Aluminium chassis design problem 1

[EsoEng]

I'm designing a chassis for a cargo trailer to be pulled by a motorcycle. It needs to be as compact as possible; there are severe limits on the thickness of the chassis, and the overall height, length, and width, and also weight, of the final trailer.

With the above in mind, the image, below, shows the first design, which consists primarily of extruded aluminium members all on the same plane. This minimises the thickness of the chassis, and thus the overall height of the trailer.

However, the problem I am having is that the design will not be strong enough, once welded, in the heat affected zones (HAZ) using standard heat treated 6000-series stock aluminium extrusions. Maintaining the geometric envelope of the chassis, and without significantly increasing its weight (such as by using a high-strength steel, which would at least double its weight), how can I mitigate the loss in yield strength in the HAZ? (I'm not talking about pre-/post-welding heating, and welding techniques to minimise the problem, as I believe they cannot prevent the significant yield strength loss I am concerned about.)I'm trying to think of ideas, but I would very much like to hear from someone with aluminium fabrication experience/knowledge as to what would be the standard solution(s), please? This is actually my first real project of the type. Pretty much, all I was taught in the classroom were the fundamentals of metallurgy, and how heat affects the strength. The easy answer (in a didactic sense) would be to fabricate the chassis, and then have it heat treated to restore its full yield strength value. However, in practise, this is awkward, and expensive (so I'm told). It makes me wonder what fabricators do to avoid the loss of strength in the HAZ, including, for example, mountains bike frame manufacturers. Perhaps, they just design-in more material in the HAZ, which is not something that would be easy to achieve for my project if I continue to use stock extrusions.

Thanks for any help.

 

[MintJulep]

Most bicycle frames use 7005, because it's relatively easy to heat treat.

 

[EsoEng]

To my knowledge, 7005 isn't available in extrusions. In fact, not many grades are. Is it standard that fabricators make most things from plate?

 

[dvd]

How big is this thing? I ask myself whether it could be a bent sheet metal piece - possibly with some riveted joints.

And, as always, something to bear in mind with aluminum is that it does not have an infinite fatigue life -

 

[EdStainless]

There are structural shapes that are not extruded.
You need to broaden your view.
And you worked hard to make sure that thin wall steel tube wouldn't work right?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[rb1957]

design is all about compromise.

You've chosen 6061T6 welded frame 'cause it "optimises" some aspects of the design, but at the cost of some bad features (HAZ).

The "best" way to fix (heat treat) is expensive and fraught with difficulties (keeping the frame square and flat).

As noted above, many of your elements don't look like extrusions; like the corners (oh, they are "bent" extrusions" ... ok, but be careful if they are ... there are limits to the extrusions you can form).

Have you considered the strength of the top plate, presumably steel. Could you make a sandwich out of the frame (ie two facings which have most of the strength) ?

What is the impact on the design of steel extrusions ? (they'd be heavier, but cheaper) Don't throw out an option without giving it due consideration.

What would be the impact of riveting the frame members ? no HAZ, better Al options (like 7075T6 ... much stronger than 6061, but more expensive) but more factory work building the frame

could you take the sandwich panel to the extreme ... make the frame out of wood or UHMW plastic sheet ? polystyrene ?? and rely on the facings for strength ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[EsoEng]

@dvd, it's one metre wide, and less than two metres in length. It will have a maximum laden mass of 200 kg, and I'm allowing a maximum dynamic load of 5-times the mass. I am very conscious of fatigue issues with aluminium, which, of course, is a big concern. Thanks for the graph. Riveting is certainly a promising solution.


@EdStainless, thanks, I do need to broaden my view, which is one reason for posting this thread. I need to get back - properly - into the engineering way of thinking after some time away. As for thin-wall steel, if it's available with the desired yield/fatigue strength I have failed to identify a source. However, I'm not keen on a tubular design for various reasons, although I would consider it if all else fails.


@rb1957, compromise is a bit of a problem, currently. Suitable material exists - such as the 7000-series - but not in a desirable form. However, I will investigate the 7075 T6-riviting idea, although previously I had largely discounted 7000-series due to cost (I will re-visit the option).

The non-extruded elements in the chassis (which are subject to revision) are fabricated from plate.

Weight really is critical with this project. The entire (unladen) trailer must be below 50 kg. Steel hasn't been entirely ruled-out, but standard stock mild steel would be too heavy.

