what do you guys use to make your own downpipe gaskets?

[aaronlosey]

what do you guys use to make your own downpipe gaskets? if you couldn't order them. also, where do you purchase the materials. thanks.

[fast4d]

you're talking about the flange? you can buy them or use high temp RTV

[slvr98 M]

I suggest the turbonetics gaskets. The flange on the exhaust manifold gets really hot and I dont think the silicone would hold up. I used to have to replace the copper gaskets supplied by AA quite often.

[fast4d]

I'm using a metal (aluminum) gasket on the header to turbine and a metal core type gasket at the outlet to DP. both from turbonetics.

[trthrrt489]

ATP turbo has a T4 downpipe gasket for sale.

[aaronlosey]

i have a custom hks turbo. however, hks failed to mention they don't sell gaskets or flanges for the hks 2835 pro s, only the non pro s. so i'm stuck making my own flange, no problem, but i'm a little confused about the gasket. high temp rtv won't do it, i'm thinking about a piece of copper that i cut into a gasket shape, or...... maybe some kind of liquid copper you brush on? don't know if the liquid stuff exists.

[igakilla]

I suggest the turbonetics gaskets. The flange on the exhaust manifold gets really hot and I dont think the silicone would hold up. I used to have to replace the copper gaskets supplied by AA quite often.

Did you use copper gasket sealant?

[igakilla]

i have a custom hks turbo. however, hks failed to mention they don't sell gaskets or flanges for the hks 2835 pro s, only the non pro s. so i'm stuck making my own flange, no problem, but i'm a little confused about the gasket. high temp rtv won't do it, i'm thinking about a piece of copper that i cut into a gasket shape, or...... maybe some kind of liquid copper you brush on? don't know if the liquid stuff exists.

Maybe I need to lay off the crack but doesn't the HKS 2835 Pro S use standard T3 flanges for the inlet and discharge? If you have an internal you could always drill out an external wastegate T3 flange a little to give the wastegate flapper a bit of room to vent(if you go with a non-bellmouth downpipe). Otherwise I think you would have to make a bellmouth downpipe? It would probably be cheaper to just go to an external wastegate at that rate...it would be more expensive but you'd also have better boost control and less headaches.

Btw, to save yourself from having to fabricate a gasket(not too hard, i know) you could just get the flange machine cut to fit the turbo and then use a sealant when you bolt it up. That's what's done on my car and it works great. If it were me I'd just get my flange machine cut...otherwise you're going to be very annoyed if you have to take the turbo off for some reason and need to make a new gasket. If you make it out of copper though you can heat it up in boiling water to have it expand again for reuse...

[Def]

Yea, I'm pretty sure the HKS 2835 Pro S is just a T3 flanged(5 bolt outlet IIRC) 2835. No other differences.

T3 gaskets should be available all over the place on the cheap.

[slvr98 M]

Did you use copper gasket sealant?

Yeah I did and let it cure for 2 days but couldnt hold up. Even the copper gaskets will deform after a while and cause a a leak.

[techno550]

gasket? you people use gaskets? if the flanges are properly machined and perfectly flat, and you torque the bolts properly (i.e. not too damn tight) then you should never need a gasket. I know this is coming from me, so therefore taken with a grain of salt for street car use, but i swear this isn't just a race car thing. Example: How many swing valve assemblies (internal wastegates) have a gasket between the exhaust housing and the swingvalve assembly? none. RTV will blow out and leak. fiber based gaskets will crumble and vibrate apart. you don't need something that will space the two metals apart creating a thermal break. thermal conduction between the flanges is key to keep them sealing. that way they expand/contract at the same rate.

[aaronlosey]

thanks for the post techno, i might try just mating the two up with no gasket. i should take a picture of the pro s, as i can't get a freaking flange to fit the discharge side. i have t25s, t28s, t3s, but none are even REMOTELY close. WAY off. its pretty wacky looking as most are very simple. i called around to tech turbos or whoever in denver, and tons of places online, traced an outline, faxed it over, talked with hks, no help anywhere. hks makes a full downpipe, but doesn't sell it without the full kit. its not a big deal though, i will just cut it out and make one myself. just kinda curious about the custom gasket as i normally just weld everything together lol.

