Fuel tank metal & welding metal

[Sat]

I have been researching metals and alloys for building the fuel tank and I came across chromoly (alloy) 4130. I was also planning to use aluminum as the welding metal. Any suggestions.

[Sat]

I have been researching metals and alloys for building the fuel tank and I came across chromoly (alloy) 4130. I was also planning to use aluminum as the welding metal. Any suggestions.

[KU_Racing]

I would suggest that anything that isnt welded to the frame be made of aluminum- save yourself a huge amount of weight that way. If you have a good enough welder, make your tank out or .015 aluminum. Super light, and strong enough to keep the fuel in.

[twpvw5]

Sat, Aluminum is what most teams use for their fuel tank, the only reason(s) I can think of not to are weldability, or if you are planning on running E85 as it destroys aluminum.

[Jersey Tom]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by KURacing:
If you have a good enough welder, make your tank out or .015 aluminum. </div></BLOCKQUOTE>

Holy crap that's thin. I'd be amazed to weld that thin of steel, much less aluminum.

[KU_Racing]

yeah.. but you should see the tank. It swells when we put fuel in it http://fsae.com/groupee_common/emoticons/icon_cool.gif

[Wright D]

You can Weld together your own plastic tank, it is the lightest way to go by far. Welding plastic is simple, and won't get eaten by E-85. Just make sure you don't get the fuel cell too hot. Custom plastics tanks are made all of the time for chemical distribution for the semi conductor industry and food industry. I have never seen one at a fsae competition, but then again I might not have been looking in the right places. I have seen custom plastic tanks on other race cars though.

You could also make a bladder type tank from silicon and fiber glass or Kevlar. Last year we made a Kevlar and black silicone crash bladder around our .02 thick aluminum tank. I think if you have a decent mold you could make a tank just from Kevlar and silicon. We used the kind of silicon that comes in a big tube; the calking gun type. Gasoline eats silicone, so you would have to use anther substance, but the lay up is just about the same as with any other composite.

Just a couple of different ideas on the same topic...

[Kyle Walther]

KUracing-
how do you secure your tank to the frame, and do you do any testing for crash worthyness(sp).

[KU_Racing]

the tank has brackets welded to the outside, which are then bolted to the frame. The .015 aluminum sounds very thin, but it really isnt- the burst strength is plenty high enough for it to be safely used as a fuel cell. In our car it is located directly behind the drivers seat, on the centerline of the chassis. Little chance it would be struck directly in a crash. The structure of the tank itself is strong enough to prevent the tank from rupturing from g-loads or from the frame bending.

As far as the other suggestions, the welded plastic sounds interesting, although I would guess that our setup is lighter. I definately like the idea of a composite kevlar tank- seems like the ultimate in safety to me. Also because I would imagine that using kevlar opens up a lot of possibilities as far as shaping the fuel tank to fit around other parts.

[Kyle Walther]

KURacing-
how do you "know" it is strong enough. do you do any physical testing from say a 20g side impact with a full tank of gas? .015 is thin but it's even thinner when the weld undercuts.

[KU_Racing]

No, we have not done any physical testing of the tank. However, we know the capacity of the tank, so we know what the weight will be when the tank is full. We know the burst strength of the material. The burst strength is significantly higher than what would be necessary to contain the fuel at 20 g's, up to a safety factor of 2 or so. The brackets are designed under the same principle, to hold the tank without puncturing it up to the necessary loads. As far as losing thickness when the material is welded, it is a matter of not using a material or process unless you can do it properly, just like any other part of the car.

[-]

If you take a look back in history, I seem to remember some publicly posted pictures of a crash that Cornell had a few weeks before the Detroit competition (2004 I believe). In the pictures their tank, which is to the right of the driver, was hit slightly and the SS welded seam at the bottom came undone and spilled all of the fuel out.

Needless to say, I have made tanks out of 0.065" or so think aluminum sheet in the past without any major headaches. I wanted to try to make one out of stainless, but I just never had the chance, I am sure it would have done just as well though.

The one thing you want to watch out for are pinholes in the welded seams, especially in the aluminum. For a good welder, this won't be a problem usually, but for someone just learning TIG it can be frustrating building a tank and then going to pressure test it (10-20 psi.. no more is necessary) and find it full of holes.

Good luck

[EliseS2]

We used .020 aluminum, I believe 6061. We ended up bonding some fiberglass along the seems, the very light stuff that is used for model airplanes. We did that because of fatigue and leak concerns.

