The antenna needs to be higher than 4m! Most SOTA activators are using at least a 6m pole and many use 8m.
That antenna almost certainly needs an ATU to get the best out of it. If you don’t want to use an ATU then convert it into a link dipole so that it is resonant.
My guess is that currently you are only radiating about half a watt!
I’m going to be doing some activating in the Welsh Boarders next Sunday/Monday, drop me an email or a PM if you want to tag along.
If it’s really “40-to-2-metres” then, unless it has links/switches/traps/whatever, it’s not resonant on most of those bands, and therefore it’s assuming some sort of ATU’s in use(*). Even when using an ATU a resonant antennna will have advantages. As you don’t have an ATU, your first step should be to make yourself a resonant antenna. That will be for just one band unless you use links or traps, of course, but a linked dipole isn’t hard to make. Plenty of instructions about on how to do that, and a handy calculator on SOTAMaps under extras. If you don’t have the bits in your junk box and need a starter, SOTABeams sell a kit of bits for making a linked dipole to which you will need to add a feeder cable and antenna wire of some sort. An antenna analyser is a handy help, too, as it’ll do a rather better job of measuring SWR than the 817/8.
Another monoband alternative is a simple vertical, but that’s not really practical for 40 metres, and even for 20 metres you’ll need a pole quite a bit longer than 4 metres. Again, you’ll find instructions around and about, including more than one place in this reflector (Lightweight Vertical Antennas, for instance).
I could rabbit on for hours, but that’s enough for now. Good luck. 
(*) Colin @G8TMV pointed out to me else-Net that it might be using the “Clansman” technique of expecting the operator to only un-wind enough wire for the band desired, which leaves the rest of the wire in a big (lossy) lump of a coil at the end…
Hi again, Pete,
In case you feel like trying an ATU, let me show you this very light one I bought as a kit and assembled + tested successfully some time ago:
This tuner may work well with the antenna you have. All you’ll have to do is bringing the wire length to something close to resonance and the minituner should do the rest.
Again, good luck!
73,
Guru
That will get the SWR to a better figure - but a dummy load has a good SWR too. The antenna design is inefficient, the extra wire on the winders at each end makes it very lossy and a poor radiator.
Agree. A linked wire for 3 bands (15, 20, 40) would be the best choice for this case.
It looks to me like you wind/unwind the antenna to the coloured markers to set the element lengths. It’s just a case of selecting the right lengths to get to resonance on the desired band.
Can you borrow an antenna analyser? The MFJ-259B is ideal for this work. Set up the antenna wound to the band markers and measure how well matched it is? You may find just a few turns more or less are needed to get to the best match. Then take it down and repeat to make sure the setup is repeatable. See what the 817 says when the analyse says best match too. You can then learn how to interpret the 817 display better.
I only use a 5m pole and have managed some 12500 QSOs for SOTA on HF using dipoles, 1/4GP and simple vertical (up and outs). Victor GI4ONL tends to use mainly a 4m pole and has logged 10500 QSOs for SOTA on HF. The longer the pole the better but short, compact and light poles do work well IF the antenna is correctly setup.
I did have a listen Pete, but there was absolutely nothing from you… zilch. I would have expected something over the 40-odd mile path down to CE-005. It all points to that antenna being the problem.
A load of good advice here! I’m sure that every suggestion here will work. My approach was a little different. I decided to use the venerable W3EDP multiband antenna, consisting of a 25.6 m wire as in inverted L and a 5 m counterpoise draped over the heather. This needs a simple tuner, just a coil and capacitor, 14 turns on a T130-2 toroid with a central 2 turn link winding, tuned with a little variable cap salvaged from on old transistor radio, gives a zero bars tune on the FT817 from 7 MHz to 21 MHz. This worked well, the FT817 giving me SSB contacts in 5 continents when conditions were decent. I also have another tuner covering 80m to 20m with the same wires but capable of taking 100 watts when I take the FT857 out. The advantage of this approach is that I don’t need to leave the operating position to change bands. Yes, I’m lazy, but it also avoids scrambling around on a rubble-strewn summit! 
Nothing wrong with stretching your legs Brian from time to time when you change bands - after sitting hunched up in one place for an hour twiddling knobs you must get very stiff… 
My advice would be to accompany an established activator to see how they do it and what equipment they use. You can then compare your setup with theirs. There’s plenty of sporadic E around this time of year to enable your 5w on 20m to be heard. A couple of weeks ago, my 5w from the KX2 put a 5/9 into Skipton from Jersey and was still 5/3 when I reduced to 100mW. I see you are in the Wrexham area so you should have no problems arranging a joint activation. Best of luck 
I activate SOTA VHF only but in other “OTA” programs,
*FT818-6w (Blank power screen display) on @12V+ power supply.
*SOTAbeams speech compressor.
*20m endfed-1m of RG58 feeding from antenna port. Antenna end ~6m off ground.
FT818 settings I recommend,
IPO…ON
ATT…ON
RF…Gain ON/SQL OFF (selected to RF Gain) menu #45.
