Extruder upgrade - e3d Titan

jhitesma

Some guy in the desert
Mentor
I've been less than pleased with my extruder for some time.

Actually, back up. Let's discuss terminology just a little first to make sure we're all on the same page. When I say extruder I'm talking specifically about the part of a 3D printer that feeds filament. This is different than the hot end which is the part which melts the filament and contains the nozzle which actually extrudes it. Though on many printers it's common to call the combination of both an extruder since on a direct drive setup (like I currently have) they're joined together and in some designs hard to tell apart. I upgraded my hot end over a year ago and switched extruders at the time but felt like my extruder change was a step back in some ways.

But perhaps I'm getting ahead of myself even with this diversion. For those reading this who haven't followed my extruder/hotend experiences since building my printer 2 years ago (almost exactly, it was on Dec 26th and 27th of 2015 that I built it) I'll give a quick review.

I have a FolgerTech 2020 i3. It's called an i3 design because it borrows a lot of design specifications from the Prusa i3 - but assembles them rather differently. The Prusa i3 uses a flat plate frame (usually of acrylic or metal though wood is not uncommon either) while this printer uses aluminum extrusions. The real i3 has it's steppers mounted on the base with lead screws rising above - while this one has them at the top and extending down. However the X gantry is almost identical to an original Prusa i3 with the same size and length of smooth rods spaced the same distance apart.

The original extruder/hotend on this printer was a clone of the Makerbot MK7/8 design (the 7 and 8 are apparently VERy similar with only some minor differences in the drive gear and nozzle.) which worked...but wasn't ideal and had a few fairly major flaws.

The first issue I had was that the fan that came with it didn't work. Folger sent me a replacement for free, and I used a spare fan I pulled off an old motherboard while waiting for it (once I realized it wasn't working.) You can see in this photo of my very first print that the fan wasn't moving:

i-vXbKf7Z-L.jpg

The other flaws are more subtle...Here it is as I was installing it for the first time:
i-Mg326pf-L.jpg

i-2HQRgLm-L.jpg

i-prdBJWb-L.jpg

The first thing that jumped out at me was that the way the fan bolted onto the heatsink was janky as all get out. You can see in that last photo how the fan would tilt forward as you tightened the screws since there was nothing between the fan and heat sink. This also meant the heat sink didn't make great contact with the heat break which isn't great since it kind of defeats the whole point of the heat sink.

See, the idea here is that you want to localize the heat as much as possible. Keep it all down in the nozzle and don't let it creep back up the filament. That's one of the nice things about the e3d v6 hot ends - the have a really nice heat break and heat sink that do a fantastic job of preventing heat creep and creating a very sharp heat break creating a more consistent melt.

This one...not so much. The big block of aluminum between the fan and the stepper motor which the hot end's throat screws into probably acts as more of a heat sink than the heat sink. One more than one occasion I had heat creep up the filament and soften it causing it to tangle around the drive gear.

I did improve the heat sink when I added my inductive sensor for auto leveling. I designed the sensor mount (the first thing I shared on thingiverse incidentally) so it would mount in that gap between the fan and heat sink allowing the fan to better tension the heat sink against the heat break:

i-Kc2Qmfp-L.jpg

The other problem I has was with the extruder body itself being injection molded ABS. It's ok...but over a fairly short period of time (a few months) I had problems with the idler that puts tension between the filament and drive gear loosing strength because the plastic where the screw it pivoted on started to wallow out. There was a brass sleeve there...but the plastic around the sleeve started to deform. Which would have been ok except for my biggest complaint about this extruder.

The tension isn't adjustable. If it was I could have compensated for the worn pivot.

Now...this extruder isn't junk by any means. I printed most of my first step of MPCNC parts with it no complaints. But as I started to do more printing with PETG and higher temps - the flaws became more and more noticable. So I got a $15 all aluminum MK7/8 clone off ebay and swapped to it:

i-L67xzk4-L.jpg

For $15 delivered you don't get much. No instructions and no fancy box. Just a few anodized aluminum bits and some hardware. But it was fairly easy to figure out:

20160312_152933.jpg

When I installed it I used some fiber gasket material between the stepper and the aluminum body to try and help insulate them so heat from the stepper wouldn't soak into the extruder body. I used the same hot end that came with my printer and eventually ended up going back to the original drive gear as well as I found it gripped filament better than the one that came with this extruder.

I was quite honestly blown away by the difference this extruder made. It was possibly the single biggest improvement in print quality anything I've done has given me. The surface of my prints became markedly smoother and layer lines were far less obvious even though I was using the same hot end and nozzle I was before. The extra tension apparently made for more consistent filament feeding is all I can figure.

In fact - I believe I actually got better print quality with that extruder and hot end than with my current setup.

So Jason, why did you change and why haven't you changed back?

Well, I started doing more PETG and really wanted to start printing Nylon. PETG requires higher temps than PLA or ABS and I found the higher I went on temp the better my PETG prints got...until I went too hot one day and melted the PTFE liner in my hot end. That made for a rather spectacular clog that was a real pain to clean. And meant my printer was down for a count for a few days while I wanted on new PTFE or a new throat.) It actually happened twice (since the thermistor in my original hot end was damaged Folger send me a whole new hot end instead of just a thermistor and I melted the tube in that one as well.)

