E3D V4 All metal hotend

[Polygonhell]

I'm running 2mm retracts at 25mm/s with no stringing. I removed all the slop from the PTFE-PTC connections on my Bowden also. Now that I can print PLA at 180-185°C again, there is mush less problem with stringing. I do use a 2mm Z-lift, which also helps.

Yes that's the biggest part of the win, and it applies to all plastics not just PLA, because you can print at lower temperature, stringing is greatly reduced without having to run a lot of retraction.
I still have the original PTFE-PTC connector on the easystruder side, which is introducing some additional hysteresis, so I need a bit more retract.
Having said that it isn't horrible at 2mm, I printed the voronoi easter bunny yesterday and there are some hairs on some of the holes, KissSlicer likely would have been string free because of wipe, Cura however wasn't.

[DavidF]

Well today I had the coupler on my bowden finally fail after 7 months. I have had the e3d hot end sitting and waiting for a reason to install it, I guess now is a good time since I don't have a spare coupler handy. I hope the new hot end works out of the box, but if it doesn't, I have drill bits handy. Thanks for doing the pilot tests for me LOL

[mhackney]

Thanks for the feedback broose. Good luck DavidF!

Polygonhell, I did end up making a fitting for the ezStruder to use the PTC that came with the E3D hot ends. The tap was a bit expensive. I am planning to make these for sale if people are interested, they simply replace the existing aluminum part. But, the other advantage is that the PTFE tube passes through the PTC all the way to the pinch roller/cobbed gear so the path is nice and clean.

cheers,
Michael

[McSlappy]

Do my eyes deceive me, or is your name now red Micheal?

[mhackney]

Yes, it's red. It actually should have been red for awhile, I was the first moderator but only had permissions to delete spam posts, etc. Sometime last year a few of us were promoted to admins and I was inadvertently not updated - until today. This actually came about as part of a suggestion I made to Steve and John to change the registration process so admins have to approve new registrations in order to eliminate spam. We met online tonight for a couple of hours and hashed out the details and updated the registration process. I run the reelsmithing forum and made this change 2 years ago and have completely eliminated spam. The key is to let newly registering members know that their accounts will beed to be approved in order to eliminate spam. Communication is the key so everyone knows what to expect. So, that's what is going to happen here starting tonight. Look forward to a spam free forum!

cheers,
Michael

[bubbasnow]

Yes, it's red. It actually should have been red for awhile, I was the first moderator but only had permissions to delete spam posts, etc. Sometime last year a few of us were promoted to admins and I was inadvertently not updated - until today. This actually came about as part of a suggestion I made to Steve and John to change the registration process so admins have to approve new registrations in order to eliminate spam. We met online tonight for a couple of hours and hashed out the details and updated the registration process. I run the reelsmithing forum and made this change 2 years ago and have completely eliminated spam. The key is to let newly registering members know that their accounts will beed to be approved in order to eliminate spam. Communication is the key so everyone knows what to expect. So, that's what is going to happen here starting tonight. Look forward to a spam free forum!

cheers,
Michael

l;alkakldjkf;alskjd;flaksdjf;alkj;ldkjf;alksjfkdls;asdwqpoiopierpoqpw

ill miss you spammers... best conversations i ever had

[mhackney]

lol

[McSlappy]

Well congrats! And I do look forward to a spam free forum.

I don't understand how the spam could possibly work for the people making them. Why on earth would I respond to some gibberish stolen from wikipedia and choose to buy a snakeskin shoe or something? Puzzles me.

[Eaglezsoar]

Yes, it's red. It actually should have been red for awhile, I was the first moderator but only had permissions to delete spam posts, etc. Sometime last year a few of us were promoted to admins and I was inadvertently not updated - until today. This actually came about as part of a suggestion I made to Steve and John to change the registration process so admins have to approve new registrations in order to eliminate spam. We met online tonight for a couple of hours and hashed out the details and updated the registration process. I run the reelsmithing forum and made this change 2 years ago and have completely eliminated spam. The key is to let newly registering members know that their accounts will beed to be approved in order to eliminate spam. Communication is the key so everyone knows what to expect. So, that's what is going to happen here starting tonight. Look forward to a spam free forum!

cheers,
Michael

I am looking forward to a spam free forum. Last weekend was a disaster with all the spam.
Your name looks good in red and I know you and the other Admins will do a good job, congrats!

