FDM for Production  

RE: FDM for Production

Before you hop in, I would suggest you hire a print farm to run some samples for you.

That way you can test if the material is appropriate and if the price is in line with what you expect.

3D printing has it's own learning curve, and your own initial prints are unlikely to be successful. Large prints have their own unique problems.

Get your own printer when the quantities involved justify the cost - and the pain of that learning curve.

RE: FDM for Production

Also you need to know the glass transition temperature of any material you plan on using, which with your temperatures limits the choice somewhat.

https://matterhackers.helpscoutdocs.com/article/109-what-material-should-i-use-in-a-high-temperature-environment

The above is an example of suitable materials.

The higher the printing temperature, also brings with it it's own set of problems, namely warping and you start to need to look at heated enclosures which may require moving electronics out of the enclosure and improving the heatbreak cooling differential, and post processing which may involve heating and cooling the finished print gently to relieve stress referred to as annealing. This can (Usually does) result in dimensional changes.

If you do get all this right, then you should have a part you can use for your application without further issues.

Not trying to put you off, just expect to have to work on this to get it perfect, to be honest, personally I like a challenge, but don't expect to just buy some filament load it and print and get ideal results first time. If you do manage it first time 😀 , buy a lottery ticket as well and post your settings etc.. 

RE: FDM for Production

@robert-rmm200

I will probably get a part printed before I buy.  But before even that I need to determine if it is feasible and find an appropriate material.  That's what I will focus on for now and if everything looks promising, then I'll buy a printer or at least get a part made for me so I can test it.  One good thing going for me is that the part is a hot air plenum so aside from some mount holes that will need to line up, it is not critical dimensionally.

I know there is a learning curve and I am comfortable with that.

RE: FDM for Production

@chocki

Thanks for the link to the high temp materials rundown.  That is super helpful, exactly what I needed to see.  When planning to thermoform or injection mold this part, I had planned on polycarbonate, but it seems that is quite troublesome, especially for a beginner.  Because of that article you linked, it appears that ABS is likely a good choice and Nylon is also a possibility.  I did notice the very low shrinkage of Nylon which makes me see it as an alternative if shrink/stress is a problem.

I happen to have a large lab oven that could handle dozens of these parts.  The oven is super precise and allows for ramp up, hold, ramp down cycles, so stress relief will not be a problem.

I too like a challenge.  I also like that our product is used for curing SLA parts, so incorporating components into the design that are made with additive manufacturing feels right.  It also allows us some design freedom and unshackles us from prohibitively high tooling costs associated with injection molding.  I think we can do this and I appreciate the help I am getting here already.

RE: FDM for Production

Here is a link to the file if anybody wants to take a look at it.  I was just looking at PETG and it may work.  It may not be quite up to the task temperature-wise, but it's not far off and I probably over estimated the temperature of the environment.  Also, there is virtually no load on this part so heat deflection forces used during testing don't apply fully.  The low shrink will make the large size less likely to fail.  If anybody has some time to pull this file up, I'd love to know how long it would take to print and how much material it would consume with standard infill settings and a high layer thickness...as high as practical because the part is not critical.  I'm going to do this myself as I have downloaded the software, but I will not be confident of my settings.  I'm going to hope PETG will work, but fall back to ASA if not.  At least that's what I am thinking at the moment.

100316-000_PLENUM

RE: FDM for Production

Also, I can't stress enough, filament drying before use, you already have a head start due to the super precise oven, and I suggest you watch this video which another member Jmone linked to a thread about drying filaments. The thread is worth reading through as well.

https://forum.prusa3d.com/forum/original-prusa-i3-mk3s-mk3-general-discussion-announcements-and-releases/drying-times-and-temps-in-dehydrator/paged/11/#post-192793

Drying does make a noticeable difference to print quality and strength for all filaments not just exotics.

RE: FDM for Production

That plenum is going to be challenging to print as is, my suggestion would be to redesign it with the large flat part as the base, avoid gentle curves where possible going up as the begining part of the curve will be at a greater than 65 deg angle from vertical so will most likely fail, think bevels instead of curves for the first part, and the flat rim unless supported will also not be possible to print. Can you make it as a bevel with the only flat parts being where the screws would go?, bridging should work across the bevel where the flat part for the bolts would be needed, so in effect cut a cylinder out of the bevel. Or can the bolts be screwed in from the other part, then create captive parts for nuts to be inserted into the print afterwards, there are various options available which may need to be considered for FDM prints which differ from other traditional manufacturing methods.

Something more to think about. 😀 

Feel free to ask, there are plenty of knowledgeable members on this forum whom have considerably more knowledge and experience than me.

Kind regards

Andrew

RE: FDM for Production

From experience with PETG, you are better off using a nickel plated copper nozzle rather than brass as PETG sticks to brass really well and is a pig to clean off, with the nickel plated copper nozzle, as it is heating up and gets towards 150 deg, any PETG residue just peels off in one piece leaving a clean nozzle. And use a silicone sock, but not if you start to experiment with temperatures above 300 Deg.