Since Nylon would require even hotter temps I decided I needed to get an all metal hot end. There are several options available and it seemed the two best were either the hexagon hot end that lulzbot uses or the e3d v6 that Prusa and a number of other manufacturers use. The v6 is a more modular design and is more popular so there's more of a community around it so I decided to go with it.

But...it mounts very differently than the original hot end. Instead of a threaded throat it has ridge that mounts into a groove in the extruder.

With the original plastic extruder from Folger I could have printed a new inductive sensor mount that included a mount for the V6...but then I'd be back to the extruder I wasn't happy with. The new metal extruder had a big aluminum chunk down there so I couldn't replace it with a V6 mount. I could cobble something together but it would make the whole stack taller and there just wasn't a clean way to mount a v6 to it other than going to a bowden setup which I wanted to avoid.

So I started looking into new extruders. I actually printed a few gregs wades derived geared extruders since I was already thinking about going geared for more torque (My extruder stepper runs really hot, but if i turn the current down I get missed steps.) but mounting them along with a cooling fan and my inductive sensor started to get ugly. And I was never happy with any of the DIY hobbed bolts I tried making as a drive gear. So I never actually used any of them.

Instead I went with this extruder:

20160317_235441.jpg

20160317_235447.jpg

I found the design on thingiverse here: https://www.thingiverse.com/thing:825133 And overall felt it did much of what I was looking for.

It had adjustable tension, it moved the stepper up and back over the top of the X rails so more of the weight was over the rails instead of hanging off of them, it's compact, it mounts an e3d v6 very nicely, it has a well constrained filament path allowing for flexible filaments to work well, and I was able to print it out without having to buy any new parts.

But I wasn't 100% happy with it right from the start. The idler arm isn't really capable of putting a lot of tension on the filament even with a heavy spring and tightening it way down. And releasing the idler was a serious pain. I made a minor change to the idler that solved the problem of it being hard to release:

3b7099b4b63a6a144c8be40038c3e9b0_display_large.jpg

But it didn't help with the less than stellar tension.

I also made a couple of sensor and fan mounts which I've been mostly happy with (the sensor mount is great...but needs supports to print. The fan mount is acceptable but did allow the fan shroud to touch the nozzle melting it one time.)

20160320_163026.jpg

7e1e6af21f75c703849c1e3ce4c6b252_display_large.jpg

And while I haven't been quite as happy with the print quality and consistency as I was with the metal extruder - it's been "good enough" since March of 2016 that I've yet to replace it. I've printed a few things to replace it with...but once I started mounting my inductive sensor and cooling fan to them...they wound up in my failed ideas pile.

I haven't gone back to the aluminum because frankly I don't want to give up my all metal hot end. I haven't had great luck with Nylon (which mostly seems to be a tension issue as tightening up the tension makes it print better for a bit...but then the tension is so high that something must be deforming because it slowly gets worse and starts to fail again.) but ABS and PETG print wonderfully now...surface finish isn't quite as nice as it was with the old extruder but it's still good enough for what I do most of the time.

Which brings me to today.

About the time I was finishing my current extruder e3d announced their then new Titan extruder to go with the v6 hot end: https://e3d-online.com/titan-extruder

It's a very compact geared extruder made from injection molded parts. It features one of e3d's "hobgoblin" hobbed shafts as a drive gear, and has a 3:1 gear reduction for increased torque - but it's done with a small acetal gear instead of a big 3D printed gear like the ones I'd tried printing. This makes for a far more compact extruder that should be much easier to mount. It's pretty sexy and I've yet to hear anyone who tried one speak ill of it.

It does have a few design aspects I'm not thrilled about. I don't like how it uses the stepper motor shaft as the pivot point for the tension idler arm. Or how short the lever point on that arm is. The acetal gear worries me a tiny bit about how well it will hold up long term running against a steel drive gear. But I wasn't sure how big of an issue these issues would be in the real world.

A month or so ago filastruder did a "Free day" promotion where they gave out a discount code good for $50 off...but it was only good for a limited number of uses. I set an alarm and had my shopping cart ready. $65 for the extruder is more than I could normally justify spending..but $15...I was on it! When the time came I went to checkout...and...was too early the code didn't work yet. Tried again immediately...and..it worked! Code accepted! Woo Hoo! I completed my checkout and then got a thanks message...that didn't include the discount?! Turned out I should have paid with a credit card stored in their system instead of paypal. Due to the delay induced by paypal all of the codes were used up between when mine was accepted and when paypal actually processed the payment. So the code was no longer valid and didn't get applied :(

Thankfully they were very apologetic and promptly canceled the order and refunded my payment. But that didn't get me an extruder.

