home made twin turbo the "right" way?

[riceburner]

A motorcycle exhaust I did a couple of months ago. GTAW purged.


[IMG][/IMG]

[IMG][/IMG]

[turbocad6]

ok, without seeing the insides of those, that welding^^ to me looks to be perfectly adequate as far as functional goes, but not necessarily beautiful "work of art" beads, which to me, at this point & with the project at hand should be good enough for now... functional welds & beautiful perfect looking welds are 2 different things, right now I'd be happy with perfectly functional adequate welds, then grind it down & polish to look like one piece. this is my goal atm.

since my goals for right now are to be able to weld these seems adequately enough to be good sound welds my questions are:

1) would I be ok just doing a fusion weld to butted seams? then, if filler was needed to get a smooth seamless finish I could go over it to fill with a bit less heat?

2) would I be much better served by grinding a v groove & filling on the first pass? for a fusion weld I'm thinking no v groove, just not too sure what the better plan of attack here would be. I think I have a better shot of mastering a fusion weld non v grooved than the later, but if it's not adiquate then I don't want to waste my time trying to perfect something that's not good enough anyway. I guess a fusion weld on a V grooved fit up with a filler pass after is also an option?


3) I'm concentrating on what the interior of the weld looks like, is it absolutely necessary to be able to see a whole bead of weld on the inside showing full penetration or would it be not so terrible if some spots are not fully penetrated enough to see a bead on the whole inner circumference of each weld? I know the answer to this can be kind of subjective, but I'm wondering if it's not too uncommon for a weld to be considered adequate without showing a full weld bead on the interior. I know this would not be adequate for a sanitary weld for like the food industry, but how is this considered when it comes to headers?

4) is there any problem heating & then reheating & reheating again as long as none of the heating is too hot to cause carbide precipitation? in other words, is the heating damage cumulative or solely a matter of max temperature on any one pass?

thank you for any help you care to give, I really would appreciate the input to help guide me as to what I should ultimately be shooting for here

[turbocad6]

During your research, had you considered rear-mount turbos? STS offers universal single and twin turbo kits that work great and reduce a LOT of the headaches of a front mount traditional setup.

http://www.ststurbo.com/universal_turbo_systems

you know, the funny thing about turbo charging is, it can be as simple or as complicated as you want it to be. a turbocharger, by it's very nature can be easy to install. I mean if you size it right, pass all the exhaust through it and then pass all the intake through the intake side then bam, your running a turbo... you will get boost & it will work, sounds simple enough & many do approach it with this simplicity. this is why even a poor design like the STS does still work, I mean it has too, hell you could put a turbo on the roof & plumb to & from it & even that will work. is it as good as it could be?

well turbo chargers are normally driven by exhaust pulses & heat, but, an STS setup works a bit different, it's more just driven by pressure alone, again, plumb it in & it just HAS to work to a certain extent, but, that in no way means that it is an ideal system, no, far from it, better than nothing? I suppose to some, but my goal here was more to build a great turbo system, designed to be extremely efficient. a gross install may still work but a properly designed efficient setup will work better. if I wanted easy then I would have just installed a jim wolf twin turbo or a greedy or a GTM, there are off the shelf options, but even those are far from ideal.

my design is built around efficiency & minimal turbo lag. when a car is idling the intake is under vacuum & the turbos are spinning very slow. when you go from this to stomping on it full throttle, it takes time for the engine to ingest enough air to raise the rpm's, then the exhaust to spin up the turbo, then the turbo to then begin to compress all the air volume between it & the throttle plate itself, all the volume of the cold piping & the intercoolers themselves, then & only then will the engine actually see boost. the combination of flow restrictions of the intake tract combined with the amount of air volume that ultimately needs to be compressed is what causes turbo lag. the more efficient the design the less lag & the less stress on everything too. it's one thing to see 9 psi at the manifold in a not so efficient setup that is actually making 12psi at the turbo but then loosing 3 psi pressure drop to the intercooler & cold piping layout than it is to have a very efficient layout that shows 9 psi at the manifold when the turbo head pressure is only ~10 psi...

