Tensile Strength Testing

[rsterne]

Sometimes I get carried away with testing, when I could have just watched a video!.... This one on extrusion width by Stefan at CNC Kitchen is really important if you are looking to improve your layer to layer adhesion to increase strength in the Z-direction....



I am just working through testing my 4th PETG-CF filament chasing this exact thing.... Stefan's conclusion was that you can go to 140-150% of the nozzle diameter without losing print quality, and he used up to 200% for structural prints.... I have already adopted 120-125% as my "standard", and am experimenting with 150%, and I will be posting some graphs shortly.... It really does increase print strength, and an added advantage (as posted by others in this thread) is that increasing the extruded width relative to the height (layer thickness) will allow you to print shallower overhangs without adding support....

Bob

[rsterne]

I have dried some of the PETG-CF filaments that I tested previously right out of the box, figuring they should be OK.... What I found was very interesing.... When dried for 6 hours at 65C, the Horizontal (X) strength increased, but the Vertical (Z) strength decreased (or at best stayed about the same), in all the ones I dried.... Here are the results for the dried filament, you can see the charts for the "as purchased" filaments in the above posts....







In my quest for increasing the layer-to-layer bonding I have been trying an increase in the line width (as in Stefan's video in the previous post).... There is no question that you get an increase in the Vertical (Z) strength as you go to wider lines.... However, the Horizontal (X) strength usually decreases a bit (but not enough to make a significant difference).... Here are my results....









From this experimenting I have decided to use a 0.25mm Layer height, and a 0.60mm line Width (150% of the 0.4mm nozzle) as my standard profile for CF filled PETG.... If you want to print really fast, you could use a 0.6mm nozzle, a layer height of 0.30mm, and a line width of 0.9mm.... maybe even a 0.333 layer height and a 1mm width (although I haven't tried that yet)....

Bob

[Baco]

i told ya 

it works even better with pla

[rsterne]

Yeah, you did!.... I just needed to find out how it works with CF-PETG....

Bob

[WhatUPSbox?]

Bob, did you see this video? Granted the printing is slow but the numbers Stefan got for Z axis were pretty impressive.

[rsterne]

I wonder how much of the strength was that there were no impurities (ie dye, particles, strands, etc.) in the filament.... I think it's pretty "clear" that pure filaments should be, and are, superior for layer adhesion....

Bob

[rsterne]

I got some Priline PETG-CF a couple of days ago, because I had heard good things about it, and the price was good.... Well, it may be cheap(er), but the performance was the worst carbon filled PETG I have tested, particularly for layer to layer adhesion.... and it was no stronger in X than a non-carbon PETG....



The prints did not show very much of that "carbon look", and I won't be buying more of it.... I'll just use it up when I want a greyish black PETG.... Further carbon filled PETG testing will have to wait until some that I haven't already tested goes on sale.... The strongest in X was the Spectrum PETG-CF10, and it had the most glittery, layer hiding appearance, but the Z strength was below the PETG average.... The second strongest in X was the Kimya PETG-CF significantly better than non-carbon PETG and most PLA, with a Z strength typical for PETG, though not as good as the best PLAs....

Bob

[rsterne]

I tried annealing the Priline PETG-CF.... tested stock vs. 2 temperatures for 4 hours each.... The horizontal strength increased about 6% when annealed at 65C, but was weaker when annealed at 90C.... The vertical strength was less on both annealed samples....

Bob

[rsterne]

I continued my tensile testing of the samples of Spectrum Carbon Fibre filled filaments today with the ASA-X-CF10.... Frankly, I see no advantage over PETG-CF filaments, and it smells quite a bit (somewhat like ABS when you saw it).... The surface finish is quite rough, particularly on the corners, where it ends up with raised ridges like tiny saw teeth, quite sharp.... I am testing all the carbon fibre samples now using a 0.25mm layer height and a 0.50mm extrusion width.... Spectrum recommend 235-260C for the nozzle, and 0-20% for the fan (I had the fan off).... The bed temperature is quite high, I used PEI with Elmer's gluestick, at 90C (100C for the first layer), the recommended is 90-110C, and I had no adhesion problems.... Incidently, the horizontal samples were breaking through the eye, so I had to reduce the test section to just 3 x 4 mm to get clean breaks there (and adjust the area in the calculations, of course).... Here are the results....



