3D printers - messin' with resin

[Steve_Cornford]

Thanks for that Martin, two tips in one!

I have just ordered the heater.
I have found that be pre-warming the resin before adding to the tank that the resin flows better, proved by the improved draining & dripping after the print has finished.
Now to find my reel of parcel tape to keep the pesky rubber seal onto the cover!
Steve

 

[Phil G]

Hi Martin,
well your ahead of me again, I am still waiting for my temp controller coming from AliExpress just a quick comment. Are you in a position to do a controlled test using resin temps of the recommend 20-25 which is a very wide band primarily intended to show it has a good working temp range. Then try if possible to increase the temp to the previously recommended 26 to 27 degrees. which is a much tighter working range.
What should happen is the viscosity will be bit lower and the UV reaction a bit faster, this results in a much lower hysteresis effect in the vat as the Z axis moves up after each exposure, you should also find the exposure time can be reduced, although I think you said your already using 2.2 seconds which is likely optimum.
this should represent very close to the best possible printing conditions, and is quite likely to create the most consistent shrinkage, from there it can be measured and offset by shrinkage allowance.
phil

 

[Martin Wynne]

Are you in a position to do a controlled test using resin temps of the recommend 20-25 which is a very wide band primarily intended to show it has a good working temp range. Then try if possible to increase the temp to the previously recommended 26 to 27 degrees. which is a much tighter working range.

@Phil G

Hi Phil,

I'm not sure I can do much in the way of a controlled test in July which would have much relevance in January. For accurate results it would need a sensor dipping in the actual resin, which is impossible to arrange clear of the Mars build plate while printing. It needs a special tank with a built-in temperature sensor.

I did think of conducting the hi-tech experiment of putting a cardboard box over the whole thing while it warms up.

The resin instructions recommend 20-25 degC so I'm a bit puzzled why you want to go to 26-27 degC? My understanding is that higher temperatures significantly affect the shelf-life of the resin, so they must do something to it. There's likely to be a reason why Elegoo suggest 25 degC as the maximum recommended. It's also the maximum I'm happy to go in externally warming the printer internals at this time of year. It got to 27.5 decC yesterday, but that was without the machine switched on or the resin tank in place.

I did get the Inkbird temperature controller while it was on special offer last week. I had it in mind for a different project. But I might get a spare sensor probe and fix it into the Mars lid to monitor the internal temperature.

The heating belt is a mains 230v device, so I'm minded to fit an earth (ground) connection to the Mars metal case, just to be sure. In the meantime I have it plugged into an RCD adaptor.

cheers,

Martin.

 

[Martin Wynne]

.
The Inkbird temperature controller is excellent. I'm glad I got it last week on special offer from Amazon. There is still a 20% voucher currently showing.

https://www.amazon.co.uk/dp/B01AWA2PTG

Very nicely made with big bright red LED displays (no UV from those). Big buttons with a proper click (no fiddly touch screen). A detailed user manual with full technical details, written in proper English. The sensor probe can be unplugged, so I can get another one to use elsewhere when I'm not 3D printing.

I drilled a 5mm hole in the back corner of the Mars lid, about an inch from the bottom, and the sensor probe is a nice push fit in it, to measure the inside air temperature. If it does work loose I shall probably epoxy it permanently in place. That means remembering to put the lid on the right way round, otherwise the probe would hit the tank. Place your bets.

I set the temperature to 24.5 degC with a 1.5 degC differential. Which means it switches the heater off when the sensor reaches 24.5 degC, waits for the temperature to drop to 23.0 degC, and then switches the heater back on. It is switching at about 10-minute intervals, so I could probably set a much smaller differential without it going mad.

I set the alarm level at a low 35.0 degC because I don't want to cook the printer internals if something goes wrong.

In practice, after switching off, the temperature continues to rise to about 25.2 degC before starting to fall back. The tank+resin has a much higher thermal capacity than the air in the lid, so it would vary a lot less and probably settle down around a constant 24 degC.