Well...a few weeks ago filastruder did a 12 days of Christmas sale with a different item getting discounted each day. I crossed my fingers and sure enough the Titan showed up for 20% off one day. It was already on my Christmas list so I let me wife know and crossed a few more fingers. But I never saw any packages delivered to the house for her so I figured she wasn't able to make it happen :(

Turns out she's gotten sneakier and had it delivered to her office. Which is good because it comes in a rather fancy and obvious box which filastruder just slapped a shipping label on and didn't bother to obscure. (and due to that label I can't really show off the nice box either...sorry.) So if it had been shipped to our house her surprise would have been obliterated since I get home first most days :D

Instead my first glimpse of that box was as I tore the paper off it this morning after pulling it out from under the tree. The packaging is quite nice, but I neglected to take any photos of it. Everything is well packed and protected, they even include all the allen keys needed for assembly.

What they don't include is a mounting bracket since that depends heavily on your printer. I printed their "Prusa" bracket they list on thingiverse and used it to assemble the little beast on a spare stepper:

20171225_203721.jpg

20171225_203724.jpg

I'm pretty impressed so far. The tensioner works better that I anticipated it seems. While I haven't been able to try printing with it yet I did push some filament through it and even with the tensioner at it's loosest setting it seems to bite better than my current extruder does cranked all the way down...even though my current extruder appears to have a longer lever arm. Not sure how they pull that off...but it's reassuring for sure! The tension adjuster is also "Calibrated" with marks so you can adjust it repeatably, and it's very easy to adjust even at high tensions.

The whole thing is also very small and light which is nice!

However, the default mounting bracket won't really work on my machine. I could make it work...but it would be a hack, and I'd rather print something nicer before tearing my machine down to install this ;)

So right now I'm debating between a few existing options:

https://www.thingiverse.com/thing:2065461 Impressive looking...and clean...but looks like it probably needs supports to print and I'm kind of wanting to switch to a blower fan for my layer fan. Also only designed for a 12mm inductive sensor and I'm using an 18mm sensor currently. (I have a 12mm on hand but it's less sensitive so I switched away from it.) So not really sold on this one.

https://www.thingiverse.com/thing:1310161 This is the one apparently semi-endorsed by e3d...and looks nice. But doesn't have a sensor mount or a fan mount. There is an optional fan that was added later...and it is a blower...but I don't have a blower fan on hand yet (have one on order now though!) No sensor mount is a bummer though...there are step files though so I could modify it.

https://www.thingiverse.com/thing:1552022 This is a modified version of the previous one, and includes both 12mm and 18mm sensor mounts (as well as bltouch) still needs a blower fan though....I'm leaning towards doing this one and just living without a fan until I get the one I ordered in a few days. Or...maybe I'll modify the fan duct to take a "normal" 40mm fan.

Or...maybe I'll just start from scratch on my own idea. Not sure I want to take the time for that.

And honestly, if my daughter hadn't woken me up at 5 this morning after going to bed at almost 2 after finishing playing Santa I'd probably have made a choice already. But I left my wife take a nap while I entertained our daughter...then had to cook dinner while she entertained...so...I never got a nap and am not quite feeling up to making a decision like this right now :)

So...next few days we'll see how it goes. What I decide as a mount and if the Titan lives up to my expectations for it. I already turned my nylon dehumidifer back on in hopes that I may finally be able to reliably print with that stuff ;)
 
Last edited:

dutchmonkey

Well-known member
I upgraded my monoprice maker select v2 to the titan areo and love this thing. If you can swap to a pancake motor add a heat sink to it you will loose alot of weight on your print head and will notice an inprovment plus you can speed up your printer as well. Good luck
 

jhitesma

Some guy in the desert
Mentor
Well, I decided to go ahead and try that last mount option I listed. My current layer fan is dying and starting to make rattling noises, and I've been wanting to try a blower style fan for awhile so with one on the way I decided to just take my time building this so I'll be ready to mount it up when the new fan gets here.

I broke out my nice Raptor Grey PLA from MakerGeeks, their Raptor series is an impact modified PLA that's designed to be annealed and is supposed to be considerably tougher and able to withstand higher temps. I haven't noticed a big difference after annealing parts made with it and haven't really tested the strength difference over normal PLA (Though I've seen a few youtube channels that have and showed large differences.) I have tested it's temperature resistance (by leaving parts in my truck along with PETG parts and regular PLA parts) and found that it does seem to be able to stay stiff at considerably higher temps than regular PLA. I may still bust out some ABS to print a final version of these parts in (I like having my printer parts in ABS or PETG so they don't deform if something gets too warm) but based on how this first print went in PLA doing it in ABS may be quite tricky.

20171227_080517.jpg

Ick. That's some major warpage. Though...looking at the skirt that printed with it I think it may be time to redo the autoleveling mesh on my printer. I can see some elephant footing even on the part that warped so it seems like it should have been a good squished first layer...but the skirt in that area was considerably thicker and less squished than the rest of the skirt. The tape on my bed was also pretty well used at this point. So...will try again tonight with a bit of hair spray to assist after doing some adjusting on my mesh.

But...also having second thoughts about this mount:

20171227_080505.jpg

20171227_080456.jpg

It looks like it will mount up to my rails just fine. But...I'm not huge on the inductive probe mount. It moves it to the other side of the nozzle from where I currently have it - not a huge deal and with the design of the titan probably unavoidable...but does mean I'll have to reconfigure my auto leveling configuration. But it kind of hangs out quite a bit...I may just redesign the sensor mount so it pulls the sensor in closer.

