boosting wideband a lab signal generator to drive a class C stage

[Opcom]

I am sure others will have an interest in this too.

Having a 50Khz to 80Mhz lab type signal generator like the Logimetrics Signalock 925, with an output of up to +20dBm into 50 Ohms (3V) as a VFO would be really nice. It's rack mount and pretty stable. Enough for me.

The frequent alternative in many people's hobby shops is the old H/P 8640B 0-512Mhz generator.

Most VFOs have 3W or less, some much less - like these generators.

I want to remove the old Kenwood TS430-S and have a dedicated VFO/driver setup. Space is not a real concern, the TS430-S and a little speaker takes up about the same room in the rack as an 8640B or the Logimetrics 925.

The question is - how to boost the output of this sort of instrument, without a bunch of annoying tuning, to a respectable level such as 25W, or even to 100W maximum, so that there is no problem to drive a class C grid like that of a 4-1000A?

The idea is to just set the frequency and attenuator on the generator, and then tune the class C grid and plate, and go!
The TS430-S can do this, no tuning needed. Why not a generator and amplifier?

It seems obvious that the booster would maybe have to be push pull, first there would need to be a low power class A stage followed by a 100W PEP stage such as that in a radio like the TS430-S or alike.

Think of it as one more step toward 'modernizing' or 'integrating' the station and not 'wasting' a transceiver to do just driver duty.. Good idea!

Has anyone successfully done this kind of thing or could share about it? Anyone taken the guts of an old HF rig to do this kind of thing?

[KX5JT]

Or you could spend like 40 bucks for a DDS VFO and dial in a source signal, same as your idea with the generator but this would be compact and dedicated to do the job.  Most of them put out like  1 to 1.5 volt signal so if you need more than that you could always build a class C amp of somesort, use a tube or fet or whatever....

I think these are pretty neat in theory...

http://www.ebay.com/itm/led-Digital-AD9850-0-55MHz-DDS-Signal-Generator-For-HAM-Radio-VFO-SSB-RIT-/171122637704?pt=LH_DefaultDomain_0&hash=item27d7b43b88


I'm interested to see what answers may come for you Patrick.  I imagine it would be similar to using the dds vfo to drive ... oh let's say.. in my case a Viking I or II.

[K1JJ]

Hi Pat,

I used one of those Chinese DDS boards as the VFO. You could use your sig gen instead.  The DDS drives a 3 watt solid state 160-40M RF board from a Greek company.  This drives a kit CCI pair of  MRF-150s. I biased it for class C and it drives the 4-1000A rig FB.

Your sig gen is probably a clean driver, but check it with a spec analyzer to be sure.

BTW, my Chinese DDS alone, was not clean enuff. It puts out spurs down maybe  -40 dB or so within 100 KHz of the fundamental.  However, once it goes thru three tuned circuits - the 4-1000A input matcher, the grid L/C, AND the final tank circuit, the overall Q and attenuation is plenty enough to drop the spurs down below -80dB on the air.

VFO cleanliness is a major concern since at -80 dB down, if on a quiet afternoon on 75M we run a KW and are +60 over S9, the spurs will be S6.    So, the overall rig's spur at -80dB or better is a requirement. Anything less and we may create noticeable trash up the band.
  
T

[RolandSWL]

Do you have a link for the 3 watt amp board?

Thanks, RSWL

[WZ5Q]

Has anyone successfully done this kind of thing or could share about it? Anyone taken the guts of an old HF rig to do this kind of thing?

Howdy Y’all,
I’ve done something along the lines of that Patrick. I know you wanted a broad banded solution so you wouldn’t have to tune the IPA, but this was the way I did it.

I’m using a T-368 VFO for an Exciter on the 160 – 10m Heavy Metal Homebrew that I’m building. The RF Final is a single 3-1000H Low Mu Triode in deep Class C. Needless to say, I will need an IPA to feed that Power Hungry Beasty.

I needed an IPA that I could supply an excess of drive for the power hungry Triode, along with tunable inputs and outputs to facilitate frequency doubling to get the required drive for 15 and 10m. After some thinking on this matter, it occurred to me that a PA section from a ricebox Kenwood TS-830s would give me all the parts I needed. It uses a pair of 6146 in parallel, has an output tank, separate input grid tuning for 160 – 10m, a power supply, etc.

I hacked up a TS-830 liberating all of this stuff, bread boarded and got it working on the bench. Rewinding the 15m and 10m grid input tuning coils and tweaking the capacitance, I was able to make a frequency doubler thereby being able to use the T-3 as an Exciter for those bands as well. The T-3 Exciter (using the 6000 tube) is providing enough grid drive (with a small reserve) to the 6146’s to be able to operate them in Class C giving me around 150watts output from 160 thru 40m. I am right at the hairy edge of enough grid drive on 20m for 150 watts output and really need to coax a little more drive from the T-3.

Working as a Frequency Doubler is FB for 15m and 10m with about 75watts output all the way across those bands. I am getting a little Red Plating on one of the 6146’s on 10m but suspect it is a funky tube. I haven’t been back to the bench to iron out these small bugs yet as a big QTH Move came around and all of my time has seemed to disappear...as well as the bench...

Take Care,

[K1JJ]

Do you have a link for the 3 watt amp board?

Thanks, RSWL

Hi Roland,

It's Voscos Electronics in Greece. Here is the web ordering info.  I just realized that the board is good up to only 10 Mhz, so it's a 160 - 40M deal.  My 4X1 rig is designed for that range, so FB. I use linear amps above 40M.

http://users.otenet.gr/~nsavvas/linear_en.html

The boards with the double heat sinks are the 3w boards.

T

[N2DTS]

I built a vfo/exciter last year, coil+cap vfo buffer, 6146 output.
40 meters only, and I want to build one for 80 and 40 meters.
I have a pto I can use, outputs on 160 meters, so I need to double and triple it to 80 and 40 meters.
The tricky part is getting the frequency counter display to work.
A dds vfo kit would be a lot better, but the dds stuff can put out hash.
A dds vfo and a no tune ips would be great, but might as well just use a cheap rice box, an icom 735 is small and works better then anything I am going to build.

I was thinking it would be cool to have TWO output stages, one tuned for 80 meters, one for 40 meters, so I could pick a band and not have to tune anything, the the frequency readout is the hard part.

On the 40 meter exciter, the vfo runs on 80 all the time, the digital readout reads twice the frequency all the time and the output stage is a doubler and is keyed.

I might just build a vfo/exciter for 80 meters, single band exciters...
The 40 meter one does seem to be very stable, no band switching to stir things up.
Makes a lot more sense to use a rice box as an exciter....

[Opcom]

For using a frequency counter on a VFO and followed by multipliers, for example,a doubler; a way is to double the counter's gate time so it counts double, etc.

What is tricky is adding or subtracting a figure to a counter when using a VFO or PTO with a transceiver that adds or subtracts an IF to/from the original 'crystal' TX or RX frequency.

either way the very simple oldest many-IC counters are best for those mods.

[Opcom]

I found this, not sure is is what I want to pay but is another source of RF amps.
http://www.italab.it/prodotti_uk.php?cat=3&scat=10

There are a lot of big amp pallets on the internet. There seem to be few reasonably priced broadband amps of the kind that take the 10-100mW range and boost it to 5-10W, to drive those bigger amps. I have not searched the -entire- internet..

[W7TFO]

Here is a little HB tube amp all worked out to do just that job :

http://faculty.frostburg.edu/phys/latta/ee/6ag7amp/6ag7amp.html

73DG