I'm not entirely sure I understand what you have in mind with the sandwich idea, although I am visualising some options. Wood is not suitable, and I would rather avoid polymers due to lack of fire resistance. For the same reason, and also due to its lack of ease with respect to field repair, and propensity to fail in irreversible ways, I'm not keen to use carbon fibre, although it remains a thought, and I could probably make it myself if I could find an autoclave. Thanks for your reply.

 

[rb1957]

generally you can get an extrusion you want in any flavour of Aluminium, but maybe you mean locally you are limited to 6061T6. Still you can rivet 6061T6 and avoid the HAZ.

you are planning on putting a facing on top of the frame ... yes ?

the 50kg includes the hitch, the wheels, axles, suspension ? how much have you allowed for these ?

how big is this ? it looks like 1m*2m ? then you'd have 3*2m +6*1m = 12m linear in the frame (roughly, = 40ft = 500in)
a "sensible" cross-section would have an area like 4in*0.1in = 0.5in2 or 250 in3 or 25 lb of Aluminium.

If weight is super critical, cut lightening holes in the webs. I've guessed a 2" (50mm) web, so you can probably cut 1" dia holes on like a 2" pitch
so in our 500 linear inches, you'd have about 250 holes, each an volume of (roughly) 0.075in3 (I'm guessing 0.1" thick flanges) *250 = 18in3 or 2 lbs of Aluminium.

repairability is a design concern ... repairing welded Al frame is also not easy.
also durability in service

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[3DDave]

How does this compare to existing products?

One area that seems odd is the lack of side and forward barriers to prevent the cargo from exiting in ways that become very exciting very fast. That and the lack of obvious suspension attachment features.

It's also the case that welders have a great deal of difficulty welding sealed off tubing. As soon as the heat is applied the air inside the tubing expands and blows out the liquid aluminum. You can leave a small breather hole, but that will just collect water and dirt inside where it's tough to get to. There are also a lot of places that will be difficult to get a torch head into, but if they are skipped will be crack initiation sites unless the loads are really low there, in which case the reinforcement isn't required.

The layout looks far more like a wood project than a welded one based on the number of corner brace pieces. Taking a longer look at similarly purposed trailers, both for motorcycle tow-behind use and for cars and trucks would be what I would do.

 

[MintJulep]

Towing 200 kg behind a motorcycle? No thank you.

50 kg empty vs 200 kg loaded? Seems disproportionate heavy empty.

Why does this need a "chassis" at all? Monocoque seems like it would be so much more efficient.

I'd be thinking about fiberglass over foam.

 

[BrianPetersen]

Look at how existing products are built. I've never seen one this complicated, and yet, where's the suspension connect?

Perimeter tubular steel frame with suspension and hitch attachment points ... simple shape, no crossmembers ... fill in the middle with a floor made of any material you choose attached in any suitable method ... done.

 

[EsoEng]

@rb1957, the riveting idea is now top of my list. It opens-up the project to a far wider selection of materials, and I won't have to rely on a welder (I can do the work myself).

If, by facing, you mean a floor, then yes. Just sheet aluminium. The trailer will also have removeable sides, and a top, so effectively a box on the chassis.

The 50 kg value does allow for all trailer-mounted components, yes.

The chassis shown in my image in the OP weighs approximately 4 kg (if memory serves. It was actually designed some time ago, and I haven't opened the file in the CAD program recently), so less than half your estimate. However, the design would only be viable, or confidence-inspiring, in the absence of the HAZ concerns. Hopefully, the final design will not require hundreds of holes for weight-saving! Thanks again.


@3DDave, this trailer is unique, which is why I'm building it. It's actually far more than just a cargo trailer; it's a camper-trailer, but one that conforms to draconian UK motorcycle trailer legislation (weight, and size limitations). As I described directly above, this is just the chassis, onto which will be further structure to contain cargo, et cetera. Thanks.


@MintJulep, compared with many motorcycle cargo trailers, 50 kg is light for its maximum load (which actually might be closer to 250 kg, I need to check - this project has been suspended for a long time). It's a chassis so that the "box" can be removed, but the monocoque idea is certainly interesting. Thanks.


@BrianPetersen, the suspension, at least in the first design, is regular torsional trailer suspension. It bolts onto struts, which in tern bolt onto chassis cross-members. However, I'll be designing an aluminium swingarm instead, and adding brakes (controlled via a thumb-actuated lever on the handlebars). Thanks.

