As I suspected my dear Watson. The cause of the slight twist in the example shown is, I believe, a product of the length of their fore / aft rods and their very thin 0.9mm wall construction.

I've had to make a few assumptions and guesses here:-

- The length of the rods is 24" - must be somewhere near.
- They state the main chassis was 0.9mm thick tubing so I have assumed the rods are too. Given their weight claim for the entire structure that's probably an accurate guess.
- I've guestimated the rods at 1" diameter from comparison to the handlebars which are usually 22mm.
- In hard cornering where the inside wheel is just lifting the rider and chassis extert a force of 150lb across the two rods. This can't be far off based on observing their video and the deflection of their system assuming points 1-3 are somewhere near.

Their long thin rods deflect 1.9" but reducing them to 12" and upping them to 1.5mm wall gives a deflection of just 0.15". Using 12" of 40x20x1.5 rectangular box deflects just 0.03" or as near as damn it no twist worth mentioning. Reducing to 12" means the axle does proscribe a tighter arc but the numbers are utterly minuscule and eminently ignorable.

Now up that force to a more porky 225lb as many of us do not appear to be as svelte as others, myself absolutely included, and you go from 2.8" deflection with the example to 0.05" with the shorter rectangular versions. Or put another way from almost 7 degrees of lean to almost none.

There will be some flex in any unsuspended tadpole trike caused as a function of main beam twist. Even using 40x20x1.5mm box in the vertical which of all the usual materials for a trike is the least able to resist twist I have found any such twist to be undetectable in use. My calculations above are designed to be in addition to any such chassis twist.

The weight you'll add to a tadpole trike to add suspension of this design are 2 foot of 40x20x1.5mm for the fore and aft rods. The weight of any pivoting parts plus the shock absorber and mounts - around 4LBS. I think that, unlike the example, you could get away with just the one central shock absorber as you'd then have removed most of the twist (famous last words!).