64 replies to this topic

[kikiturbo2]

first, let me just say how much I enjoy this forum.. some frighteningly smart people frequent these areas...


now, I am building my lancia delta integrale from the ground up, up to grA spec... Instead of buying a roll cage I decided to build one, as this project is as much about self build as it has to do with driving the bloody thing afterwards...

I am not sure what tubing size to use, and material to that mattter.... All the rule books say aprox. 44 mm 2.5 mm wall thickness, cold drawn seamless... but that is for Mild steel....

I wanted to use 4130 / 25Crmo4 steel, which is much stronger.. While searching for material (I can not get 4130 locally) I also found 1.7734 / 15CDV6 which is supposed to be even better than 4130... Sorry for mixing the various standards....

my question is the following... What type of material and what tubing size do the WRC / tourning cars use.?.. I want maximum stiffness/torsional rigidity, with the lowest weight penalty of course...

also, anyone familiar with 1.7734 / 15CDV6 ? I can not find welding instrucions for that steel....


cheers

vlado

[275 GTB-4]

Originally posted by kikiturbo2
now, I am building my lancia delta integrale from the ground up, up to grA spec... Instead of buying a roll cage I decided to build one, as this project is as much about self build as it has to do with driving the bloody thing afterwards...
cheers

vlado

Vlado....if you are going to use the integrale in anger....then I would seek specialist advice from a reputable roll cage maker (how much is your safety worth?).....I would also try and find a grA spec example to see how it was done as it raced (in particular, the mounting points...and yes the tube diameter, cross brace arrangements etc

Now if you are ever in Fremantle, Australia....there is this Museum...and....an ex-works lancia delta integrale lives there!

Good Luck

[kikiturbo2]

Hmm I am a bit far away from Australia...

I know how the original grA cage was done, it's design and mounting points.. I will be doing it a bit differently annyhow, more like the designs they run now, I just wanted to know how it's done now...
cheers
vlado

[McGuire]

Unless this is a hardcore, true professional racing application, mild steel tubing (such as SAE 1020) is probably a better choice. It's far cheaper, does not require heat treatment after welding like 4130 and will be faster and easier to work with in general. All chrome-moly allows is thinner wall thickness for a structure of comparable strength. For a car this size chrome-moly will save around 20-25 kg I'm guessing, but at several times the cost in materials. There will be no weight savings at all if 2.5mm minimum wall thickness is actually required.

A tip: with production sedans, often it is easier to cut off the roof, fit, fabricate and install the cage, then weld the roof back on and refinish than to try to work around the roof. This is especially true in cases where the cage is made to fit around all the production passenger car trim. If anyone was wondering how that ship-in-a-bottle was accomplished on some of the factory concept cars etc, that's how.

[Lukin]

Originally posted by McGuire
Unless this is a hardcore, true professional racing application, mild steel tubing (such as SAE 1020) is probably a better choice. It's far cheaper, does not require heat treatment after welding like 4130 and will be faster and easier to work with in general. All chrome-moly allows is thinner wall thickness for a structure of comparable strength.

Plain mild steel will be fine I think. Like McGuire said, CM will be strong, but do you need the extra strength it will bring compared to mild steel? If not then there is no gain from using CM (apart from a lighter wallet and bigger headache). CM and MS have the same modulus of elasticity so there will be no difference in the torsional rigidity of the cage with either material.

[Ben]

I have also heard that 4130 being less ductile than mild can split in an impact and is therefore not a good idea for roll cages.

In the UK people use T45 which is a chrime moly steel but slightly more ductile than 4130 and doesn't require heat treatment of welds. You can also weld it to mild.

Ben

[BorderReiver]

My last few briSCA's have all had T45 cages, nice to work with, safe and reliable.

[kikiturbo2]

well, I have heard about T45 a lot, but is seems to be a british phenomenon, as I can not get it anywhere but the UK....

luckily... 1.7734/15CDV6 looks to have the same properties... I just can not get any cross reference to see if it is the same material.... i.e. it can be welded with no post weld heat treatment, it is stronger than 4130, etc. etc...... It seems that German race car builders use it for the cages....

cheers

vlado

[Greg Locock]

I'd go for mild steel. However T45 has a UTS of 800 N/mm^2, and about 20% elongation, which makes me think it is very similar to a heat treated 0.45% carbon steel covered by DIN 17200.

