Yeah, but then it wouldn't fit in the space between the SSR and the power supply any more. As it is right now there's just a couple millimeters of clearance on one side. It might have been smarter to mount the power supply against the back of the exstrusion, turned sideways, and then just integrated the plug cover into that.

Not a big deal though because every one of prints ever has been under 100mm tall which leaves plenty of room.

 

Did anyone has experience to share on adding some JST connectors for the x carriage electrical parts?

i cant think of any potential issue with the cooling/hotend fans by adding connectors to it, just not too sure will there be a risk for Hotend and thermistor wire? especially on thermistor since the connectors will affect the resistance slightly?(read somewhere in the net)

 

JST connectors (assuming you're talking about those poxy little white PCB mount style connectors) are rated at approx 1.5A if memory serves. That's good for: Thermistors, fans, LED's and endstops.

You're technically right, there is a contact resistance when two connectors are mated. But the resistance seen when mated, will be so low that it won't affect your measurements. Your contact resistance will be in the realms of hundreds of milliohms, and your thermistor, at its lowest resistance will be about 500ohms, down from 100,000 at room temp.

Most all other parts, on the other hand, are much higher current, and will likely heat the connector housing causing it to warp/melt. Typically, people use crimps, if they want to connect to copper wire close to the hotend, or, use long nichrome wire (usually already attached to the heater) and solder & heatshrink if connecting away from the heater.

I personally use these connectors:
HILP-02V-3-E - JST (JAPAN SOLDERLESS TERMINALS) - Connector Housing, HIL Series, Receptacle, 2 Ways, 3.3 mm, HIL Series Socket Contacts | Farnell element14
HILP-04V-2-R - JST (JAPAN SOLDERLESS TERMINALS) - Connector Housing, HIL Series, Receptacle, 4 Ways, 3.3 mm, HIL Series Socket Contacts | Farnell element14

These use crimps to terminate the wire, and have connector current rating of 5A per conductor. Plenty for fans, (hotend) heaters, thermistors, steppers, LED's. Everything except the heated bed. (although, as i use a mains bed at 240V 1.5A, i could even use these connectors for the bed, Only reason i havent done so, is because I dont like the idea of potentially being able to touch mains if I disconnect them)

It bugs me that the connectors you're suggesting (and are commonly used) aren't meant for cable mount connections, and are intended to be fixed into a PCB! I just stock a few of the male & female versions of the 2 and 4 pin connectors i linked to, and that covers everything except the heated bed. Although they're quite a lot larger, Much easier

 

So the motor mount bracket, bottom corner brackets, and perpendicular bracket have been relatively successful (after about the 5th reprint due to stupid mistakes).. Got it printing at 150mm/s with 3000 mm/s^2 acceleration with my 370g stepper and it stays pretty stable. Never did get the Z sliders to work though, and I decided to abandon the X slider for now -- though I may come back to it later.

Anyway, here's some pics and a video.. I'm tired and have work in the morning, but I'll put up the scad and STL files tomorrow or over the weekend. I'm not sure it was worth all the plastic it took to get it right, but it was fun

(note: the video is 120mm/s with 150mm/s non-print movement.. I upped it to 150 all around on the subsequent print)

 

Another solution to cooling the filament. This time using two 50mm blower fans. I found blower fans to be far more effective at moving the air with my Prusa I3 so wanted to use them again. Published on thingiverse - thing 1643230. To be used with fan ducts thing 387301.

 

Could always make the forum publicly available to celebrate hitting 100 pages!

 

Has anyone tried GT3 belts on their C-Bot? I had some slippage on my belts last night after 20 minutes at 150mm/s print speed.. I'm wondering if my steppers are too weak and I need to move to NEMA23 (maybe these? US Ship Nema 23 CNC Stepper Motor 1.26Nm(179oz.in) Bipolar DIY CNC Router Mill), or if my belt is too shallow.. I have some steel reinforced GT2 belts coming over on a slow boat from China that I'm hoping will fix the problem, but I'm wondering if I should just hedge my bets and order a roll of GT3 (or some other belt profile)..

 

I've tested up to 180mm\sec and not had any slippage. That's not saying the print quality was stellar, but there was no slipping. My first guess would just be loose belts?
Edit: That being said, my tests were probably less than 10 minutes.

 

The belts are pretty tight.. As tight as I could get a pair of pliers anyway. Is there a trick to getting them tighter?

Also, did you test that high with a direct drive extruder? I'm using a 370g (~0.8 lb) motor, which makes the carriage 3 or 4 times as heavy as a bowden carriage.

 

Yep, the direct-drive design that Carl made, plus my big-arse fan on it. Honestly my belts aren't uber-tight, They have some play when I press on them, but not sloppy in any way.