[SanjayM]

Just to say - that we're actually going to take a huge road trip off the island, going to our manufacturers, do a bunch of measuring, etc etc. We're absolutely looking into this whole nozzle orifice length as a possible avenue for combating issues. No verdict as yet, but much thanks to Mr Hackney for all the investigatory work.

[mhackney]

Good news Sanjay! I've collected info on nozzle length for 25 nozzles and all of them were well over 1mm. Most were 2mm and a few were 2.5 to as high as 3mm!

The 3mm filament nozzles seem to be the same but because of the geometry, the "fix" with a standard drill bit is too risky. A bit with a steeper angle will be required to reduce the bore length on these.

Cheers,
Michael

[SanjayM]

I've just gone through a few hundred nozzles with a set of callipers and a drill bit as a depth probe. Including some from older batches. All of which measured around 1.05-1.10, in a single batch they vary less than 0.03 or so. My method was to insert a 2mm drill bit into the jaws of the calipers, zero the calipers, and then pop a nozzle on the drill bit and re-measure. When using a hot-pull method to take an impression from the nozzle it seems like it would be quite likely that the small nozzle orifice portion would be subject to some stretching/elongation, especially considering it has much more surface area per unit volume compared to the rest of the piece being pulled out, so it is more subject to being soft all the way through. I'll do a hot-pull on some of my machines later and measure the impression to see what it looks like compared to the inside of the nozzle.

Looking at the way in which the parts are handed from one turret of the CNC lathe to the other and assessing the GCODE used to create the parts it doesn't seem very likely (impossible?) for the machine to create a nozzle orifice of more than 2mm as the final drilling process doesn't involve the small diameter bit encroaching on the part by more than 2mm at any time. The only way I can see of the orifice being lengthened is for the nozzle to slip in the chuck of the second stage of the lathe, but if that did happen you'd get big very noticeable errors on features of the nozzle such as the cone, the tip-flat, and the underside of the hex portion.

With that said I don't have access to every nozzle ever shipped to run my calipers over.

If you wanted to try a similar thing with a 3mm drill bit, you could try using something like a 1-2mm drill bit to peck out some material, instead of the full 3.2mm (not 1/8th" as many here are talking about). Would save the hassle of grinding.

[mhackney]

Sanjay thanks for the info.

I use both measurement techniques and the results are nearly identical (taking into account the tip bevel on the drill which allows the drill tip to enter the office bore a little - this gives a slightly shorter bore reading).

Please note - the method I use to measure the plug involves freezing the nozzle with the PLA in it. There is no elongation of the PLA extrusion in the bore when I remove it to measure. This gives a very accurate profile and length for the tip and bore. The secret is to freeze the tip and PLA.

All 5 of my nozzles are right at 2 mm measured both by the frozen plug method and the caliper-drill bit-nozzle method you describe. The first nozzle I got last year came with my E3D, the other 4 came with the Kraken. Several members here have measured 2, 2-1/2 mm and Polygonhell (who knows what he's doing!) has one that is 3mm, one at 2mm and one at 1mm in his sample of 3 (look back a few pages in this thread).

So, it sounds like the design spec in 1.0mm +/- a bit. I didn't shorten one of mine to that to test but I would think it would be much better than 2mm! This could explain the variability in performance some see vs others. For me, all of my nozzles had very long 2mm bores. Following the lead of J-Head and other nozzles, I went to .5mm and am VERY happy with the performance now. The difference is obvious extruding into air. With the short bore, extrusion starts immediately and ends very shortly after after the extruder stops (less than a second), with the 2mm bore, it takes a bit of time for extrusion to start and then it continues for well over 3 seconds after the extruder stops (extruding 10mm at 50mm/s into air). A very marked difference.

Cheers,
Michael

[Polygonhell]

In my case the 3mm one was the old style nozzle from one of the original V4's, the 2mm one was from the V5 I got in my latest order with the Kraken and the ~1mm one is from the Kraken packaging I haven't bothered measuring the other 4 I have.

[McSlappy]

Thanks Sanjay for checking in we're all e3d enthusiasts here and it's always nice when you show up to comment.

Anyway, here's my testing - please critique my methodology if I'm incorrect.

