[Cooling] Candlelight Vigil, please pay your respects.  

Re: [Cooling] Candlelight Vigil, please pay your respects.

@ jltx,
first I think your design is definitely good for two reasons: the main the clearance on the nozzle view that is invaluable, the latter the design with a longer intake pipe because I think that the small fan suffers too much for back pressure due to the limited outtake space. The flow is very uniform (even less pronounced) as result.
As I said before the problem is the backside design. Mainly in my case the blower scratches the bed cover and not the rear pins and unfortunately not only during the calibration but also in the normal printing if the nozzle has to print in same rear position. I've eliminated the screw overflow but is the plastic cover (also the low profile) that simply does not fit.
I've used the latest version with the clip that keeps the blower at correct height perfectly parallel to the bed.
Also I do not like very much the too tight PINDA hole that make difficult the correct placement of the blower, a little more clearance should be appreciated.
I think the only way is to redesign the back side of the blower rising it to be sure do not touch anything in various printer builds.

Re: [Cooling] Candlelight Vigil, please pay your respects.

Also I do not like very much the too tight PINDA hole that make difficult the correct placement of the blower, a little more clearance should be appreciated.

^^---- THIS

I used an exacto knife for 20 minutes opening that hole up.

Your low profile cover had the same issue as mine, and did not sit flush on the board. It's an easy fix.
jltx is going to fix it I think. (I sent him a picture of where it was touching. I simply cut some plastic away with an exacto knife, and got it to fit fine)

So I printed a 10x10x10 cube, in Vase mode. (Ignore the bottom (top in the video) I just did a full reset on the printer, and my live Z was WAY off. (0.06mm) I have since fixed it.
Check the corners!! The diagonal waves are due to Vase mode. (I printed another one without vase mode, and they were gone)

0.2mm first layer, 0.15mm layer height... Really, REALLY cheap PLA filament. ZYLTech.

I can't tell any of the sides apart.Should have marked it on the printer. I'm printing a benchy now. Will look at it when I wake up in a few hours.

Re: Candlelight Vigil, please pay your respects.

**snip**

^^--- is an abuse of the notificaton system. Just ignore it.

I know the files attach here.

The pic of the printed part, has the places I cut to allow a perfect fit.
The pic of the solder shows what it looks like. Both of the solder blobs have a "dent" from putting it on the first time.

Re: Candlelight Vigil, please pay your respects.

**snip**

^^--- is an abuse of the notificaton system. Just ignore it.

Let me know if I did something wrong.

I know the files attach here.

The pic of the printed part, has the places I cut to allow a perfect fit.
The pic of the solder shows what it looks like. Both of the solder blobs have a "dent" from putting it on the first time.

Since I hadn't received your pictures (yet) I took mine apart and imagined where things could go wrong. Interestingly, for the new cover Prusa turned the hex so the flat was forward, which totally makes sense. But now the points can run into over-aggressive solder blobs. Also, the flat has essentially no clearance to the LED, possibly even conflicting. So I trimmed these back and additionally slightly notched the right side where is hits the heatbed since there is less clearance. I uploaded V2.1. The spacer is also a hair thicker to help the front screw tip from poking through too high.

Now that I see your pics, I see I fixed the correct problems. Give it a try and let me know. When the cover is properly installed, it should be the same height as the back pins, no taller. Sorry this caused trouble.

Re: Candlelight Vigil, please pay your respects.

Let me know if I did something wrong.

Nope, you are fine. I just wanted to give you a notification because of the failed message. So I abused the quote system which sends a notification.

Since I hadn't received your pictures (yet) I took mine apart and imagined where things could go wrong. Interestingly, for the new cover Prusa turned the hex so the flat was forward, which totally makes sense. But now the points can run into over-aggressive solder blobs. Also, the flat has essentially no clearance to the LED, possibly even conflicting. So I trimmed these back and additionally slightly notched the right side where is hits the heatbed since there is less clearance. I uploaded V2.1. The spacer is also a hair thicker to help the front screw tip from poking through too high.

Now that I see your pics, I see I fixed the correct problems. Give it a try and let me know. When the cover is properly installed, it should be the same height as the back pins, no taller. Sorry this caused trouble.

I'm guessing you are using a longer screw than mine. I had to cut the spacer in half to engage the nut.
I actually got the old one on. 😉 I'll print out a new one later today and measure my screw also.