I'm tempted to splurge a bit and try a differential IR sensor instead of my inductive sensor. I could tuck it up under the stepper very easily to make for a more compact setup. But that would run me $30 and I'd rather put that cash towards a pancake stepper to drop some weight. So...we'll see.

I'm also worried about clearance for the 40mm noctura fan I replaced the stock 30mm e3d v6 fan with. That little 30mm fan is just crazy stupid loud. Way louder than the rest of my machine combined. The 30mm noctura is silent and I love it. But it is bigger...and the mount I'm using puts it at an angle (for screw clearance issues) so it may not fit well on this mount. Unfortunately hard to say for sure until I pull the hot end out of my current extruder and try it on this one - which I'm not ready to do just yet.

Will think about it more today and try printing a different mount (or the same one) again tonight....really itching to try this thing! (though not looking forward to redoing my mesh and sensor settings! May as well upgrade to a newer Marlin while I'm at it as well.....
 

LitterBug

Techno Nut
Moderator
Really great to see your thorough explanations with pictures on here! I really hate the hot end and extruder on my XYZ. There are quite a few tutorials and mounts on thingiverse for using an E3D on the XYZ even with the stock firmware. But I think my best route would be to scrap the junk stock firmware first, switch to Repetier, do the hot end swap, followed by the extruder. Easier to track problems down when doing one step at a time. Currently have a direct drive extruder, but either the drive gear has no bite, or there is not enough pressure from the idle wheel. Really like the looks of the extruder you have here, but definitely would have to do the firmware first in order to calibrate the feed rate with that gear reduction.

Cheers!
LitterBug
 

jhitesma

Some guy in the desert
Mentor
I upgraded my monoprice maker select v2 to the titan areo and love this thing. If you can swap to a pancake motor add a heat sink to it you will loose alot of weight on your print head and will notice an inprovment plus you can speed up your printer as well. Good luck

Curse you for planting this idea in my head dutchmonkey :D

Ok, I'll admit I was thinking about the pancake motor all along. But figured it was a bit much to ask for along with the extruder. If I had asked for both my wife probably would have just gone lower on my list instead so figured only asking for the titan I'd have a better chance of actually getting it.

I'd really love to do the aero...but that will have to be a future upgrade - I don't really NEED the extra Z it would give me though I do like that it uses a 40mm fan.

Anyway, just pulled the trigger on the e3d 13nm pancake motor. And since I still had some bonus money left over...tossed in a differential IR sensor as well so I can ditch the inductive probe as well.

The differential IR sensor intrigues me....I'm still not sure what I think about it. $25 is pretty expensive for what it is IMHO, $10 more than the stepper! But...if it works as claimed it should be pretty sweet.

https://miscsolutions.wordpress.com/mini-height-sensor-board/

The developer claims it will even work on plain glass (but not on the mirrors I've been using since I have so many of them.) though Tom's review of sensors seems to indicate it's not very reliable against glass:


I'm currently using an LJ18A3-8-Z/BX which according to Tom's tests is more accurate than the IR sensor:

Screen Shot 2017-12-28 at 9.42.29 AM.png

But...this tests are idealized. He's testing against a big chunk of aluminum. He also found that aluminum tape didn't help...but my own testing showed that a sheet of HD aluminum foil under my glass made a large difference.

I also switched last year from my glass plate on the PCB heater to a 3mm aluminum printbed hoping to get my inductive sensor working more reliably. But I actually found it seems to be less sensitive against that (triggers at a lower distance) than it was against my sheet of HD foil under glass. Which...surprised me.

So I'm not huge on the aluminum bed. I don't want to print right on it which has left me using blue tape which I don't really care for. I'd rather go back to glass.

So...we'll see. I'm hoping that glass with hair spray or PEI will work well with the IR sensor. It's also small enough I can mount it under the stepper and make my new extruder setup even smaller and lighter.

Just...going to add a few more days of delay while I wait for the new stepper and sensor.
 

dutchmonkey

Well-known member
:applause:;) sorry but if your going to do it might as well get the most out of it. its nice when you can print at 80mm/s :cool: (not for everything but it is fun) also i put a heat-sink on the back of my stepper not sure if it really needed it but better safe then sorry on a long print i have a few extra if you need one. you can see it here i thermal epoxied it onto the stepper. the aero also increased my print area from 200x200x180 to 210 x210x215.
 

Attachments

  • 20171211_201348.jpg
    20171211_201348.jpg
    517.2 KB · Views: 2
Last edited:

LitterBug

Techno Nut
Moderator
Curse you for planting this idea in my head dutchmonkey :D

Curse YOU jhitesma, for planting these ideas in MY head. ;) Ordered an E3D...

After playing with the MP Select Mini for a while, I am more than convinced that the majority of my problems are in the hot end on the XYZ.

Cheers!
LitterBug
 
Last edited:

jhitesma

Some guy in the desert
Mentor
I actually have the heat sink from my original extruder which I've been meaning to mount on my current stepper. That was one of the things I liked about this extruder design is that the back of the stepper is exposed so I could add the heatsink. Not sure if I'll add it on the e3d pancake or not yet...depends on how hot it gets and how well it works :)

High speed would be nice...but I'm not hurting for it. I still want to do my X axis linear rail upgrade as well...but that would be too much change in one go. I'll see how this works and then once I'm happy with it will work on a new mount to match up to the linear rail stuff and swap it all over. Then with both X and Y on rails I'd feel a lot better about running higher speeds ;)

Then I can think about the aero upgrade....

