Sadly, these were a disappointment because the impellers just don't spin fast enough to build much static (and the noise isn't great, either). Turns out a WS7040 does a much better job and does it in a tiny package by comparison. Oh, well, that's experiments for you!

@@jamespray Ah that's a shame. I could really do with something compact that can achieve 0.5-2 LPM with good static pressure that isn't a conventional pump as they would need an inline HEPA filter so it becomes expensive quickly.

@@ZozobraDoom You really have to look at the specs vs your requirements in these decisions ... while it depends on what you mean by "good static", there's probably nothing reasonable in the centrifugal compressor space that has such low flow at any meaningful static rating. That is really the realm of positive displacement designs. My impeller-based compressors worked around 9-15 CFM but only 0.075-0.15 PSI, which really makes them more blowers than compressors (though I don't think there's a hard definition of what's what).

If I do another project at this level of complexity, I will DEFINITELY not skip slow-cure epoxy, lol!

​@@jamespray well if you like carbon, next time you could go all in and print the compression mold for a forged carbon fibre part. I dont know anything that could be stronger stiffer and lighter than that. Anyway, I really like the video and that you included your engeneering struggles.

You're welcome! It's the worst when you hear a good song in a video and there's no dang credits. Like the one midway through the Dakar 2018 Cars video. Such a banger and I'll never know where to find it

@@jamespray you talking about "Best Of Car - Dakar 2018" ? I just tried using shazam for like 10 minutes, and got 4 different songs that were wrong :(

@@PureRushXevus Yeah, that's the video. The song that cuts in at 2:30. Gosh I want that track!

I haven't done anything for the SR, but it's probably a design space with similar constraints, so I could see such a duct looking similar to this one. The only handles you'll see my designs under (legitimately) are this one, the one in the Discord sequences here, and Kiolia.

I have tried a flex shaft with the Nimble in the past, and while it makes mounting and stuff much easier, extrusion quality goes out the window. Past a certain, very small amount of bend/misalignment, Flex shafts cause periodic over/under extrusion because they aren't equivalent to CV drives. As soon as I traced my horrific woodgrain to the flex shaft, that whole setup went in a box and I never looked back

The IGUS joints didn't have the preload necessary for delta joints, and were just too sloppy ... though that may have just been my implementation, which pins the joints on rods instead of compressing them with mounting screws. I currently use ball joints from MP Jet, which are great quality, properly preloaded, and showing no signs of wear so far.

@@jamespray Thanks, that's great to know, as I did consider buying a set for testing, but... one set of 12 is like 50USD and that's not a small amount. Have you tried adding springs to pull the two rods together to take away the slop? That doesn't fix it's inaccuracy though

@@jerrymk6846 Springing the arms together is a bad bandaid that only wears the joints out faster (I did try it with the IGUS joints, but the print quality improvement was marginal except at very low speeds -- enough to use them to print the next iteration for the Jets, but nothing I'd want to run long-term). Also, NO, the MP Jets are actually really affordable -- they come in sets of 6, which isn't necessarily clear from the listings. I know somebody who thought he was getting 24 and bought 144 I would buy 3 packs to have 18 so I could be extra-picky about picking 12 for a new set of arms.

@@jamespray No no, I know MP Jets one are cheap, and so does those ball joints for RC car models with nylon housing (or alu) and brass ball. I mean the IGUS ones are not cheap, and a set of 12 will run me 50USD, so I guess... thank you a lot for saving me from buying them, to find out that they are not suitable for the job. I did see IGUS sells Delta robot kits, so I'd assume it would be good for it...

@@jamespray And you mentioned that springs make them wear out faster, what's your view on the half-sphere plastic sockets that ride on stainless bearing balls, with only the springs holding it on. I know that FLSUN has been using this type for quite a while, and is also quite common in the delta space.

I might, but the BOM is still pretty long, and currently I don't have an option that doesn't depend on a Nimble stepper bracket (which isn't a thing you can just buy anymore). If I do update the drive end to not use the bracket, I'd be more likely to share the heavy one :)

@@jamespray Thank you so much for your reply! Do you have a tip/coffee link? I would like to support you!

@@_Nashtii I didn't, but now I do, since you asked! Cheers! (Added a link to the description)

🤜🌟🤛

Thanks, I'm glad you got something out of it! The uneven airflow coming out of blowers is a result of the air trying to flow outward all the way around the impeller and getting accelerated around the housing wall as a result. Blowers can actually have forward swept, straight, or backward swept vanes, without changing that aspect; the vane direction has other implications like the balance of static pressure vs CFM (straight vanes also accumulate less dust over time).

