Backer plates, Ti or stainles steel?

[ChrisA]

Reading about backer plates and how they counteract bowing, I read the very best best performeringf backer is to use a literal rail with no carage. While this works it is slightly impractical. But assuming I am using stainless steel rails, you'd think a stainless backer would outperform titanium and cost only 1/3rd as much.

Am I right or is there some reason to prefer Ti? No, please don't say it removes 10 or 11 grams. In terms of percent reduction of moving mass it would be very small.

I'll have to check but I assume the backers are designed to have the same cross sectional area as the linear rails, that is how I would make them. If this is the case then you'd want the same material.

I'm really asking to see if I understand the theory. I don't mind spending a little more to get a better result, but I'd hate to spend more to get a worse result.

 

[ark]

Weight savings are only practical if you use a backer on the X axis (which is in my opinion unnecessary with a single front rail). I personally think titanium is a waste of money, unless corrosion of steel is an issue in your environment.

The backer options are titanium or (non-stainless) steel. There is a reason stainless isn't an option: Most stainless steel (300 series) is austenitic, and austenitic stainless steel has too high of a thermal expansion coefficient (CTE). There is also martensitic stainless (e.g. 400 series), which is almost identical in CTE to regular carbon steel, but it is not as common a material. By the way, stainless rails are made from 440C which is martensitic - this is why the same backers work on both stainless and non-stainless rails. In principle you could use something like 410 stainless, but I'm not aware of such a thing being made.

The two key material properties in a backer are thermal expansion coefficient and stiffness (Young's modulus). The CTE must be significantly lower than that of aluminium, because it is this difference that leads to a change in unloaded length between the extrusion and backer. The material must also be stiff enough, so that this change can actually force the extrusion to bend.

Here are some material properties for example:

Material	Coefficient of thermal expansion α (10^-6 K^-1)	CTE difference from aluminium	Young's modulus E (GPa)
Steel (1018)	11.5​	10.3​	200​
Stainless steel (304)	17.3​	4.5​	193​
Titanium (Grade 5)	8.6​	13.2​	114​
Aluminium (6063)	21.8​	-​	69​

304 stainless has a CTE difference from aluminium less than half that of steel - you would need extremely thick backers (I think it was close to 10mm? can't find my graph) for it to work. Titanium has a larger difference than steel, but it is less stiff, so overall it turns out that the needed thickness is about the same for the two.

By the way, I personally use some junk linear rails as backers on my own V2. (I had to slightly modify the A/B drives to get the full length to fit.) It's definitely practical, especially if you use an umbilical cable instead of drag chains on the gantry.

 

[ChrisA]

Weight savings are only practical if you use a backer on the X axis (which is in my opinion unnecessary with a single front rail). I personally think titanium is a waste of money, unless corrosion of steel is an issue in your environment.
...

Thanks, The bit of information I did not know was that 400 and 300 stainless had such different CTE. I figured I could use 400 type SS rails and 300 type backers and be perfect.

I live near Ocean and the printer will live in an unheated shop space. Plain carbon steel rusts quickly. so it looks like I'll be using Ti.