There is some logic in going for a high yield stress steel - in a roll you don't want the roof to crush. That's different to a frontal collision where you want crush to absorb some energy.

[kikiturbo2]

There is some logic in going for a high yield stress steel - in a roll you don't want the roof to crush. That's different to a frontal collision where you want crush to absorb some energy.

true, but quite a lot is accomplished with proper design..... Nowdays, there are customary two pieces of tubing going from the top of main hoop to the floor.. giving more strength to the main hoop in case of roll over... I did not use to see that in old rally cars...

roll cage is connected to the rear side of the front turrets, leaving everything from there on, to the front of the car , practically OE... same thing in the back... so basic crush structures are kept.. .Having said that, I would not want to be in a serious crask in an integrale, it is not the best put togeather vehicle, plus the shell is, basically, from 1300 cc, 1980's Delta....

having seen the complexity of the new cages I can only conclude that nowdays they run some seriously thin, high strength tubing with added links trought the car, giving it added shell stiffness (torsional rigidity..)

I mean, just look at the focus WRC, now that is a cage you could not see in the late 80's..

cheers

vlado

[kikiturbo2]

another thing.... I appretiate the "mild steel" idea.... but as I can not get even mild steel in seamless CDS... I have to shop outside... so I just might as well buy something better.....


still waiting for some quotes from germany... then we will see...


v.

[McGuire]

Though they are both suitable for this application, T45 and 15CDV6 are different materials. T45 is a chrome-manganese steel that is mainly/only available in the UK. 15CDV6 is a chrome-moly steel, sort of an EC version of SAE 4130...rather different composition and properties, but used in many of the same aircraft/motorsports applications as 4130 in the States. It does have one advantage over 4130 in not requiring normalizing after welding. I presume the price would be similar...in other words expensive, and hard to justify for a roll cage or space frame (for me anyway) but excellent for uprights, wishbones etc.

The vast majority of race car chassis and roll cages here in the States are fabricated using the SAE 1010/1020 range of mild steels. The material is inexpensive, relatively easy to weld and work and produces a very rigid yet tough structure. A roll cage for an application like this would use approximately 1.75" x .125" wall thickness 1020 or 1026 seamless tubing. With proper construction techniques safety is beyond question. The NASCAR stock cars you see taking multiple flips and barrel rolls at 190 mph are constructed just this way.

A word to the assembled about 4130...if you do not normalize the welds, the product will be effectively no stronger than mild steel and all the extra $$$ basically goes down the drain. This especially applies with TIG due to the narrower heat zone. Also, make sure you are never using chrome-moly hydraulic tubing in structural applications. It's fully annealed (for mandrel bending) and may as well be pasta.

[JtP]

I don't go for the cutting the roof off bit.

I have a friend who had the same problem of completing welds on a BMW cage. So you could use his technique, tack weld the cage aligned in the car and then cut holes in the floor around the base of the cage and drop the cage downwards out the car giving enough room on top of the cage to finish the welding.

[kikiturbo2]

So you could use his technique, tack weld the cage aligned in the car and then cut holes in the floor around the base of the cage and drop the cage downwards out the car giving enough room on top of the cage to finish the welding.

I wanted to do this first... but, as my shell currently has a sunroof I am thinking of closing the hole and removing the mechanism etc... in which case, if the normal roof does not cost an arm and a leg, I might change the whole roof skin anyway... also, roofs on integrales are a common rust problem area....

[J. Edlund]

Tubes made from for example T45 is also a lot more difficult to bend than mild steel tubes. This means that just bending alone can increase cost quite a lot, infact in some cases a mockup is made from mild steel, those are then sent to whoever can bend T45 tubing which then make the T45 tubing in the same dimensions.

To take full advantage of the higher strength the tubes must also be cut for a prefect fit, which requires some special tools.

[McGuire]

Originally posted by J. Edlund
Tubes made from for example T45 is also a lot more difficult to bend than mild steel tubes. This means that just bending alone can increase cost quite a lot, infact in some cases a mockup is made from mild steel, those are then sent to whoever can bend T45 tubing which then make the T45 tubing in the same dimensions.

To take full advantage of the higher strength the tubes must also be cut for a prefect fit, which requires some special tools.

This is an excellent point. I have never worked with T45, but chrome-moly alloys such as 4130 are much harder to work than mild steel. They will wear out your tools faster too.

