3D printers - fun with FDM printers (Marlin-based)

[Steve_Cornford]

Bingo!
I have bed release!
In the end a sharp tap with a hammer on an offcut of planed oak did the trick, not enough oomph with the the wooden steak mallet!
STeve

 

[Steve_Cornford]

Hi Martin,
Is the filing jig profile referenced by:- https://85a.uk/templot/club/index.php?threads/3d-printers-fun-with-fdm.277/page-3#post-9472
the one to use?

Steve

 

[Martin Wynne]

Hi Martin,
Is the filing jig profile referenced by:- https://85a.uk/templot/club/index.php?threads/3d-printers-fun-with-fdm.277/page-3#post-9472
the one to use?

Steve

@Steve_Cornford

Hi Steve,

Yes.

Cura makes it difficult to see the full file name and profile name. We need a convention for keeping track of different profiles while everything is experimental. I will devise something before I get in a muddle -- I've got about 20 different profiles in Cura at present.

cheers,

Martin.

 

[Phil G]

Hi Steve,
I found it easier to get the jig off the bed if both parts are printed at the same time.
That way you can get a scraper under the thin joint piece which seems to lift easier.
Clearly only any good, when you get get both parts on the bed, which get harder as the switch size increases.
cheers
Phil,

 

[Steve_Cornford]

B switch blade back filing jig
Steve

 

[Martin Wynne]

B switch blade back filing jig

@Steve_Cornford

Hi Steve,

Looking good!

Did the second part free itself from the glass as it cooled from 60degC ?

Is there much shrinkage difference between the two parts? Was there much time difference between the two Cura profiles?

p.s. for anyone reading this who hasn't seen the jig instructions, they are at:

https://85a.uk/templot/club/index.php?threads/using-the-templot-rail-filing-jigs.728/

cheers,

Martin.

 

[Steve_Cornford]

Hi Martin,
No, it did not free itself, but only a gentle tap was needed.

Shrinkage.
I will measure & report back.

Time (cura estimated as i was not in room when either actually finished!)
South part printed using brick profile estimated 4 hours 30 minutes
North part printed using jig profile estimated 3 hours 50 minuts

Do you think it will be ok if i use the two patrs printed with different profiles?

Stev

 

[Martin Wynne]

Do you think it will be ok if i use the two parts printed with different profiles?

@Steve_Cornford

Hi Steve,

Should be fine -- they wouldn't have bolted together if there was any significant problem.

p.s. for the switch jigs, don't forget my note about the sticky tape at:

https://85a.uk/templot/club/index.php?threads/using-the-templot-rail-filing-jigs.728/

It has only just dawned on me that I still haven't yet released that update. I really must try to get 244a out soon. I wonder if I dare risk releasing it now with the chair heaving still not working?

cheers,

Martin.

 

[Phil G]

Hi Martin,

It has only just dawned on me that I still haven't yet released that update. I really must try to get 244a out soon. I wonder if I dare risk releasing it now with the chair heaving still not working?

In my opinion a release of 243b with all the little gems and the bug fixes is fine for now. I don't see a need to rush chair heaving until your happy with it, thus putting you under pressure for the 244a release as being supper urgent. Although I am awere folks have enquired about it.

Just thinking out loud here, would it be hard to make the SC chair option avaible before the competition of chair heaving? Fully understand there directly related to each other. My thinking is if the SC was an option for custom chair design, it would allow people keen to move forward, with something to work on whilst waiting for chair heaving to be released.
cheers
Phil

 

[Steve_Cornford]

Using the B Switch blade - back of blade filing jig
First a refresher copy of Martin's Illustration

Steve

 

[Martin Wynne]

Using the B Switch blade - back of blade filing jig
First a refresher copy of Martin's Illustration
View attachment 8789

Steve

• Don't be misled by the 88mm dimension, it doesn't play any part in actually using the jig when locating the rail in the slot. It's just to explain the design of the jig for the back planing angle.

Martin.

 

[Steve_Cornford]

B switch blades continued

I cut two lengths of rail 145mm long for the left and right blades.
I marked the head of each rail with permanent marker, then laid the two lengths on their sides with the rail heads touching and then marked one side of each rail with permanent marker for about 50mm at one end. This identifies the back side of each blade.

I then took one blade and inserted it into the jig so that the marked side was uppermost and was inset about 19mm from the end of the jig.

1with the jig on a flat & level surface, tightened the wing nuts to hold the rail firmly, then clamped jig to table. First used bastard, then no 2 and finally smooth file to take the back of the rail down to the surface if the jig.
You can see your progress as the permanent marker disappears. First on the edges of the head and foot, and the on the web at the blade tip.