Maybe they're too tight? I've always just pulled on the ends with my bare hands to a 'reasonable amount' before tightening up with the other hand. However, I've had issues where I over-tightened the bolt pressing the idler wheel against the filament on the extruder shaft, and would actually make the extruder skip. Maybe there's too much friction in your belt system based on the tightness?

 

You may be on to something.. I tend to try to balance the tension by alternating tightening the belts... I found the issue from last night -- a pulley's set screw had come loose. But I *know* I had it tightened down.. So maybe it is the belt tension being so high it's knocking the set screw out of position..

Edit:

It's the motors.. I was watching this time.. They made a grinding sound and didn't move for a sec. The hotend has a hard time keeping up too -- drops pretty low about 5 minutes in.

 

I bought some of those grey polyurethane belts with the steel strands in them too, thinking they would be less prone to stretching when rounding corners and what not. Installing them you can tell the difference immediately, as you realize that the rubber/fiberglass belts stretch a lot as you pull them tight, these you pull taut and that's about as far as they're going to go.

There is a problem with them and I'm not sure I can wrap my mind around exactly why. Some prints were coming out ok, but others were horribly warped.


This is the orientation they were on the bed and you can see the bottom left corner, those are supposed to be straight lines. It is visible on the right side of the right print as well but not as obvious. Jogging the head from X=275 Y=150 to Y=250, the head very visibly does not go in a straight line. At different X values the deviation is more or less noticeable. For some reason around this point the belt is being pulled tighter by one motor than the other. It would seem that there's a "hard spot" in the belt and when it rolls over one of the bearing surfaces, it doesn't conform to the radius of the bearing and therefore takes a slightly longer path and produces bowing around that point. Visual inspection revealed nothing obvious.

The fact that it only seems to occur close to X=0 and X=300 would seem to indicate it has something to do with the proximity of the carriage to the XY bar ends perhaps? That might be a red herring, as maybe whatever is causing this just happens to manifest itself there due to where it is occurring along the belt.

I'm wondering if maybe these belts are designed for larger bend radii and they're not conforming to the F623 radius (5mm vs 6mm for a GT2-20)? They definitely don't bend as easily as the rubber belts. Of course I could have just gotten crap eBay-quality belts, which they are.

 

Hey man sorry it took so long for me to reply, but thank you very m uch for the links. I'm going to print them later and reassemble the printer tomorrow I think.

 

Man, that's disheartening.. Maybe using idler pulleys on the bearings to make the turn radius uniform would help?

 

I must be cursed.. Now that everything else is finally working, my power relay went and did this. Any reccomendations on a 12 V relay? Preferably one that holds up better than this POS from Makerfarm.

 

Ouch, I've got the exact same one. It just dawned one me that I build my C-Bot one year ago this month (which means that relay of mine has had pretty regular usage for a year). Luckily that hasn't happened to me yet though...
I'd be interested to know what people think are good SSR's: I feel like there has already a been a discussion on this...

 

I remember one on SSRs for AC output... I ordered a Fotek 25A DC to DC SSR... Hopefully it works out. I'm wondering if I should order a heatsink for it too... Do they really get *that* hot?

 

I've finally got my initial build 'finished' & am working on optimizing print quality.
Overall the print quality is excellent - but as this is my first bowden setup & as a result i'm struggling to get rid of surface blobs/zits.

I'm using Simplify3D to slice & have been playing around with some of the settings that they suggest here: Print Quality Troubleshooting Guide | Simplify3D
So far none of them seem to make a difference. It looks like the issue I have is due to too much filament glooping out on the start of each perimeter - with 3 outlines, the inner prints first, then the outer 2 smoosh around the initial blob.

Currently i'm retracting 4.5mm at 40mm/s & have the extrusion multiplier turned up to 1.05 (I had previous issues with clogging & this seemed to help)
As it stands i'm not completely sure if its an over extrusion or retraction issue.

Can anyone give a little guidance here?

 

Ditch the bowden... The standard C-Bot build can print at 90mm/s direct drive with no problems. I learned from my kossel that bowden is just not worth it.

Edit: it occurs to me that I forgot to ask, have you actually calibrated your extruder? Also, do you have an all-metal knockoff V6 hotend? If so check that everything there is tightened down and the PTFE tube goes all the way in. If those things are good and you still want to try the bowden approach, you can set the retraction restrart to a negative value and/or set a coast distance.

 

Measure your filament at 3-4 places on the spool at least. Check both linearly up and down the roll and also axially around the same linear location. If the the numbers are not all within .1 to .15mm of each other, double check or perhaps look at getting different filament. Anyway, average the numbers out put that into your slicer as the filament size. Run at %90 - %95. If you need to run at %105 then your filament diameter average and filament diameter setting in the slicer are off. The quality of the filament manufacturer matters as well.