Also an excellent point on the importance of fit in assembly, no matter what the materials. You want what is called a "water-tight fit" on all joints prior to welding -- no visible gaps whatsoever. For mitering round tubing you really need a milling machine, or a lot of patience. There are cutting jigs available which mount on bench or vise and use a hole saw and power drill, which work pretty decent with mild steel. If you are doing much chassis fabrication in chrome-moly, a CNC plasma cutter is an unbelievable blessing. Of course not many home shops can afford one, but prices on this stuff are coming down all the time. I believe there is now a kit out there for around $6000.

[JwS]

Someone previously said that the weld joint of non normalized 4130 would be the same strength as mild steel, I don't have alot of direct experience but I do have a degree (ok a minor) in metallurgy. What I understood was that if you don't normalize the welds they will actually be relatively brittle? I think Carroll Smith talks about this also.
JwS

[McGuire]

Originally posted by JwS
Someone previously said that the weld joint of non normalized 4130 would be the same strength as mild steel, I don't have alot of direct experience but I do have a degree (ok a minor) in metallurgy. What I understood was that if you don't normalize the welds they will actually be relatively brittle? I think Carroll Smith talks about this also.
JwS

Believe it or not, this is a rather controversial subject with some divergence of opinion. 4130 was originally developed for the aircraft industry, and specifically devised for gas welding. Fully normalized, it has a rated tensile strength of 90,000 PSI, while the mild steels are in the 30-50,000 PSI range. (Somebody fix me if I recall the numbers wrong.) It is said that an un-fully normalized 4130 structure may have something like a half to a third of its potential strength.

The question is: what is fully normalized? Just as you say, the problem is with hardening or brittleness in the weld and adjacent area. Some authorities say the weldment must be preheated to 600 F and then reheated to 1200 F post-weld then air-cooled. Others say 300-400 F preheat. Others say one or the other (but not the other) is required. Others say these procedures cannot be successfully controlled in the field, especially in large structures or limited access, so there is essentially no point, while others say 4130 is fully air normalizing in ordinary conditions, therefore it's not an issue. In the commercial aviation industry, they MIG weld without any additional pre-or-post steps at all and appear to have no problems. That may be a result of the relative stress levels as anything else. Also, it should be noted that different filler materials are recommended for different normalizing procedures.

The only wise choice for any individual is to study the issue carefully and decide for oneself who is telling it straight. Since you have some metallurgical schooling I expect you can sort out the issues AT LEAST as well as I can, and I would be interested to know your take should you choose to wade through all that.

There is general agreement that for material thicknesses under .060", it is not super critical. Also, gas welding tended to mollify all these issues somewhat with its wider, more unfocused heating zone and more gradual heating/cooling rate. A lot of aircraft were gaswelded from 4130 in the second world war without normalizing, with no trouble at all. But with TIG welding you get a high, quick heat concentration over a narrow area, and if you have a number of welds concentrated in one small area, around mountings etc, you can see how a real structural problem could very well result.

Smith was in the must-normalize camp, and I am too. As I see it, where you need 4130 (control arms, radius rods, etc.) you have a high concentration of welds clustered together, but you also have a workable component size and the normalizing processes can be precisely controlled. And while there is some dispute as to the need for normalizing, I have yet to see any claims or evidence of adverse effect. On larger components with lower specific loadings, like the chassis/rollcage itself, you can avoid trouble altogether and use 1020. Then you always know what you have.

[Greg Locock]

Good.

I'm glad to see some discussion of this.

We spend a lot of money (which pays for itself so maybe it isn't a lot of money at all!) on high yield strength steels, say 400 or 500 MPa.

Mostly we spot weld, but also MIG. Typical section thickness is 1.8 mm say on a subframe.

Obviously we use steels designed for this job, and equally obviously we don't do any post welding heat treatment, and equally obviously we do check (or rather our customers check) that the resulting structure is strong enough.

Here's the Greg theory: I suspect that ANY normal high temperature welding op removes all the benefits of the steel. It softens it back to 200 N/mm^2.

But, this is an extremely localised soft zone, and will rapidly work harden in service, by yielding. Since it is only a small area the resulting distortion is slight.

In some ways it is a bit like Gregg's femur - if it isn't stressed it stays soft, if it is stressed it hardens up.

But tis only a theory, someone with a Vickers machine could test it, and I've no doubt someone at work has, but I wouldn't be able to tell you about it if they have!