I ordered my head about a month ago, so these are (I assume) pretty recent nozzles. I have a .6 a .4 and a .25mm
I measured just the .25 and .6 since the .4 is being used right now since I reinstalled the head last night (man I missed it!)

I did this same technique with my original bit and it came out at ~2mm

So I began with this drill bit - a 2mm x 48mm bit. The extra hi-temp silicone in the flutes is a special feature that improves the accuracy of it. Yep.
[img]http://i.imgur.com/jk29UcT.jpg?2[/img]

Here's the .25mm nozzle - spot on 1mm which is nice.
[img]http://i.imgur.com/QMUVluh.jpg?2[/img]

And the .6mm nozzle at 1.1mm (it looks like 1.2 in the picture, but I checked again and it's 1.1mm)
[img]http://i.imgur.com/4sqyFiD.jpg?2[/img]

This is just an approximation due to the shoulder on the bit and the nozzle being different. But I can assume that these are from a different batch to my original nozzle.

I have yet to stress test a PLA run. I've been running PLA for a few hours with no issues, but nothing terribly taxing yet.

Anyway just thought I'd let you know that my recent ones seem to be correct in their nozzle length. Perhaps there have been some off batches?

[Generic Default]

Sanjay,

I'm not sure where you get your nozzles made, but make sure that whoever does the CNC machining is setting the tool offsets on their swiss type CNC lathe.
Sometimes the operator will get lazy when changing out drills and they'll just slide a new drill bit into the tool holder and eyeball it to make sure it's close. Eyeballing gets stuff to within a millimeter or two, which is the error on the nozzles we're seeing.

You should also consider specifying that they use sharper angle drill bits (might even be in the chamfer end mill category) on the main bore of the nozzles. Experiment more with orifice length and different materials, hopefully you will have the same results that the users here have.



Also, I just saw some Chinese knockoffs of your E3D hotend for about 50 dollars, and they looked identical (except for the PTC connector). If you're getting your parts machined in China, make sure the factory isn't making extras and selling them to competitors. Stuff like that happens a lot!

[SanjayM]

Mike
I'm totally inclined to believe you, especially if you're freezing out the PLA (which I misread as you doing a hot pull at glass transition temp to release the filament - apologies). I think it's possible that there may have been a batch, or portion of a batch with slightly offset nozzle lengths. I've measured from a whole range of batches, but it's perfectly feasible that sub-portion of a batch may have had a drill bit offset slightly - I just can't find any myself.

Polygonhell
I've just gone back and dug out a v4 and that 1 comes in at 1.05 (sample size of 1 is a bit lame I know).

McSlappy
That's pretty much exactly my methodology - except with digitals and zero'd the calipers on the length of the drill bit so I didn't have to bother with maths! I'm interested to know how you get on.

Generic Default
That's my current best theory as well. I just need some hard data/example to go back to them with.

We're totally redesigning the nozzles for v6, with all this feedback in mind. They'll be backwards/forwards compatible with v4/v5 nozzles so you can bring old nozzles forward and new nozzles backwards if that takes your fancy.

We've seen those too - we don't manufacture any machined parts in China, parts like these need a level of care and supervision by us that we just can't get by going thousands of miles away. Being able to drive over there and discuss things is totally crucial to our business.

On a side note we've purchased some of these and the first arrived in the post yesterday. I'm obviously biased here, but I can tell you without a shadow of a doubt that they are hilariously bad, to the point of being dangerous in some ways. We'll be doing a review of sorts on them once they all arrive.

[mhackney]

Thanks Sanjay. This might be one of those situations where the squeaky wheel gets the grease - all of the data that's been sent to me the past week or so have been 1.5mm bore lengths or longer - but these are all folks that have PLA issues. The ones with the correct bore probably aren't compelled to measure since they don't have a problem! It is odd that all 4 of my Kraken and the months earlier V5 are all much longer but that could just be the luck of the draw.

When it comes time to get some testing on the v6 and v6 nozzles I'll throw my hat into the ring - I think you can tell that I push these to about the limit and if there is anything that needs to be tweaked I'll find it!