Bowden is fine: if it's setup correctly. 4.5mm is about what I was use for PLA. I run 60-80mm/s for retraction speed. Check your firmware can handle that in the accleration, jerk and max speed for the extruder. By default those setttings are usually low in Marlin, if you are using it, and won't allow for really quick retraction which is what you want. Quick and fast, so jerk values above movement speed is good and / or at least high acceleration on the extruder.

 

It is possible. I have no way of determining exactly what the problem is, but swapping back to regular GT2 belts made it go away. If I run the steel belts through a little test rig made out of the 3mm ID bearings I can definitely feel there is a variable amount of resistance where some portions of the belt go around more easily than others. Visually, I can not see a difference in the belt at these points. The rubber belts can fold in on themselves with a radius of like 2mm max but the poly/steel belts are easily 2-3x that radius.

See here with a tensioned belt, the white belt naturally wants a larger radius around the corner. I think in some areas, the belt is slightly more pliable and the radius decreases enough to pull things one way or the other (or maybe the opposite, the belt is more stiff in some areas). This is with roughly 8 lbs of tension force on the belt. Perhaps with a lot more tension you might be able to reduce it further, but you'd probably need to strap down the motors to the extrusion because this really starts to bend them outward under the belt tension.

 

Is that similar to how it looked around the motors? I'm wondering because that's the only place where it does a "U" turn.. The rest are 90 degree angles or very close to it...

I haven't re-implemented the rear corners or the Y bar yet.. Maybe when I get to that part I can make the pulley profile variable so I can try a wider separation.

I can't cancel my belts order at this point.. They're already on their way. I'll try them out and see if mine have the same trouble.

 

Around a GT2-20T pulley it looks to be just about right with regards to the radius. That's a diameter 12mm - 2x0.75mm tooth engagement, so it is effectively only 0.5mm smaller than the 10mm diameter bearings.

Running a quick 90 degree test, which I can't photograph without making some sort of vice rig to hold everything, it looks like there's still a tendency for the white belts to want to go a little wider around the bearing until a very unnatural amount of tension is applied to them and even then I'd question if they're really going completely straight. It is more force than I'd feel comfortable applying to the stepper mounts though unless there was a way to keep them from tilting inward.

I wish there was a way to objectively confirm this though. I mean my printer was clearly swerving and I could remove and re-tighten the belts and it would still do it. However, there were plenty of areas where it worked just fine so it would be difficult to test for in a general case unless I could figure out what was happening.

I do note that the belts at my carriage are over 1mm further out from the extrusion than they are at the XY bar end bearings which might be creating extra tension as the carriage gets very close to Xmin and Xmax which could account for an elongation along the X axis as it nears that point, but that does not account for why it only happens in a certain Y range (between 0.5 * Ymax and 0.75 * Ymax).

I would say give them a try anyway. I mean the dimensions of our printers are quite different even if they share a common design. Maybe strap a pencil to the carriage and drive it around in concentric rectangles and see if any of them have deformed edges once you have it built?

 

When I was trying my PLA carriage sled I had made it a bit too wide, and it did exhibit a similar "bowtie" effect.. Even with the standard carriage mine is a little bit off center.. I think it's plausible the stiffness of the white belts would exacerbate that. I'm thinking I might make a slight mod to add a hole through the carriage belt connectors so the belts can enter from the underside.. No idea if it'll work, but it should at least test this theory.

I've noticed the tension is very different on specific lengths of the belt depending on the position.. So it does seem likely that the behavior would only occur when the tension on one side of the carriage is different from the other.

And the pencil idea seems good.. If nothing else it should show me where exactly the problem occurs (if it occurs on mine).

 

Can someone point me to some springs that will be good for the bed. I have some now that work good, but they were part of my makerfarm kit and I'd like to order more but on their site the only one they list is 2 bux a piece and I feel that is a bit over priced. Thanks!

 

K, hell has frozen over and I am switching to direct drive, I think....

To those of you with the titan extruder, asking for your thoughts on it.

• Performed as expect?
• Reliable?
• Flexible filaments?
• Bronze / Wood / other infused filaments?
• Use of pancake Nema 17? Which one?
• Any artificats in prints? Ringing? Rounded corners? Inconsistent layers?

I am finding that the more exotic I get with my filaments, the more issues I am having with clogging. I've tracked it down to just simply not having enough pushing force due to the bowden. I love my bowden, but love being able to print these filaments more. I can get detail still, with lower speeds and firmware tweaking, it was just nice to be able to superb detail at highers speeds ala bowden...

On a side note, I was already working on the generator, but have even more reason to incorporate the changes folks have asked for. So here's my short list, please let me know if you have more.

• Effective Fan duct. - Going with a twin perpendicular rail duct, but might switch to a four duct fang style turned 45 degrees so it hits the primary infill route.
• Titan extruder mount.
• Carl's direct drive.