I have no way of checking the accuracy of the sensor, but the results match as far as I can tell on my old photographic thermometers. I don't know the accuracy of those either.

The controller also has a cooling output, which could switch on a fan or a cold-water pump or whatever if the temperature goes too far above the set value, but I can't see that being needed with the 3D printer -- global warming permitting.

It's been running now for 4 hours and seems entirely happy. Of course it's hardly serving any purpose in July, the test will come in the winter. It would be possible to run a second heater belt below the first one -- just about, without obscuring the printer display too much. That would take the heater to 50W which would make a significant difference in very cold conditions. The controller won't mind -- it is rated 2kW.

cheers,

Martin.

 

[Steve_Cornford]

I must stop reading your tips!
But at leat I have saved £6.60

Your parcel tape tip is an absolute winner, especiall if the tape is aligned with the tank thumbscrews.
It is so much easier to remove the lid without catching the ruber seal, thanks again for that one.
Steve

 

[Martin Wynne]

.
A possible use for your well-used first-wash "dirty dunk" IPA. Working away from daylight/UV brush it over paper, card, cork, ply, etc. Take it out into the sun for an hour. The IPA evaporates, the resin cures, and you have a nice rigid sheet of sealed board for modelling, etc.

Not sure if there is any advantage over traditional shellac, sanding sealer, etc. -- except that you have lots of it, and it's effectively free.

Like any other sealant, it needs to be on both sides to prevent curling.

And the further thought is -- would it work like a solvent to fix resin chairs to plywood? I know you can get the UV adhesives, but this would be brushable and penetrating like other solvents. Perhaps we would need to make a stronger resin/IPA mix, but still thin enough to be brushable.

The alternative is to try to recycle used IPA by curing it in the sun to precipitate the resin, and then filtering. As I've been finding out, that's a messy business because the resin precipitate stays very sticky and slimy while in contact with the IPA.

cheers,

Martin.

 

[AndyB]

Oh dear! It's well over two years since I got my resin printer and it's hardly been used at all. I did get as far as fitting a small fan in the workshop's wall where I was going to put the printer in a sort of fume cabinet but that's about it. Maybe I'll get it going in earnest over the Winter

The current project is rail welding, just not that sort of rail. The rails are 1.5 inch square steel tubes and I'm arc welding 5/8" square steel tube palings to them. I'm making railings to replace the wooden railings around our deck. They've been there for almost forty years and are falling to bits. The steel I'm using is quite thin which makes it a bit challenging. By the time it's all finished I should be an excellent welder

 

[Steve_Cornford]

That's good if your deck has lasted forty years!

 

[Martin Wynne]

.
This looks interesting:

Martin.

 

[James Walters]

Hi Martin & co,
This company (Geetech) have just contacted me, they're sending me one of these printers to try out and review. I'll make a video about it in the fullness of time, and obviously give it a thorough test with plug track components and report back on here as to its usefulness.
At $99 one wonders whether the technology could become any cheaper.
It only has a 2k screen, but that is more than capable of producing detail which none of us can actually see.

 

[Martin Wynne]

@James Walters

Hi James,

I'm very interested to see what you make of this, compared with results for the same STL from the Mars.

One thing I'm very keen to see you confirm is that this printer has a CLEAR FEP film, and not the ludicrous FROSTED ACF film which some printers are now being supplied with. A frosted film obviously kills the sharpness of fine detail, the excuse being that it allows the machine to run much faster by releasing each layer from the film much more easily.

Also be sure to switch off the anti-aliasing in Chitubox if it is on by default with the supplied version. That again softens detail and changes dimensions, the excuse this time being that it gives a smoother surface finish.

I wish these printers would have a single button labelled "Engineering" for those of us who want accurate dimensions and sharp corners, and not a nice smooth Toby Jug printed at breakneck speed.

cheers,

Martin.

 

[Martin Wynne]

@James Walters

Just arrived -- yet another 3D printer!