[slvr98 M]

gasket? you people use gaskets? if the flanges are properly machined and perfectly flat, and you torque the bolts properly (i.e. not too damn tight) then you should never need a gasket. I know this is coming from me, so therefore taken with a grain of salt for street car use, but i swear this isn't just a race car thing. Example: How many swing valve assemblies (internal wastegates) have a gasket between the exhaust housing and the swingvalve assembly? none. RTV will blow out and leak. fiber based gaskets will crumble and vibrate apart. you don't need something that will space the two metals apart creating a thermal break. thermal conduction between the flanges is key to keep them sealing. that way they expand/contract at the same rate.

Like you said if the flanges are properly machined and perfectly flat. What are the odds of that.

[techno550]

Like you said if the flanges are properly machined and perfectly flat. What are the odds of that.

Got a file? If you know how to use a file, making something the size of a DP flange or T3 flange flat is extremely easy. On the turbo side of things though, usually pretty good right out of the box.

All of the flanges between the header on a even Formula Ford and the head of said vehicle were run without gaskets too. They'd come in from wherever, and all we'd do is oven them to strain relieve them and then hand file the flanges flat. No leaks ever. Same goes for basically every race car I've ever touched.

With as thick as people are making the turbo inlet and outlet flanges these days (1/2 inch??) its going to be hard for them to warp in the first place. Strain relieving would be nice for the manifold, but probably won't affect a flange that thick. so just file the flange, do the same to the mating surface on the turbo, and bolt them together. (though I think flanges for that side of things are very easy to come by for people that are a little lazy.) Then on the DP side, same goes there. The outlet is likely pretty smooth already as they tend to machine those. (same with the turbine housing inlet.) run the file over it to make sure its smooth, and then move on to the DP flange. People tend to not make these as thick as the turbine inlet flange on the manifold, so they have a bigger chance of distorting when welding. Again, a file can fix this. (unless of course they warped the hell out of it, but if thats the case, a gasket wouldn't seal it either. Gaskets want even pressure...

Gaskets are mainly there to reduce costs in mass production. If both ends are properly machined, no gasket is necessary. Both ends machined nicely is much more expensive than both ends roughed flat plus the cost of a gasket.

[igakilla]

gasket? you people use gaskets? if the flanges are properly machined and perfectly flat, and you torque the bolts properly (i.e. not too damn tight) then you should never need a gasket. I know this is coming from me, so therefore taken with a grain of salt for street car use, but i swear this isn't just a race car thing. Example: How many swing valve assemblies (internal wastegates) have a gasket between the exhaust housing and the swingvalve assembly? none. RTV will blow out and leak. fiber based gaskets will crumble and vibrate apart. you don't need something that will space the two metals apart creating a thermal break. thermal conduction between the flanges is key to keep them sealing. that way they expand/contract at the same rate.

Thermal conduction between two flanges is key when you *aren't* using a gasket...but thats really an oversimplified take of it. Correct me if I'm wrong but I am almost positive race cars don't use gaskets because 1.) their EGT's are always hot enough to keep the flanges sealed to eachother with the help of some bolts and 2.) a poor quality gasket will blow out. Personally I would be weary to bolt two flanges(moreover two different metals for the manifold and downpipe) that are made of different materials with different expansion rates together(making it more difficult for them to seal properly). Also, there is a huge heat differential in pre-turbo gasses point-to-point that will pull on the turbo and push it in unpredictable ways...which is another reason why a gasket(or at the very least a durable sealant) is a good idea--unless of course you would rather have your manifold or downpipe crack or warp over time. I can see the benefits to not using gaskets(you can get by with a lot less in terms of a high quality downpipe design too..no need for a potentially leaky flex section when you're not worried about putting stress on you gasket and blowing it out...but then the stress is going straight to your welds)...but I see a lot of downsides to it too.

Btw, I still recommend you don't use a gasket becaue it's going to be a needless pain...but I would use a sealant.

[slvr98 M]

I appreciate you saving me all of the typing. Alot of your points were my thoughts exactly.