[riceburner]

I hope to someday arrive at the same level of understanding of turbo charging as you have been able to achieve.

But, if you wanted to be really efficient you would be running a split tangent turbo with a collector that has been properly designed to maximize the exhaust pulse and minimize the effects of any back-pressure in the pipe.

[turbocad6]

that would be a twin scroll turbo, would be the plan if I was using a single larger turbo driven by both side exhaust banks, but by splitting the turbo into 2 smaller units I'm achieving the same thing as a twin scroll... if you notice with my collector design, the collector is done in such a way that it creates an anti reversion chamber at the point that the 3 pipes meet

[Rick V]

During your research, had you considered rear-mount turbos? STS offers universal single and twin turbo kits that work great and reduce a LOT of the headaches of a front mount traditional setup.
http://www.ststurbo.com/universal_turbo_systems

I don't know much about turbochargers but... it seems to me that for quick response and highest efficiency you would want the turbocharger as close as possible to the exhaust manifold and the engine air intake. That rear-mounted turbo is several feet away from the exhaust manifold (compressed warm but not 'hot' air) and several feet away from the engine air intake (a lot of air to compress before the engine will feel it).

It's like an air horn trumpet to replace an electric car horn. It's very responsive if you keep the compressor/trumpet distance short. When I first installed an air trumpet with three feet of hose to the compressor, I hit the horn button and waited........ duh?.... for the horn to finally sound.

Great post!

[blown99gt]

heres a couple i made at my part time job, its fun building exhaust for the most part.. purging is very important! you dont want to get that black crap sucked into the impeller of the turbo..

[jman]

Hey, Does SOLAR FLUX and similar products help with this kind of thing? I was told that it can be BETTER than purging due as it 'seals better' and comes away easily with a tube brush, not too mention that it shouldn't harm the TURBO even IF some was left on by accident... Was I mis-informed?

Cheers,

[blk mk8]

what are the red nozzles on the inlet of the turbos?

A motorcycle exhaust I did a couple of months ago. GTAW purged.


[IMG][/IMG]

[IMG][/IMG]

[jakeru]

Jman - Before welding, solar flux is a light gray powdery substance that can be readily wiped off, but after welding where it gets heated, it turns into a dark, "glassy" hard substance, and it not easy to mechanically remove. I managed brushing it off with a power cup brush with stainless bristles, and heavy pressure, but it leaves the surface "mottled" (I'm sure there is a better word... describing "tearing" the surface on a small scale), and you won't be able to access between tubes on headers with this mechanical removal method either. If you have some heat discolored stainless, and come up with a way of removing the heat discoloring, that method will likely remove the solar flux residue as well. But try and remove stainless heat discoloring mechanically sometime... (I'll bet you will find those oxides are pretty tenacious stuff!)

There are special acid mixtures I've heard can be painted on that will remove the solar flux, but I've never tried them. They seem to be extremely potent, potentially dangerous substances, with ingredients such as nitric acid. They will passivate the stainless steel (remove surface oxides and all heat discoloration) also. "Wonder Gel" is name of one such product.

I've only used solar flux with oxy acetylene stainless welding, and that combination actually worked really, really well, however it was necessary to coat the backside and frontside of the weld. With TIG's inert gas shielding the front side, it may work out OK to just put solar flux on the back side, where perhaps it won't mar the front side surface finish. I intent to try solar flux TIG stainless sometime, next opportunity I have to experiment with stainless steels. I had no adverse affects leaving solar flux residue ont he insides of automotive exhaust tubing, however I haven't given much thought before about if it would be wise to leave in there upstream of a turbine inlet. The solar flux and its after-welded residue doesn't seem to corrode anything, like aluminum welding flux does, so that's a really good thing. I don't think solar flux could be used in any food grade applications however.

