I'm sorry but this is how that turbo flange should look even after being welded to the header.

http://www.atpturbo.com/Merchant2/merchant.mvc?Screen=PROD&Product_Code=ATP-FLS-068&Category_Code=ATP-FLS3

The OBX flange is poorly done. The first picture of the headers Dyna posted are a crappy hand cut flange that will in no way, shape, or form line up properly with the inlet of the twin scroll housing on the turbocharger. The second picture that Voiture posted shows a standard single volute flange that has been welded up in the center with another poor fitment divider plate. It too will not fit the way the flanges should. It doesn't even look like that flange would seal off the volutes from each other at the turbine housing?!?!??!

Both of those flanges will cause some unwanted uneven reversion in the header system due to overhanging edges on the turbine housing. You only want the natural tuned reversion of the exhaust pulses in the header, not any other out of time/uneven reflective pulses going back up the primaries that will affect the benefits of a divided setup.

Dyna....in my experienced and informative way of discussion on this type of work.... I'd buy a new flange and cut/weld it on there in a proper fashion to get the most out of the system. anyone who has enough TIG welding experience WILL be able to weld it on by hand very cleanly.

o2 sensor should be post-turbo as others have stated. Pre-turbo pressure will damage the sensor and give false readings.


Dyna I have priced out the SE 40mm gates and the Tial 38's as well as the drop in fitment F46. If you would like to use something else(synapse 60 vband for instance) just let me know and I'll get prices on those components. I talked with Peter from Synapse Engineering about his drastic increase in price over the past few months and it's based around different material use and different manufacturing companies for the various components. Their prices used to be cheaper than Tial but have now increased on par with Tial's units.

I'll try to get you a message later on here, gotta go grab some grub with my wife right now :)

The obx headers also appear to be stretched and formed at the turbo flange. If you have the time and really want to make those headers work very well, I have some suggestions for you to re-do the collector primaries going up to the turbo flange coupled with a new flange all together.

If they're stretching the collector primaries that much it should have been dubbed over to a T4 flange, instead of jacking a smaller T3 flange and porting the crap out of it to the point where it'll cause major flow issues at the juntion of that flange into the volutes of the turbine housing.

again as you stated though....cheap off hand ebay headers :)

Thanks Adam,

I'm interested in hearing your suggestions. It's no good the way it is so while cutting it apart I might as well do it right.

interested in your suggestions I like seeing thoughts explained. I recently developed a collector for the 993 heat exchangers and it was so confined that it makes it hard to get 2"X2 into a t3 flange but It worked out well
http://forums.pelicanparts.com/uploa...1235222635.jpg

als my pictures are decieving the y is split 100% down the center and is not coming in at 45 degrees like it looks

Very nice Ben, Great Workmanship.

You can tell when a person puts pride in their Work

Ben,

Sure thing, not a problem :) Just remember you opened the floods gates to my long winded posts here, hahahaha!

I have seen these pictures of the setups you make and by just visualization the quality is far better than that of OBX. Just note that none of this in here is directed against the quality or function of your product as all, just straight line facts for reference for those interested here in this thread.

For reference TS = twin scroll in this post

Is this a true TS setup add-on you offer for the 993 HE's, or just a 2x2 merge collection into a standard single volute(scroll) t3 flange?

I only ask because a t3 TS flange is a bit different in how you end up welding the primaries to it(they don't actually merger before the flange). You can take a T3 single volute flange and add a divider plate to make a "t3 divided flange", but the openings are still going to be physically different in some aspects and create overhangs on the turbine housing flange itself that face into the oncoming flow. If you purchase a true t3 TS and overlay it to a standard t3 flange(non ported) you'll see the slight difference in port configuration + the divider.

Non-ported true t3 TS flange and "merge" setup(supra manifold)

http://brewedmotorsports.com/1assets...nigen2twin.jpg

Merge collectors into a standard t3 single volute flange

http://i238.photobucket.com/albums/f...s/HPIM3260.jpg

You can see the difference in port and primary spacing leading into the laser cut t3 TS flange there, and the volute inlet on the t3 TS turbine housing matches that exactly so any sloppy hand cut/overly ported TS flange, or standard T3 flange converted to TS use, is going to cause flow issues into the volutes on the turbine housing itself.

The whole point of using a TS setup is to isolate exhaust pulses and concentrate their energy to the turbine wheel. Any small disruption or induced reversion before that happens and you just start to negate the benefits of the setup.

