Hi Guys,
So I finished my design of wave guides for tXXrp's, They are available for download from this link: https://sketchfab.com/zgliszcz
I believe that each drivers opening should have own funnel that will guide sound wave in both axis. If guides were similar to the one from LCD or Abyss they would work only in one axis as shape of the holes arent wide but round. All 4 of them are variations of same concept but with different profile. G4 have some sort of "focus" profiles. Similar one but angled to the back I'm planning to try in my upcoming project (for different casing). I might make variant with funnels guiding waves to the sides (opposite to focus) which should reduce back waves and standing waves in fostex original casing. They will need to be printed in laser 3d printer or you will need to prepare support frame for them by yourself. Before using on internal side it might be a good idea to sand corners on driver casing or on bottom sides of wave guides. I also believe that by removing white paper from internal side of the driver and applying it (paper dampener) straight on top of metal magnets frame and tighten it with wave guides on top might give nice results (although ventilating will be pretty weak as dampener paper will breath only in places where are holes). Also you will need to remove two thin supports from driver and leave only thicker two. (same like in SeEnCreaTive design)
This is side part of my design work. Currently Im sculpting my new headphones which should be available in near future. Although for few reasons they will be available only for sale (if they will prove to be worth it) all wave guides are for free to anyone interested in using them for non commercial projects or as base mesh for their own design (but that will mean you should release them for free under same license) I hope they will help you push even further sound horizons on your tXXrp's.
Before printing please make sure that sizes are fine (one of the sizes should have exact 38mm and second one about same, thickness is different on each guides about 4-5mm)
If I wrote some crazy nonsenses I will edit it tomorrow as my language skills needs rest
2 axis prifiles and top view:
bottom side (same for all variants)
π Update
Recently I made 11 new variants to my wave guides. Some of concepts (I believe) will work good only after special treatment inside enclosure. Some of them might not work at all, who knows

some of them should work only on one of the sides of the driver. Some might require bit of sanding or cutting to shape driver frame to avoid wave distortions (places marked as "D" have deeper geometry that driver frame). It is very likely that they will require sanding external sides on the driver side (previous G designs might require that also). Some of the designs are based on BMF idea and were made for him to test if his concept work, but I couldn't resist to modify them a bit as I was disagree with some of his assumptions (mostly because of specific fostex vents design and because of that it was also hard to implement it exactly like he wanted so they are something between his thoughts and my interpretation/thoughts but I know that he will be able to modify/improve designs easily without modify printed elements). Another thing is that Im still not 100% sure if I didn't screwed up sizes. I did precise measurements but as everyone who measured Fostex driver know they are for some reason fractional in some cases... and its pain in the back... Another thing is that I batch export all of Pi drivers in different than usual way (to save time) and Im not exactly sure if sizes were exported nicely. On the bottom of each image I put information about "X" size they should have, but keep in mind that websites that offer 3d printing usually dont show that precise dimensions and they may vary (but shouldnt more than 1mm), 1mm is big difference but even if dimension will vary it should be printed right (they might show rounded dimensions but 3d printing process is still more precise). As you can see there is lot of "if" and "should". The most safe and closer to original idea design is guide Pi10 (the most boring looking isn't? ) and previous G designs. Other guides might cause distortions and uneven sound wave distribution. I believe that all those designs can help reduce standing waves and hinder backwave. Just like in G designs there might be issues with thin parts marked as "A" and "D" on bottom image. BMF asked me if I can make design without two ditch for thicker frames but it will require removing all of magnet frames and issue with thin elements would still occur in outer edges (marked as "A"). Some of the guides have thinner funnels holes (marked as "B") to make more balanced/flat diaphragm move (based on part of BMF concept but reduced as I couldn't make radical enough diversity because of technological limitations). I believe that those design can push bit more progress and diversity in modifying tXXrp headphones, but they might require new approach in modding or require custom enclosure to really show its potential. Another thing I would like to mention is that they will cost less than previous G designs because I further reduced geometry (printing process should take less time) and made pair as one element (big thanks for SeEnCreaTive for figuring this out) you will need to cut part (marked as "C") and sand the leftover. If there will be interest in that I can reexport G guides in same way so they will be cheaper also). If anyone will decide to print any of those design I will appreciate your comments and thoughts, and that will help me improve them). I would suggest to do not smooth them (there is feature like that on some websites) because it might damage all those thin elements. you can apply thin layer of lacquer later to make them extra smooth (just don't melt them

).
All guides (also one for pre Fazor LCD2) are available at this link:
https://sketchfab.com/zgliszcz/models
Pi1

X size:36.819 mm. As far as I remember (and as I can see) defocus on external rows of funnels and focus on internal rows. I wouldn't use them on ear side
Pi2

X size: 36.819 mm. External focus rows, smooth internal rows (6 funnels in centre have wide geometry), rows between looks like "defocus" but its because geometry of center rows overlapping their geometry. Can be used on both sides
Pi3

X size: 37.2616mm. strong defocus geometry with aggressive sharp edges of funnels. Probably wont work on ear side.
Pi4

X size: 37.1478mm. Soft geometry slightly focusing, "BMF peak" in centre and thinner funnels in middle rows, to increase effect funnels in centre can be coated with layer of lacquer or partially cover by fibre (or completely sealed by plasticine or my favour bee wax). Should work on both sides
Pi5

X size: 36.8837mm. Should work on both sides, have variation of "BMF hump"
Pi6

X size: 36.2724mm, focus effect but smooth wide funnels in the centre with thinner holes, should equate a bit diaphragm movements. Should work on both sides.
Pi7

X size: 35.989mm, strong defocus effect on one axis. Just like all defocus wave guides I wouldn't use it on ear side (but its worth checking as maybe it will work better than focus or normal geometry because of small distance to ear)
Pi8

X size: 35.6743mm, defocus on middle rows
Pi9

X size: 36.5319mm. "BMF hump" and strong defocus, I'm not sure if it will work on ear side because of defocus geometry
Pi10

X size: 37.2702mm. Classic approach probably the most safe design from Pi family

should work on both sides. I would suggest to start with this design if you are curious to try my concept.
Pi11

X size: 36.5366mm. Random geometry, and probably same effect;D
Its hard to tell which design will work in what situation/mod before trying them, some of them have quite complicated geometry so I just write "focus/defocus" as my English is average, so just please let me know if you have any questions or something is confusing.