At under £90 delivered this Geeetech Alkaid printer is less than half the cost of the Mars. I expected it to be a bit cheapo, but it isn't. Not actually tried it yet, just unpacked it, but at first sight it's excellent, well made and packaged, and arrived within 2 days from the UK.

The actual printing specs are identical to the Mars, and the build plate is the same size. No blue snips, but instead you get a tiny pair of angled tweezers which look to be just perfect for the loose jaws.

I will try it later and compare the results with the Mars, and report back.

James has one too and is planning to do a proper review and video soon. edit: now released:

cheers,

Martin.

 

[Steve_Cornford]

Hi James,
Well done.
I like the resin volume marks in the resin tank, also the 4 screw levelling of the build plate might give better results than the 2 screw ball joint of the Mars 2 Pro.
It would be interesting to print a resin timber base and check the dimensions, especially the thickness of the timbers at the four extremities.
ie if your timber thickness is designed to be 3.20mm (the default is 3.24 but lets ensure it is in multiples of the slice layer size) what does it come out as?
Steve

 

[Phil G]

Hi James,
Great video, the Alkaid certainly looks a great entry level option.
What is interesting, is Elegoo seem to have responded they are now selling there Mars 3 (4K printer with carbon filter to reduce the odour issue). for $149 USD. Clearly 50% increase on the Alkaid but for that you do get the 4K ( 35micron Print resolution) and the carbon filter, which does make a huge difference to the odour (until the carbon is saturated) along with a full start kit ( gloves, mask, filters scraper etc)
worth noting however this is a Mars 3 runout sale. so they will not be available for long.
https://www.elegoo.com/products/elegoo-mars-3-lcd-3d-printer
Cheers
phil,

 

[James Walters]

Thank you Martin and Steve,
Steve, I will do a test print of some timbering just to see how it does and check accuracy.
I'm thrilled with this little printer, I'd expected it to not be up to much but now I'm thinking of buying a second one.

 

[Martin Wynne]

@James Walters

Hi James,

Great video! Which resin did you use for those test wagon prints? Was it Geeetech's own resin? Did you use their default settings for it?

There's a question on your channel about how the resolution scales to the model. It's 0.05mm in all 3 axes. That's the maximum resolution of finished model detail possible. It is equivalent to 3.8mm full size in 00/EM, i.e. a bit more than 1/8".

In practice there will be some light spill from one pixel to the next, and it will also be affected by the type of resin and the exposure time, and the condition of the FEP film. Let's say on a good day the real world model resolution is 1/4". In other words, the printer can print the difference between a wagon plank 6" wide and a plank 6.1/4" wide.

As a practical example the chair screw heads are 1.1/16" square full size. If you spray them with a bit of paint and look at them with a magnifying glass you can see that they are square and not round. The corners are not sharp enough to cut your fingers on, but there are definitely 4 of them.

For parts which need to be accurately dimensioned with sharp edges we turn the anti-aliasing off. Where that's not so important, such as on a curved model surface such as say a locomotive smokebox door, it would be worth trying turning anti-aliasing on. This helps to create smoother surfaces and disguise the layer lines on the model.

cheers,

Martin.

 

[Phil G]

There's a question on your channel about how the resolution scales to the model. It's 0.05mm in all 3 axes.

Hi Martin,
Is that right 0.05 in all 3 axis, I though it was .0.050mm in X and Y and 0.02 mm in Z

I guess if you went the Paul route and got the Mars 4 9K with its 0.018 mm in X and Y and 0.02 mm in Z you should be down to a full size 2.00mm resolution if you factor in light bleed that should be accurate to about 1/8 of an inch full size.
cheers
Phil,

 

[Paul Boyd]

Hi Martin,
Is that right 0.05 in all 3 axis, I though it was .0.050mm in X and Y and 0.02 mm in Z

I guess if you went the Paul route and got the Mars 4 9K with its 0.018 mm in X and Y and 0.02 mm in Z you should be down to a full size 2.00mm resolution if you factor in light bleed that should be accurate to about 1/8 of an inch full size.
cheers
Phil,

Hi Phil, I’m really looking forward to playing with the Mars 4 9K when I’m back home towards the end of next week! Once I’ve got the hang of it I’ll be printing plug track in 3mm scale to represent 4mm scale narrow gauge bullhead chairs. There’ll be all sorts of horrible shenanigans in Templot to do that

 

[Martin Wynne]

@Phil G

Hi Phil,

The Z-axis resolution is claimed at 0.02mm, i.e. lift per motor step. But the layer height in the settings is 0.05mm, as recommended.