This is why I'm such an advocate of using the proper flanges on TS system. If you're going for response on a mid size motor(under 3.5L) you're probably using the T3 flanged setup. Running 2" primaries into that TS flange on tight radii can be difficult as you've seen but if you can't do it to a true TS flange without porting the openings then you're better of ramping down to a smaller primary size.(see notation below) Yes that will limit total flow a small bit but the benefits of using the proper flange far outweigh it and remember, you're using a T3 flange so top end flow isn't your concern. Now if flow and maximum power are what you're after then you're better off stepping up the a T4 flanged housing as that will facilitate easier fitment of larger primaries into the volute opening on the flange without distortion or porting. GT3582R's are offered in a T4 TS option from a few places(ATP for example) in a 1.06 a/r sizing.

Note.....if you're REALLY good with porting and have a good steady hand, or have a CNC, you can perform a taper port of the t3TS flange on the leading side(primary side) with a step recess that will allow you use a slightly larger primary than normal and still have a good flow transition. You don't want to go nuts with this, no more than a 10% taper per inch of port length.

I love TS systems for what they're worth. It's just my thought and opinion that if you're going to build and use this type of system that you'll want to maximize all it's benefits and minimize the draw backs of it as well(flow restrictions up top vs single volute)


Another little trick for you while we are on this subject of your 993HE pipes here...... looking at your picture there with the wastegate pipes leading off the primaries T-ing into the primaries at 90 degree works, but can create some turbulence and flow/choke issue at high flow rates. Now I haven't done what you've done here for the 993's and I don't know the space constraints that you've been working with but what I would do(or at least attempt) is to merge the wastegate lead pipes into the primaries at a minimum of a 45 degree into the flow. In fab terms, instead of a 90 bend I'd use a full tight radius 180 and begin the flow transition into the wastegate lead pipes right at the HE flanges, wrapping back around an terminating in the same fashion as you now have them. This promotes better flow into the gates at higher flow rates(higher RPMS). It doesn't make a big difference to most until you start messing with EBC electronic boost ramps and then the benefit comes out. It allows you to ramp up the crack pressure rate on the EBC much higher, without boost spikes and boost creep from showing it's ugly head. As with the exahust flowing itno the merge collectors, you want the flow transition going into the wastegates to be as smooth as possibly

That last topic is discussed in detail in Corky Bells Maximum Boost book as well if you need more info on it :) Nothing you MUST do. would certainly add a bit more cost and complexity to the fabrication of the system. Maybe offer it as an option to your setups though in the future. Most systems for cars don't do this though just because if space and fitment issues.


Post is getting long, I'll stop here for now :)

Not to rain on anyone's parade here. :D

But to have have an effective twin scroll turbocharger, shouldn't you be working with equal length tubed primaries? I'm not saying you aren't going to see some gains with the OBX headers, but it probably won't be what you expect on low to mid-range rpm's.

Ben, your work, as usual, is beyond reproach. Excellent craftsmanship.

For maximum benefit your absolutely correct. Even OEM applications of TS don't always use equal length headers/manifolds. The Mitsubishi Evo is a prime example of this with runner variances in a cast manifold setup. Some of the aftermarket bolt on EL tubular headers for that car have actually shown to lose power over the stock manifold. Just an example.

The discussion is good in this thread. Keep it up! :D

Yeah thanks adam I am aware of all that you posted and no this is not a twin scroll setup though I could if needed but space is a huge issue to work with stock 930 turbo locations along with using stock brackets and muffler so I had a bunch of issues mostly the fact of packaging. the dual wastegate setup I use will cause a bit of turbulence But in reality nothing that will be noticed in 99% of applications it would be the super high output cars that really should have a T4 flange anyhow.

as to the OBX headers I really can't figure out why anyone would chance there turbo to thses but ultimatly thats up to those folks the price is hard to beat. my system with wastegates and heat exchangers is $1600 approx so considerable more then the OBX

Adam I suppose I hi jacked the thread on this so Thats where the confusion came in I would appreciate any and all feedback on my systems as I would like to mainstream them as an econimcal solution while stronger then and lighter then the others.

So original poster for hi jacking continue on:)

Nah you didn't hi-jack, it's all along the same lines of conversation here.