I removed this blade and repeated operation with its partner.
You can cut the rails longer if you wish but there is a danger of snagging and subsequent bending of rail whilst filing, so I preferred them just long enough to poke out the rear end
Steve

 

[Martin Wynne]

I cut two lengths of rail 145mm long for the left and right blades.

@Steve_Cornford

Hi Steve,

Looking good.

145mm seems a strange choice of rail length? In a typical B turnout in 4mm/ft scale there is no prototype rail joint at that distance from the blade tip. The prototype switch rail length is 22ft-6in, i.e. 90mm from the tip to the rail joint. I think most modellers would take the rail in one piece to the wing-rail front joint, in which case it requires about 230mm of rail for a B-8. Then the prototype switch rail joint becomes a dummy joint with fishplates at 90mm from the tip. Or the finished rail can be physically cut at that point if desired (but do the front of the rail first).

cheers,

Martin.

 

[Steve_Cornford]

Hi Martin,
Well I would prefer to use the longer length as you suggest and have a dummy joint. That is what I did for my test OO-SF resin based turnout.
I did start with a longer bit in the FDM filing jig, but while I was filing I snagged it and bent it!
So I decided to go all prototypical a d have a real rail joint. I chose 145mm so that a little bit sticks out to hold and make positional adjustments
I think I will knock up an anti-snagging guard young a couple if bits of timber so that I can revert to the longer bit if rail & dummy joint.
Just an old Dodger trying to avoid accidents!

Ps I have received the borosilicate glass & currently have a B switch front left blade jig on the go. Should finish about midnight!

It's a bit like watching the old black & white television, Interlude program of the Potter's wheel, strangely addictive.

Steve

 

[Phil G]

Hi Steve,
I was going to comment about your rail length but Martin beat me to it, Personally I have found the best solution given I am working in 0.5 meter rail lengths is to use the full 0.5 meter rail, as long as you slide the rail to be filed either 19 or 20 mm from the end to be filed I use 20 mm and have carful marked the jig accordingly. The other end slides though and is no issue at all.
The clear advantage is simply if your not happy with the filed rail (mainly back jigs,) but the same does apply to the front filing jigs to a lesser extent. You can just file a bit more off so only losing a couple of mm of rail at worst case.

Your right about making the bottom rail with a marker pen as that also required to ensure you get the right rail in the right front filing jig.

Again personally I found it better not to use the tape, as I found its better to use the front clamp to nip the rail section to be filed. the tape makes it more of a hard stop. That said its horses for courses, and I can see why Martin included it as an option.

I get about 25 pairs of blades from a single back jig before I find its better to remake the Jig, I.e. I am starting to see clear filing damage to the jig by then, so better to remake the jig.
As a concept its fantastic.
cheers
phil,

 

[Phil G]

PS I forgot to mention the vast majority of my switch's are C D and E sizes so its the D size that are getting on for over 25 pairs of blades. Also n the D and E size find I have to also add a 25 mm g clamp midway along the jig, that's because the two M6 bolts get a bit too far apart to do effective mid point clamping, especially when the filing end is only lightly nipped.
cheers
Phil,

 

[Martin Wynne]

Hi Martin,
Well I would prefer to use the longer length as you suggest and have a dummy joint. That is what I did for my test OO-SF resin based turnout.
I did start with a longer bit in the FDM filing jig, but while I was filing I snagged it and bent it!
So I decided to go all prototypical a d have a real rail joint. I chose 145mm so that a little bit sticks out to hold and make positional adjustments
I think I will knock up an anti-snagging guard young a couple if bits of timber so that I can revert to the longer bit if rail & dummy joint.
Just an old Dodger trying to avoid accidents!

Ps I have received the borosilicate glass & currently have a B switch front left blade jig on the go. Should finish about midnight!

It's a bit like watching the old black & white television, Interlude program of the Potter's wheel, strangely addictive.

Steve

@Steve_Cornford

Hi Steve,

I'm a bit puzzled by this snagging and bending issue. Did you put the jig in the vice or G-clamp it to the bench? What did the rail snag on?

cheers,

Martin.

 

[Steve_Cornford]

Snagged on my jumper
Steve

 

[Phil G]

Hi Steve,
I see To be honest its summer here, and hot. So I am shirtless cause its so dammed hot. That may of cause be too much information.

cheers
Phil,

 

[Martin Wynne]

Snagged on my jumper
Steve

I keep saying it's experimental.

Martin.

 

[Steve_Cornford]

Oh so that's where I am going wrong, no Lab coat!
I will add it to the shopping list.
Monday evening I successfully printed a B switch left blade front filing jig.
Martin, you are a magician! I don't know how you do it, but the holes lined up perfectly even though it is printed on the slant!