At least I'm not at all interested in the new aqua aero :) I do like higher temps...but water cooling just seems silly expensive for that!
 

jhitesma

Some guy in the desert
Mentor
Well, the new fan will be here Saturday...and I could get this going with my inductive sensor then...but...with the new parts on their way I'm not in any rush. I'm also still a bit concerned about how my 40mm fan will fit and am wondering where I can put the IR sensor so figured I better fire up onshap and do some modeling....

Here's the onshape project for those who want to play along:
https://cad.onshape.com/documents/2...434753e59a5d278079/e/0607b4dd88a014432f64a7d8

Screen Shot 2017-12-28 at 10.07.04 PM.png

Full disclosure...I cheated. Big time. Almost all of this is from existing models.

The basic titan mount is gyrobot's original since he was kind enough to provide a .step file: https://www.thingiverse.com/thing:1310161

The titan, v6 and stock v6 shroud are all models I found on thingiverse as step files (the v6 is actually an igs but onshape can handle it.)

The 30mm to 40mm adaptor I'm currently using only had a STL so I had to re-create it...and I didn't bother to recreate the whole thing - just the flanges and enough to hold them together.

I'm using the adapter flange instead of a custom 40mm shroud because all of the ones I found and tried didn't fit as well as the stock shroud. But the adapter flange needs to mount the fan at an angle so you can access all 8 screw holes without them overlapping.

Unfortunately as you probably noticed in that first photo...the flange doesn't clear :(

Screen Shot 2017-12-28 at 10.20.40 PM.png

So I've got an issue to figure out. Even before I get to deciding where/how I want to mount the IR sensor I need to sort out my cooling fan.

I see two obvious options here. I could try and make my own custom shroud or my own custom adapter. Hmmm.

I suspect an adapter will be easier to design so it's printable...but I may try a shroud first.

Getting late tonight though so I'll have to figure that out tomorrow.
 

jhitesma

Some guy in the desert
Mentor
Got all my new bits today! The fan was supposed to come Friday...but USPS slacked off and said they'd have it here Saturday...then decided they'd just take their time and wait until today.

The new stepper and IR sensor were both expected today and arrived today.

I'm not sure how I feel about the new stepper. It's not THAT much smaller than my current one, only a bit over 5mm:

20180102_134957.jpg

And if I'm doing the unit conversions right then it's quite a bit weaker than my current one. I know the titan gives a 3x increase but a big part of why I'm doing this is I want more power out of my current stepper. The e3d pancake stepper is rated at 12.7N-cm while my current stepper is rated 2.7kg-cm. which according to this page: https://www.convertunits.com/from/kg-cm/to/N-cm is 26.5N-cm..over twice the torque of the pancake motor.

I'm not really concerned about weight...but the e3d motor comes in at just 162g while my current stepper is 197g so 35g lighter.

I'm not sure losing 35g of weight is really worth losing 50% of the holding torque. I'll try it...since it is what e3d designed around....but it seems unlikely to please me.


I still need to design and print a mount that will hold the new IR sensor. I've seen a few that mount to the v6 fan...but that doesn't seem like a good location to me since the fan shroud can wiggle. I'm going to add something to the new bearing mount that it can attach to - I have to re-print that anyway since the previous attempt warped. This IR sensor is TINY though. Dropping the inductive sensor for this is going to give me quite a bit of weight savings - not that I'm really concerned about lightening up my extruder :D

20180102_134846.jpg

For size comparison:

20180102_134858.jpg

So now that all the parts are in hand maybe I can have this thing ready tonight!
 

jhitesma

Some guy in the desert
Mentor
So I went ahead and modeled up a mount for the IR sensor yesterday:

Screen Shot 2018-01-02 at 4.31.37 PM.png

Nothing fancy. But I finally got to try the onshape in-context editing and it was pretty nice. Since I already had everything else modeled up I found a model of the IR sensor (It was only a STL...but small enough I only took a minute or two to manually re-create it) then brought it into the assembly so I could position it where I wanted it:

Screen Shot 2018-01-02 at 4.31.59 PM.png

Screen Shot 2018-01-02 at 4.32.08 PM.png

Y axis it's even with the nozzle, Z it's adjustable from 4.5mm to 2.1mm from the tip of the nozzle, in X it's offset 14.7mm.

It's supposed to have a sensing range of 2.5-3.5mm to glass on top of black paper so that should give me plenty of adjustment.

Only problem was I forgot to leave room for the pins on the IR module:

20180103_080959.jpg

Not the end of the world...the print warped on me again:

20180102_224454.jpg

Which wasn't surprising because I never did re-create the mesh on my bed - my first layer is pretty ugly and getting zip ties in is almost impossible even after removing the elephants foot:

20180102_224425.jpg

So...gotta print it one more time. But definitely going to recalibrate my auto leveling first. I also made the zip tie slots 1mm wider to better fit my zip ties, and added a cutout in the IR mount to clear the pins.