[McGuire]

I neglected to mention that if very much bending or forming is required with 4130, for example in a roll hoop or side bars or in sheet metal fabrications, the material must be obtained in annealed state for working. In that case heat treatment is absolutely required, no question about it.

[ZoRG]

Have anyone tried to tig weld some thinwall 4130 typically used in roll-cage construction and see how "strong" or "brittle" it is, it's all in using the correct filler material. You will be surprised, if you use the correct filler you will have no brittleness at all, and proper strength. I think it comes down to the fabricator in the end.

[Greg Locock]

I don't think anyone has suggested that it would get brittle. If anything we're expecting the opposite - excessive ductility and softness.

[dosco]

Originally posted by ZoRG
Have anyone tried to tig weld some thinwall 4130 typically used in roll-cage construction and see how "strong" or "brittle" it is, it's all in using the correct filler material. You will be surprised, if you use the correct filler you will have no brittleness at all, and proper strength. I think it comes down to the fabricator in the end.

Well, you say "thinwall" and that's exactly the point.

If, for example, you are using 1/8 inch thick wall and welding it to a 1/4 inch thick plate, post-weld heat-treat (PWHT) may be required, depending on the design requirements (anneal versus anneal and "age").

I deal with stainless, nickel alloys and titanium where I work, and they each have their peculiarities.

Stainless 321 and 347 require PWHT if they are going into an corrosive (acidic) environment....otherwise the HAZ will have carbide precipitates that will dissolve when in contact with the acid. Another possibility is post-weld machining.....generally a full anneal is a good idea if you are going to machine the part after welding - particularly if there is high residual stress due to poor fitup (and I see crappy fitup ALL THE TIME - poor fitup is UNACCEPTABLE).

Nickel alloys like 625 (that are not age-hardenable) generally do not require PWHT. Age hardenable alloys like 718 usually require a PWHT to put the "hardeners" back into solution. Otherwise undesirable structural effects such as strain-age cracking can occur when the part is loaded b/c the HAZ is hardened but the remainder of the base material and weld reinforcement are still "soft."

Ti alloys usually get a PWHT, but it depends a lot on alloy and part design.

As far as the importance of filler wire, it's pretty much not as much of a big deal as the welding suppliers tell you it is - as long as the filler is matched to the weld (especially in dissimilar material welds).

I'm not all that familiar with the weld characteristics of 4130 otr 4140, but both materials are considered "alloy" which means there are constituent elements that may come out of solution after welding, which may require PWHT to restore mechnical properties. For a space-frame, PWHT would most likely require a heat treat fixture to hold everything in the correct alignment during HT otherwise you may have a pringle coming out of the oven that is useless and will require substantial rework to fit into your car.

That said, many of the materials with a higher yield point may not be desirable due to brittle fracture characteristics, but the charpy impact data should explain the suitability of the material to the purpose. Also I'd say that if a 1000-series material will do the same job for a few pounds of weight but cheaper, stay away from the alloy stuff due to difficulty in fabrication and the possibility of post-weld cracking....

People tend to get stars in their eyes about these high-strength materials.....4140 is great for an airplane where weight is critical, but if you can pump out a few extra HP the weight of 1000-series may be tolerable.

[JwS]

I think the problem actually is that the weld zone is brittle if it is not normalized after welding. This is less of a problem if you use a torch because more heat is put into the part and the weld zone cools more slowly. If it is Tig welded the welding is quick and much less heat is put in, so the surrounding metal is still cool and it quenches the weld which makes it brittle.
The normalising procedure involves heating red hot (not sure what temp) and then cooling in still air or a sand filled box.
If it were aluminum the work hardening concept works better, Al alloys (most) are soft when quenched and then gain hardness with age and HT, or cold working. Steels are generally the opposite, they are quenched to a hard brittle state, and then HT to soften them to a workable state.
If a speciallized filler metal is used you can avoid some of these problems, but you probably wont really have full strength, mild steel is often used when welding 4130, so you have a mild steel joint in the end.
JwS

[dosco]

Originally posted by JwS
[B]I think the problem actually is that the weld zone is brittle if it is not normalized after welding.

I'm not familiar enough with the phase diagram of 4130 or 4140 to know if it is a precipitation-hardenable meterial, however, the chrome-moly steels (from what I've read) are generally considered "austenitic" and therefore (I would think) the 4000 series steels are not age-hardenable (i.e. PWHT aging will not "harden" the material as a function of carbide precipitates coming out of solid solution). But then again, not having a phase diagram in front of me, this is only educated guessing.