Claimed stepper-motor resolution per step is always a bit iffy because it relies on electronic micro-stepping the motor. That can be affected by the load on the motor and resonance effects. The dimensional accuracy is restored when it gets to the next full motor step.

cheers,

Martin.

 

[Martin Wynne]

Hi Martin & co,
This company (Geetech) have just contacted me, they're sending me one of these printers to try out and review. I'll make a video about it in the fullness of time, and obviously give it a thorough test with plug track components and report back on here as to its usefulness.
At $99 one wonders whether the technology could become any cheaper.
It only has a 2k screen, but that is more than capable of producing detail which none of us can actually see.

@James Walters

James has now posted his video review of the Alkaid 3D printer on his Bexhill West channel:

I know several other members have obtained one of these printers.

In the UK it's available from: https://www.geeetech.com/alkaid-lcd-light-curing-resin-3d-printer-p-1210.html

To see the UK price, click Ship To at the top of the page and enter your location and currency, and click Save. Then click GB for Ship From.

When you get it showing Ship from GB and Ship To GB it will be showing UK prices, currently £84 + £7 delivery.

In very tiny lettering on the box label, it appears that it is actually sent from:

4PX Express Online Seller (phone 02081 278528)
Leicester Commercial Park Unit 1
Dorsey Way
LE19 4DB

with 2-day delivery by Yodel if you are given a tracking number with no other information.

Bumping this topic so any comments or questions about using it can go here.

Martin.

 

[Phil G]

But the layer height in the settings is 0.05mm, as recommended.

Hi Martin,
That's a very good point, I had forgotten I had set my Z to 0.035mm each layer, to match my stated X and Y.
To be honest I have no idea if that really makes any difference, other than 0.05 clearly slices quicker then 0.035.
there is also the shrinkage to factor in. So a lot of guess work.
At the end of the day, I agree we are splitting hairs here.
cheers
Phil

 

[Martin Wynne]

Hi Martin,
That's a very good point, I had forgotten I had set my Z to 0.035mm each layer, to match my stated X and Y.
To be honest I have no idea if that really makes any difference, other than 0.05 clearly slices quicker then 0.035.
there is also the shrinkage to factor in. So a lot of guess work.
At the end of the day, I agree we are splitting hairs here.
cheers
Phil

Hi Phil,

If you change the layer height you need to change the exposure time to match, because you are changing the volume of resin per pixel in each layer. The recommended layer height is presumably based on how far the UV light can penetrate the liquid resin, and on ensuring each layer is strong enough to pull off the FEP.

cheers,

Martin.

 

[Phil G]

Hi Phil,

If you change the layer height you need to change the exposure time to match, because you are changing the volume of resin per pixel in each layer. The recommended layer height is presumably based on how far the UV light can penetrate the liquid resin, and on ensuring each layer is strong enough to pull off the FEP.

cheers,

Martin.

Hi Martin,
Attached my Mars 3 pro settings, I got the exposure time by tweaking it using the cones of calibration. STL attached, the idea is the success side has full contact on the cones and the failure side does not. by changing the exposure time you influence the result.

cheers
Phil

 

[Steve_Cornford]

Hi Phil,
Can you post a mug shot of your "advanced" settings screen?
Steve

 

[Phil G]

Hi Phil,
Can you post a mug shot of your "advanced" settings screen?
Steve

Hi Steve, as requested
cheers
Phil,
Thinking about it I can't remember if I has anti aliasing on or off when I did my cones of calibration.
If I get time I will redo my cones tomorrow

 

[Phil G]

This company (Geetech) have just contacted me, they're sending me one of these printers to try out and review.