[techno550]

Boy I hope this doesn't turn into another thread like the tire width one.

Thermal conduction between two flanges is key when you *aren't* using a gasket...but thats really an oversimplified take of it. Correct me if I'm wrong but I am almost positive race cars don't use gaskets because 1.) their EGT's are always hot enough to keep the flanges sealed to eachother with the help of some bolts and 2.) a poor quality gasket will blow out.

Race car's usually don't see EGT's above what you would want to see in a street car. They are still piston based IC engines. Especially the endurance engines with which I have plenty of experience. They have to last. Any quality of gasket will blow out. Unless perhaps the gasket is a machined piece of steel? We don't bother with gaskets. Gaskets are for people who can't do it right.

Personally I would be weary to bolt two flanges(moreover two different metals for the manifold and downpipe) that are made of different materials with different expansion rates together(making it more difficult for them to seal properly).

The differential expansion will not crack the manifold or the turbo. At least not because of not using a gasket or any crap like that. I don't see how any stresses at the flange itself would do that anyway unless its a bending force. When the turbo inlet flange on the exhaust manifold gets hot so does the turbo. Both will expand. They will expand outwards and will perhaps slide against each other. Not a bad thing as both still being flat will still seal. They will also swell, which will increase the tension on the bolts that hold them together. Thats why you are careful in selecting the grade of the bolts in those locations so that they will have more elasticity and not snap. This is still an issue even with a gasket. Though the gasket will likely take most of the load and itself go through severe compression cycles instead.
Again, them sliding against each other will not cause a leak. Warping causes leaks.

Also, there is a huge heat differential in pre-turbo gasses point-to-point that will pull on the turbo and push it in unpredictable ways...which is another reason why a gasket(or at the very least a durable sealant) is a good idea--unless of course you would rather have your manifold or downpipe crack or warp over time.

Pre turbo manifold temps and pre-exhaust-turbine exhaust housing temps will be the same. They are exposed to the same exhaust gasses. its not until the turbine that the EGT's change significantly. At that point they drop. Expansion will be roughly the same with the flanges bolted together or spaced off each other. So how would a gasket (or sealant???) prevent a downpipe or manifold from cracking and warping??

I can see the benefits to not using gaskets(you can get by with a lot less in terms of a high quality downpipe design too..no need for a potentially leaky flex section when you're not worried about putting stress on you gasket and blowing it out...but then the stress is going straight to your welds)...but I see a lot of downsides to it too.

The welds should be stronger than the pipe. If you break a weld itself, it was a bad weld. if you break the pipe at the weld, thats another thing. that falls into the "breaking the pipe" category. Gasket isn't going to prevent that though.

Btw, I still recommend you don't use a gasket becaue it's going to be a needless pain...but I would use a sealant.

sealant? something to blow out? also a bad idea. spray copper like for head gaskets *sometimes* works, but good flat flanges are the real answer.

[Def]

My factory Nissan stamped/folded metal gaskets are holding up just fine around the turbo/manifold/turbo outlet. I did blow out one cheap arse 3" exhaust gasket between my "cat" and exhaust - but that's because the "cat" has cheap ass thin flanges and once it let some exhaust gasses out it hardened the gasket. So no sealing from that sucker - but that's at least 3 feet past the turbine so it doesn't exactly get blazing hot back there.


I don't think ultimate temps are the worry with flange to flange sealing without gaskets, but rather thermal cycling - which a street car sees MUCH more of than any type of racecar. You're not exactly going to fire up the racecar to grab a gallon of milk a mile down the road then back again after a quick cool down period, and that can eventually cause the mating surfaces to warp just slightly and then you've got a leak. No gasket sealing can work, but from what I've seen it isn't exactly common for street cars. Then again, my car has OEM parts available for the turbocharger assembly, so I get that reliability aspect - thus no one in the Nissan FI forums has any trouble with gasket sealing.

Why don't you guys try to source some OEM metal gaskets? OEMs aren't going to put up with craptastic gaskets blowing out all the time, so they put some time, $$$ and experience into making them last. Do a little research and reap those benefits.