Volume knob all the way up and use RF-Gain (outer knob) to adjust volume/noise.
I have done 10-12 activations over the last month with this setup and have had amazing results. Mostly early evening. 2 SSB QSOs 4500+ miles Italy and Slovenia from Ohio, USA.
***Morse/CW is the way to go for reliable QRP, I’m getting there!
Erik
KE8OKM
The logic behind the yaesu menu items always confuses me…
No bars on power setting for maximum setting; IPO on to turn-off the pre amp.
I am surprised that att off is the setting to turn off attention!
It doesn’t turn off the preamp.
The IPO and ATT are used only to attenuate nearby loud signals… otherwise both off. Noise blanker on, narrow filter on with 500 Hz CW filter fitted.
I’ve tried lots of 817 settings to use for weak signals, and the best is to confirm that full RF gain equals no bars on the S-meter, then back off RF gain until three bars show, then turn on the outboard bhi DSP filter to your favorite settings (you do have one, don’t you?), then advance the AF gain to taste with headphones. Voila… best ears, ever, other than a KX3 on FT-8.
If full RF gain shows some bars, move elsewhere in the AZ or give it up and go bowling.
Elliott, K6EL
Hi Pete,
Sorry to hear of your HF SSB difficulties. There’s been loads of good advice on this thread! I don’t have an FT-817/818 but I do use a uBitx transceiver, which is similarly low powered and even more basic. I’ve used a simple link dipole (actually it started out as a single band 40m dipole, then I made ~10m extensions to bring it to resonance on 80m. So now it’s a 2 band link dipole. 
) and as long as I put out a spot or two I’ve always managed to qualify my summits (so far, thanks to the dedicated chasers).This is in Aus though - might be different in the UK.
I think that’s how it works, but I’d say it should work fine at least on 40m, as even if the coil of wire is lossy, on 40m there won’t be much coiled up.
But the main reason for this post is to agree with the previous posts that the antenna SWR may well be the problem. It sounds like a bit of experimenting with the antenna and SWR meters (external and built in) may be a good idea. As has been mentioned, a basic analyzer would be a help.
Obviously, I don’t know what your meter at home is, but
(A) if it’s not a cross needle type meter, is it calibrated properly?
(B) are you reading the right scale? Sounds obvious, but it’s not sometimes 
and
(C) they’re not always accurate.
I have an MFJ tuner with a built in cross-needle meter and if I set up the tuner to read 1:1 SWR on the built-in meter, the SWR at the rig is often around 3:1… I’ve found this with different antenna, feed lines, and so on, and have confirmed the accuracy of the rig’s meter several times with my analyzer. I probably could recalibrate it but it doesn’t worry me too much.
But if the SWR is up, the rig will not only cut back the power to protect it’s finals, but there’ll be reduced power transfer to the antenna as well. I’ve seen this quite dramatically with my uBitx - it’s got no SWR protection, so it will try to output full power in to any load, until the smoke comes out, anyway… but I’ve had reports of my signal dropping fairly dramatically when the antenna is mismatched.
Just a few thoughts - which probably doesn’t add much to the previous posts…
Good luck!
I could be wrong but I believe IPO-ON turns off the RF-Preamp.
Curious as to what others think about this hams FT817 settings?
“IPO (Intercept Point Optimization) – Bypasses RX Preamp on HF/50 MHz for improved performance during strong-signal conditions.”
From the FT-817ND page at Yaesu: Welcome to Yaesu.com
wunder
There is no HF/50 preamp as such. IPO, Intercept Point Optimisation bypasses the first RF stage and applies the signal to the the first mixer on the HF/50 path.
If there was a preamp, there would be a switchable gain stage in front of a gain stage, from the word “pre-amplifier” an amplifier in front an amplifier.
You should write to Yaesu and put them right on this, according to the handbook “The IPO feature bypasses the receiver RF preamplifier,…”!
They can call it what they like Brian. The whole thing is marketing madness. Just like calling car seats Naugahyde, the use of a word sounding like hide was to disguise it wasn’t really leather but PVC.
As I have said, you can only have a preamplifier if you have an amplifier, the clue is the name. There is a single gain transistor for HF and that is bypassed by the IPO control. There is a 50MHz preamp but it’s not controlled by the IPO, it’s always active. So for 50Mhz the signal path goes preamp then switchable HF RF amp, 1st mixer. For HF it’s just the switchable RF amp then 1st mixer.
All of it is marketing led nonsense. If you say RF amp on/off people who are not too clued on RF design will want to know why would you want to turn the RF amp off? Calling a preamp sounds like you’re controlling an extra feature and is a more understood concept. They use the term IPO (intercept point optimisation) to sound fancy. If you lower the gain before the main selectivity in a receiver of course you’ll improve the IP3. They’re not alone, Kenwood have AIP Advanced Intercept Point which is an alternative lower gain RF amp selected instead of the higher gain RF amp.
I’ll bet you a pound of Wasabi the Yaesu engineers wanted to called it RF amp on/off!