these range figures exactly match the range of our RF power out.

 

Let’s say we’re not experts but fairly experienced and can deliver audio into a mic at a range of only 9db and we operate a 5 watt QRP rig.  With a mean output at 9db down from the full 5 watts, we’ll be sending out peaks of 5 watts, but only 625 milliwatts mean power.  This same calculation has a 40 watt transmitter putting out a mean power of only 5 watts for the same reasons.  Using an RF speech clipper at a conservative 85% or so clipping factor is like having an FT817 with a 35 watt linear amp attached.  Except you are able to do it with only 5 watts!!

 

Conclusion

In summary, this little RF Speech Clipper has been no less than fantastic in increasing my readability, especially in high QRM/QRN conditions.  I no longer use the switch, unless someone asks me to demonstrate the difference.  It’s on all the time and operates as if it was simply part of the rig.  I’m ordering another one to use with in my desk mic.

 

For QRP phone work, it’s the best thing since sliced linears!!            JJJJJ

 

 

72 de VE3TNW, Tony    bravhart@cogeco.ca