Re: [Cooling] Candlelight Vigil, please pay your respects.

@ jltx,
As I said before the problem is the backside design.
...
I think the only way is to redesign the back side of the blower rising it to be sure do not touch anything in various printer builds.

With full understanding that this is your design and follows your design criteria I just want to add input in support of modifying the rear of the design to clear the bed stand offs during calibration. I'd like to try this design because nozzle visibility is very appealing but I'm not willing to risk forgetting to remove the assembly for calibration.

Steve

Re: [Cooling] Candlelight Vigil, please pay your respects.

@ jltx,
As I said before the problem is the backside design.
...
I think the only way is to redesign the back side of the blower rising it to be sure do not touch anything in various printer builds.

With full understanding that this is your design and follows your design criteria I just want to add input in support of modifying the rear of the design to clear the bed stand offs during calibration. I'd like to try this design because nozzle visibility is very appealing but I'm not willing to risk forgetting to remove the assembly for calibration.

Steve

I second that.. I regret that it puts more work on jltx, but I think this design is going to end up being the best alternative going forward. I hope the challenge will be accepted! 😉

Re: Candlelight Vigil, please pay your respects.

I'm guessing you are using a longer screw than mine. I had to cut the spacer in half to engage the nut.
I actually got the old one on. 😉 I'll print out a new one later today and measure my screw also.

Damn it. You are right. Somehow I got an m3x12 in my kit and didn’t check. I’ll fix that too.

Re: [Cooling] Candlelight Vigil, please pay your respects.

@ jltx,
As I said before the problem is the backside design.
...
I think the only way is to redesign the back side of the blower rising it to be sure do not touch anything in various printer builds.

With full understanding that this is your design and follows your design criteria I just want to add input in support of modifying the rear of the design to clear the bed stand offs during calibration. I'd like to try this design because nozzle visibility is very appealing but I'm not willing to risk forgetting to remove the assembly for calibration.

Steve

I second that.. I regret that it puts more work on jltx, but I think this design is going to end up being the best alternative going forward. I hope the challenge will be accepted! 😉

Trust me, the back pins drive me crazy. But it’s really from a purity of design standpoint. In practice I found that I never do full XYZ after the initial build. I did once because I replaced the X ends with R2 and that probably wasn’t necessary. I’m curious if I am out of the norm here.

In any case, I appreciate the ask and wil look into it, but I’m in the middle of a massive project that I need to get to alpha phase first. I will be looking for advanced testers.

Re: [Cooling] Candlelight Vigil, please pay your respects.

Trust me, the back pins drive me crazy. But it’s really from a purity of design standpoint. In practice I found that I never do full XYZ after the initial build. I did once because I replaced the X ends with R2 and that probably wasn’t necessary. I’m curious if I am out of the norm here.

Even the possibility of a collision due to forgetting to remove it because I had to run the xyz calibration months later and forgot all about the possible issue isn't an acceptable risk 🙂

In any case, I appreciate the ask and wil look into it, but I’m in the middle of a massive project that I need to get to alpha phase first. I will be looking for advanced testers.

Of course. I, as well as most I'm sure, very much appreciate the design efforts of those who choose to share their work. I just spoke up to provide feedback that may be useful in your future efforts. I'll keep an eye on your thingiverse page in case you ever revisit this.

Thanks again,

Steve

Re: Candlelight Vigil, please pay your respects.

I'm guessing you are using a longer screw than mine. I had to cut the spacer in half to engage the nut.
I actually got the old one on. 😉 I'll print out a new one later today and measure my screw also.

Damn it. You are right. Somehow I got an m3x12 in my kit and didn’t check. I’ll fix that too.

I updated with a shorter spacer.

I also updated the nozzle to increase the PINDA hole and adjusted airflow based on CFD sims. Would be nice to get tested with anemometer (hint, hint).

Re: [Cooling] Candlelight Vigil, please pay your respects.

Trust me, the back pins drive me crazy. But it’s really from a purity of design standpoint. In practice I found that I never do full XYZ after the initial build. I did once because I replaced the X ends with R2 and that probably wasn’t necessary. I’m curious if I am out of the norm here.

Even the possibility of a collision due to forgetting to remove it because I had to run the xyz calibration months later and forgot all about the possible issue isn't an acceptable risk 🙂

Fair point.

I took a quick look and remember now the problems. The pins almost come up to the heat block which means I have to raise most of the back half. This creates restrictions on the airflow and I didn't want to compromise that. I'll think about other solutions.

