PTFE coupler - Theory and practical details

Discussions concerning the Ultimaker 2 / Ultimaker 2+ series of printers, including the Ultimaker 2 Go
Iltacitoduca
Reactions:
Posts: 51
Joined: September 14th, 2016, 6:39 am
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by Iltacitoduca »

Anders Olsson wrote:
Iltacitoduca wrote: I never bought a hotend from E3D for the simple reason that I never thought that an interchangeable nozzle as proposed by E3D could work without leaks...besides the fact that achieve the precise connections between the steel part and the aluminum side and the nozzle side of so reduced size, it would result in a high precision mechanical processing. and from the proposed price I highly doubt that this is the case
For the same reason I never purchased an Olsson block, no offense, but I really don't like your solution, which more than anything else is a compromise to have an interchangeable nozzle

I not even take into account the ruby nozzle ... in the printing of thermoplastics I do not see the use of nozzles Sapphire / Ruby, and certainly not for the cost since the synthetic rubies cost nothing (more than anything else could be costly one initial investment to get a custom made ruby) nor for the dimension or anything else ...
So all I see behind the use of the nozzle with ruby is unconvincingly ...

Instead, it would be helpful to draw experience from what already exists in the field of injection molding... don't you guys?
Well, the E3D design certainly has it issues, technically it works but I much rather change nozzle on my own heater block than on the E3D.
The E3D hotend appears mainly to be designed to be cheap to manufacture on simple machines with cheap labor, that is my opinion.

My "Olsson block" on the other hand is a rather complex product, as I had no budget to change the printer or to manufacture my own nozzles and as I never thought we would make more than 200 pieces in total :-)
When I only had the prototypes Ultimaker were not interested. When we had sold a few hundred we had done all the R&D, testing and lots of marketing and it was for that reason suddenly hugely interesting for Ultimaker and they did not want to change a thing on it as it was working good enough. And luckily this rather complex block can still be manufactured relatively cheaply using complex machines and well paid labor.

It is often like this in product development, things are rarely optimized but there are often various unpredictable chain events, sourcing issues, assembly considerations and random things that ends up forming a product that is good enough for a particular task.

The ruby is mainly a project for my boron carbide filament, which requires something of that magnitude to be printable at all.
It has advantages for more common filaments too though and I am hoping that solving the wear issue once and for all will have other effects. For example filament manufacturers might start selling more interesting composite filaments if they know there is a nozzle that surely can handle it.

The PBI that gudu used is a typical example of injection molding materials, but you can not always compare injection molding with 3D-printing.
You for example get a much more challenging wear issue on the tip of a 3D-printer nozzle than inside an injection molding machine.
Hi Anders,
WHAT?? :-D sorry but I not agreed with your statement about the complexity of your "Olsson block"... seriously there is nothing complex about it...
nor in the design, nor in the manufacturing process.
And what at UM choose for their 3D printer is under the eyes of all ... They saw a market opportunity and have ridden the wave... opinable but it's the law of market, nothing more nothing less

Printing B4C with a material of a lower hardness, will be tricky... especially dispersed in a fluid medium... even the inner wall before the nozzle tip will wear out rapidly...

Advantage over other materials, perhaps...
Yes absolutely sure that you can not always compare the two procedures/processes and I never wrote that are comparable I wrote that from similar processes we can take advices.

Instead, about the "much more challenging wear issue on the tip" I have many doubt ... first because the tip operates at pressures which are at least 10 or 20 times lower than that in an injection molding process
and second because the material that came out (unless it is not B4C) may, at most, have a polishing effect... (but I doubt about it, due to the fact that it's brushing without pressure over the nozzle tip)
Kindly regards
Claudio
User avatar
Neotko
Reactions:
Posts: 1142
Joined: February 7th, 2016, 7:02 pm
Location: Madrid
3D Printer(s): UMO+ x2.5
Contact:

Re: PTFE coupler - Theory and practical details

Post by Neotko »

You should have made it and sell it them.