Additionally, I would agree that the chrome content would tend to cause precipitates at the HAZ and cause brittleness....therefore requiring some type of PWHT (although I find the term "normalizing" to be somewhat misleading.....I think "solid solution annealing" is probably a better term). With that said, I'd be suspect of "torch annealing" because of the lack of process control as well as localized (as opposed to global) heating.............an oven (along with HT fixture) would be the best bet.

Conclusion: kikiturbo2 should use 1000-series steel for his roll cage.

[dosco]

Originally posted by JwS
If it is Tig welded the welding is quick and much less heat is put in, so the surrounding metal is still cool and it quenches the weld which makes it brittle.

Also....generally in the thinner sections I've seen under the microscope, the middle of the weld has nice, large, "equiaxed" grains while the HAZ is filled with smaller "quenched" grains. In the transition area the grains are usually "dendritic" and are a kind of cross between equiaxed and small.......

This phenomena also explains cracks that are often seen in the middle of "high aspect ratio" weld beads (thin/narrow weld compared to thick section). E-beam welds that are not preheated are generally susceptible to this (and one of the parts I was involved with suffered this problem).

[JwS]

I agree that the term normalizing is kind of misleading, but it is the term they use for the condition of 4130, so I guess we have to use them.
Torch methods of HT are of course not the best, but what I remember, I think, came from instructions for 'field repair' of aircraft frames.
Yes, a phase diagram would answer many questions, and I don't doubt that if you found the right person they would be able to tell you exactly what to do. I am almost sure that 4130 is not an age hardening alloy.
The metallurgy of a weld bead and HAZ is a study all to itself, I took a course in welding, and when you see all of the problems you can get it makes you scared to weld anything but low C steel or the like.
Now do you see why we recommend low C steel! ugh...

JwS

[JwS]

I just did a quick search and found this

http://www.lincolnel...chrome-moly.asp

Which seems to cover the welding aspects of 4130,
JwS

[dosco]

I'm somewhat suprised that Lincoln would call it "TIG" as opposed to its AWS designation of "GTAW."

Hm.

[kikiturbo2]

well, firstly, I plan on using MIG, and use gussets where ever I can. Gussets are time consuming, but my years of experience riding and breaking thin walled aluminium and steel MTB frames has tought me that gussets work.

From what I have read in books dealing with car frame construction it looks to me as if post weld heat treatment is not as important as it seems given propper gusseting....

most open problem in connection with CrMo tubing in roll cages is the lack of "toughness" in the material... as in it's fondness for breaking and not bending under stress.

also, my major concern is bending the stuff, so I will first have a good look around for people that are able to make a clean bend in CrMo.....
mild steel is still on the list... :-)

cheers guys

vlado

p.s. as for Lincoln using "TIG"..... I guess it is just to make it more popular... usually big welding machine manufacturers use both in the same sentence....

[Engineguy]

Originally posted by dosco
I'm somewhat suprised that Lincoln would call it "TIG" as opposed to its AWS designation of "GTAW."

Hm.

I suspect that if you gave me a penny every time the words tig and mig are spoken by a welder, and I gave you a dollar for every time gee-tee-A-dubba-you or gee-emm-A-dubba-you are spoken by a welder.... I'd be a rich man. And they are selling their product to welders, of course, not the AWS.

[dosco]

Originally posted by kikiturbo2
well, firstly, I plan on using MIG, and use gussets where ever I can. Gussets are time consuming, but my years of experience riding and breaking thin walled aluminium and steel MTB frames has tought me that gussets work.

From what I have read in books dealing with car frame construction it looks to me as if post weld heat treatment is not as important as it seems given propper gusseting....

1. Thin wall bicycle frames are what - 1 or 2 mm wall thickness, max? I would imagine that you'd be using something like .125 inch wall for a car (don't know off the top of my head what that is in mm). I would agree, though, that gussets are important for good load transfer between tubes.

2. PWHT isn't as important if there is enough material cross-sectional area in the built-up region of the gusset (ie the design margin between max design load and max material stress is large). So if you have a joint with few gussets or structural members, you could have a weak spot in the frame.