Great question! It's *almost* like SketchUp doesn't hold me back, even though it's not "actual CAD software". Weird (Oh, and don't tell the DIY Perks guy; he's a SketchUpper, too)

Welcome to the club! Don't forget narrower tape, thinner zip ties, and speed holes in the standoffs 🤣

I do use input shaping for high-accel printing, but CarbonShrike doesn't really need it below about 10k and I have done a 100k ringing test without IS that didn't look bad at all. It's not that there is no ringing, it's just that the amplitude is extremely low due to the minimized mass.

I press X to doubt, but hey, you can always try it!

especially de-scaffolding part

I'm not quite sure what you mean, but I'm not currently planning any new tutorials...

The setup shown (with the Pico hybrid hot end) is very hot end-limited and good for burst flow of about 20 cubic, but only ~12 sustained. I am testing something new and much better now, and the extruder is still not the limiting factor. More details will come out eventually, probably

It's Arya, borne & Lynzz - Need U [NCS Release], linked in the description music credits 🙂

Mostly just availability. I can drive to the hardware store and restock my 5-minute epoxy in 15 minutes, whereas I'd have to do actual research and order longer-curing stuff online and wait til tomorrow or next day! It's really no big deal for less critical work (or smaller glueups, say 1-2 pieces per like when gluing delta arms), but it's, well, a poor excuse with something like this 😅

Thank you! The CF rods themselves are actually pretty good reamers, being slightly abrasive. I chased every drilling pass with CF, but I cut out a LOT of that process for the sake of time, lol.

@@jamespray 👍

5-6g more, I think. Not much by itself, but it was the only way to drive the weight substantially below 40g once I figured out the hardware reductions. This is what I mean about capturing that "last 1%" -- it's often easy to get "so far", and then pushing farther gets much harder. I should mention, though, that this was not a great part for doing this kind of CF work. When you can just have a simple truss with a few pieces of CF, it's really slick and straightforward!

Thanks for the advice -- I'll have to try triggering the dive reflex next time! And no, not a stupid question. It is linked to the print head by a short bowden tube, yes, but I also sheath the bowden in Kevlar cable mesh anchored to the print head and extruder (bypassing the PTFE fittings). This is extremely effective at eliminating stretch from the equation. As for the bend, even at extreme bed edge, it's not as severe as one might imagine without testing (this was a big part of why the Nimble prototype ever existed). Maybe 15-20 degrees at most, measured from vertical, spread over 3-4cm of bowden tube. So far as I can tell, it has no appreciable impact on the extrusion rate. That's not to say it's perfect, but most of what I'd like to improve from here is in the overall mass and the damping/spring rate of the drive end and flying gantry ... it never ends!

It's a super-helpful development resource to know what things weigh. I have it tied to a list I've slowly compiled by putting every hardware item I buy on a scale

"Thanks for sharing, Tyler. I think we've all been there. Let's recite THE PRINCIPLES, shall we? Good enough is GOOD ENOUGH! MAKE -- A -- LIST! Define dependencies! Close it out or close it down! Saying NO to yourself is LOVING yourself! ... Excellent, everybody. Who's next?"

Thanks for the feedback! It's tough to balance pacing with cramming in everything I'd like to communicate, sometimes. Or most of the times

bro pls dont make me have to recreate this design

ill pay for your design

Messing around with a printer that sucks is the first step to messing around with a printer that doesn't suck!

You did -- in the "metal" clips (full disclosure, 1900x90k used in those clips, but it's moved faster)! It's just really hard to catch on camera so I put dwells between the moves 😅 If you mean moving that fast with the new extruder, I can't currently film that because of some NDA parts now on the machine, but the day will come eventually ...

@@jamespray my bad, i guess i wasnt paying enough attention watching while desining parts 😂

@@spayrex_ lol, happens to me too. No worries! I have a couple of motion testing videos on the channel too, although they're older and don't showcase the full capabilities of the current fully armed and operation battle station I actually haven't done a full testing suite in a while because it's absolutely terrifying to skip steps at these speeds!

thanks for the tip, I was trying to get a good result with a screw modifier and adjusting the origin but wasn't getting anywhere. using a bezier curve makes a lot of sense