Re: [Cooling] Candlelight Vigil, please pay your respects.

First of all, please don't think less of me for this... It looks like a 8 year old drew it in MS paint... trust me... I wish I could do better... And It would take me 2 weeks to draw a "concept" to share with one of you guys who knows what they are doing in any of the CAD programs.

The basic idea is, mount the cooling fan up high like Jltx, then almost immediately split the air flow into 2 paths, one will go down the left side and one on the right side of the nozzle. My concept is, that much like a leaf blower and a beach ball, the air will flow around the nozzle, and "close" behind it. However we KNOW that air from a single direction is bad. Why? Because by the time the air reaches the back of of the nozzle it's HOT. I would think there are 2 ways to resolve this. 1. Blow the air FASTER. 2. Increase the sources of air. (Jltx took this to the extreme) The issue is, if you have air coming from all sides the air will create a "pressure" at the nozzle, and thus decrease the air flow in the center. (recall, where will the air go? on large prints, the cooling shroud is only 1mm from the part, not allowing air under it at all. So now, rather than the traditional theory of "Air flows where you point it"... The entire "cavity" of the "ducting" is acting as ducting itself, and now the air no longer wants to blow on the nozzle, but out the front.

My concept is to either have 2, or 3, "high pressure" nozzles. If there's two, have them placed at roughly 150 degrees from each other. If there's 3, have them at 120 degrees from each other exactly, but duct the air slightly "around" the nozzle, rather than straight at it. (To keep the air moving faster by not directly blowing into the path of another nozzle's air flow)

Just a theory, and I'm curious if anyone else agree's if it might work?Air flow is weird... ALSO, the purpose is to cool the printed material. 3 nozzles would provide superior cooling, creating a full 360 degree envelope, with minimal flow work, nozzles, and obstructions, thus maximizing flow volume and velocity!

ALSO
@jltx I printed the new part for the heat-bed connection... it works without modification. 🙂

Re: Candlelight Vigil, please pay your respects.

...
I also updated the nozzle to increase the PINDA hole and adjusted airflow based on CFD sims. Would be nice to get tested with anemometer (hint, hint).

Sorry, I'm to busy to test right now. (Of course not to busy to read forums and whine 😮 )

I'm not sure what to make of my measurements anyway as I can't do it with the machine printing. A better test would be some kind of test part that's sensitive to part cooling. Would that be towers, bridges or what? It would be neat to see a change in nozzle directly effecting something on a print.

We need proof! ( Unfortunately probably statistical proof )

To start with just changing fan speed in the g-code would be the easiest. The test part would repeat a feature maybe 10 times and have 2 different fan speeds edited in the g-code. If a difference was found, then orient the test part in four different directions to see if that mattered.

Then see if the new nozzle matters or is the same as PRUSA's with better visibility.

Re: [Cooling] Candlelight Vigil, please pay your respects.

all this for a better... benchy?

I hate that a little boat has become the gold standard for what constitutes acceptable print quality.

Seems a very low return on invested time and effort. I'm probably just the odd-man out on this one.

Good luck though. 🙂

Re: Candlelight Vigil, please pay your respects.

...
I also updated the nozzle to increase the PINDA hole and adjusted airflow based on CFD sims. Would be nice to get tested with anemometer (hint, hint).

Sorry, I'm to busy to test right now. (Of course not to busy to read forums and whine 😮 )

I'm not sure what to make of my measurements anyway as I can't do it with the machine printing. A better test would be some kind of test part that's sensitive to part cooling. Would that be towers, bridges or what? It would be neat to see a change in nozzle directly effecting something on a print.

We need proof! ( Unfortunately probably statistical proof )

To start with just changing fan speed in the g-code would be the easiest. The test part would repeat a feature maybe 10 times and have 2 different fan speeds edited in the g-code. If a difference was found, then orient the test part in four different directions to see if that mattered.

Then see if the new nozzle matters or is the same as PRUSA's with better visibility.

Once I get my easter-prints done. I should be able to test this. I will upgrade my 8 sided cooling test I designed a while back.

75 degree overhang, on all points. (basically pushing the limits. I know with prusa nozzle I could get away with a 75 degree overhang on a benchmark, with minimal sagging. 80 was a bit of a mess.. (and this overhang is on the OPTIMAL direction) so 75 degrees should be a good. "Does it work, or not."