Hi James,
You made a very good point in your video comments. Could you do me a favour and weigh the Alkaid, complete with packaging if you still have it? I can then find out what the post cost out the UK to NZ actually is. It could still be the most cost effective route.
Cheers
Phil,

 

[Martin Wynne]

@Phil G

Hi Phil,

Page 14 of User Manual:

Net weight: 4.2Kg

Gross Weight: 5.8Kg (I assume that includes the packaging)

Package Size: 308 x 352 x 458 mm

Martin.

 

[Stephen Freeman]

Hi all,
I have just taken the plunge with the Geetech Alkaid, what is the consensus on the type of resin to use and what other bits do I need?
Sorry if this has been covered before but if I can be pointed in the right direction, that would be helpful.

 

[James Walters]

Hi Stephen, I've been getting great results with Anycubic ABS like resin using the standard slicer profile settings loaded directly from the USB stick supplied with the machine. You may need to experiment, but these have worked for me and produced excellent prints.
I've not used water washable resin with this machine but I'd expect equally good results.
Link here:
https://amzn.to/3FiYmDR

 

[Martin Wynne]

Hi Stephen, I've been getting great results with Anycubic ABS like resin using the standard slicer profile settings loaded directly from the USB stick supplied with the machine. You may need to experiment, but these have worked for me and produced excellent prints.
I've not used water washable resin with this machine but I'd expect equally good results.
Link here:
https://amzn.to/3FiYmDR

@Stephen Freeman

Hi Stephen,

For plug track chairs the important detail there is ABS-Like. The Anycubic water-washable version works fine:

https://www.amazon.co.uk/gp/product/B0C7BFQZ2C

In addition, you will require at least:

1. a bucket of water

2. a sunny day

3. a room without any daylight

4. an empty jar of pickled beetroot, covered in kitchen foil (to store filtered resin if you need to clean out the tank). Mixing used resin back into the original bottle is not recommended (reduces shelf-life). Empty pickled onion jars also work.

For IPA-washed resin the Elegoo ABS-Like also works well. All my original testing was done with it:

https://www.amazon.co.uk/gp/product/B07P84PQCF

Other types of resin may be preferable for other modelling jobs. Buy spare tanks, or more pickled beetroot to store resin changes.

cheers,

Martin.

 

[Stephen Freeman]

@Stephen Freeman

Hi Stephen,

For plug track chairs the important detail there is ABS-Like. The Anycubic water-washable version works fine:

https://www.amazon.co.uk/gp/product/B0C7BFQZ2C

In addition, you will require at least:

1. a bucket of water

2. a sunny day

3. a room without any daylight

4. an empty jar of pickled beetroot, covered in kitchen foil (to store filtered resin if you need to clean out the tank). Mixing used resin back into the original bottle is not recommended (reduces shelf-life). Empty pickled onion jars also work.

For IPA-washed resin the Elegoo ABS-Like also works well. All my original testing was done with it:

https://www.amazon.co.uk/gp/product/B07P84PQCF

Other types of resin may be preferable for other modelling jobs. Buy spare tanks, or more pickled beetroot to store resin changes.

cheers,

Martin.

Thanks, unfortunately we do not indulge in Pickled Beetroot but we do generate lots of empty glass jars. I have ordered the ABS-like, so we will see how we go. Apart from track I can see it having its uses for signalling items as well as other model railway related jobs.

 

[Martin Wynne]

Thanks, unfortunately we do not indulge in Pickled Beetroot but we do generate lots of empty glass jars.

@Stephen Freeman

Hi Stephen,

The reason for suggesting pickle jars is not just whimsy.

They have a wider brim than most which makes it much less messy to do this than with narrow-neck bottles:

You may need a supply of disposable filters:

https://www.amazon.co.uk/dp/B09Y1HTV7Y

p.s. all this is only needed if you have failed prints leaving bits stuck to the FEP film.