You want to complain about every aspect, sure go ahead.
User avatar
ivan.akapulko
Reactions:
Posts: 389
Joined: February 11th, 2016, 11:27 am
Location: Ust-Kamenogorsk, Kazakhstan
3D Printer(s): Ultimaker 2
Contact:

Re: PTFE coupler - Theory and practical details

Post by ivan.akapulko »

Injection molding in my opinion much more expensive, if we are talking about making hotends. Turning metal is worth a penny, and the level tolerances of the product not so much affect the final cost, at least where I live. I admit that in Europe/USA, the situation may be radically different from mine. Much more interesting may be, in my opinion, more serious study in the field of metallurgy in relation to heat treatment of finished products to modify physical characteristics of used metals and alloys, and optimization of internal and external geometry, and then applying heat-insulating coatings to reduce loss of efficiency.
User avatar
ivan.akapulko
Reactions:
Posts: 389
Joined: February 11th, 2016, 11:27 am
Location: Ust-Kamenogorsk, Kazakhstan
3D Printer(s): Ultimaker 2
Contact:

Re: PTFE coupler - Theory and practical details

Post by ivan.akapulko »

In fact, thermodynamic modeling of the existing hotends always remains behind the scenes, and results of tests no lights. Although it would be worth it ;-) The use of bronze in products with price-level like 3d systems or UM3 ultimately determined by the cost of production and methodology of earnings on spare parts. It's like the automotive industry, you drive 100k miles, and then she begins to rob your wallet. And is inherent in the design of the car is still in the design phase. Because the manufacturer wants to eat, and the risks and volumes of loans are high and competitors are not asleep.
Anders Olsson
Reactions:
Posts: 415
Joined: February 8th, 2016, 8:37 pm
Location: Uppsala, Sweden
3D Printer(s): UM2 UM2E, UM2Go, UM3, Delta Tower, Form 1+, Form 2
Contact:

Re: PTFE coupler - Theory and practical details

Post by Anders Olsson »

Iltacitoduca wrote:WHAT?? :-D sorry but I not agreed with your statement about the complexity of your "Olsson block"... seriously there is nothing complex about it...
nor in the design, nor in the manufacturing process.
Without explaining manufacturing complications in detail, let's just say that there is a reason why other companies like E3D or even Ultimaker (on UM3) would split the detail in one square and one cylindrical part... :-)
Iltacitoduca
Reactions:
Posts: 51
Joined: September 14th, 2016, 6:39 am
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by Iltacitoduca »

Neotko wrote:You should have made it and sell it them.

You want to complain about every aspect, sure go ahead.
that useless comment
just dealing with every aspect, which is the problem? I am touching your preferred toy? don't like to read some critical comments? did it? really? LOL
Iltacitoduca
Reactions:
Posts: 51
Joined: September 14th, 2016, 6:39 am
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by Iltacitoduca »

Anders Olsson wrote:
Iltacitoduca wrote:WHAT?? :-D sorry but I not agreed with your statement about the complexity of your "Olsson block"... seriously there is nothing complex about it...
nor in the design, nor in the manufacturing process.
Without explaining manufacturing complications in detail, let's just say that there is a reason why other companies like E3D or even Ultimaker (on UM3) would split the detail in one square and one cylindrical part... :-)

Without arguments is hard to argue :-)
So, or you bring some concrete fact or there isn't progress in the discussion and accordingly in the search for a better solution that can involve more people :-)
User avatar
Neotko
Reactions:
Posts: 1142
Joined: February 7th, 2016, 7:02 pm
Location: Madrid
3D Printer(s): UMO+ x2.5
Contact:

Re: PTFE coupler - Theory and practical details

Post by Neotko »

Iltacitoduca wrote:
Neotko wrote:You should have made it and sell it them.

You want to complain about every aspect, sure go ahead.
that useless comment
just dealing with every aspect, which is the problem? I am touching your preferred toy? don't like to read some critical comments? did it? really? LOL
I can't care less. But your constant bashing is annoying. If you gonna bash on other people work about cost or quality without taking into account that humans need to make a living and nobody leaves from air then yeah is annoying.

Also most of the time I see people doing that they just theorize and they don't release anything interesting. So if you are gonna make something so precise as you want, please go ahead.
Iltacitoduca
Reactions:
Posts: 51
Joined: September 14th, 2016, 6:39 am
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by Iltacitoduca »

Neotko wrote:
Iltacitoduca wrote:
Neotko wrote:You should have made it and sell it them.