[Jersey Tom]

.063" aluminum sheet..

[Hardrocker Motorsports]

.063" aluminum sheet, with internal foam, for E-85. We have had the alchy sitting in a "test" tank for over 3 months now and no signs of corrosion.

[Steve Yao]

Hardrocker,
Is the test unit grounded or insulated in this tank? Temperature?

The fuel tank will eventually be grounded to the chassis and probably somewhat near some headers, correct? Both of these issues should be considered in corrosion testing i think...

[Hardrocker Motorsports]

Steve,

We usually dont insulate our tank mounts. Our test subject is just sitting at room temp in one of our labs on a metal shelf. Not a test I'd show off to the judges, but gives us some peace of mind.

[Bill Kunst]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Posted March 06, 2006 10:26 AM
I have been researching metals and alloys for building the fuel tank and I came across chromoly (alloy) 4130. I was also planning to use aluminum as the welding metal. Any suggestions. </div></BLOCKQUOTE>

???? Did you come across a supply of 4130 in a dumpster???? Where you going to then weld this together with aluminum?

I worry that you are trying to throw this out as if you have done some research. Go with SS or Al or etc, design it, test it, run it. Tell judges why you did what you did, save world, go home and watch tv.

As for those welding the massless tank systems, may I suggest safety over "really freakin' light". I am a teacher, who was once a FSAEr, and have both work in sheet metal and taught it. Any tank that bulges when filling would give me the puckered anus effect while driving that vehicle. I would love to see the FEA of this tank and the forces on it. Also, unless it receives heat treatment, we can assume a stout scissors may just cut you tank in half. Sorry for the digs, but I have been Alumi'n it lately and have been running in to the "safety factor of five.317 give or take 4 1/2 within in +or- .0000001mm" and this seems to piss me off.
Bill

[Jersey Tom]

Every ounce of weight counts. But I wouldn't skimp on the thickness of a fuel tank.

[Steve Yao]

HR,
I agree your test is better than nothing. I am just concerned by the excluding of factors like electrical conduction, heat, and constant fluid movement. All of which could hasten corrosion from months to weeks. Although, I admit I don't recall my chemistry in all that detail.

The only anecdote I really have is an old aluminum-body fuel regulator we had on a shelf back at New Mexico that evidenced significant pitting after a season or two running on ethanol.

Seems like your tank would last through competition at least, I have no idea beyond that.

[KU_Racing]

the tank doesnt really swell when there is fuel in it. jeez. we also line the tank with the same industrial latex crap that lines the tanks of all of ur hondas and toyotas and whatnot.

[packtim]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jersey Tom:
Every ounce of weight counts. But I wouldn't skimp on the thickness of a fuel tank. </div></BLOCKQUOTE>

From reading these posts, do you know how to determine the minimum thickness of the fuel tank. For the purpose of this, im going to assume the tank is a cylindrical vessel? Im doing my dissertation on optimisting the fuel system, and was going to model it as point load at the middle of the cylinder (should be the weakest point), but im not sure how to work out if it will rupture under a given load or not. Can anyone suggest anything to me?

[sourplum]

cylindiral tank..erm quiet a nice idea.it can help a lot in encounter sloshing during hard cornering.but u should consider on placing the tank too..does it enough space to put such a cylindrical thing..erm i'm wondering about that

erm use .015 aluminium..don't make it too thin or u'll get into trouble when welding it..trust me..

[packtim]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by sourplum:
cylindiral tank..erm quiet a nice idea.it can help a lot in encounter sloshing during hard cornering.but u should consider on placing the tank too..does it enough space to put such a cylindrical thing..erm i'm wondering about that

erm use .015 aluminium..don't make it too thin or u'll get into trouble when welding it..trust me.. </div></BLOCKQUOTE>

Thanks for that. The problem is, I need to try and justify the thickness of aluminium I am using, i.e calculations. I have only learnt stuff on cylindrical pressure vessel failures so far. This is a redesign for next years car, but I have already worked with the person who designed this years car, and has gone with a cylindrical tank with a sump on it.

Thanks for your help

[sourplum]

what if you try with catia.i've never done that actually but that's what i'll do if i'm in your place..btw,i'm also designing fuel tank and fuel system for my university's fsae car..



tronoh,malaysia

[KU_Racing]

packtim, I dont think that the pressure vessel analogy is quite the same. The fuel is not pressurized in the tank. Any loading that you get from cornering, or from an accident, will not be applied the same as a 'pressure vessel' load with even distribution. The load will depend greatly on how much fuel you have in the tank, and the axis along which the load is applied.

