I am posting about headers here because I can use this website. There is a long discussion of headers on groups.io. I don't know how to attach files on that blog.
Anyway, I have used on my tcs both the RD Enterprises and what I think are the Dave Bean headers. I don't know for sure because my car came with headers made by RS Engineering in California. They look exactly like the headers in the photo Grumbebuns posted in this topic. I called Ken about these headers and apparently Dave Bean worked with RS Engineering on the header design, so I think they are the same as the Dave Bean headers offered today. Ken said they could make them but they were something like $2500 in S.S. Mine are just steel. Dave Bean also sells another header design for the Europa TC. It looks like a 4-1 design. I don't know the price, but I attached a photo of it which Ken sent me.
I attached a printout of the design and the analysis. There is a header calculator at the website shown on the pdf file. I am not sure the basis of this model but there are a number sites online which reference the same header design formulas. It is probably based on transient gas flow dynamics and empirical data.
Anyway, the Dave Bean design matches the prediction very closely for putting the peak torque around 5500 RPM. Also, the DB header primary length were matched within 1". The only disadvantage I see is that the DB headers are much more expensive than the RD ones. Another advantage of the DB header is that the 2-1 merge collector is horizontal to the ground so the ground clearance is about 1.5" more with these headers than the RD design.
The RD design has somewhat unequal length primary tubes, (approximately 3-6") and the merge collector is perpendicular to the ground, reducing ground clearance. Also, the secondary tube length is very short. The tube diameters are similar. The DB secondary pipe ID is slightly larger at 1.375 vs 1.35 for the RD header. The overall length "P" in the analysis is about 6" shorter than the DB design, so the peak torque will be at a higher RPM
I attached my chassis dyno run showing peak torque at 5000 RPM. I think chassis dynos skew the torque curve to lower rpm because the transmission losses are not constant with speed.
For reference, my car has sprint cams and 10.3:1 compression ratio.
From the dyno data comparing favorably with the header analysis program, it may be a good design tool.