3. Using MIG? Make sure the joints are well fitted, clean, and your MIG setup is correct. MIG units are easy to "point and shoot" but I have seen plenty of crappy MIG welds (cold lap, lack of fusion, oxidation, rampout cratering, etc). There are universal rules to welding, regardless if the method used to weld is stick, MIG, TIG, , gas, SAW, etc etc. Remember: well fit and clean.

most open problem in connection with CrMo tubing in roll cages is the lack of "toughness" in the material... as in it's fondness for breaking and not bending under stress.

Exactly. Thus the need for PWHT (anneal) to remove all work hardening from tube forming operations as well as recrystallizing the HAZ areas in the welds.

mild steel is still on the list... :-)

Seems like the best/simplest option to me.

p.s. as for Lincoln using "TIG"..... I guess it is just to make it more popular... usually big welding machine manufacturers use both in the same sentence....

TIG is a common acronym used for the GTAW process......in retrospect TIG is thrown around more than GTAW, so perhaps I've had my nose in the AWS books for too long.

[dosco]

Originally posted by Engineguy


I suspect that if you gave me a penny every time the words tig and mig are spoken by a welder, and I gave you a dollar for every time gee-tee-A-dubba-you or gee-emm-A-dubba-you are spoken by a welder.... I'd be a rich man. And they are selling their product to welders, of course, not the AWS.

You're right.

Sadly, the level of technical competence of welders is pretty low. Many are craftmen, but the majority are hacks.

Since I've become involved with welding (and weld inspection, weld qualification, etc etc) in the last 4 years, I'm constantly amazed to see horrible welds of all kinds, on all kinds of products. I was especially suprised to see some nasty looking welds on a Space Shuttle Main Engine that was on display at KSC. Amazing.

Ah well.

[Greg Locock]

Maybe its just rockets. The Ariadne museum in Toulouse is a terrific display case of dubious optimisation and workmanship.

[ZoRG]

Originally posted by dosco


Well, you say "thinwall" and that's exactly the point.

If, for example, you are using 1/8 inch thick wall and welding it to a 1/4 inch thick plate, post-weld heat-treat (PWHT) may be required, depending on the design requirements (anneal versus anneal and "age").

Hi

Not sure if you are aware of this, but in roll cages, they don't use thick wall 4130, only in the tubes, the plates/boxes are mild steel.

No weld in cage that I know of are normalised/post-weld heat treated, maybe for rally/dtm type teams with unlimited budgets that can afford a oven big enough to fit a entire car in, but according to SFI spec's post weld heat treatment is not required.

[ZoRG]

Originally posted by dosco


1. Thin wall bicycle frames are what - 1 or 2 mm wall thickness, max? I would imagine that you'd be using something like .125 inch wall for a car (don't know off the top of my head what that is in mm). I would agree, though, that gussets are important for good load transfer between tubes.

Actually the thickest for 4130 is .08 inch for the main beams and thinner for the rest. Mild steel is .125 inch. This is what I meant by thin walled.

Also, you are not allowed to mig it, afaik all forms of racing requires it be tigged.

EDIT: I should mention that either ER-70S-x or ER-80S-x are used on roll cages, I am almost 100% sure that ER-70S is used on just about every dragster and funny car out there.

[kikiturbo2]

when people talk about post weld heat treat on 4130 cages, it invlolves only using oxyacetilene flame to heat up the weld area to a certain temperature and then let it slowly cool off... I have never seen any mention of heating whole cages/chasis in an oven... The problem lies in the focused nature of TIG welding ,as the heated zone is small enough to be rapidly cooled by the surrounding metal....

vlado

[ZoRG]

I see.

Still not a very common practice nevertheless it is actually quite uncommon, if you follow the link posted earlier, you will notice that post weld heat treatment is only required with 4130 filler, of which I have yet to hear someone actually use in the construction of roll cages.

[kikiturbo2]

true,

I just opened my post after writing that and received some data sheets from the people selling tubing.... anyway, I quote: 25CrMo4, welding: without pre or post heating, MIG or TIG with CrMo1 or higher strength 1.7734.2 wire........ 25CrMo4 being european equivalent to 4130....

these people, Tennant Metall, are europes biggest sellers of CrMo for motorsport aplications.....

so, if the price will be right, I just might bite the bullit and save myself 10 kg on that cage after all.... :-)

vlado

[ZoRG]

If you decide to do it, would you be so kind to post pics of it when you get a chance?

[kikiturbo2]

i will have a build diary website, do not worry.... :-) but do not expect it anytime soon.... :-( I am planing to do the cage in the next 4-6 monts... too much other work I am affraid...