Screen Shot 2018-01-03 at 8.58.39 AM.png

The only other issue is that this is a 4 bearing design and I'm currently only running three and don't really want to switch to four. The way it's designed I should be able to just put a single bearing in the center and still zip tie it in...but I may adjust the slot positions to match a center bearing better before I print this.

Ahh, what the heck:

Screen Shot 2018-01-03 at 9.23.56 AM.png

That wasn't so hard.

Screen Shot 2018-01-03 at 9.31.50 AM.png

Tempted to add some skylights over the zip tie passages to help make them easier to install. Also tempted to remove some material from the IR mount...it doesn't have to be anywhere near that beefy. But...I like how having it tied into the same plane as the stepper/extruder mount should make it stable and easy to print accurately. This is probably good enough...and the shape change may even help with the warping.
 

LitterBug

Techno Nut
Moderator
I started working on the E3D V6 migration for my XYZ. Seems every design I have picked on Thingiverse is based on specs for generic E3D rather than genuine and has fitment issues. Really glad I have a 2nd printer to print parts with....

Cheers!
LitterBug
 

jhitesma

Some guy in the desert
Mentor
I started working on the E3D V6 migration for my XYZ. Seems every design I have picked on Thingiverse is based on specs for generic E3D rather than genuine and has fitment issues. Really glad I have a 2nd printer to print parts with....

Yeah, the published specs are nice...but designing based on them doesn't seem to work well. Both the real e3d parts and the knockoffs are slightly different in different ways.

Slowish day at work so keep taking a look at this and having ideas.

Realized it won't have the nice cable management of my current extruder...so...:

Screen Shot 2018-01-03 at 2.30.27 PM.png

They'll be kind of tricky to print well without support...but I think my printer can handle them. They're just so I can zip tie the hot end and IR sensor wires out of the way.

With the fan turned 60 degrees it's out of the way and I won't have to modify it yet:
Screen Shot 2018-01-03 at 2.32.35 PM.png

Though I'd still prefer a cleaner fan mount that lets me have it off the front.

I'm half tempted to just work it into this design...add a 40mm mounting plate and then just duct the air from there. Would be impossible to print without supports though so probably better done as a second piece.
 

jhitesma

Some guy in the desert
Mentor
Well...almost there.

I'm lazy. I've probably mentioned that before. So of course I still haven't re-calibrated my auto level. But I tried printing my new modified X-carriage anyway...and it warped. Even worse than the original design:

20180103_204541.jpg

Nuts.

On the upside the IR sensor mount and the wire tie spots printed just fine:

20180103_210654.jpg

20180103_210706.jpg

Unfortunately I don't have any M2.5 nuts. Thought I did. I have 2.0 and I have 3.0...and I have 2.5 washers....but no 2.5 nuts. So while the sensor mounts fine and the pins clear and I can use the full adjustment range....I can't really mount it yet and need to hit the store for some nuts.

The zip ties all fit perfectly this time - so that's nice. And the tripod design looks good to me.

I haven't swapped the actual hot end over yet so not 100% sure the IR sensor is in the right location. But that's something I may have to fine tune after the swap. If I need to raise it more I can sand out the slots, and if it needs to go lower I can make some extensions.

I'm not 100% happy with the shape of the IR mount still. It looks unintentional right now. But I had it with the bottom "flat" and it looked goofy that way too.

So...probably not going to make any further changes to the design...but I am going to finally recalibrate my autolevel and do one final "better" print.

Was just really hoping to avoid having to do that since I'm just going to have to do it again when I get the new sensor on. The current print would probably work. But that warping worries me enough I'll reprint it again.
 

jhitesma

Some guy in the desert
Mentor
So...how far off is my autolevel mesh....let's do a mesh verification to find out!

20180104_180257.jpg

Um...yeah. That's....that's pretty far out. The right side the nozzle is too close and it can't even extrude anything (you can hear the stepper clicking as it tries in vain to push the plastic) and the left side the nozzle is too far so the plastic won't even stick. Yuck. My bed has definitely shifted and I'm overdue to rebuild this mesh.

Re-ran the auto mesh generation and then ran a new verification:

20180104_212514.jpg

That's a LOT better. There are still one or two spots that could be improved (though I think at least one of them was due to the initial purge being dirty.) but overall it's much more consistent.

Unfortunately by the time I finished that it was too late to print my new extruder mount :(

But...hopefully tonight it will go smooth now that my first layer should be much cleaner!
 

jhitesma

Some guy in the desert
Mentor
Arrgh. You'd think I'd be rewarded for finally doing the right thing and leveling my bed.

Got home from work and started the print. Looked good...first layer is a bit too short...but it's uniformly too short. Got about 1cm tall and was looking good...the one troublesome corner was maybe just starting to try and lift. Then...the machine just froze.

No clue what happened as I wasn't there in front of it...but it just stopped. Still indicated it thought it was going - but nothing happening. Steppers energized, but not moving. Sigh.

Ok...cleaned it off...and since I reset the machine went and did a full mesh creation/validation to make sure I'd be in good shape before starting again. Even put down fresh blue tape.