[techno550]

Nissan used gaskets for the reasons I mentioned above. Its cheaper to seal something with a gasket than to machine the surfaces flat. You using an internal wastegate? Is there a gasket between the swingvalve assembly and the exhaust housing? If its a garrett, there isn't. Its common for them to use a gasket on the turbine inlet side of things, but again, thats because its cheaper. When using gaskets, tolerances are much higher.

Stamped metal gaskets are likely the best bet for a reliable seal with a gasket. Metal doesn't care nearly as much about thermal cycling as most composite gaskets. The silly little stamped metal bits deform greatly when sandwiched between the flanges, so the flanges would only need to be reasonably flat ... and then only in the area of contact with the gasket.

Again though, its the difference between the cheap way and the right way. I guess you have to choose one.

The bit above about breaking downpipes or manifolds still irks me though. Its rather annoying when people make up crap like that. just my $0.02

[igakilla]

The differential expansion will not crack the manifold or the turbo. At least not because of not using a gasket or any crap like that. I don't see how any stresses at the flange itself would do that anyway unless its a bending force.

I don't think I can tolerate this thread for much longer so I'm going to add my last 0.02. Two pipes, with different expansion rates, being exposed to different temperatures *will* expand differently and pull/push on the turbo when expanding/contracting. Pre-turbo gasses will be hotter than discharge gasses. As air moves through a tube it cools...that's basic physics.

Pre turbo manifold temps and pre-exhaust-turbine exhaust housing temps will be the same. They are exposed to the same exhaust gasses.

EGT's do change significantly pre-turbo. A good manifold design will limit this...but this is a reality as true as the fact that the sky is blue. Even assuming that you are right and that in this guy's car he uses two materials with identical expansion coefficients for the downpipe and manifold since one is longer than the other the *actual* expansion will be different...especially when you consider that the pipe isn't expanding at the same rate the entire way through due to dropping temps. A gasket absorbs the pressure of the expanding pipe rather than making the other pipe warp or crack over time.

This debate has really gotten out of hand...the only reason I responded one last time was that the accusation was made that I'm "making things up." What works on race cars doesn't always work on the street and vice versa. This is arguably one of those cases...

[Def]

Nissan used gaskets for the reasons I mentioned above. Its cheaper to seal something with a gasket than to machine the surfaces flat. You using an internal wastegate? Is there a gasket between the swingvalve assembly and the exhaust housing? If its a garrett, there isn't. Its common for them to use a gasket on the turbine inlet side of things, but again, thats because its cheaper. When using gaskets, tolerances are much higher.

Stamped metal gaskets are likely the best bet for a reliable seal with a gasket. Metal doesn't care nearly as much about thermal cycling as most composite gaskets. The silly little stamped metal bits deform greatly when sandwiched between the flanges, so the flanges would only need to be reasonably flat ... and then only in the area of contact with the gasket.

Again though, its the difference between the cheap way and the right way. I guess you have to choose one.

The bit above about breaking downpipes or manifolds still irks me though. Its rather annoying when people make up crap like that. just my $0.02

No gasket on the swingvalve - but I wasn't exactly saying gaskets are the best way to go about it, merely that OEM gaskets tend to hold up just fine. I don't see what the advantage flat flanges with no gaskets offer over a robust gasket that holds up just fine... There is no performance benefit either way, and I think people just don't want exhaust gasses blowing all over the place(along with the performance hit that comes with that).

So two ways of approaching the problem, and one is cheaper(a good OEM gasket) with neither offering a performance advantage. I'd go with the cheaper solution everytime.

[techno550]

I don't think I can tolerate this thread for much longer so I'm going to add my last 0.02. Two pipes, with different expansion rates, being exposed to different temperatures *will* expand differently and pull/push on the turbo when expanding/contracting. Pre-turbo gasses will be hotter than discharge gasses. As air moves through a tube it cools...that's basic physics.

EGT's do change significantly pre-turbo. A good manifold design will limit this...but this is a reality as true as the fact that the sky is blue. Even assuming that you are right and that in this guy's car he uses two materials with identical expansion coefficients for the downpipe and manifold since one is longer than the other the *actual* expansion will be different...especially when you consider that the pipe isn't expanding at the same rate the entire way through due to dropping temps. A gasket absorbs the pressure of the expanding pipe rather than making the other pipe warp or crack over time.