Your product is good no questions or doubts.. I never jumped on the 930 header production as I felt it was a market that was too saturated for me to invest time into making mass produced units. I have build 930/965 header setups myself but they're all one-off based pieces for each cars individual setup. I've installed all the big name prefabbed pieces(B&B, GLH, OBX, etc) out there and I can say that I've never been truly happy with any single setup in terms of fitment, build quality, durability, etc.....

Packaging ALWAYS proves to be a nuisance when trying to incorporate a mass produced piece to work along or in conjunction with OEM components. 90% of the time I don't like dealing with that and opt not to produce a part because if it, so kudos to you for going the extra mile. General rule if you make a good product that performs better and/or is cheaper($$) than the competitors then you'll get ahead soon.

Your $1600 option comes very enticing to 99% of the general public who is looking for something like this especially when confronted with an average $4500 price tag for someone like me to make a one off, tuned, 625 system destined for a race car. Marketing is key and you'll get there with a quality product the likes of which you fabricate there

Thanks for the kind words. I hope that something sparks with these. I have thought about doing headers as well but at much larger price point then the merge could could be optimized but My major market tends to be N/A guys not turbo guys so I will wait for now

Your not raining or anyone's parade:) Its a great question and something that have much time thinking about.

Ideally I would prefer to have equal length primaries with an i.d. matched to the flow requirements but with the the least amount of tubing volume. From what I have learned; There seems to be a trade off between equal length primaries and the increased volume (adding to turbo lag) that they add in which also increases the surface area potentially reducing the heat charge to the drive the turbo.

To me the ideal equal length is much harder to build in a single turbo design and you end up with compromising results. There is a trade off between length and volume with a turbo header design. It is a shorter run and to me easier to build a equal length side mount header for dual turbos.

So the Question becomes, Does the gains from equal length primaries out weight the gains from a lesser volume system?

Perhaps the ideal system would be one that can most effectively use pulse timing of equal length primaries with a volume that contributes the least amount of lag.

Personal opinion: I think that thermal barrier coatings would benefit either design when done properly and that's why I'm willing to experiment with TBC.

Hi Ben, your input and questions are relevant and add to the bigger picture of header design. Your comments are always welcome. I do not consider this a hi Jacking :)

As far as why anyone would chance using a poor quality header. I would of preferred to build my own from scratch, but for the price $379.00 it seemed like a bargain and I thought to myself how bad could it be? Well like I have said before " I was not expecting much when I bought it and I have not been disappointed" :)
Though there is much to correct it still provides me with a few good pieces to work with in building a header that suits my needs

As with all things in Life, I will make the best of what I have and hopefully learn from my mistakes.

Unfortunately it can be pretty bad; as in destroyed turbo or blown engine bad. You don't hear much about these disasters, likely due to the embarrassment factor, but they have occurred. The LEAST amount of issue you will have is fitment. None of these things bolt up and fit perfectly. I have dealt with some absolute disasters in fitment.

I have designed a header for the 930 that can also be used on the 911. Hope to have a prototype in a few weeks. Equal length primaries and secondaries. Will start out with 1.5" primaries and then do bigger if they are successful. Will expand to include a heat exchanger if successful. As always Ben will manufacture them for me. Options will include twin waste gates and split volute Y- joint. Many things to work out and time is an issue but it will be done.

Yes I agree 100% with your comment and that is why I'm sharing the quality of this brand so that everyone interested in one will see what they are like.

I plan to clean up and inspect all internal tubing before using it

Those pulses will hit same turbine sooner or later. But longer wastegate pipe means bigger cavity to be filled, so some of pulse-mode energy that would be used to drive the turbine would be damped in that cavity. At high revs/high boost I don't think it matters so much.

With other words, I believe potential gains of twin-scroll are to be seen lower down. Probably as slightly quicker spoolup/lower boost treshold. I don't think it will influence the upper portion of power band a lot.

It's worth a try, but if you have expensive turbocharger it might be a good idea to do some testing first. You can also wrap the headers with insulating blankets (but it usually erodes them quicker).

Judging from pictures headers look OK, especially considering the price. I'm looking forward to read the results.

Best regards,
Goran

I did some testing today, here are my results.

Disclaimer, this testing was done to satisfy my own product knowledge and in no way represents a scientific critic of this product. The opinions are that of my own I am not affiliated to this company in any way and have nothing to gain. I am sharing my results so that others here can learn of my personal experiences.