The textured build plate is ideal for PETG but in my opinion less so for anything else and I find using PLA on it, just contaminates it leaving it difficult to print on, once you start to get into exotics, it becomes challenging with bed adhesion and I would look towards buying a plain spring steel bed and using something stuck to it as you will be replacing the said stuck to it layer frequently, you will also need to modify your printer to successfully print at higher than 300 deg temperatures.

RE: FDM for Production

@chocki

I understand what you are saying about the design not being ideally suitable for FDM.  It was designed for thermoform.  The good news here is that I can redesign it.  I'll do just that.  You mention a maximum of 65 degrees from vertical.  Is that a generally accepted rule of thumb or is that your personal recommendation?  Either way, I will work with 45 degrees for my redesign, just curious.

I'll alter the design and post the revised model.

RE: FDM for Production

I've revised the design to make it more FDM compatible.

100316-000_PLENUM FDM DESIGN

Edited to include picture, by Joan.T

RE: FDM for Production

17hrs 15min , 78.39m of filament for 4 wall thick 20% gyroid infill using a 0.4mm nozzle with some supports for the flat bits of the bolt mounts.

RE: FDM for Production

With PETG:

If surface quality is not that issue, using a nozzle with 0,4 mm, a layer hight of 0,35 mm, 20% infill (cubic), 4 contours without support (on a MK 2.5) takes approximately 9h and 30min, with support (I would not use) approximately 11h.

Printing parts like this it could be an advantage to use a nozzle with 0,6 mm diameter without support approximately 7h 10min and with 8h and 20 min.

I print molds for parts of fire resistante concrete - as these parts are bigger I often use nozzles with 0,6 mm diameter.

Best regards, Clemens Mödlin

RE: FDM for Production

True, I used settings that give a good surface finish, but for functional parts where surface finish is not so critical, switching to a 0.6 mm nozzle results in both quicker and stronger parts. Change your hot end to a volcano and you can ramp up the speed even more.

I think we can say that my figures are for the longest it would take to print this part with a decent finish

RE: FDM for Production

@clemens-m

You would not use supports?  Is that because you think it will print without them or you would further rework the design to not need them?  When you look at the part what is your gut feeling, am I likely to succeed in PETG with an out of the box Prusa MK3?  If I were to use supports, would I need the multi material upgrade for that?  I would think so.  Thanks for your advice.  Regarding the larger nozzle, whatever it takes to speed up the process, the part is not fussy.

RE: FDM for Production

@pmcgarr

Whatever it means: "the part is not fussy"

I would recommend the 0,6 mm nozzle as your part is functional and not a model part where details and surface quality are that important.

Why without supports? Well - the first layer @ the screw holes will have an awfull surface, because they are printed  in "free air", but allready the 2nd or 3rd layer will be ok. And when you mount your part to the machine or whatever it is, the screws will plane the surface. So nobody will see it and you save time and material during printing. I would start with a design like this and than would optimize it.

Best regards, Clemens Mödlin

RE: FDM for Production

Sorry - have forgotten it - no you don need the MMU (as long as one color a part is enough - and with some tricks it is also possible to print in more than one color (but that has its limits)).

Best regards, Clemens Mödlin

RE: FDM for Production

Supports with PETG can be problematic since they stick so well, but there are a few tricks, firstly increase the pattern spacing, as long as you have good bridging settings, then the increased pattern spacing is not a problem, and the most important one, make sure your support interface layer is not in the same direction as your first supported layer else the two will fuse together and then your in trouble.

In support material in slicer you can alter the pattern angle, I find 45 degrees to the first supported layer with a pattern spacing of 2mm works every time and leaves an easy to remove support. Granted it will never be as good a finish as a first layer or top layer, but for bolt hole flats, or hinge flats, it is fine.

Another plus with wider pattern spacing is that it uses less material.

RE: FDM for Production

@chocki

Forgive the beginner question, but when you refer to supports, are they made of the same material or a separate support material?  I saw something about support material that will dissolve away in post process.  That's why I thought I would need the multi-material upgrade, so I can feed both PETG and support material.  But, possibly I am missing that you can also print supports from the same material you are printing the part from.  I have a lot to learn. 🙂

RE: FDM for Production

You can use dissolving supports for some complex models which need supporting from the inside, But I'm only using supports of the same material I am printing with, and only where needed, also if you look closely (You do need to check each layer using the layer view slider), you will see that the top support layer is approx 90 deg to the first supported layer (45 is better) this minimises the contact points between the two layers and makes it much easier to remove the supports, if both layers are running the same way, then the filaments will fuse along the whole touching lengths, if they are at an angle to each other, they will only fuse at point contacts where they cross over each other.

In the picture below, the green is the support material but is of the same material as the rest of the print and is only in contact on the one layer which would have been printed into fresh air, it does not touch the print in any other place, and since there are only crossing contacts, it is easy to remove with little surface scarring.