[barefooter14]

riceburner.

those welds look a little to dark or "hot".. when you over heat stainless you lose the beneficial properties of the stainless. also it looks like you went with free hand and added a lot of filler. you should try and walk the cup (takes a little practice but once you get it your welds look much better and the heat input is a lot less because you move at a constant speed. either that or if your machine has pulsing you can try that to help keep the temps lower..

sorry for the off topic post.

also to the OP when welding something like this you should really be aiming for tig'in it(takes more time but in the end a near discontinuity free weld once you get the technique down and backpurge). watch some youtube videos on walking the cup practice it and with in a few hours or maybe a few tries you will get it down.

[riceburner]

Thanks for the input.

[turbocad6]

I did buy some solar flux, but haven't tried it yet, I think back purging is the way I'd prefer to go though because of the required cleanup with the solar flux, but I probably should at least try it too, gotta pick up some drygas to mix some up to try it.

I also bought a stainless steel pickling compound from bradford derustit called wondergel (LINK). just brush it on & let it sit 20-30 minutes & rinse it off, pipes just looks like perfect stainless steel with no discoloring left at all after. one of the pictures I posted above in the last picture set I posted of my practice welds showed this, the top pipe in those shots looked just like the bottom one at first, but then I put wodergel on it for just 15 minutes & you can really see the difference there, just looks like silver stainless with no discoloring



I've been practicing & practicing, I'm almost getting to the point of not sucking too much on these pipes, so far I think I can almost get a halfway decent pretty solid weld for the most part with backpurging, but not so beautiful looking just yet & not as consistent as I'd like.I'm not really overheating it too much I think, but I am getting enough heat to have the filler suck all the way in to wet the seam on the inside for the most part. still having some areas not get fully hot enough for full inner seem wetting but getting better & closer to being able to do the whole seem with that much penetration, getting better at using the pedal but at first this was giving me a lot of trouble & I'm still not very comfortable with it yet, but getting better I think.

I know that this tubular stainless pipework is one of the most difficult things to learn to tig properly, but I think in a way it's good that I learn on this, once I can get this down then other stuff will just be easy in comparison I think.

I went to a lincoln tig demo this week at my LWS as they were having a lincoln promotion going on, got to see someone actually tig weld something in front of me for the first time. he was using a scratch start with the argon valve on the torch instead of a setup like I have for this demo. I watched him start a bead on 2 pieces of mild flat butt-welded, I watched him start & then just move straight & continuous from beginning to end, all the while leading the weld bead with the filler rod & just continuously tapping it at the leading edge of the weld. when he was done I looked down to see a nice roll of dimes weld & a huge light bulb went off in my head. all along I've been thinking that the movement of the torch was what did this & what gave it that look. I think I was thinking too much about my torch movement trying to do this while welding before I realized that it was more a result of the filler itself, also noticed that he held the tungsten extremely close, almost touching but not quite.I then picked up the torch & did the same as I just seen him do & after looked down & I saw my first roll of dimes easy as crap on flat mild steel, not so easy to do when the seam wraps around in a 1.5" diameter curved pipe though, that's for sure


I also ordered & received a 4'x4' 1/8" thick sheet of aluminum, I got 5052 H34 I think? maybe it's H36, memory fails me at the moment, I just asked for something that could be bent without cracking & is weldable & this is what they recommended. gonna cut some coupons & start welding stuff up, gotta get up to speed with aluminum too to build my intercooler tanks from this stuff. & should have plenty left over for lots of other stuff too,

[turbocad6]

here are a few shots of the intercooler cores & there placement,

I custom ordered the exact sized cores that I wanted so I can build for the best fit & flow for my overall design & available space,

these cores flow enough cfm's to comfortably support 850hp+, although I probably won't ever see much over 500hp
unless I eventually wind up doing something serious about beefing up the AWD drivetrain a bit ,

airflow & efficency is going to be very high because of the side mount short tube vertical cores,
2 1/4" cores & no rear airflow restrictions like having a radiator & a/c condenser behind them mean very high heat transfer
then there's the very short pipework in plumbing these up because of the close proximity to the turbo & throttle bodies


these are good bar & plate cores, and with the 1/8" 5052 custom tanks these are gonna be a sweet intercooler setup,
now I just gotta learn how to actually weld aluminum


the bumper of course will be custom for these intercoolers, with properly ducted openings for them, that much I already know how to do

.