 

[rb1957]

I think you should rethink having an Aluminium sheet as your loading tray/floor ... Aluminium isn't very durable. I'd suggest Steel sheet, if it is 2m x 1m = 2m2 = 20sqft = 3000 in2, at 0.032" thk, steel would weigh 30 lbs ... but would be very effective in carrying load. An option might be kevlar ... probably more expensive and more processing but lighter.
4kg = 10 lbs ... humm, very light, maybe 0.04in (1mm) thick flanges ...

Sure add sides to this base, hinged maybe to fold onto the base ? or removable. Humm, UK motorcycle rules ... have you talked to a UK bike hobby forum ?

Are those triangular prisms separate pieces, or welds ?

I think Brian's point is that the suspension points should be reinforced in the chassis. I guess I'd have a heavy cross member to pick up the suspension (maybe two) and a heavy longitudinal member to pick up the hitch, and then lighter sections to support the edges (to mount the sides) and to support the base.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[CWB1]

Looks like a ton of welding (read: labor/cost). Personally I'd design it as a unibody rather than tube-frame structure. Most motorcycle trailers are fairly small, so stamping your floor and a second stiffening subfloor sheet on a press brake should be easily doable. The two "floors" can then be quickly overlaid and attached with fewer welds and less fitting, saving a ton of time/labor/cost as well as weight.

 

[SWComposites]

is this a one off trailer or intended for quantity production?
for minimum weight I recommend a composite (fiberglass or carbon fiber or combination) integral design - floor, walls, etc with hardpoints for suspension attachment. you don't need an autoclave; many resins can be cured in an oven or even at room temp.

 

[EsoEng]

@rb1957, the Kevlar floor idea is interesting, but steel seems too heavy. As there are so many cross members, I feel an aluminium floor would be well supported, and well suited for filling-in the gaps between structural members. Heavy objects will not be small, so the floor (sheet) will not be under a lot of stress from cargo.

When I first had this idea, I spoke with a commercial motorcycle cargo trailer manufacturer, from whom I purchased several components (including the wheels, which, along with the tail lights, might be the only parts that will actually end-up in the final design). He gage me plenty of advice, and the legislation is clear so no need to ask for further help in this respect. Personally, I think the legislation is largely counterproductive. It limits the distance between trailer wheels and hitch, which will make towing the trailer less stable. Motorcycle-towed trailers in the US, as far as I know, are not affected by this limitation (or the others), and so are more stable under tow despite weighing far more.

Everything shown in the image of the chassis is welded. It's effectively a single object. The triangular pieces are just stiffeners cut from L-section. Further thanks!


@CWB1, I've been talking with a welder. He's seen the plans (as they are), and assures me his costs will not be excessive. However, I like the stamped double floor idea but fear it would require the machining of a large die, or similar, which would be costly. Or, do you have in mind something different to what I'm thinking? Thanks.


@SWComposites, nothing is final, of course, so I will think some more about the composites idea, although I am concerned about spectacular, un-repairable failure scenarios, and also lack of resistance to fire. It's good to know that curing composites can be simpler than I thought. This is a one-off, but if it's successful I would consider small-scale production. Thanks.

 

[dvd]

I think that a well-equipped sheet metal fabricator could create the component(s) required with an open setup. There are corrugating dies, and roller dies, and finger dies, and wiper dies, etc. which could form a laser-cut blank into a stiffened deck. Combining with some riveted stiffeners and attachments would make a stout and simple platform.

 

[rb1957]

I think you're not hearing the advise ... none of which seems to say "here's a way around your HAZ problem". Nearly all seems to say ... scrap the proposed layout and most propose a sheet metal or a composite solution.

How did you determine 50kg is the appropriate weight for this ? How much to the components you already have weigh ? Have you done a weight budget ? What do similar commercial trailers weigh ? How are they made ??

Not sure what those triangle blocks achieve ?

GL

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Stick.]

Common Mistakes Made in the Design of Aluminum Weldments

This resource gets posted semi-regularly when the issue of welding aluminum comes up and it's a pretty good primer on the subject.

 

[EdStainless]

Using one solid sheet with folded edges and then stiffening it with "C" or "Z" formed shapes would be a lot easier to execute.
You could still weld, or rivet, or even adhesive bond for that matter.
Using a honeycomb sandwich for the floor could add most of the stiffness that you need.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed