DIY E39 M5 headers - The Unofficial BMW M5 Messageboard (m5board.com)

GT2 Evo · 25th May 2006, 21:28

Has anyone here attempted to build their own headers for the E39 M5?

I would love to have a set of Supersprint, or better yet, Dinan headers on my car, but there is no way that I'm shelling out $3k ($8k for Dinan). I have built headers for other cars before, and I estimate that I could build a top quality set for my M5 for roughly $1500 (using Burns stainless tubing and collectors).

Does anyone know what diameter the primaries are on the Supersprint and Dinan headers? Primary lengths? Collector?

Burns has a header design program that could pretty much nail these items down based on myriad engine data points (bore/stroke, compression, cam specs, valve sizes, intake and exhaust runner lengths and desired powerband range). No substitute for real world testing though.

Anyway, this would be an interesting winter project for me and may be something I could build for people (say $2250) if it yields good results.

KirklandM5 · 25th May 2006, 21:54

I'm sure it can be done, but I'm sure it would talke a lot of work to do using mandrel bends. I would not be surprised if it took 40 hours of work to do. I would think if a race fab shop did this it would probably be close to $5,000 taking into account fabrication time and materials.

Quote:

Originally Posted by GT2 Evo

Has anyone here attempted to build their own headers for the E39 M5?

I would love to have a set of Supersprint, or better yet, Dinan headers on my car, but there is no way that I'm shelling out $3k ($8k for Dinan). I have built headers for other cars before, and I estimate that I could build a top quality set for my M5 for roughly $1500 (using Burns stainless tubing and collectors).

Does anyone know what diameter the primaries are on the Supersprint and Dinan headers? Primary lengths? Collector?

Burns has a header design program that could pretty much nail these items down based on myriad engine data points (bore/stroke, compression, cam specs, valve sizes, intake and exhaust runner lengths and desired powerband range). No substitute for real world testing though.

Anyway, this would be an interesting winter project for me and may be something I could build for people (say $2250) if it yields good results.

GT2 Evo · 25th May 2006, 22:10

My $1500 estimate was for me to build them for myself. Burns 4 into 1 collectors cost $350 each in 321 stainless and 321 u bends are between $70-$85 each. I figure maybe 8 ubends plus a couple feet of strait tubing. Gets me to a total cost of about $1500.

I'm pretty good with a TIG welder, but you are correct, the fabrication time would be maybe 30-40 hours the first time. Once you have a proper jig made up though, and know how to construct them, I'm sure the construction time would be half that.

KirklandM5 · 25th May 2006, 22:19

yeah, I think that's bascially what Dinan did...if you look at their headers they look like custom fabbed headers...and I'm sure they did a jig for later models. I think what you are paying for is all the testing they did to get their design right. What I've heard is that the Dinan headers with a Tri-Y design seem to do better at the lower end than the SuperSprints. In anycase, I think if you used the Burns modeling software you could still make improvement vs the stock design.

Quote:

Originally Posted by GT2 Evo

My $1500 estimate was for me to build them for myself. Burns 4 into 1 collectors cost $350 each in 321 stainless and 321 u bends are between $70-$85 each. I figure maybe 8 ubends plus a couple feet of strait tubing. Gets me to a total cost of about $1500.

I'm pretty good with a TIG welder, but you are correct, the fabrication time would be maybe 30-40 hours the first time. Once you have a proper jig made up though, and know how to construct them, I'm sure the construction time would be half that.

cosioboy · 25th May 2006, 23:59

Here is one of the many wrong things that can happen if a header is not made right, and this is by a company that to this day makes headers (this was their generation 1 which as you see failed miserably)

http://www.m5board.com/vbulletin/sho...30&postcount=7

palantirion · 26th May 2006, 01:23

You can make headers, but you can't make up all the testing Dinan has done to optimize flow charateristics on each cylinder. That's why they charge what they charge. I have yet to see any proof (or even a claim) that anyone makes a better M5 header (more HP with no loss to the bottom end).

E55AMG2 · 26th May 2006, 05:35

If you're willing to build a set of equal length headers, and sell them for a price of $1500/set, you could easily cover your expenses on volume (once you found someone to produce them). I'd be down to test a set.

Machtig · 26th May 2006, 06:13

Do you guys have any idea how much work it is just to install a set!? This isnt a 30 min job like throwing a set of headers on a Chevy would be.

KirklandM5 · 26th May 2006, 06:19

Quote:

Originally Posted by Reverend73

Do you guys have any idea how much work it is just to install a set!? This isnt a 30 min job like throwing a set of headers on a Chevy would be.

It's a big job, and BTW, it takes more than 30 minutes installing a set of headers on a Chevy...but probalby not as tough as an M5. Try isntalling a set of headers on any camaro.

EM5PWR · 26th May 2006, 06:31

Quote:

Originally Posted by KirklandM5

It's a big job, and BTW, it takes more than 30 minutes installing a set of headers on a Chevy...but probalby not as tough as an M5. Try isntalling a set of headers on any camaro.

yeah , the big block ones were the worst....