This material was about 13% stronger than most PETG-CF filaments in X, but more than 14% weaker in Z.... but is was significantly weaker than Spectrum's PETG-CF10 in the horizontal (X) direction (but that was one of the strongest I have tested so far).... When I did a final test with a 0.60mm extrusion width at 270C, it was NOT as strong at the 0.50mm width, quite a different result than I had with the PETG-CF.... With the smell and rough finish, I won't be using it again.... If you want what I have left (~ 150g), just pay the postage and you can have it!.... My wife says I am not to use it any more because of the smell!....

Bob

[rsterne]

I dried my Nylon filaments overnight at 90C, and retested them.... They were stronger (more about that later), but I was still having the occasional coupon breaking through the eye where the pin is.... I can't make that part any bigger, so I redesigned the whole system for new, smaller, faster to print coupons that use half the material....



The vertical coupons are round, using a hyperbolic center and conical ends to anchor them.... The horizontal ones are similar, but use the same rectangular test section I did before, at 5mm x 3mm.... I made them that way (with a flat bottom) so that I could print them flat without supports....  Of course this means I had to design and print fixtures to hold the ends of the coupons....



I printed them from Polymaker PA-612-CF carbon filled Nylon, which has been the strongest material I had tested so far.... They are printed horizontal, with support under the thinner tab, where the pin goes through to mount them on my lathe based testing equipment.... This is what they look like with a coupon in place....



I have done a lot of testing to insure that the data is consistent with my previous results, and I am delighted with the correlation.... All the coupons break in the middle, in the thinnest section as they should.... The fixtures don't flex, and are showing zero distortion or wear.... They may as well be machined from aluminum, considering the loads they have to stand, which is no problem....

Bob

[rsterne]

Here are the results on the retesting of the two carbon filled Nylon filaments I tested a while ago, after drying at 90C for 12 hours.... First the 3DX Tech PA6-CF-G3....



The gain in strength in the X direction (horizontal) is about 1000 psi, but the vertical (Z) direction had a big jump, from 2400 psi to over 8000.... The layer bonding was stronger up to 270C, but at 280C the quality went in the trash, and the Z-yield dropped a lot.... Do not use over 270C for this filament!....



For the Polymaker PA612-CF, it also gained a bit in X, and also made a large (30%), gain in Z, ending up at nearly 7000 psi.... After proper drying, both these carbon filled Nylons exceeded even the best of the PLAs that I tested in Z and in X.... Don't think that because they are vacuum packed with dessicant from the factory that you can skip drying them before use, you can't....

Bob

[rsterne]

I tested the last of the Spectrum sample pack today, this time it was the PA6-LW-CF15S.... The "LW" stands for "low warp", and it contains 15% carbon fibre in the Nylon base.... Here are the results....



The horizontal (X) strength was nearly 12,000 psi, which is right in there with the other Nylon filaments in the previous post.... The vertical (Z) direction was the strongest I have tested, at over 9,500 psi, which is 80% of the filaments printed in line (X).... That is quite astounding layer-to-layer adhesion.... Not only that, but from 250-290C the performance in both directions was pretty consistent.... At 300C, however, the strength in both directions dropped off rapidly.... The narrow part of the vertical coupon was shiny and irregular.... At 290C there was lot of stringing, and a bit at 280C, but at 270C the print quality was quite good.... This filament definitely is on my short list.... It doesn't print quite as nicely as the Polymaker PA612-CF, but a lot easier to print than the 3DXTech PA6-CF-Gen3.... I am finally seeing just how good a good quality carbon filled Nylon can be.... 

Bob

[TorqueMaster]

Bob,
How's the print quality at 250C?  I dare not go higher on my stock hot end, but these numbers truly impress.  Are you using an enclosure?  and has it lived up to the "LW"?   Where did the sample pack come from?  A quick google doesn't show many vendors for this brand in the Americas.  Thanks!

[WhatUPSbox?]

Those are nice numbers. Apparently Nylon looses strength as the moisture is reabsorbed. It may be interesting to set aside several coupons and let them reabsorb moisture for several weeks
Example: https://www.degruyter.com/document/doi/10.1515/SECM.1995.4.3.199/pdf

[rsterne]

Print quality is good at 250-270C.... I haven't seen any warping with the Nylon, but I have been using a Garolite bed at 100C.... No enclosure on my Ender 5-S1, but it is tucked into a corner in a draft free, unoccupied room I keep at 21C (70F) when printing.... I got the sample pack from 3D Printing Canada....

https://3dprintingcanada.com/products/carbon-set-multi-pack-1-75mm-spectrum-carbon-filament-5-x-0-25-kg

I am still collecting various types of PETG-CF, you can print it at lower temperatures than Nylon and it's less hygroscopic.... The Spectrum version was as strong as the Nylons in X....