Started it again. Looking about the same...first layer even but a little too squished (need to fine tune my z offset on the sensor a hair.) but should be fine.

And...it froze again at almost but not quite the same point :(

Grrr.

Thinking this filament may be cursed. May load up something else tomorrow and give it a go. Too late tonight for another attempt :(
 

jhitesma

Some guy in the desert
Mentor
This is getting frustrating....I did 4 prints of this thing over the week iterating it and it's looking pretty good. Finally got my z offset nailed and my mesh dialed in so 1st layers aren't squished anymore and I'm not having any bed bonding issues.

But suddenly I can't get this darn thing to print :(

5 tries failed today. 3 of them just suddenly froze with no explanation again, 2 had layer shifts.

I switched to a different filament which is part of the issue. It was overextruding when I first switched...so had to dial in the extrusion factor for this filament - that's what led to the first layer jump.

The second layer jump...no clue what led to it.

The 3 freezes....again no clue. My only guess is it could be static which seems bad right now, the other night my machine locked up when I touched the table it was on.

So, not sure what's going on but it's driving me crazy at this point :( So close to having this installed...I could go with the last "successful" print...but I really want one with no warp on this....
 

LitterBug

Techno Nut
Moderator
Thanks for sharing your frustrations. It's good to see others going through some of the same growing pains in builds and mods. So many times we edit out the problems and just show the stuff that goes right which makes it seem as though everything is always easy.

Cheers!
LitterBug
 

jhitesma

Some guy in the desert
Mentor
Yep. I have a lot of projects go really smooth and just come together. Then some just drag.

It's actually been a few weeks of dragging projects. First having to pull the engine in my truck to replace the freeze plugs. Then to get back into fun projects I've been very slowly working on putting a seppuku into my Explorer - though that's just slow due to lack of space to work on an airframe that size comfortably.

I also got an itch to finish something I started almost 13 years ago. A couple of synthesizers and a midi sequencer. The first synth is based on an old Commodore 64 SID chip - I actually had it working back when I first built it. And I have all the buttons, led's, and rotary encoders to build the full "Step C" control surface: http://www.ucapps.de/midibox_sid_csC.html Actually I only have 2x40 LCD's (got a pile of them cheap) so mine would be a little different and probably can't fit into a C64 case anymore (though I do have one on hand if I can work it out.)

So...it works...but it's not pretty:
20171231_165536.jpg

The synth is the pile of boards on the table on front of the keyboard :) The synth only has traditional MIDI connections, but the keyboard only has USB MIDI. So I have to use a software patchbay on a computer to send notes from the keyboard to the synth. The hardest part of getting it working again was finding an appropriate adapter to get it plugged into one of my cheap amps. It only has one SID chip wired up right now but I have a second SID board built up and a separate Core for it (though with an upgrade to the PIC on the first board it could handle two SID's by itself.)

I also started an FM synth based on the Yamaha OPL3 chipset used on early sound blaster cards. I knew from old photos ( http://www.dunephotos.com/Electronics/MB-Projects/ ) that I had it mostly built, just 3 chips left to install in sockets (after testing the power circuits) and one more DAC for the main synth chip. I never got it powered up back when I built it because it requires +12, -12v, and +5v and I didn't have any good source of -12v available at the time ( I was out of old both PC power supplies and funds to order regulator chips.)

And I bought some parts for the sequencer...but it was about to undergo major changes by it's designer so I didn't order all the parts, just a few of the PCB's.

So as I mentioned I had the SID synth up and running - that was last weekend. I decided I also wanted to see if I could upgrade it to the latest specs if I was going to build it into a case. The main reason I stopped building these was the price of cases, panels, and knobs. All of which I can now either 3D print or fabricate on my CNC at home. So it's about time I finished these. Plus...if I build the sequencer my daughter may have fun with it. Lots of lights, buttons and knobs to play with and create music with.

I looked into upgrading it...but it's on MIOS 1.7 with the old 1.1 bootloader. So...while it shows up as a MIDI device on my computer and I can interact with it the special MIOS software for flashing the MIDI programs doesn't recognize it anymore. And upgrading from this old of a version says "Contact author for help". Oh boy.

Well, I did so and he replied back quicker than I anticipated saying since I have a Pickit3 the best bet is to just flash the new bootloader myself then use the PC tool to flash the SID software. I can't run the latest SID with the chip I have but there is a newer improved version I should upgrade to. I hooked up my Pickit3...followed his instructions on what to flash...and...now get nothing. The chip says it verified on programming but I get no response from it when I put it back in the board and power it up. I had forgotten what a pain PIC flashing is compared to AVR or STM. I flashed my first PIC in the mid 90's with a parallel port board that used two 9 volts to provide the high voltage needed to put a PIC into programming mode. The Pickit3 should be a joy to use compared to that...but I've had no luck with it :(

So all week I've been fighting with that, flashing it 2-3 times a night trying to figure out what I'm doing wrong. I've also been going crazy because I couldn't find the FM synth. I'd actually done a remarkably good job of keeping all the parts for this project together this long and not poaching from them for other projects. But the OPL3 board and it's corresponding CORE board were both missing.

