The KT34A at W5XD was upgraded in September, 2022 using the m2inc kit.
Here are four snapshots of the SWR curve as measured in the shack. The legend is in chronological order as I modified the antenna.
As the SWR curves above show, the wornout 1985-vintage KT34, as measured in 2019, had reasonable SWR on 10m, but
its 15m and 20m low SWR points had drifted out of those bands, lower in frequency and higher in frequency, respectively.
The upgrade (the final, green curve) clearly fixed 20m and 15m but 10m never gets below 1.5:1 in the shack (and never below 2:1 at the balun: raw analyzer files linked below.)
The "upgrade of only rear DE" was applied in 2019 (because I could do it from the tower and without a lift.) It helped on 20m, hurt on 15m and moved the 10m least-SWR point about 500KHz lower in frequency (maybe this 10m shift to lower frequency is a hint?).
The "m2inc upgrade" curve is with the antenna dimensioned per the m2inc "Upgrade Dimension Chart"
in their upgrade kit. The 20m and 15m SWR is great, but something is clearly amiss on 10m. After corresponding with m2inc, they suggested moving 10m shorting bars, which (after many intermediate steps not presented here) resulted in the green curve when I decided to quit moving straps.
The DXEngineering square balun plugged into the round balun hole on the KT34. The resin balun shelf available from DXEngineering is what the balun is attached to (but with the shelf's aluminum angle piece discarded.) The 3D printed piece (using PETG filament) is fastened to the DXE boom-to-element clamp and holds the resin shelf. 
Below are the RigExpert antenna analyzer files of the final three shorting strap configurations as measured at the balun (i.e. the 150ft of coax is not involved.) The antenna at 25ft height.
The final link is the data measured with the dimensional configuration that is now cranked up to 88ft. I did not record at-the-feedpoint curves prior to these last three cut-and-try's but the progression was
similar. As the RDE and FDE shorting straps were shortened, the 10m SWR very slowly improved. C17 and C18 also include lengthening the RDE 20m tip to 56.5".
- kt34-upg-c15-at-feedpoint.asd. RDE 10m: 25.0", FDE 10m: 24.25", DIR: 23.25"
- kt34-upg-c17-at-feedpoint.asd RDE 10m: 24.5", FDE 10m: 23.75", DIR: 22.75", RDE 20m: 56.6"
- kt34-upg-c18-at-feedpoint.asd T-Match width to 18" on each side.
Notes about the the published dimension charts.
Because I am convinced I followed the dimensional instructions correctly for both the original KLM assembly and for this upgrade, I am suspicious that the upgrade dimension chart might not work for my situation on 10m. But cannot make any
claim as to how it might be causing the issue. I record here that its nontrivial
to take the dimensions in the KLM manual and compare them with those in the M2INC, at least when you think you must quibble over 0.5" differences.
There are two major differences between their mechanical configurations and their instructions:
- The original and the upgrade dimensions are not measured the same way as each other.
My 1985 KT34A manual says that "adjust to the dimensions ... by hooking a tape measure on the type "A" capacitor straps and pulling .... to the outer edge of the "B" and "C".
On the other hand, the M2INC upgrade manual says "measure to the inner edge"
That KLM "outer edge" is 0.5" further away than the M2INC "inner edge." Therefore, when the numbers in the two charts match, the
midlines of the straps in question are 0.5" closer to each other on the KLM than the M2INC.
- The pair of straps in the middle of each element, called out as straps(A) in the KLM picture, cannot be placed in same location in the upgrade as in the original without drilling new holes
in the fiberglass rod. The closest the upgrade strap midlines can get to the KLM positions involves using some force to butt the new strap up against the 8-32 nut whose center defines the midline
of the original 0.5" wide strap.
(My calculation is that the upgrade straps' midlines are about 5/16" outboard of the KLM straps centerlines.) I chose to hammer the butts and not to drill. And what does "same location" even mean when the new straps are 0.25" wide and the old ones are 0.5"?
I am not an antenna designer, and will not offer a conclusion about how to translate the mechanical differences (which I can confidently report as above) into the electrical differences in the way the antenna handles RF.