Our tank was FEA analyzed with a mechanical load on the bottom half of one side of the tank- a close simulation of a 7 m/s side impact with a half tank of fuel. From that point, or something similar, it is just analyzing oyur diplacement and stress concentration numbers to determine what thickness you need. One more thing- dont forget to analyze in the as-welded condition, especially if you are using aluminum.

[packtim]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by KU_Racing:
packtim, I dont think that the pressure vessel analogy is quite the same. The fuel is not pressurized in the tank. Any loading that you get from cornering, or from an accident, will not be applied the same as a 'pressure vessel' load with even distribution. The load will depend greatly on how much fuel you have in the tank, and the axis along which the load is applied.

Our tank was FEA analyzed with a mechanical load on the bottom half of one side of the tank- a close simulation of a 7 m/s side impact with a half tank of fuel. From that point, or something similar, it is just analyzing oyur diplacement and stress concentration numbers to determine what thickness you need. One more thing- dont forget to analyze in the as-welded condition, especially if you are using aluminum. </div></BLOCKQUOTE>

Thats what i was implying, i.e its different to the pressure vessel analogy. No you mention it, FEA definately seems the way to go with this problem. Its going to be more work than I thought! Thanks for your help.

[Garbo]

In dynamic/transient situations (ie, all the time), the tank will get its loading from dynamic hydraulic pressure. The pressure will certainly not be uniform or uniquely on the bottom of the tank. In ships, it is quite common for dynamic loads to blow the top off tanks if they are partially filled. Many classification rules prohibit ships from operating with partly filled tanks.

While I realize that these are small cars, not large ships, the loading will be similar. Naval Architecture texts will have pertinent calculations but the loading is certainly not a simple as applying load to the bottom half of the wall of a tank.


Cheers
G

[KU_Racing]

I understand that this case is oversimplified, but the loading I used was the closest approximation I could find to how fluid would operate in the tank. I seriously doubt that the top of the tank would blow off- I would guess that the reason that would happen in a ship would be from the tank changing shape and compressing the fluid inside when the ship flexes in the ocean swells. Our tanks are vented, which would relieve air pressure.. from there I dont see 2 gallons of fuel generating enough pressure to blow the top off of the tank.

[packtim]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by KU_Racing:
I understand that this case is oversimplified, but the loading I used was the closest approximation I could find to how fluid would operate in the tank. I seriously doubt that the top of the tank would blow off- I would guess that the reason that would happen in a ship would be from the tank changing shape and compressing the fluid inside when the ship flexes in the ocean swells. Our tanks are vented, which would relieve air pressure.. from there I dont see 2 gallons of fuel generating enough pressure to blow the top off of the tank. </div></BLOCKQUOTE>

So if you think I have over simplified, how do you think I should model it? If I use FEA, its not necessary to do a simple cylinder I guess, as I can just draw the tank design as normal.

[KU_Racing]

I dont think you should try to analyze the tank by some crazy method- I was implying that I dont think it is possible to model every possible force and hydraulic pressure the tank will see- you have to get as close as you can to optimum, add your safety factor, and build your tank. Simple as that.

[Wright D]

A way to minimize loads on the fuel cell would be to mount it into the chassis using straps, as apposed to tabs welded on to the fuel cell itself. This helps by distributing loads over a large area. It also eliminates the risk of a crack around the welds of a mounting tab due to fatigue from fuel sloshing and vibrations.

I like to mount radiators the same way, no tabs, just nice wide straps and saddles.

[packtim]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Wright D:
A way to minimize loads on the fuel cell would be to mount it into the chassis using straps, as apposed to tabs welded on to the fuel cell itself. This helps by distributing loads over a large area. It also eliminates the risk of a crack around the welds of a mounting tab due to fatigue from fuel sloshing and vibrations.

I was considering this, but was worried about centrifugal forces moving it when the car corners, althought it would also make the tank more accessable if need be.

I like to mount radiators the same way, no tabs, just nice wide straps and saddles. </div></BLOCKQUOTE>

I was considering this, but was worried about centrifugal forces moving it when the car corners, althought it would also make the tank more accessable if need be.

[KU_Racing]

If you want to keep the tank from sliding, just use weather strip. Not the crappy thin styrofoam kind, but the industrial grade stuff that is really dense. It will give you more than enough friction to keep the tank in place.