Pre-turbo gasses will indeed be hotter than post turbo. TIT is always hotter than TOT. Two materials with different expansion rates will expand at different rates. Duh. The runners in the manifold should not be exposed to drastically different temperatures though. A good manifold will get the gasses to the turbo as quickly and efficiently as possible. The EGT's change a bit based on distance from the head, how much heat is lost through the manifold, and also on what sort of reactions are still going on inside the manifold.
That being said, if they were drastically different temperatures in the runners, it would have NOTHING to do with the gasket (or lack thereof) between the manifold and the turbo, or the turbo and the downpipe. That is the only way you are going to create internal stresses in the manifold which will crack it. The manifold will expand, the downpipe will expand, and the exhaust housing on the turbo will expand. It happens. Things shift, and the exhaust system after the downpipe is likely hung on rubber in a street car. (solid mounts on a race car, and we STILL don't break downpipes, manifolds, or exhaust housings from it.)

I think you're confusing the term *gasket* with *magical joint*. They are very different.

This debate has really gotten out of hand...the only reason I responded one last time was that the accusation was made that I'm "making things up." What works on race cars doesn't always work on the street and vice versa. This is arguably one of those cases...

The debate is out of hand because once again those who are in over their heads choose to argue. (see: tire threads, crank harmonic threads, etc...)
This works on the street as seen by many a street car with no gaskets in certain places on the exhaust. Suggesting that the exhaust expansion is either absorbed by the gasket or cracks the manifold is rediculous. What gasket is used between the swingvalve assembly and the exhaust housing on turbochargers? None? Lets say a metal gasket (such as nissan OE ones that Def has mentioned) is used. How much do you think a torqued down metal gasket is going to "absorb" in terms of compression based load? That much? really? You might want to put some thought in to these things before making a fool of yourself.

[techno550]

No gasket on the swingvalve - but I wasn't exactly saying gaskets are the best way to go about it, merely that OEM gaskets tend to hold up just fine. I don't see what the advantage flat flanges with no gaskets offer over a robust gasket that holds up just fine... There is no performance benefit either way, and I think people just don't want exhaust gasses blowing all over the place(along with the performance hit that comes with that).

So two ways of approaching the problem, and one is cheaper(a good OEM gasket) with neither offering a performance advantage. I'd go with the cheaper solution everytime.

There are definately gaskets that will hold up. Same sort of gaskets can be seen on any part of the exhaust. Usually an aluminized bit with a bunch of little dimples in it.

The thing is though, if the flanges are flat, you'll never need to worry about a gasket. If the flange warps drastically, a gasket won't seal either, and the only way a flange will warp is if it is drastically overheated. Thermal cycles alone won't warp them. Thermal cycles will however slowly tear apart some types of gasket. (especially composite gaskets that try to bond to the metal on each side.) Metal gaskets are reisitant to this as they don't rely on a bond to the metal on either side. They slide with the expansion/contraction just like any other metal on metal surface would. (sound familiar? )

Gaskets are cheap, but its also a time versus money thing. Gaskets are cheap when time is critical (such as in manufacturing). but when you have the turbo off the car, the manifold just sitting there, the downpipe just sitting there, whats *cheaper*, metal gaskets for the turbine inlet flange and DP flange? or you taking 15 minutes with a good file?

Looking to manufacturers (who are known for doing things like switching to paper washers instead of brass in gearboxes to save $0.015 per car) will tell you the cheap*er* way of doing things, but not always the better way of doing things.

[Hova]

Bumping a 14 year old thread, but there was a lot of information here, and Michael seemed like adamant on his stance.

I’m about to throw on my new T4 turbo in, should I use a gasket or just go flange on flange? Both mating surfaces measure nearly zero runout.

[Def]

Use a gasket. No gasket with a FILED flange will leak all over the place. There's just too many leak paths. A gasket is going to at least offer some compliance to help close up some of that leak path.

It also helps when you've got two items with a different coefficient of thermal expansion bolted to each other through various mechanisms.