Rolf Ekstrand

Greetings y'all,

Just received my QDX-M this morning.    Wow ordered it on March 13th.

Anyway, I ordered it as a 10 m unit, but my intention was to put it on 20  running FST4W.   I assumed I could use the filter data from the High band QDX manual.  It that a correct assumption, or?

Rolf K9DZT

Paul WB6CXC (tech-blog: wb6cxc.com)

You will probably find that the QDX transmit frequency is not stable enough for most FST4W modes, due to heating if the internal TXCO during the transmit cycle,  FST4W-1200 has a tone spacing of 1.465 Hz (same as WSPR), and the QDX 20 meter transmit frequency drifts about 1.4 Hz during the transmit cycle.  It might be OK with FST4W-60, but it's going to be marginal.  I've used my QDX for 20 meter WSPR and i was received well enough, so I guess it's possible,

I've got a lot of blog postings about the QDX frequency stability issue, and how I adapted the QDX to use an external reference clock: http://wb6cxc.com/?p=307 (and there are other related posts).  Even an external cheap TCXO should provide adequate stability for FST4W, you don't need a GPSDO for that.  The issue is the heating of the QDX circuit board caused by the transmitter output transistors.

Hans and others are looking into various methods to improve the stability, but for the moment this is where we stand.
Paul Elliott - WB6CXC

Brad Morris


you may have already addressed it elsewhere, but I’m presently too lazy to go searching…are you selling your external ref board? That’s a neat addition to the QDX. 


Rolf Ekstrand

I am well aware of  the thermal issues with the QDX and I  intend to operate with a GPSDO.  However If you modify the cooling of the PA and move the board out of the case it will get better, but not as good as need to be.  In Feb. when down at Orange Beach, AL I did operate one of my modified QDX on FST4W-120  20m and got plenty of reports (even from down under) but the spread was too much.   


Gregg Myers

Lucky me, I just received two TCXO from Qrp-Labs. One went into a U3S where it did a remarkable job of stabilizing drift on 10m WSPR. But that is a different thermal situation and entirely easier to tame than the thermal situation inside a QDX. So I plan to try either mounting the TCXO outside the QDX box via a SMA connector (plus I would need the voltage pins and the transient suppression as Paul suggests), or... maybe just try mounting the tiny TCXO board inside the case but off the main circuit board, possibly insulating it some. Paul mentioned he believes the heat conducts through the main circuit board so maybe that would help. Of course all this is entirely un-necessary for quicker modes like regular WSPR. Not that I would mind having a GPSDO, but I don't currently and they are expensive compared to the little TCXO board from Hans.

Gregg W7GRM

Paul WB6CXC (tech-blog: wb6cxc.com)

Brad, I could be persuaded to sell the external ref interface board if more than a few people actually wanted one.  I would probably include a replacement QDX back-panel that had the extra hole for the SMA jack (a custom circuit board as I have done for other projects).  There would be two versions, one with the multiplier PLL, and the other with just the protection circuit.  Remember that this requires a slight modification to the QDX -- removing a capacitor and soldering a few wires -- and an external reference of some sort will be needed.  The combined price would about $20, perhaps less if I get JLCPCB to assemble most of it for me (they probably wouldn't do the SMA jack).  Compared to the basic QDX this isn't particularly cheap!

I would probably add a tiny U.FL connector to the board, in case Hans eventually adds an external ref option to the QDX PCB.

As I mention in my blog, you don't need a clock multiplier if you have a 25 MHz reference handy, and the Bodnar GPSDO units can provide 25MHz directly.  I use the 10MHz to 25MHz multiplier because I use my 10 MHz GPSDO as a general lab reference (sig gens, spectrum analyzer, counter, etc.)

Rolf, as for FST4W spreading, I'm running a referenced QDX on 20 meters within near Line Of Sight to the KPH receiving site (Point Reyes, California), and we are seeing spreading of 5mHz (which is low as the system can measure).  I've got an identical transmitter in Friday Harbor (about 900 miles to the north of KPH), and as you would expect, the spreading varies between 5 and 500.
Paul Elliott - WB6CXC

Rob Robinett

As spreading goes up, the probability of decoding the message goes down, that is true even for WSPR-2.
So if you want your transmissions to get reported more often, GPSDO your transmitter.
If you want to report more receive spots, GPSDO discipline your receiver.
The QLG2 is not a GPSDO, but on transmit it will be a big improvement over the internal oscillator in the QDX.
For receive the QDX internal oscillator if pretty good up to 20M, but the unique benefit of the QDX is that it is a stellar FST4W transmitter when driven by a GPSDO
A 5 watt QDX produces a 14 dB stronger signal than a 0.2 W U3S or other WSPR beacons.  Your would need to put a much bigger antenna very high in the air to get the same increase in signal strength from a 0.2 W beacon.  Of course increase both power and antenna size and height if you can!