[Steelwill]

Nice thread. I eventually will learn to tig. The term back purge what is that ?

[Rojodiablo]

Nice thread. I eventually will learn to tig. The term back purge what is that ?

When tig welding SS tubing, Ti& mag anything, and in many cases aluminum tubing, you will want to weld in an inert environment. What backpurging means is the inside, or the backside of a weld is purged with usually argon; sometimes an argon/ helium mix depending on what you are tig welding with. If you do not backpurge, or seal the weld from outside air, you will have what is called 'carbon precipitation'. It will contaminate a weld, and it's effects can range from a slight blackening on the backside, to a pot-holed weld that is weakened.
With regards to headers on a performance motor, a lot of flow and essentially power is lost because the welds will not be smooth on the inside of the exhaust. As for sanitary food piping.... do you want food or drinks that are dragged across slag??!!!! And in pipeline assemblies, the pipe needs to be as smooth as possible to allow a cleaning 'pig' to pass thru and clean the pipeline.

Often, a really well built tubing weld will be extremely smooth on the inside, and look almost like they welded the inside of the pipe.

[Steelwill]

So you feed argon to the inside of the pipe while you weld the tubing ?

[Rojodiablo]

So you feed argon to the inside of the pipe while you weld the tubing ?

There are several ways to back purge. The easiest is to cap the tubes with some masking tape, and put a small feeder line into the tube. You don't need nearly as much flow as you do from the torch, so it's usually the norm to use a second tank, or a split regulator setup with 2 regulators; one for the welder, one for the purge line. Maybe 3-6CFH. Not the 12-18 or so for welding.
Another way is to do an 'inert box'. This is where you have literally a metal/ aluminum box that is deep enough to set your project down into a bit, and you have the feeder line into the box. Argon is heavier than air, so you start the purge line, run it for say 10-20 seconds to 'fill' the box with argon, and reach down into the box and weld. (This is a good way to weld a bunch of twisted stuff like headers. You don't need to tape everything off and move the purge line constantly. Also, brackets, hangers, etc. that will benefit from back purging can be done this way.
The third way is primarily how one would weld Ti or magnesium. A sealed welding booth, similar to a sand blasting booth. Loaded with argon, you reach in with the sealed arms, and weld away. It would not be easy to do headers like this, but it would guarantee an excellent environment for full purge welding.

[Bgbkwndo]

Ideally, the purge gas should be fed from it's own bottle and flow meter, as opposed to a
"Y" on a single bottle that feeds the TIG torch and the purge line. This prevents starving either the purge system or the TIG torch. I learned this method while welding Titanium, which is very prone to contamination when exposed to atmosphere while welding.

Just my $.02 worth. Bgbkwndo.

[jakesporting]

I wonder if the extrude hone process would have worked for
the original headers .

You might want to look into it for some future projects as it is a deburring process
that uses a putty like substance that has media in it .

I've see intake manifolds done with this process and i was really impessed at how it followed the contours of the inside.

Do a search on the web for extrudehone to learn about the process.

[turbocad6]

I've got a smith dual regulator & there is no danger of starving one for the other at all, the flow of each is easily verified with the balls in each tube & one doesn't affect the other, they work independently just like 2 bottles would & it works great, but, what sucks is having to manually open & close the purge each time you weld.

I don't open & close the flow from the regulator itself, I just adjust that once & it stays adjusted to the 7-8 cfms I set, then I mounted a ball valve in line at my welding table & can easily turn the flow on & off from there. even with this there are times that I could forget to turn it on or even off too, which sucks. in a lot of my practice sessions I've had times here & there when I did a practice weld while forgetting to turn the backpurge on. it takes a bit of time to notice the difference in the welding feel when forgetting, & at that point it's too late, ruined, if that happened in the middle of welding up an assembled header tube it would really suck, all my practicing is to develop the ability to do ideal welds with no mistakes consistently, so now I'm thinking about maybe rigging up a solenoid & timer to sync it with the weld. either that or moving the dual regulator directly to the torch gas fitting out of the dynasty, might try this but I'll really have to butcher up the torch & the regulator too, may be worth it though, but a solenoid & timer would be less chopping up of equipment. is there anything commercially or commonly available to sync the back purge with the weld?