[GBarclay]

good luck with the cage.

Seems like a lot more metallurgists on this thread than welders.

I build cages in the USA mostly to SCCA specs.

My first question would be what series and what are the current rules?

One problem I notice in the rally world. Canada requires FIA specs but will sanction cars that meet SCCA specs. FIA specs mean the rollcage tubing needs to be 1 5/8" with a .120 wall thickness. 1 5/8 tubing is pretty rare in the US and is priced accordingly. 1 1/2 and 1 3/4 are popular and the price reflects it.

Unless the budget is very generous, or the class is extremely competitive, I'd stay away from Cro-Mo. Not purely because of the normalization required ( I am in the Caroll Smith corner).

With Cro-Mo you need to have the welds done with TiG rather than Mig. Cleanliness and tube fitment are even more important when dealing with Tig, and require a comparable increase in the preparation time. tig is also much slower than laying a few Mig beads. So your labor/welding budget is comparably higher.

We have used powdercoating ovens for normalizing large chassis structures, but an oxy/acet torch works fine. Heat the weld area to the desired temp, we use welding crayons (markers that change color at a certain temp), but a good cherry red is fine. Then let cool in still air. Can be very time consuming which eats into a budget quickly.

Perfect world - CAD designed, FEA tested, CAM cut and bent, Cro-mo tube, tig welded, normalized chassis.

real world - designed to fit car, DOM tube, Mig welded, fits your budget.

Everything eventually leads to how much budget you have to work with.

If this is your first cage, I'd seek the advice of an experienced builder/fabricator

[kikiturbo2]

hey, nice to have a fabricator finally.... welcome to the tubing party.. :-)

realistically speaking.... the car is to be a track "toy", and maybe I will use it for hill climbs and in an ideal world, I will be using it on tarmac rallies but as a "zero" car... the car is too old to be homologated for serios work, and my budget will not allow me to race seriously...

the primary purpose of the cage is chasis stifness, and that is why I choose to do it myself as cages I can buy from Sparco and OMP are inferior in that respect... also, i will connect the body to the cage in much more places... The car, delta integrale, has a seroius problem with torsional chasis stifness, so I am striving for a major improvement in that respect...

Quite frankly, right now I am planning for 45 mm Cro-Mo, 2.5 mm for the main hoop, two bent tubes going from the main hoop to the front and down the A pillars + the tube going across the windscreen opening... the rest, and that includes X on the doors, across the roof, connections to the front and rear suspension turrets, a big X in the back, and other smaller tubes, I want to do in 1.5 mm.... Gussets everywhere of course....

I have time, and I want to learn to do it right, so DIY is the way for me....

[GBarclay]

I have a limited knowledge of Lancia's but assume your car is similar to the Delta Ingtegrale's that used to compete in WRC.

Given your "track toy" with hillclimb use, here would be my take.

Main hoop - behind driver/codriver - 1 3/4" x 0.120 wall DOM or equivalent (sounds like 45mm tube dia). So you are very close using 45mm x 2.5mm Cro-mo

all other tubing - 1 1/2" x 0.120 wall DOM or equivalent. approx 38mm x 3mm

all straight tube not on perimeter of car - 1 1/2" x 0.095 wall. (like x-brace in main hoop) approx 38mm x 2.5mm. Using CroMo you could go as thin as 1.7mm

This would meet or exceed specs for rally/hillclimb cars in the US. as long as the car stays below 2800 lbs.

I usually exceed the specs in the name of safety unless weight is the primary concern. Good triangulation, good anchor points, good gussets - all help improve any cage.

Cro-Mo is twice as expensive as DOM, but even in a full cage you are only talking about a $500 difference in material cost. Labor cost will also be more.

As most of your concern, aside from safety, seems to be chassis stiffness, I would try to tie in as much of the suspension as possible. Certainly tie into the strut towers front and rear if possible. I also attempt to tie the base of the main hoop into the suspension pick up points for the rear axle. Again, assuming independent rear with some combo of links.

Good luck and keep us posted.

[kikiturbo2]

thanks....

the car is the Delta Integrale that dominated rallying for years... :-)

thanks for the tubing sizes, that is exactly what I needed....

the thing that led me to Cro-Mo is that I do not have mild steel DOM tubing readily availabe locally, so I will have to import stuff from germany anyway, in which case, considering that transport alone will be in the region of 200-400 $ I can go and buy higher quality tubing.... I am trying to get as much info as possible, as I will be getting some other tubing (for manufacturing suspension arms etc...) at the same time, to cut transport costs....

I will try to keep the car under 2400 lbs.

vlado

[dosco]

Originally posted by GBarclay
With Cro-Mo you need to have the welds done with TiG rather than Mig.

Why?

Cleanliness and tube fitment are even more important when dealing with Tig, and require a comparable increase in the preparation time. tig is also much slower than laying a few Mig beads. So your labor/welding budget is comparably higher.

Again, why exactly? I could see skipping the cleaning if you were going to stick weld some mild steel....generally speaking, though, cleanliness is "next to godliness" in welding fab.

We have used powdercoating ovens for normalizing large chassis structures, but an oxy/acet torch works fine.

This depends on exactly how you want to control your post-weld, post-heat treat mechanical characteristics.

If you're going to engineer the crap out of your fram (ie go for minimal wall thickness), do it correctly and go for the oven HT. BTW, all this jabber about "an oven big enough for XXX" is silly. Ever see a vacuum heat treat furnace? We've got 3 of the smaller/cheaper ones where I work and they're mighty big. I could easily fit a roll cage in one of them....and they're not exactly a rarity.

If your going to "swag it" and go with a "commonly accepted wall thickness" that leaves you some margine, then the torch process is fine.

Perfect world - CAD designed, FEA tested, CAM cut and bent, Cro-mo tube, tig welded, normalized chassis.

real world - designed to fit car, DOM tube, Mig welded, fits your budget.

Well, if "real world" means "make something and hope it saves your ass in a crash," feel free to solder up some copper pipe and bolt it to your car with some SAE 1/4-20 UNC bolts and don't worry about it.

If this is your first cage, I'd seek the advice of an experienced builder/fabricator

Good idea.

[dosco]

Originally posted by ZoRG
Not sure if you are aware of this, but in roll cages, they don't use thick wall 4130, only in the tubes, the plates/boxes are mild steel.

Uhm, I was giving an example. Thanks for the enlightenment, though.

No weld in cage that I know of are normalised/post-weld heat treated, maybe for rally/dtm type teams with unlimited budgets that can afford a oven big enough to fit a entire car in, but according to SFI spec's post weld heat treatment is not required.

As I mention in another post, it all depends on what you want out of your structure.

And as far as "an oven big enough to fit a car in," there are lots and lots of HT furnaces around, and the majority of them *are* big enough to fit a car in. And budget isn't really all that much of a big deal - you weld your structure and send the weldment out for HT as required to a HT shop that is certified to the necessary specification. Not a big deal - really.

[dosco]

Originally posted by ZoRG
Also, you are not allowed to mig it, afaik all forms of racing requires it be tigged.

EDIT: I should mention that either ER-70S-x or ER-80S-x are used on roll cages, I am almost 100% sure that ER-70S is used on just about every dragster and funny car out there.

This is the second time I've heard this.

Any particular reason for TIG over MIG? Doesn't make any sense to me.

[dosco]

Originally posted by ZoRG
you will notice that post weld heat treatment is only required with 4130 filler, of which I have yet to hear someone actually use in the construction of roll cages.

So "normalization" is only required for what, exactly?

And why is there any discussion of "normalization" if its only a requirement for 4130 filler?

Seems like much ado about nothing.....

[ZoRG]

Originally posted by dosco


This is the second time I've heard this.

Any particular reason for TIG over MIG? Doesn't make any sense to me.

I have no idea, you will need to speak to either NHRA or SFI to ask them why only TIG is allowed.

Originally posted by dosco


So "normalization" is only required for what, exactly?

And why is there any discussion of "normalization" if its only a requirement for 4130 filler?

Seems like much ado about nothing.....

From lincoln's site:
Q. Why is 4130 filler metal not recommended?
A. 4130 filler typically is used for applications where the weld will be heat treated. Due to its higher hardness and reduced elongation, it is not recommended for sporting applications such as experimental airplanes, race car frames, roll cages, etc.

Q. Do I need to heat treat (stress relieve) 4130 after welding?
A. Thin wall tubing normally does not require stress relief. For parts thicker than .120", stress-relieving is recommended and 1,100ºF is the optimum temperature for tubing applications. An Oxy/Acetylene torch with neutral flame can be used. It should be oscillated to avoid hot spots.