Tuesday, I left the printer producing a B switch right blade front filing jig and went to the cinema for an 11:30 viewing of "Wicked Little Letters". Thoroughly enjoyable....

Upon my return however, disaster had struck!
The printer was still trying to produce the filing jig, but had the nozzle had clogged and so there was about 3mm of invisible layers!
Of course I mean "missing" layers".
I immediately aborted, and investigated the cause.

There was some filament dust on the filament drive gear wheel and surrounds.
There was "caramelised" filament on the nozzle, and in the nozzle etc.
I believe a combination of the spool snagging in the rollers and the fact that I had obviously not adjusted the filament drive wheel tension correctly (if at all, I can't remember) had caused the problem.

After reviewing a Youtube video about nozzle changing, I found it quite easy to follow the instructions to disassemble the print head mechanism, and replaced the nozzle, as well as cleaned out the heater block and the filament feed tube(the aluminium finned item).

I replaced the nozzle rather than attempting to clean it, as the filament had become a hard brown enamel stuck to the outside & inside of the nozzle.

I cleaned up the filament drive wheel.

Having put it all back together again, I fed the filament in and this time made sure that I did adjust the tension before trying the print again.

I must say that it was all relatively simple to take apart & put back together again, much easier than changing the headlight bulb in my wife's Ford Fiesta!

The B switch right front blade filing jig came out perfectly yesterday evening
Today I have printed a 1:6 vee filing jig, and the printer is 91% of the way through a 1:6.5 vee filing jig.

These jigs are really good, thank you Martin.

Now I just need to use them properly, and also get down to actually designing a layout!

Steve

 

[Martin Wynne]

I believe a combination of the spool snagging in the rollers and the fact that I had obviously not adjusted the filament drive wheel tension correctly (if at all, I can't remember) had caused the problem.

@Steve_Cornford

Hi Steve,

Sorry to hear you had problems. I haven't had any cause to dismantle the Kingroon extruders so I'm pleased to hear that it is quite straightforward if necessary.

Are you using the supplied spool rollers, filament break detector, and bowden tube? Do you think they caused the problem? I didn't like the look of any of them -- I have the filament feeding straight into the top of the extruder from a spool hanging on a holder on the wall above the printer. It just needs a short bit of filament in the break detector to fool it. I use the same arrangement on the smaller Kingroon too (there is no break detector on that one), and I have had no problems with filament feeding on either of them. I put some furniture polish on the spool holder so that the spool rolls easily.

For the spring tensioner on the drive wheel, I have it turned fully clockwise, although I never found any instructions for the proper setting. I think the setting can be eased off for use with softer or flexible filaments, but not needed with PLA.

One thing I learned early on with filament is how easily the loose end can get tangled in the other wraps of filament on the spool if you happen to leave the end free and allow the tight wraps to unravel. Sometimes it is not obvious that it has gone under one of the other wraps while your back was turned, and it only becomes obvious that the filament is tying itself in knots after it has been running for an hour or so. Nowadays whenever I have a free end, or immediately when opening a new spool, I put a large crocodile clip on the free end before letting go of it. If it does ever manage to escape, I unwind several yards of loose filament and carefully rewind it to ensure there are no tangles. Some makes of filament provide a small plastic clip to put on the free end, so obviously it is a known issue.

Glad the jigs printed well in the end.

cheers,

Martin.

 

[Steve_Cornford]

Hi Martin,
Its all part of the learning experience, quite enjoyable in a masochistic kind of way.

Unfortunately in the room I have temporarily commandeered for the Kingroon I am not able to devise a version of your overhead spool suspension. Management won't allow it!

The pades on the bottom of the spool roller thingies make them a bit unstable. I might try removing them and mounting the roller bases on a bit of plywood instead, although once they are in the right position they seem to work ok.
Its just that I must have disturbed them after I had set the print running.
At the rate I am going with making jigs I will probably need to purchase another spool!

Perimeter squares
Perhaps we should have a competition to see who can come up with the best idea for utilising all the perimeter squares?

I hope you have been able to get out & about in the odd hours of sunshine we have been enjoying over the last couple of days.

Steve
ps You will be glad to know that there is a scene in the film "Wicked Little Letters" that includes a boiled egg!

 

[Martin Wynne]

Perimeter squares
Perhaps we should have a competition to see who can come up with the best idea for utilising all the perimeter squares?

@Steve_Cornford

Hi Steve,

In 4mm scale they are about 2.1/2" x 5/8". Might be useful for wagon strapping or brake rigging or similar. Or perhaps for wooden fencing. Or wagon loads. Stacked on edge they might make gratings or grilles. The difficulty is that normal solvents don't work on them, it needs to be superglue.