To avoid your first print failing, I suggest giving the build plate a good scrub to remove any traces of factory grease, dirt, etc., before the first use.

cheers,

Martin.

 

[Phil G]

Apart from track I can see it having its uses for signalling items as well as other model railway related jobs.

Out of interest has anybody ever tried 3D printing wheels?
I was thinking wagon wheels, not the tyres, just the wheel body. 8 spoke of 3 hole type.

The reason of asking is a I have an old Unimat 3 lathe, and I just read an article on converting it to a CNC lathe. Would be perfect for machining tyre profiles, just need a way to make the wheels.
cheers
Phil,

 

[Steve_Cornford]

Hi Phil,
Have a read through of this:-
https://www.rmweb.co.uk/topic/122847-any-wheel-any-size/

steve

 

[Phil G]

Thanks Steve,
Having been down the Alice in wonderland, RM web site rabbit hole for over an hour now. It certainly looks doable with ABS like resin,
I will start with wagon wheels.
I think its going to come down to getting the right and consistent resin shrinkage, to ensure it results in both a press fit good enough to hold the tyre in place but not so tight as to damage the resin wheel hub.

Pity there is no way to model a Gibson ring. (no reference to Alan Gibson in that comment) otherwise driving wheels maybe also doable.
cheers
Phil,

 

[Martin Wynne]

Thanks Steve,
Having been down the Alice in wonderland, RM web site rabbit hole for over an hour now. It certainly looks doable with ABS like resin,
I will start with wagon wheels.
I think its going to come down to getting the right and consistent resin shrinkage, to ensure it results in both a press fit good enough to hold the tyre in place but not so tight as to damage the resin wheel hub.

Pity there is no way to model a Gibson ring. (no reference to Alan Gibson in that comment) otherwise driving wheels maybe also doable.
cheers
Phil,

@Phil G

Hi Phil,

If you can turn a wheel rim, you can turn down a 3D-printed wheel centre to fit?

Martin.

 

[Steve_Cornford]

Hi Phil,
Here is another rabbit hole for you
https://www.rmweb.co.uk/topic/12400...peckett-p4-em/?do=findComment&comment=4144673

I purchased some if these from Dave, and some Gibson wheels to get the rims.
Just waiting for my cataract op. Before making progress with the chassis & wheels.
But others have succeeded with this venture (3d printed wheel centres).
Steve

 

[Phil G]

@Phil G

Hi Phil,

If you can turn a wheel rim, you can turn down a 3D-printed wheel centre to fit?

Martin.

Hi Martin,
Yes that's quite right, I was however hoping that I could get the resin to press fit the tyre, and the open up a 1.98 mm hole with the the tyre on the resin wheel thus ensuing its all running true to the axle with just one operation.
cheers
Phil

 

[Phil G]

Hi Phil,
Here is another rabbit hole for you
https://www.rmweb.co.uk/topic/12400...peckett-p4-em/?do=findComment&comment=4144673

I purchased some if these from Dave, and some Gibson wheels to get the rims.
Just waiting for my cataract op. Before making progress with the chassis & wheels.
But others have succeeded with this venture (3d printed wheel centres).
Steve

Does he explain what he means by oven baked resin anywhere in the post, I cant find any detail.

By the way, have you decided if your going 00-SF or 18.83? what about 18-SF almost the perfect compromise i.e. close to scale track and can still use RTR wheel standards. well that's my excuse anyway.
cheers
Phil,

 

[Martin Wynne]

@Phil G

Hi Phil,

18-SF ? Not currently in Templot but happy to add it if you can provide the details.

EM-SF is there, but not suitable for RTR wheels -- EMGS and other kit wheels only. Same back-to-back as standard EM, so can be mixed on the same layout, interchangeable rolling stock with EM layouts. 18.0mm track gauge, 0.8mm crossing flangeway. 17.2mm check gauge (same as EM). Improved appearance for EM models, but with more restriction on radius.

If you want closer to prototype with RTR wheels the usual option is standard EM (16.4mm back-to-back with RTR wheels).

cheers,

Martin.