You want to complain about every aspect, sure go ahead.
that useless comment
just dealing with every aspect, which is the problem? I am touching your preferred toy? don't like to read some critical comments? did it? really? LOL
I can't care less. But your constant bashing is annoying. If you gonna bash on other people work about cost or quality without taking into account that humans need to make a living and nobody leaves from air then yeah is annoying.

Also most of the time I see people doing that they just theorize and they don't release anything interesting. So if you are gonna make something so precise as you want, please go ahead.
LOL!
User avatar
Izzy
Reactions:
Posts: 604
Joined: February 12th, 2016, 2:29 am
Location: England
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by Izzy »

Nothing is impossible, but may be highly improbable.
As an engineer you would be aware that It is possible to mass produce a high precision heater block CNC refined and pared down with intricate filament heating and guide tubes, at a cost..
If there are only a few hundred the cost would be very high perhaps too high for us mere mortals to afford and test on our home hobby machines. And cost is the biggest factor, these machines are aimed at the high side home hobbiest and I can't afford a €600 print head, so Anders solution is a simple effective solution giving flexability and serviceability for the Ultimakers.
From the Olsson block and now the matchless I have a range of nozzles from 0.1mm upto 2.0mm to play with.
As Engineers we can learn more from our mistakes than our successes, it is easy to over engineer or under engineer things, we will never get it perfect, we will always see errors in our own work, but we may come close, and we can always guarantee criticism from others and if we are lucky perhaps a pat on the back.
Anders Olsson
Reactions:
Posts: 415
Joined: February 8th, 2016, 8:37 pm
Location: Uppsala, Sweden
3D Printer(s): UM2 UM2E, UM2Go, UM3, Delta Tower, Form 1+, Form 2
Contact:

Re: PTFE coupler - Theory and practical details

Post by Anders Olsson »

Iltacitoduca wrote:Without arguments is hard to argue :-)
So, or you bring some concrete fact or there isn't progress in the discussion and accordingly in the search for a better solution that can involve more people :-)
Well, you make a lot of odd claims that suggests that you have very little experience in machining, so I have to somehow test your knowledge by giving you leads as in my previous reply, to know what level to start at :-) (I thought the cylindrical and rectangular part lead would be enough)
Anyone with experience in machining would immediately understand why a E3D-block is simpler to machine and start questioning why the Olsson-block has to look like that. The complicated part mostly has to do with the precision needed in the positioning of the different features!

At both the original UM2 block and the Olsson block, it is critical that the screw that fixes the heater and sensor is positioned with precisely the right distance in relation to the heater and the sensor, or the fixing will not work!
The same thing goes for the heat break thread in relation to the nozzle screw thread.
Both these things are critical for the functionality and relatively time consuming to do quality controls on.

If you want to make let's say 1000 blocks you are not going to measure all of them.
Then you need a machine that can machine all sides of the block in one go, and that can measure in it's own tools, so you can guarantee that the holes are in the right position in relation to each other.

That means a CNC lathe with powered tools that can machine from different directions, material feeding and removal function for ready pieces.
An additional complication is that you have to cut the ready part from the raw material, The raw material would typically be round, so your machine needs to be able to grip the ready square part and hold it while it is being cut of from the raw material.

For the the seat for the nozzle, which is wider than the nozzle thread, you need to get a custom made tool that can go inside the nozzle hole and mill that pocket, without crashing into the threads.

All these things can be avoided by making a through threaded M6 hole like in the E3D-block and on the E3D block the relative positioning of the holes is not critical for the functionality.

So while the E3D-blocks can be made with relatively low precision in any 3-axis milling machine, with cheap manual labor handling the pieces between the different machining steps, the Olsson block needs a machine that might not be available in any workshop in a small country like Sweden.

The Olsson block has advantages over the E3D though such as that you don't need to hold the block when tightening the nozzle and that plastic will only leak downwards in case of a leak. Also you don´t need to modify the UM2 to fit the Olsson block (a more simple to machine design would have needed more modifications to the printer).