I hope you have been able to get out & about in the odd hours of sunshine we have been enjoying over the last couple of days.

Yes thanks, although I'm not climbing over many gates at present.

See: https://85a.uk/templot/club/index.php?threads/fresh-air-photos.1/post-10561

cheers,

Martin.

 

[bainin]

@Steve_Cornford

Hi Steve,

In 4mm scale they are about 2.1/2" x 5/8". Might be useful for wagon strapping or brake rigging or similar. Or perhaps for wooden fencing. Or wagon loads. Stacked on edge they might make gratings or grilles. The difficulty is that normal solvents don't work on them, it needs to be superglue.



Yes thanks, although I'm not climbing over many gates at present.

See: https://85a.uk/templot/club/index.php?threads/fresh-air-photos.1/post-10561

cheers,

Martin.

The solvent for PLA (and ABS, as well as most modelling plastics) is dichloromethane. Available from chemical suppliers, and as EMA Weld from some modelling suppliers. Also as SameStuff from MicroMark in the USA, I believe. Quite fast acting, and comes with the usual warnings about solvents.

 

[Steve_Cornford]

I have some EMA plastic weld in stock so will give it a go.
Steve

 

[Steve_Cornford]

EMA plastic weld seems to work ok!

No just need to persuade Martin to add rivets to the perimeter square

Well Martin, wagon strapping was your idea!

Sorry Martin, not another distraction from chair heaving

Lovely bit of sunshine at the moment in Brighton.
Steve

 

[Steve_Cornford]

What would the guide look like for just one rivet?

On a more serious note, I find the perimeter square a really good visual indication that the levelling & bed adhesion is ok, and a suitable place to abort.

I am driving the printer via USB and the Cura monitor has a 100mm movement optio whereas the built in lcd screen only goes up to 10mm.

When print is finished, very easy to click 100mm, then click +Z, to raise the X bar and make it easy to remove bed, replace bed & apply PVP if needed, before -Z, and starting next print.

This really is a neat little printer.
Steve

 

[Steve_Cornford]

Hi Bainin,
Thanks for the info re: EMA plastic weld and PLA.
Really useful tip.
Steve

 

[Phil O]

I bought 2 1/2 litres of Dichloromethane from a company in the Welsh Borders, I can't check to see where I bought it until I return home, towards the end of next week.

I bought it around 12 years ago, I have used around half.

 

[Phil G]

Hi Guy,
FYI, Dichloromethane,(Ch2 Cl2) Also called Methylene chloride or MC for short, used to be a very common chemical, it has been the active ingredient in paint strippers for many years. Also for a long time it was the main auxiliary blowing agent in the manufacture of flexible and ridge foam. In these two application its was used in hundreds of thousands of tons globally.

These days not so much, as it's now starting to be phased out in industrial application's. Already one of the worlds largest manufacture of MC has stop making it.
It is quite likely this chemical will get harder and harder to access over time.

It has quite a low boiling point of 39.6C so will evaporate (especially in summer) if the lid is left open,
Please ensure all the correct safety procures are followed correctly with this chemical.
cheers
Phil,

 

[Stephen Freeman]

Dichloromethane is banned in the UK as far as paint stripping is concerned and unless you are a professional lab user. MEK is still available quite easily and works well on ABS, not so sure on PLA though.

 

[Steve_Cornford]

EMA plastic weld still available from UK model shops and online.
I will give a go with an Exactoscale chair and a socket less PLA+ sleeper.
Steve

 

[Steve_Cornford]

When it came down to it, rather than waste an Exactoscale chair, I used an offcut of Exactoscale chair spruce,.
It works, a nice strong bond, which means we can use a hybrid approach whilst awaiting for the more exotic chairs from Templot.
This is using the E-sun PLA+ filament and the Exactoscale plastic (ABS ?)
Steve

 

[Paul Boyd]

Dichloromethane is banned in the UK as far as paint stripping is concerned and unless you are a professional lab user. MEK is still available quite easily and works well on ABS, not so sure on PLA though.

I’ve used butanone (aka MEK) successfully with Anycubic PLA and 3mm Society chairs. It forms a good welded bond.