Cheers,
Michael

[mhackney]

McSlappy - there is a relationship between diameter and length of bore for an extrusion nozzle. If the bore is too short with respect to diameter, you get pig tailing and swelling (aka die swelling) as material comes out. If it is too long, it requires a lot of pressure to push the material through. Sometimes this is desired - like in a particulate filled material, as it can along the particles in parallel with the extrusion flow. For us, this is an issue and is what I've discovered to be the case with the nozzles I have. I spent a few years in a research lab extruding ceramic polymer precursor fibers. I vaguely recall that the optimal diameter to bore length was in the neighborhood of 1 to 1.5 for those materials. Extrusion speed and other factors affect this too. Every material is different and nozzles on 3D printers are a compromise to support multiple materials.

That said, a 1:1 to 1:2 diameter to length seems to me to be reasonable. So, for a .4mm orifice, .4mm to .8 mm bore is probably a good target for very viscous materials like molten PLA and ABS. The larger the orifice diameter, the longer the bore and vice versa. I would anticipate your .25 mm nozzle with a 1mm bore to be problematic. I haven't measured my .25mm nozzle yet nor have I attempted to print with it. I guess that will be next on the list!

[Generic Default]

It seems like PLA causes all of the problems with hotends. I've been printing almost exclusively trimmer line nylon for around a year now and I haven't had a single jam, not even once. If making the orifice ratio shorter fixed PLA problems, will it cause more problems with other plastics? I'm mostly concerned about oozing. It seems intuitive to me that having a longer orifice ratio would reduce oozing, but your results indicate the opposite.

And Sanjay, hopefully you will be the only customer that knockoff extruder company sells to! More realistically though, their cheaper prices will attract newer 3d printer hobbyists who don't do research before buying products. And when they have terrible experiences with the cheap knockoffs, they will be hesitant to buy anything that looks similar (your hotend). Also be prepared to deal with the knockoff hotend giving your legitimate hotend a bad reputation since they look nearly identical.

It sucks when people take your design/idea/invention and make low quality copies of it. It sucks more when you lose money from it. It sucks even more when you get blamed for other people's failures! Just sayin'.

[Nylocke]

This whole nozzle study thing is making me wonder about my Schools Makerbot 2... I'm wondering what kinda nozzle profile could be making it print with PLA so poorly... I'm going to do a study with my CAD teacher on it, and maybe see if we can modify those nozzles to get better results. Ill also be looking around for a 70 degree ish bit to mod my 3mm E3D nozzle

[mhackney]

Generic Default - "if out ain't broken, don't fix it!" If you are getting good results I wouldn't mess with it. The reason why decreasing the bore length DECREASES oozing for PLA is that it significantly reduces back pressure. Every plastic is different and behaves differently at melt temperatures. PLA is more persnickety than others.

[Polygonhell]

I agree with the if it aint broke don't fix it comment.
I've been printing Taulman Bridge a lot with mine recently, and the modified nozzle actually helps with stringing because the decreased back-pressure allows me to print at a lower temperature. Having said that there are other contributing factors with the batch of Bridge I have, notably, the filament is very oversized 1.8 - 1.95mm and requires extra extruder pressure to get it into the nozzle.

[Eaglezsoar]

I agree with the if it aint broke don't fix it comment.
I've been printing Taulman Bridge a lot with mine recently, and the modified nozzle actually helps with stringing because the decreased back-pressure allows me to print at a lower temperature. Having said that there are other contributing factors with the batch of Bridge I have, notably, the filament is very oversized 1.8 - 1.95mm and requires extra extruder pressure to get it into the nozzle.

Wow, are all the Taulman products that far out of spec for the diameter?

[Polygonhell]

I agree with the if it aint broke don't fix it comment.
I've been printing Taulman Bridge a lot with mine recently, and the modified nozzle actually helps with stringing because the decreased back-pressure allows me to print at a lower temperature. Having said that there are other contributing factors with the batch of Bridge I have, notably, the filament is very oversized 1.8 - 1.95mm and requires extra extruder pressure to get it into the nozzle.

Wow, are all the Taulman products that far out of spec for the diameter?

Bridge is the only one I've seen this far out of spec, the wider sections actually jam in the extruder body which is specced at 2mm.
It's unfortunate, I really like the material, it looks and feels like the other Nylons, it prints well at 230C, seems to be mostly immune to the worst of the Nylon moisture issues and I can actually print large flat areas with it that stay flat.