The whole thing is also about volumes and quality. If I would have made no more than 200 pieces in total, as I thought when designing it, we could have used a normal 3-axis CNC and simply tested all blocks before delivery.
When volumes are much larger, many different people are involved and the blocks are installed in brand new machines for all customers rather than being bought by tinkers, you have to deliver a product that does not need quality control and that is never outside the specifications.
Iltacitoduca
Reactions:
Posts: 51
Joined: September 14th, 2016, 6:39 am
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by Iltacitoduca »

ivan.akapulko wrote:Injection molding in my opinion much more expensive, if we are talking about making hotends. Turning metal is worth a penny, and the level tolerances of the product not so much affect the final cost, at least where I live. I admit that in Europe/USA, the situation may be radically different from mine. Much more interesting may be, in my opinion, more serious study in the field of metallurgy in relation to heat treatment of finished products to modify physical characteristics of used metals and alloys, and optimization of internal and external geometry, and then applying heat-insulating coatings to reduce loss of efficiency.
Yes! That's all true.

Is where I find a lot of interesting info is right on injection moulding nozzles, use of metals, finishing treatments, shapes, heating methods, thermal conductivity study, there are a lot of manufacturer that live on the production of nozzles for plastics moulding.
ivan.akapulko wrote:In fact, thermodynamic modeling of the existing hotends always remains behind the scenes, and results of tests no lights. Although it would be worth it ;-) The use of bronze in products with price-level like 3d systems or UM3 ultimately determined by the cost of production and methodology of earnings on spare parts. It's like the automotive industry, you drive 100k miles, and then she begins to rob your wallet. And is inherent in the design of the car is still in the design phase. Because the manufacturer wants to eat, and the risks and volumes of loans are high and competitors are not asleep.

Agree. (the problem is that in some cases they start to rob your wallet just after 100 hours... or less... imagine your car stop working after 100 hours or every 100 hours! will be unacceptable...)
The fact is that there are no Competitors.. I mean there are "competitors" but they are all at the same level! and this is quite shameful
Iltacitoduca
Reactions:
Posts: 51
Joined: September 14th, 2016, 6:39 am
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by Iltacitoduca »

Anders Olsson wrote:
Iltacitoduca wrote:Without arguments is hard to argue :-)
So, or you bring some concrete fact or there isn't progress in the discussion and accordingly in the search for a better solution that can involve more people :-)
Well, you make a lot of odd claims that suggests that you have very little experience in machining, so I have to somehow test your knowledge by giving you leads as in my previous reply, to know what level to start at :-) (I thought the cylindrical and rectangular part lead would be enough)
Anyone with experience in machining would immediately understand why a E3D-block is simpler to machine and start questioning why the Olsson-block has to look like that. The complicated part mostly has to do with the precision needed in the positioning of the different features!

At both the original UM2 block and the Olsson block, it is critical that the screw that fixes the heater and sensor is positioned with precisely the right distance in relation to the heater and the sensor, or the fixing will not work!
The same thing goes for the heat break thread in relation to the nozzle screw thread.
Both these things are critical for the functionality and relatively time consuming to do quality controls on.

If you want to make let's say 1000 blocks you are not going to measure all of them.
Then you need a machine that can machine all sides of the block in one go, and that can measure in it's own tools, so you can guarantee that the holes are in the right position in relation to each other.

That means a CNC lathe with powered tools that can machine from different directions, material feeding and removal function for ready pieces.
An additional complication is that you have to cut the ready part from the raw material, The raw material would typically be round, so your machine needs to be able to grip the ready square part and hold it while it is being cut of from the raw material.

For the the seat for the nozzle, which is wider than the nozzle thread, you need to get a custom made tool that can go inside the nozzle hole and mill that pocket, without crashing into the threads.

All these things can be avoided by making a through threaded M6 hole like in the E3D-block and on the E3D block the relative positioning of the holes is not critical for the functionality.

So while the E3D-blocks can be made with relatively low precision in any 3-axis milling machine, with cheap manual labor handling the pieces between the different machining steps, the Olsson block needs a machine that might not be available in any workshop in a small country like Sweden.