Cheers,
Paul

 

[bainin]

Dichloromethane (aka methylene chloride) was banned from paint strippers by the European Union in 2010, when the UK was still an EU member. But it is apparently available for other uses. I was recently able to order I litre (advertised in quantities of up to 5 litres) by for delivery to my home in Ireland without any issues. It is still advertised in the UK, including as EMA Weld.
MEK (butanone) and other chemicals used for ABS will at least partially dissolve some components of PLA filaments. AFAIK no “PLA” filaments are pure PLA.
My own tests with MEK, acetone, butyl acetate and one other (cannot recall the name) showed that none fully dissolved any of the PLA and PLA+ filaments I had on hand. They all left a solid core of off-white material.
They probably dissolve enough of the material to allow adhesion, although I am not convinced they provide the best bond in the way MEK does with ABS. Dichloromethane did dissolve all the material.
I did find that at least one glossy PLA had a matt surface after a light coat of acetone — possibly because of the partial effect — which might help some modellers. I plan to test that more when I can.
Mike

 

[Phil G]

My own tests with MEK, acetone, butyl acetate and one other (cannot recall the name) showed that none fully dissolved any of the PLA and PLA+ filaments I had on hand. They all left a solid core of off-white material.

Hi Mike,
Your comments make total sense given MC is a stronger solvent than all the others mentioned.

Just for clarity, I was not suggesting paint stripper should have been used for the same purpose merely pointing out that was a major industrial use of the substance in the past. (It would not have worked even if it had not been banned) as there are other chemicals in paint stripper, mainly designed to stop the MC evaporating, but always leaving a nasty skinning type residue.

MC should be handled with caution, as it has a UN number 1593 for transport purposes. It is classed as a 6.1 Hazardous substances 6 meaning its toxic.
Its also on the Californian proposition 65 list, as suspected of being a carcinogen.
Personally having handled this product in bulk quantities for well over thirty years. I would not be too concerned about either of the above as long as all the right safety and PPE are used.
cheers
Phil,

 

[Martin Wynne]

.
A fly has appeared in the ointment.

The custom FDM profiles I have posted for plug track include a very high setting for the Flow Equalization Ratio in Cura. This is needed to get the thin undercut socket walls for the clip-fit plug tangs to print, to the correct dimensions without stringing.

This works fine in Cura.

Cura is designed for use with the Marlin G-code interpreter on many FDM printers.

But there is a new breed of high-speed FDM printer which use Klipper firmware instead of Marlin. This requires some changes in the Cura settings, see:

https://www.obico.io/blog/cura-with-klipper-firmware/

It seems to me very unlikely that the Flow Equalization Ratio setting will work as intended in Klipper. I haven't been able to find any sensible information about this in the Klipper reviews. As always the only way to find out would be to get one and see.

But I already have 6 FDM printers. A 7th one is surely madness.

On the other hand, I can't post suggested profile files here if they don't work on the latest printers, and the Klipper-enabled ones are likely to become increasingly common. They also print much faster than Marlin and I know some users are put off using FDM because of the very long print times. It might be possible to speed up significantly the production of timbering bricks on the Klipper printers. The only way to find out is to get one...

Martin.

 

[Phil G]

Hi Marin,
Do you need a 7th Printer? My understanding was kipper started out as more of a retro fit software system. I also though its biggest advantage is WIFI interconnectivity to your computer (I don't think you even need a new main board to install it), you simply have to download and install kipper on your current main board. I am sure some manufactures are now supplying it as a standard though.

Out of interest did you design the flow equalization ratio values on Cura version 5.4 or 5.5? only asking because V5.5 contains the new Arachne libraries which greatly change how Cura interprets thin walls.
It also opens up the question, if you are using V5.5 it would require everybody else to be running V5.5 to get everything out of the feature.
cheers
Phil,

 

[Martin Wynne]

Hi Marin,
Do you need a 7th Printer? My understanding was kipper started out as more of a retro fit software system. I also though its biggest advantage is WIFI interconnectivity to your computer (I don't think you even need a new main board to install it), you simply have to download and install kipper on your current main board. I am sure some manufactures are now supplying it as a standard though.

Out of interest did you design the flow equalization ratio values on Cura version 5.4 or 5.5? only asking because V5.5 contains the new Arachne libraries which greatly change how Cura interprets thin walls.
It also opens up the question, if you are using V5.5 it would require everybody else to be running V5.5 to get everything out of the feature.
cheers
Phil,

@Phil G

Hi Phil,

I think Klipper needs its own board with its own processor chip. But any such upgrade to an existing machine is not going to provide the extra speed -- which needs a higher-rated nozzle heater, faster extruder, much more powerful cooling fans, probably a different power supply and possibly uprated stepper motors.

It was back in Cura 4 that I started adjusting the flow equalization ratio, and it has worked the same way through all the subsequent Cura versions to the current 5.6. The issue is not in Cura but with the change to Klipper instead of Marlin for the interpreter on the printer.

cheers,

Martin.