Thermal Barrier Coating Destruction Test # 1

A small piece of SS tubing was used as a test piece simulating a section of header pipe.
It was prepared by abrading with a # 80 flapper wheel touched up by hand with # 80 emery cloth.
The instructions recommended a film thickness from .0008”to .0015”
Using an Air Brush a coating of .001 was applied and confirmed by measuring after curing.
A section of the coated tubing was directly heated with the flame of a propane torch until the coating failed. The temperature was in excess of my measuring device when the coating started to blister up. The SS tubing was a bright red almost yellow from the heat of the flame when the coating failed. The blistering occurred rapidly but was contained to the area that was overheated, the area immediately adjacent showed no damage.
The blistered damaged area flaked off in very small particles (see photo) when scrapped with a fingernail, the adjacent undamaged area was intact and I don’t suspect any damage to the bond. In other words the damage seems very confined and contained only to the over overheated section. Unfortunately with out the proper measuring device this test was not able record the temperature that this product failed at.

Dip Coating: I noticed air entrapment on the edge of the spoon that was used to transfer the liquid product from the can to the air brush cup, which did not flow out and left a hardened slightly bubbled area. The spoon was not abraded prier, but because it was dip coated in the process of transferring material, I decided to post cure it and also use it in testing. The dipped coating on the spoon produced a glossy slick surface. However the cured coated surface could be marked with a sharp tool and scrapped off with sufficient force.
With some effort it can also be abraded off with 3m red Scotch Bright.
I used a heat gun to heat the underside of the spoon while measuring the topside. Again not very scientific, but I noticed approximately 200 deg/f difference between the gun temp and the top side of the spoon (remember the spoon was dipped and coated on both sides) I would appear from my crude testing that the product does contain and reduce heat transfer.
http://forums.pelicanparts.com/uploa...1235531596.jpg
http://forums.pelicanparts.com/uploa...1235531698.jpghttp://forums.pelicanparts.com/uploa...1235531786.jpg
In the photo above you can see the results of the destruction testing and the damage from overheating the coated surface
In the photo below You can see the small particles that flaked off below on the white paper. As you can see the tubing is very thin walled and was easy to heat to almost melting through it.
http://forums.pelicanparts.com/uploa...1235531882.jpg
here is the spoon that dipped and ended up being included in the testing
http://forums.pelicanparts.com/uploa...1235532591.jpg

Part 2 of today's testing
TBC Dent Test.
After the first test an undamaged area of the tubing was selected and hit with a ball peen hammer to simulate a ding to the exhaust header pipe.
The hit deformed the test pipe and left a crater of approximately ¾” diameter and ¼” deep. The coated area of the dent seemed unaffected, there were no signs of cracking, blistering or peeling and the material bond was as good as other undamaged areas.

Though these test were fairly crude, I was able to determine the flow characteristics when sprayed and brushed. I was also able to determine the rate of application to achieve the proper film thickness.

Conclusion: The material sprays on well and produces a slick tough glossy finish. It also brushes on and flows well, but flashes off fairly fast which will make it difficult to do larger sections of the recommended film thickness and get the desired smooth slick finish.
I suspect that the product bond will not be affected by minor damage such as sustaining a dent. With its ability to stretch and conform to the dent, I suspect that it will also with stand expansion and contraction of heat cycles well.
Its pretty amazing when you consider a coating thickness of only .001” effectively reduces the transfer of heat. I was skeptical of a .001” coating actually doing much but seeing is believing and it was nice to see this first hand.
This coating might help reduce heat cycle cracking. Unfortunately I could not yet determine the maximum safe working temp of this product and as such I am not yet convinced of its ability to withstand turbo exhaust. Further testing is needed with proper measuring equipment before I will be comfortable using it on the inside of a Turbo header. It could however be used externally with thermal benefits to contain the heat energy to the Turbo. However I suspect that this might exacerbate thermal cracking of the header tubing.

In my personal option, this produce would be of benefit in a NA header, providing a slick surface, reducing external heat transfer, lessen potential cracking, reducing heat discoloration on SS or Chrome headers and you would also benefit from its corrosion resistance properties.
Another suggested use from the manufacture is to use this product externally on calipers to reduce heat transfer.

This photo shows the coating on the inside of the dent that still seems ok
http://forums.pelicanparts.com/uploa...1235533419.jpg

This one shows that you can abrade the coating offhttp://forums.pelicanparts.com/uploa...1235533494.jpg