Bob

[TorqueMaster]

Thanks for the link -- ~$60 a half kg -- pricey but worth it if the application calls for it.

PETG-CF that's as strong in the X as this nylon is well and good -- but I'm interested in a strong Z -- the weak link.  This latest filament is gangbusters in that regard.  Since the CF mostly works in the XY direction, the non-CF version of this PA6 may also be excellent in the Z, but with a hit to the XY performance.   Won't look as cool, though, the more matte CF appearance is very appealing.

[rsterne]

The fusing of Nylon between layers is impressive.... Most of the good test coupons look rather "crystalline" in appearance with few or no layer lines visible.... That is obviously the secret to their above average "Z" performance....

Bob

[rsterne]

I needed some green PLA for a project, and purchased some of the "house brand" from 3D Printing Canada.... It printed beautifully, and when I was finished my project I tested the leftover filament.... Here are the results....



It was right in the ballpark for strength in both directions, a little stronger in X and a little weaker in Z (layer bonding) than the "brand name" PLAs I tested previously, and wayyyyyyyyyy better in both directions than the "Econo PLA" from Overture.... so good value for money.... The highest strength in X was at 220C, and in Z it was at 210C, but so close in layer bonding that I decided to do some additional testing at 220C to find out what happens with different Extrusion Widths, both staying with a 0.2mm layer height (and varying the widths), and increasing the layer height as the width increases, maintaining a 2:1 ratio of width to height.... In other words, 0.20mm layers with 0.40mm width, 0.25mm layers with 0.50mm widths and 0.30mm layers with 0.60mm widths (all done with a 0.4mm nozzle)....  Both sets of results are plotted below....



As I have seen before, there isn't any advantage in the horizontal (X) direction from increasing the extrusion width (or the layer height), but increasing the width of the strands increases the vertical (Z) strength significantly.... The gain was pretty linear with a 0.20mm layer height, but when you increase layer height in proportion to line width, the gain was greatest with a 0.25mm layer height and 0.50mm line width.... I have seen this before with other filaments....

Bob

[rsterne]

Well, I made a big mistake on this post, which I am now correcting.... I had the wrong number in my spreadsheet for the width of the horizontal (X) coupon, so the yield values I reported for the horizontal (X) yield were lower than they should have been.... The vertical (Z) yields were correct, however.... I have corrected the charts, and I am editing the text to correct this error.... For anyone who read the original version, or to the Manufacturers of the two filaments involved, my apologies....

 I recently purchased four more brands of PETG-CF filament (well, once was PET-CF), to continue my testing to find a good one.... Neither of today's candidates qualify.... First I tried the BAMBU Labs PETG-CF, and in both the horizontal (X) and vertical (Z) directions it was weaker than the non-CF filled PETGs I haver tested.... Here are the results....



So I figured after that disaster, I would get a reprieve when I tested one of the highly touted 3DXTech filaments, their PETG-CF.... Here are the results for that....



It was about average for the carbon filled PETGs in X.... but easily the weakest of all the filaments of any kind I have tested in Z, with the exception of the Econo PLA I tested.... 3DXTech recommend using 230-260C for this filament, and it was under 1000 psi for layer-to-layer adhesion, which means the coupons were breaking at less than 30 lbs. of pull.... Even at 280-290C it didn't reach 2000 psi.... 

I am getting quite a collection of PETG-CF filaments that look pretty but are weaker than regular PETG.... Oh well, still a couple more to go tomorrow.... 

Bob

[rsterne]

While I was preparing today's data, I discovered an error on the charts (and text) for the previous post, which I have now corrected, please check out the changes above....

Today's data is for the remaining carbon filled PETG I have, made my 3D Printing Canada, and for the Bambu PET-CF (note, not PETG, and over double the price per kg.).... First for the 3D Printing CF-PETG....



It was pretty typical in the horizontal (X) direction for the CF filled PETGs I have tested, but a bit below the average in the layer-to-layer adhesion.... None of these CF-filled PETGs have been stronger than non-CF filled PETG in the vertical (Z) direction so far.... Next I tested the BAMBU Labs carbon filled PET filament.... This is not PETG, and has a noticable odour when printing.... It costs about twice what their CF-PETG does by weight.... Here are my results....



The printing temperature was higher, with the strength peaking at 280-290C.... It was significantly stronger than their PETG in the horizontal (X) direction, reaching nearly 12,000 psi.... However, it was disappointing on layer-to-layer adhesion.... performing much like a typical PETG in that regard.... For my next tests I am putting on a 0.6mm nozzle, and do some experimenting to see if I can increase the strength in the vertical (Z) direction on a few filaments....

Bob