I finally found them Friday night...in a spot I'd looked 3 times. Of course. I have to see if I have the 3 missing chips, they're common/cheap TL074's I probably have them just didn't install them because they don't get installed until after you confirm the power circuits are working correctly and I never got that far. The CORE powers up and over MIDI I see messages from it. Don't think I'm going to try and upgrade it just yet. The spare CORE I have for the second SID also powers up and sends MIDI commands announcing it's awake, and I have one more CORE chip with an unbuild CORE board to go with it. The sequencer can no longer fit on a CORE, it now needs a new CORE32 which uses a STM32F4 Discovery board.

Well, that's good news! Because it's the same F4 discovery board I used to use on my knuckle copter as a FlyingF4 running TauLabs - one of the first things dRonin did after forking was drop the FlyingF4 platform due to some issues we found on mine...the DiscoveryF4 just isn't a great choice for that particular project. So my old Disco board is just sitting here gathering dust. Flashed the MIOS bootloader to it...and...it came up and shows up as a USB MIDI device. Nice, so much easier than PIC!

I also found another YAC512 for the FM board so was able to install that giving it a full 4 channels:
20180106_105455.jpg

I got that soldered on last night. Today I need to see if I have those TL074's and wire up an old PC supply - I may be able to get this thing going. And just like the SID I have all of the LCD's, LED's, encoders and buttons to wire up the full control surface. But building those will probably take me most of the year. I just want to get the electronics on this one up and going ASAP.

So that project has been making progress...but slower than I had hoped.

Just like this one...it's printing again now...from SD to avoid the freezing issue I think is between the CHIP and the RAMPS.

I also think this black filament may be part of the problem. I did a mesh verification with it and it printed gloopy there too. Just doesn't seem to print great. So I may have to switch to a more obnoxious color since that's all I have on hand right now.
 

jhitesma

Some guy in the desert
Mentor
I'm getting desperate here. If changing one simple thing each time isn't helping then let's just change everything and do stuff I've never done before. This is the "I don't care anymore I'm just going to try stupid stuff" stage. At any point I could install the carriage I currently have the extruder mocked up on. So why not toss caution to the wind.

I took the tape off my bed. I hate that stuff...and I suspect that what's happening is the print is staying stuck to the tape, but the tape isn't staying stuck to the bed. And that's allowing the print to warp while staying attached. So...off with the tape. I did give the bare aluminum a light coat of good old hairspray.

I also finally remembered to put the changed Z offset from -3.00 to -2.80 into EEPROM storage so I don't lose it randomly. Though with the loss of the tape I was worried that I'd now have to go back to -3.00 but even at 2.8 I was thinking about trying 2.7 or 2.75 so...we'll just try it.

I didn't bother to redo my mesh - since my inductive sensor reads the metal plate and not the tape anyway that should be the same, only the Z-offset should make a difference.

I didn't bother to run a mesh confirmation - I was just running out of time.

I did do an extrusion test on each filament I've been using and adjusted my extrusion multiplier as appropriate.

I sliced with Cura instead. Cura put the part rotated 180 on Z so the short end faced away from me. Meh, why not. At one point yesterday I tried putting the part on a 45 degree angle. Still layer shifted in Y. (though I've also had X shifts...and I haven't had shifts in either in a LONG time.) Anyway..let's give Cura a try this go. This hideous green filament is at a 1 extrusion multiplier so should work just fine with defaults in Cura.

Looked good...until the last few minutes :(

20180107_214541.jpg

Arrrgh.

What's odd is it looks like it skipped on multiple layers in a row consistently...that doesn't seem as likely to be mechanical.

Also a few layers up on the bottom you can see a poorly extruded area. I haven't oiled my filament in a few months and know that this maker geeks will bind up in my all metal hot end if I don't. Just glad it didn't keep happening. Will have to put the oiler back on next try.

It also looks like Cura didn't handle the 3mm loops for attaching my wiring harness as well as slic3r does. Though there's one more that's not visible (which was my final modification) at the end of the side going away from us on the right in the photo. That one is there for me to tie off the power wire to the layer fan. Oddly enough that one DID print well.

Cura also did a better job on the bridged areas where the bearings install. Slic3r always tried to run the bridge in the longest dimension no matter how I positioned the part. There's probably a setting to tweak that but I've never had to so didn't look for it. Cura instead did the first bridge layer on a diagonal. The shorter runs had far less sag and it came out much nicer overall.

I was really nervous about the first layer - you can see that the skirt didn't stay in place. Also not visible is one of the fan mounting holes on the face against the bed - it's inner perimeter didn't stick and is partly blocking the hole. But the rest of the 1st layer looks gorgeous. It's not quite as nice as glass...but it's way better than tape. No squishing and other than that one spot everything stuck great - but released nice when the plate cooled down.

I checked the Y axis a few times this weekend since it seems to be jumping the most on this print...but it seems fine. And sometimes X has jumped. Just crazy that this was printing fine for me all week (other than a less than stellar job sticking to the bed in one small spot) and now that I've finalized my tweaks and want to do a final print it's been nothing but failures. (I even went back and tried the last version that printed well without my final changes....and it just keeps failing for me now as well.)

Oh well at least I got to waste more filament by getting closer :p