so i finally got around to building a simple 40m dipole out of some 14 gauge stranded wire i had laying around the garage. used the 468/f formula, cut each leg to about 33.4 feet, fed it with some RG-8X i had left over from a previous project. center insulator is just a piece of PVC i drilled two holes through, nothing fancy.

hung it as an inverted V off a 25 foot mast in the backyard, legs going out at maybe 120 degrees or so. hooked up my NanoVNA and the resonance is coming in around 7.4 MHz instead of down closer to 7.150 or wherever i was aiming. so either i cut it short or the velocity factor on this coax is messing with something or my NanoVNA is off. antenna is maybe 8 feet off the ground at the ends if that matters. been scratching my head on this for a couple days now.

anyone dealt with this before? do i just add a bit of wire to each leg and retrim or is there something else going on here

yeah the ground height is almost certainly your issue, or at least a big chunk of it. when the ends are that close to the ground the antenna loads differently than the free-space formula assumes. the 468 constant kinda presumes the thing is up in clear air, not 8 feet from dirt. try pulling your ends up higher if you can, even another 5 or 6 feet makes a noticeable difference. also the inverted V configuration itself tends to shift resonance a bit higher compared to a flat top because the angle changes the effective electrical length somewhat.

if you cant get more height just add a few inches to each leg at a time and retune. i usually leave myself a foot extra when i first cut and trim down rather than the other way around but its not the end of the world, splicing a bit on works fine too just solder and weatherproof it properly.

NanoVNA readings at that low a frequency can be a little sketchy if you havent done a proper OSL cal recently or if the cal was done with a short lead and then you added a bunch of coax after. not saying thats the problem but worth ruling out. i had mine reading resonance almost 200 kHz off once and it turned out i just needed to redo the cal at the end of the feedline instead of at the device port.