Beginner question here: Using Steel bolts in a threaded aluminium hole - can it work? 1

[swiftarrow]

Hi! Beginner engineer here...
My company wants to get rid of a costly aluminum welding step and replace it with a bolted-on joinery with rubber flanges. I want to get your (more informed) opinions on whether this will work? I'm concerned that there will be bi-metallic corrosion between the bolts and the aluminium.

The details:
We have a large (2m long, 150mm wide, 20mm tall) extruded aluminium heat sink, with holes along it's length for coolant to flow. Currently, the two ends are welded to aluminium end-pieces manually (long, costly, etc).

We are designing a new end piece, which will be mated with the end of the aluminium, and bolted into it from the end (two holes will be tapped to hold the bolts). The dimensions are such that the largest bolt we can use are M3.5. I've attached a STEP file of the extrusion with the two ends we currently have.

The joinery must be strong enough to withstand internal pressures up to 10 bar, according to our production manager. Operating pressures are going to be around 2 to 4 bar, and for certification, pressure is 6 bar.

Per my calculations, three M3 bolts of 8.8 grade will satisfy the strength requirements. However, I am concerned about the potential for bi-metallic corrosion of the threads.

The bolts will be separated from the coolant flow, but there may be contact at some point of the life of the product (20 years). There will be daily thermal cycles from below ambient to around 100 deg C (and up to 200degC in exceptional cases). The extreme temperature range is -20degC to +200degC, not a normal occurrence.

My questions are:
[ol 1]
[li]The aluminium surface area (anode) is sufficiently large that the overall bi-metallic corrosion should not occur. BUT, will the aluminium threads corrode, in close contact with the steel bolts?[/li]
[li]How long should the bolt be inside the aluminium receiver? (I guess I can calculate this based on shear strength of the aluminium used, but if you have any tips / rules of thumb, that would be helpful).[/li]
[li]Does anyone here, with their much greater experience than I have, foresee any big problems with this?[/li]
[/ol]

Thanks alot for your help!

 

[CoryPad]

To avoid bimetallic corrosion between the screws in the extrusion use a coating like Magni 565 (ISO 10683). The engagement link should be two times the fastener diameter.

 

[IRstuff]

Can't see STP files, so can't tell if you are trying to seal fluid flow or not, but two bolts does not sound like enough for that sort of applicaiton

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[swiftarrow]

@CoryPad, thanks for your recommendation! Still, since the fastner diameter (M3 bolts) is 3mm, the engadgement link would be 6mm, which really doesn't seem like enough to me, given that the bolts need to withstand 3120.81N tensile force.

@IRstuff,
Strange you couldn't see the stp files. I can't re-generate them right now as I'm in the middle of drafting.
You are right, we are trying to seal the fluid flow. Two M3 grade 12.9 will withstand the tensile stresses, but I don't know about the aluminium thread shearing, or temperature cycles, etc. Lots of factors to consider here.

On a side note, does anyone know about these "Servo Screws"?

They would be better than M3 for the reason of Al thread shear, if I could get them in three (or max 4) mm OD. I can't figure out what to search for though!

 

[CoryPad]

Engagement length of 2 times diameter will allow the aluminum to withstand stripping from 10.9 screws, so you should be fine.

 

[Screwman1]

What you are wanting to do is not a difficult problem. Cory is right about the coating and the length of engagement. You don't need any special fastener designs for this, it is a common type of application in the commercial marketplace.

 

[Compositepro]

I doubt any seal will last through that kind of cycling with material of such differing CTE as steel, aluminum and rubber. But I can't view your file either, whether it is a .STEP or .STP.

 

[EdStainless]

You don't need more than 2 or 2.5x the bolt diameter, more threads does not increase the holding strength.

Are your bolts going to be symmetrical about the fitting? They need to be, or else you will also have bending and flex to worry about.
You bolted end pieces will be Al also, right?
As long as you use coated steel bolts this should work.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[IRstuff]

" we are trying to seal the fluid flow."

I just don't have any software for STP files. But, I was actually concerned by having only two bolts compressing whatever perimeter seal you're using. Typically, a bare minimum is bolts on corners, so at least 4, but because you have a 150 mm side, you probably need more like 2 more bolts in the middle to keep the seal properly compressed so that it can seal properly. Just consider that a typical head gasket on a car engine requires something on the order of 8 to 10 bolts to keep the gasket compressed, and that's with a relatively mild pressure requirement.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[moltenmetal]

You'll find LOTS of German process equipment sealed with 2-bolt flanges, handling conditions that are much more strenuous than this. If the mating flange is stiff enough and the bolts and receiving thread strong enough, you'll be fine. Rubber makes all the difference- hard gasket materials require a vastly greater sealing stress as well as significantly stiffer flanges to seal reliably.

 

[swiftarrow]

Hi everyone!

Thanks alot for your replies! You've cleared up a lot of questions I had.

My apologies that you couldn't use the step file. Right now I'm in the middle of editing the files, once I can I'll generate a screenshot and post back here.

With this input, I guess that we should use three screws, symmetrically mounted, two on either end and one in the middle. I tried to create a layout with four screws, but it wasn't working properly due to the geometry.

We will probably use more than 2.5X dia of engagement length. Since the diameter of the screw is 3mm, 2.5X comes to 7.5mm, which "seems" quite small. We'll go with about 12mm engagement length.

We will also use rubber (or similar material) o-rings with proper channels so that they can compress to create a seal. We're trying to work with o-rings about 1.5mm cross-section, as the dimensions are so limited.

The flange is quite stiff, but the pressure (10 bar) is quite large, and the life time (20 years) quite long, and the environment (large temperature variations daily) not to mild. The only steel is the bolt, the rest is aluminium, so with thermal expansion, the seal should just get tighter. With the dimensions we are working with, the difference in thermal expansion ratios between the steel and the aluminum comes to less than 0.1mm at the extreme case, so it should be OK.

The only big question here is the aging of the seals, I guess.

Thanks a lot for your input, and always welcoming more!

 

[cloa]

You could at least say what the coolant is?

http://www.eng-tips.com/threadminder.cfm?pid=1529

Use translation assistance for Engineers forum

Note the rules include No Student posting

 

[swiftarrow]

@cloa,

Good point. The coolant is Tyfocor L (or LS... whichever one has the Al inhibitor in it). The coolant is likely to be in touch with steel or stainless steel heat exchangers somewhere else in the circuit.