GW3UEP 630m Transmitter Tweaks

DF8UO's GW3UEP 630m Tx

I recently received an interesting note from Daniel, DF8UO, describing his GW3UEP transmitter modifications:

Hi Steve,

I have seen on your Website, that you are also a MF/LF fan and builder of GW3UEP QTX. I build the version with 2 IRF540 as you proposed and the same version with 2 IRF630. As I bought some time ago a PSU with can be regulated from 1V to 60V and max 10A (it is a Manson HCS 3404) I made some measurements on both version. Find attached an Excel with the measurements I made, if you are interested in.

If you use the IRF540 version you can see that if the power exceeds 100 W, the efficiency goes down, as the IRF540 is only specified for 100V.

Currently I am using the IRF630 version running at 300W without problems.

From Daniel's test data, it looks as though the IRF540's are quite happy as long as you keep the transmitter at 100W or less ... anything higher would certainly benefit from the switch to IRF630's as the high efficiency is maintained at least to the 300W level. Something at this level would no doubt benefit from higher-voltage capacitors and heavier inductors as well as a little more heatsinking, enough to shed at least 25 watts of heat between the two FETs.

Daniel drives his amplifier with an Ultimate3, followed by a 10db attenuator, a squarer and into the totem pole driver stage. His own version uses toroidal inductors rather than the original air wound coils.

A pair of GW3UEP amplifiers using IRF630's at ~40V into a small power combiner, should provide ~550W of useful output at >90% efficiency ... more than enough power to do some serious work on 630m.

630m Resources - Part 2

Using the transverter approach offers the advantage of providing a higher-stability signal generated by the main station's transceiver oscillator system. Most of the digital modes require a higher stability than would be needed for straight CW.

If your main interest is normal speed CW, then a digital VFO, such as the one devised by GW3UEP, would do the job.

If you are content operating on 475.0 kHz, then an inexpensive 7.6MHz crystal, in a divided-down oscillator will work well, and with good enough stability for some QRSS work as my earlier "GW3UEP Transmitter In QRSS Mode" report indicated.


The other option for frequency generation is a DDS.

N3ZI DDS

For several years now, I have used an inexpensive N3ZI DDS to drive my 2200m transmitter and, more recently, my 630m system. The N3ZI DDS is an affordable option should you decide on a non-transverter style of transmitter.

For those wanting something a bit more robust, W1VD describes a very nice high-power transmitter. His website is packed with useful circuits and ideas for both 2200m and 630m.

W1VD Dual-Band Kilowatt : http://www.w1vd.com/

Another practical way of generating moderate power, certainly enough to meet the Canadian 5W EIRP limit, is to combine two or more 100W transmitters, such as the GW3UEP transmitter. It is reasonably simple and inexpensive to combine several modules with the use of a homebrew power combiner, as previously described.

Homebrew 630m two-Port Power Combiner / VE7SL

Of course, none of this would be of any use without thinking about suitable 630m antennas...ones that will fit in your backyard!

GW3UEP 630m Tx Complete

I was finally able to put the finishing touches on the new GW3UEP Class-E transmitter by mounting the hardware on a small 19" panel.

Added to the panel were a 5A DC meter to monitor drain current, a keying jack and a key shorting-switch for keydown testing. This is the third version I have built using the design on Roger's page, with all three performing pretty much as advertised. It is an inexpensive, easily reproducible design.

 

I made one small change in the third version (above) by adding an additional FET in parallel with the single IRF540 thus cutting heating dissipation as described in an earlier blog. At normal CW keying speeds, no heatsinking seems needed on this particular version, when operated on 12VDC. Anything a little higher or any QRSS modes would require heatsinking. I did eventually add two small heatsinks in case I chose to run QRSS at some point.

 

 

When operated at 12-13V, the measured output power is around 25W while at 22-23V, the output is around 70W. I suspect that you could get in excess of 100W out, at higher voltages and larger heatsinking or by blowing the sink with a small fan.

 

The main function for the new transmitter will likely be as a semi-permanent beacon as well as for a small signal source when tuning the antenna. I may also use it in the WSPR mode with a modification to the oscillator section. One particularly nice feature with the Class-E design is that with no load at all, the current drops to almost zero, should the antenna suddenly fail.

 

If you live in VE7 or across the border in W7 land, please have a listen on 475.0 KHz and let me know if you can hear the "VVV" beacon over the next few days!



 

Smoke Testing The GW3UEP 630m Transmitter

Drain (top) vs Gate (lower) on testbed Class-E GW3UEP TX

I've just completed the Muppet-styled printed circuit version of my previously breadboarded GW3UEP 630m transmitter. The earlier version was built "ugly style" in order to optimize part values and measure circuit parameters.

Testbed (Ugly-Style)

The "ugly version" performed well at 12V and during overnight beacon testing was aurally copied as far east as Kansas. Although the final version has yet to be mounted on its small 19" rack panel, along with a meter to monitor final amplifier drain current, all indications show that it too works well.

Final Version (Muppet-Style)

This version, based on the GW3UEP design, has a few small changes, the main one being the addition of a second parallel-connected FET ....described in an earlier blog.

Running the TX at 12.8VDC on the drain(s) at 2.3A produces an input power of 29W. The measured power out, after the LPF, is 23W into a 50 ohm load. This represents an efficiency of 80%. When run in the normal speed CW mode, the FETs run cool enough that they would probably not even need a heatsink but if run in any of the QRSS (long keydown periods) modes, would certainly benefit from  heatsinking.

Running the TX at a higher voltage of 22.6VDC (on the amplifier only) yields a current of 4A for an input of 90W. Measured output power is 71W for an efficiency of 79%.

Heatsinking would be required at this power level, even for normal speed CW but the finals seem to run just slightly warm. A larger heatsink or possibly a small fan as well would be required for any QRSS CW modes.

I suspect that the efficiency could be further improved yet with very fine tweaking of the output circuit L/C network but the extra few watts gained would not be significant.

At either power level, this easy-to-build transmitter would make a great "first 630m transmitter" for anyone wanting to get started on our new band.

630m Trans-Pacific WSPR

Courtesy: https://www.google.com/maps/

It looks like the new 630m band may hold some surprising opportunities once the winter DX season is upon us. A recent posting to the down-under 600m Yahoo Group by David (VK2DDI) in New South Wales, Australia, set off a flurry of excitement when he announced his August 25th (0950Z) reception of the 475kHz beacon signal from WG2XIQ, operated by John (KB5NJD) near Dallas, Texas.

This is particularly noteworthy in view of the relatively low power used for John's beacon....around 200W. With the typical backyard antennas being used at these frequencies, efficiencies are very low and John's actual ERP is less than 5W. The transpacific reception of John's signal by VK2DDI confirms what most LF'ers already know....that small suburban lot amateur installations can have positive results on 630m without the need for huge antenna systems.

The WG2XIQ beacon was operating in the WSPR mode, which has become very popular amongst 630m experimenters as well as those just interested in listening-in. WSPR is not a QSO mode but strictly a one-way 'beacon' mode. Although two stations may each spot each other, it is not considered to be a valid two-way QSO. A check of evening  WSPR activity will often reveal dozens of stations actively spotting what they are hearing.

Like most LF stations, John's is mostly homebrew.

WG2XIQ/KB5NJD

I'll let him describe the details:

"I have a few ways of making RF in the shack. I can do CW with a very nice waveform using the GW3UEP VFO/Driver coupled with a GW3UEP 100w amp with waveform shaping. The other way is via the MF Solutions transmit downconverter, developed by John Molnar, WA3ETD/WG2XKA. I have two of those boards, one is a backup. I use a GPSDO for the LO and use that signal to drive two parallel GW3UEP amps with max power at 125 watts each. The W1VD Ø degree hybrid combiner brings them together in phase for close to somewhere between 200 and 250 watts TPO depending on how hard I drive and how close I match the TX levels entering the combiner. I filter the output with the W1VD KW LPF that was built by Dave Robinson G4FRE (ex WW2R). I power the amps with a pair of BK Precision 30V 6Amp variable power supplies (variable current limit threshold also). Scope match is used to resonate and match the the impedance. IF Rig on 630m is typically a Yaesu FT920. These days RX antennas are the VE7SL multiturn loop or the TX vertical, both of which have their own merits depending on the conditions at the time."

John's 630m Vertical

"Antenna is an 80 foot asymmetric T-top marconi with 100 foot and 200 foot legs....radial system is almost 3 miles of radials connected via various busses. 26 ground rods around the property. I monitor current in the shack and sample via a Bauer current transformer from an AM BC ATU."

630m Radial System

630m Antenna Loading Coil & Variometer

"As far as my system is concerned, I am the poster child for "If I can make it work, anyone can!" 

John's system does indeed work well...just last year at this time, his 630m signals were copied by KL7L near Anchorage, Alaska.

Of course, equal credit must be given to VK2DDI for having a system good enough to hear John's signal all the way down on Berry Mountain, New South Wales, Australia! It is there that David has set up a fine LF station, 500m above and overlooking the Tasman Sea...an ideal location for weak-signal LF work.

VK2DDI - Berry Mountain, NSW

David's receive antenna at the time, feeding and SDR-IQ receiver, was a simple non-resonant 90' vertical wire, with no ground radials and no tuning. It seems that the old real-estate adage, "location, location, location", can also be applied to LF reception!

 

WG2XIQ Signal As Heard in VK

David's screen capture of John's WSPR signal, although very weak, is clearly visible at 09:50 and apparently, strong enough for a solid decode.
David also runs the Berry Mountain Grabber, providing other VK and ZL experimenters a handy way of checking their system progress or propagation conditions.

If you have been doing any WSPR work on HF, you might be surprised at what you can hear down on 630m, even without a dedicated antenna for that band. Surprisingly good results can often be had with a non-resonant antenna as the signal to noise ratio can often be better even though signals may sound weaker. Give it a try and spot what you hear!

If you are interested in learning how to receive WSPR, here is a nice tutorial by ZS6SGM. 

Should you be interested in knowing more about obtaining a Part 5 licence to transmit on 630m, John will happily guide you through the process. He can be contacted via email or you can find him hanging-out most nights on the ON4KST kHz (2000-630m) chat page.

To keep on top of what is happening or who is on-the-air, most LF'ers rely on three sources:

• the RSGB LF Group reflector
• the Lowfer list
• the LWCA Message Board 

Radio amateurs in Canada have had 630m as an amateur band since May of this year but unfortunately are not allowed to contact any of the experimental stations. Hopefully the U.S. will also obtain 630m as a ham band some time soon. In the meantime, a Part 5 licence for any U.S. amateurs would be a good way to be  all set when that day eventually comes!

Parallel FETs

When I built the GW3UEP 630m transmitter, I made one small modification by adding a second FET in parallel with the first. Normally when dealing with RF devices (tubes, RF transistors) in non-switching modes, adding a second device in parallel with the first, always guarantees a significant increase in output power. This is not the case with FETs when parallelled in switch-mode amplifiers.

Second FET added to GW3UEP 630m transmitter

Adding the second FET reminded me of an interesting e-mail that I received last year from Minto, PA3BCA, who uses parallel FETs in all of his LF transmitters.

"Adding (parallelling) FETs in an existing TX configuration will indeed not lead to a large increase in power output. It is easy to explain why. The FET (in class D or E) acts as a simple on/off switch, with a little bit of series resistance (the Rds_on). And parallelling switches will increase the current (and thus Po) only slightly by halving the Rds_on, but this is negligible.

A calculation for instance with my PA: 50 Volts, 600 Watts.
Say the Rds_on of a FET=0.2 Ohm, so for a single FET the dissipated power is 70Watts. Now with two FETS in parallel the Rds_on is only .1 Ohm. Dissipated power is then only half that of the one-FET configuration, ie. 35 Watts. So 35 Watts less loss in the FETs. But even if all of the 35 Watts so gained becomes output, this is only a 5% gain in output (600 + 35 Watts). Not easy to see on amateur class output meters.

The big advantage of parallelling FETs therefore is something else. I did it to protect myself from sloppiness, clumsiness and stupidity. FETs are very easy to parallel. When heating up, the Rds_on increases so dissipated power is nicely distributed over the FETs. By parallelling the FETs, the total dissipated power halves (in this case from 70 to 35 Watts). This means that the heatsink can be smaller or stays cooler. The dissipated power per FET decreases by the number of FETs squared. With two FETs, the dissipated power in a FET is only 1/4 of the single FET configuration. In this case only 17.5 Watts instead of 70 Watts per FET.... 70 Watts dissipation per FET is too much for comfort, there is no headroom and if the temperature gets too high, the FET will die quickly.
And secondly, my PS delivers 20+ Amps when short circuited. A single IRFP360 that is already hot will then certainly die (for instance user error by applying forward bias to the gate). Two IRFP's in parallel can easily handle the 20 Amps.

So user error (no antenna, variometer way off, screwdriver dropping from fingers and connecting the gate to + 12 volts, had all this happen) until now has never resulted in a FET dying.

The flip side is that the Input capacitance and the reverse transfer capacitance also double. That is the reason I added an extra high-current totem pole to drive the FETs. The BC337 / BC227 just were not up to the task. I found it a cheap price to pay! "

In my particular case (30 Watts in / 25.5 Watts out), using a single FET resulted in 4.5 Watts of total dissipation while going to two FETs could result in a total dissipation of just over 2 Watts, or 1 Watt per FET. Even when keydown for long periods of time, the FETs are cold.

While parallelling FETs would certainly be more rewarding when applied to a high power switching amplifier, even small amplifiers can enjoy some benefits of the extra addition.

630m QRSS30 Last Night

Before dismantling the little 25W GW3UEP test board in order to repackage it, I tried one more night of beaconing on 475.050 KHz, this time at a little faster speed compared with the previous night. Apparently the thunderstorm activity was at a much lower level as well, resulting in improved reception reports all over.

From Garry, K3SIW, in Illinois:

Andy, KU4XR in Tennessee, set up a 630m grabber to watch for my signals which quickly popped-up shortly after my local sunset:

JD in Kansas also had much improved reception over the previous test:

Thanks for the reports and the fun!

It sure would be nice to see some more Canadians getting on the band!!!

Overnight 630m QRSS Beacon

After completing the QRSS tests with the GW3UEP 630m crystal-controlled transmitter, I decided to let it stay running overnight to see how it would hold up and if 25W was enough to 'be seen' on 475KHz. In the morning I received three reports...from Kansas, Illinois and Alaska.

As well as sending along an Argo screen capture, Garry, K3SIW in Elgin, Illinois reported:

"Your signal was in and out here the whole night until local sunrise neared "



courtesy K3SIW


The QRSS60 signal is weak but discernible in the thunderstorm QRN at a distance of 1300 miles / 2100km.


John Davis in Kansas was battling even worse QRN but managed to catch a small glimpse of the weak signal just before his sunrise.

courtesy John (JD) Davis

Laurence, KL7L north of Anchorage in Alaska also sent a nice screen capture after setting up one of his many Alaskan Snapper low frequency screen grabbers to watch for my signal.

 

courtesy KL7L

Laurence commented:

"Nice signal on both arrays - a little weaker on the K9AY ast its suffering from the tx loop coupling at the moment -; this shot taken on the omni and shows occasional small Au doppler spread not seen on the beam - nice signal - AOS 0712 LOS 1105Z"

The path from Mayne Island to KL7L is almost exactly the same as the one to Illinois (1300 miles / 2100km) but unlike the eastern path, this is mostly 'over water' and one with little thunderstorm activity.

It seems that the 25W transmitter can do a credible job when run at QRSS60 (something it was never intended to do) if not a bit chirpy.