The Olsson block has advantages over the E3D though such as that you don't need to hold the block when tightening the nozzle and that plastic will only leak downwards in case of a leak. Also you don´t need to modify the UM2 to fit the Olsson block (a more simple to machine design would have needed more modifications to the printer).

The whole thing is also about volumes and quality. If I would have made no more than 200 pieces in total, as I thought when designing it, we could have used a normal 3-axis CNC and simply tested all blocks before delivery.
When volumes are much larger, many different people are involved and the blocks are installed in brand new machines for all customers rather than being bought by tinkers, you have to deliver a product that does not need quality control and that is never outside the specifications.

Anders you are talking as if software like Solidworks or Solidcam doesn't exist! :lol:
Perhaps you or at UM have designed the hot-end at a valuable drawing table, but it takes less than an hour to reproduce your/UM hot-end with Solidworks and even less to pass it to Solidcam... so..
I Still Not See The Complexity of Your or UM design!
Raw material are typical round? :lol:
But please!! Stop kidding! you can find raw material in round, square/rectangular and exagonal shape as easy as buying a red/green or yellow apple!

But really ? Are you trying to sustain that the complexity is this? the raw material? the precision on design? The Machining ?

WHO cares about E3D!!! I NEVER compared those toys, right? I NEVER wrote E3D is better!
Seems more You have a fixation with that hot-end! :lol:
I already state That I NEVER had an E3D and I will NEVER buy it !! Fullstop
Anders Olsson
Reactions:
Posts: 415
Joined: February 8th, 2016, 8:37 pm
Location: Uppsala, Sweden
3D Printer(s): UM2 UM2E, UM2Go, UM3, Delta Tower, Form 1+, Form 2
Contact:

Re: PTFE coupler - Theory and practical details

Post by Anders Olsson »

Iltacitoduca wrote:Anders you are talking as if software like Solidworks or Solidcam doesn't exist! :lol:
Perhaps you or at UM have designed the hot-end at a valuable drawing table, but it takes less than an hour to reproduce your/UM hot-end with Solidworks and even less to pass it to Solidcam... so..
CAD software has very little to do with machining. What I am trying to explain to you has nothing to do with the drawing itself but how a CNC machine works.
I Still Not See The Complexity of Your or UM design!
The complexity lies within that the functionality of the block relies on accurate manufacturing, where things has to be accurate relative to each other in all three dimensions. That is makes things very difficult if you are going large scale on simple machines with manual handling.
One would normally design components so that they don't require this type of precision, but in this case it was not possible to make it simple to manufacture.
Raw material are typical round? :lol:
But please!! Stop kidding! you can find raw material in round, square/rectangular and exagonal shape as easy as buying a red/green or yellow apple!
You might think that for example a nozzle is made from a hex-shaped rod?
It is not, and there are very good reasons for that!
It has to do with, among other things, how the feeding mechanism of a CNC lathe works.
The same goes for the block, it might sound contradictory but it starts life as a round brass rod. :-)
WHO cares about E3D!!! I NEVER compared those toys, right? I NEVER wrote E3D is better!
Well, E3D is a very good example of something that is designed to be easy to manufacture in simple machines, the opposite to the Olsson block in other words.
As far as I know it is also the most common hotend on the market, so I thought it is a rather good example in this discussion, I still don't like E3D hotends though :-)

I hope that others reading this catch some interesting insight in manufacturing by the way, so that I am not just trying to explain things to this one person..
avogra
Reactions:
Posts: 4
Joined: May 30th, 2016, 3:58 am
3D Printer(s): Ultimaker 2

Re: PTFE coupler - Theory and practical details

Post by avogra »

Absolutly, Thank you for the insight!
@iltacitoduca: i don't understand what you complain about. If all the existing hotends are so bad or too expensive and you know how to make it better, then you should go and sell your hotend and become a rich person. Or you could make a prototype and demonstrate its performance. At least you could tell people like Anders how exactly they can improve their hotends. At the moment you only say it is bad or too expensive. That isn't helping anyone.
Post Reply

Return to “Ultimaker 2 / Ultimaker 2+ / Extended & Go”