I kind of want to build an inert chamber too, but really the backpurging I'm doing now is pretty effective when I don't forget to turn it on. now with everything I'm doing here it's really winded up just being a project in itself just tooling up for this turbo project & I'm thinking the effort I would have to put into building a nice chamber might be better just put towards getting these done adequately with just the back purging, but I do have to come up with a reliable automatic backpurge solution at least.

today for the first time I tried to weld aluminum. I fired up the plasma for it's first time & I cut a few squares & strips from the 1/8" aluminum sheet that I got. I figured aluminum would also take me a bit of time to learn & planned on having to do a lot of practice & maybe even then not being good enough to do really nice looking welds for my radiator support, well thankfully the aluminum welding turned into a piece of cake, very easy to do, very easy to control, I was getting beautiful rolls of dimes from almost the very beginning.

they say an aluminum pool is very hard to see & hard to tell when it's pooled but I was able to see it fine, I was able to control it easily. if the weld bead is a little too high then you know you need a little more heat & if it's too deep then you need to ease up a bit but it's very easy to control on this 1/8" plate at least. I was able to run long straight perfect beads quickly & going quickly winds up giving beautiful near perfect roll of dimes bead so easily, def good enough for my intercoolers & radiator support & anything else I need to do in aluminum. I was able to weld up seams even with minor gaps & poor fit up, I love how easy it is to control.

when welding aluminum I can see the reflection of the tip of my tungsten glowing in the mirror looking weld pool & just run it continuously tapping my fill rod to the tungsten reflection, surprisingly to me my beads look as nice as many of the nicer looking beads I've seen, I'm really happy that aluminum is not going to add weeks & weeks to this TIG learning process at least, although of course I'll play some more to see how different settings & adjustments affect the weld. today I just did straight 50/50 AC at 125 hz & felt no need to adjust anything, just welded like butter...

[turbocad6]

and jake, I did speak to & even sent pictures to extrude hone, but it would not have removed the slag, just polished over it. this also would do nothing for the fact that not only is the slag a problem itself, but the resulting contamination at the back of those welds also really compromised the welds themselves & they are considerably weaker, there was not much hope at that point...

[turbocad6]

ok, been practicing & practicing welding this SS schedule 10 pipe for my headers,
cutting piece after piece & then welding it back together, over & over again..
I think I'm at the point where I can get pretty good penetration throughout my welds fairly consistently,
but not 100% to the point of absolute sanitary conditions, as there are still minor voids throughout.
how perfect does this need to be to be considered ok for headers? is this good enough to be considered good? adequate?
or do I really need to keep doing this till I get it flawless with no voids at all in the whole interior of these welds?
how good is considered good here? don't know how much help I'll get with this buried in this thread but I;'ll try here first at least,
any help or advice would be much appreciated by anyone who cares to share there input














am I there yet? am I at least close? what should I be shooting for & how far off am I from good enough for headers?

[Jason280]

Great thread!

[i4sillypwr]

Still looks like you're cooking the **** out of it. Are you trying to weld it in one pass? It can be done...full pen with good root reinforcement and a decent fill on the cap side but it's tough and I've never seen or heard anyone knowledgable recommened it.It's just a pain in the *** imho. Try it in two passes. One root pass with a feather edge land, 1/16 gap, 1/16 filler and waaay less heat or waaay more speed. Then cap it off in another pass. Your progress might be discouraging at this point but you have kind of jumped straight into the deep end of the pool with this one so keep your head up. I would try to perfect thebutt weld process on 1/8in flat stock first and then tackle the insanity that is changing your torch angle while going around small pipe. Don't worry after a few years of doing it every day it gets easier :P


This is a sick project and I love it. I would totally weld the **** out of that entire project for free just to have a hand in making it a reality man. I salute you sir!