Rebuilding my Ord Hadron

[geoffs]

After some work getting prin ting again (heat bed and temperature issues),, I had a failure 3 hours into a 4 hour print.
It seemed that the Y axis slipped by about 4 mm so it started to deposit layers on thin air, never successful.
THe belts seem a bit loose and I can't seem to tighten them enough, perhaps they are just worn out after 3 years?
Not having easy access to MXL belting, I'm going to use GT2 which means replacing the pulleys and another recalibration.
New ACME threaded rod can go in on the Z-axis as well.

I'll get to finish a print one day ...

[geoffs]

Preparing for surgery. Parts to replace are the metric threaded rods with ACME screw threads, GT2 belt and pulleys.

GT2 belt and pulleys, ACME thread rods and nuts.

X-axis carriage comes apart easily, thanks in part to the use of Molex Microfit connectors for the extruder. I want to make sure that the extrusion settings in Marlin are correct so I'll remove the hotend. (what's the secret to getting Marlin to extrude when the extruder temperature shows too low? Must be a way to do this for calibration.)

Molex Microfit 3.0

Genuine J-Head, 0.4mm

X-axis motor removed and there I have to stop, I just can't get the MXL pulley off the motor shaft. With the aluminium pulleys it's easy but the plastic ones stick - in this case unmovable. I have a small bearing puller in the shed but it's cold, dark and wet outside so it will have to wait until tomorrow. The Y-axis pulley is also stuck enough so I can't get it off either..

Stuck fast

[geoffs]

Small bearing puuler I have is too big to remove the stuck pulley so I had to make one. It's a bit rough and bent after one use but it did the job without damaging the motor.
I gave the motor shaft a rub with emery paper and the new pulley is an easier fit, still tight enough though.

Home made puller, guaranteed good for one use.

ACME rods installed easily enough although with the new motor couplers, they flex a lot so I'll need to make something to hold the top of the rod steady.

Still using half of the old anti-wobble mod.

New belts were less pain than I had expected, once the pulleys were replaced. New foot for the Y-axis motor.

Another thingiverse goodie.

Pluggable thermistor, end stop and heateer wires make removing the bed a lot easier. The arduino/RAMPS mount needs to be redone in a better colour, as do the cable routing mounts. As soon as I get it printing again ...

[fulg]

Hey Geoff,

For the top of the ACME rods I used this to stabilize them: http://www.thingiverse.com/thing:119997

These limit the print height a little (cannot get higher than 150mm) but they could be modified a bit to give back an extra 10mm or so.

I am also using the same "cable management" solution but I do not like it very much, it was annoying cutting that shower hose and this solution makes it hard to do any wiring modifications later. Out of laziness I tie-wrapped extra cables to the outside of the shower hose, which kinda defeats the point. I plan on replacing it with 1/2" corrugated flex tubing eventually, which should look just as good and make further wiring changes easier.

Cheers,
Ben.

[geoffs]

All back together and it still work!
I had to adjust steps/mm which, with the GT2 belt, came to exactly 80 for X and Y. Z took a bit of research to find out that my ACME rods are 2mm pitch with an 8mm lead so steps/mm were set to 400.
I took the opportunity to check the extruder steps/mm as well but they were spot on.
I replaced the bulldog clips I was using to hold the glass to the heat bed PCB and I think that made a difference to levelling.

Despite all this, I still can't print on a heated bed (hair spray on glass). The layer thickness seems OK but it just doesn't stick. I've tried temperatures from 60 to 75 in 5 degree increments but it doesn't make any difference. Printing on tape with an unheated bed is OK so I have the extruder temperature setting right (215).
Next step is to try PVA or some similar glue instead of the hair spray.

I might give up on PLA and give ABS a try and see how that sticks.

[geoffs]

Success!

Gave a PVA solution a try and it worked, first layer sticks nicely and is properly squished. You've never seen someone so happy to turn out what is actually a quite crappy print

The extruder was clicking a bit during the run and that, combined with the apparent under extrusion shown in the image, led to me thinking the stepper current was too high/low - turns out to be too high. Dropped that and ran the same print job, this time successfully. More successful anyway.
Now to print some calibration items to do a bit of fine tuning.

The print sticks REALLY well, I may have to break the glass to get it off which could end up being a bit more expensive than painters tape ...

[fulg]

Awesome. Your next print should probably be those rod stabilizer things I linked previously, because your layers are not uniform (apparent in your first picture). This MAY also explain why you're having issues with first layer adhesion, since any whipping of the rods will affect the resulting height of the X gantry and result in inconsistent layer heights. When I switched to ACME rods, my vertical surfaces suddenly became very straight, almost perfect. What you are getting looks like the same bad layers I was getting with the crooked threaded rods that came with the kit...

I am surprised that too much current on the extruder caused your under-extrusion, perhaps you were printing too fast and grinding the filament? More power means more torque means hotter steppers, it should not cause missing steps, only wasted power (and more damage when things go wrong). When extruding by hand (disengaging the tensioner), is there a lot of resistance? On my E3D v6 there isn't much resistance required to get a steady flow of melted plastic to ooze from the nozzle. If there is, then you should probably raise the extrusion temperature a little to help with that. Every color/brand/type of filament is different...

Cheers,
Ben.

[geoffs]

I printed a couple more of those parts (fan bracket I think) with different temperatures, adjusting extrusion multiplier (+5%), adjusting extrusion diameter setting and they came out with much more even layers and top surface - still not perfect.

Despite a few good prints, something has changed and I now get what I think is serious under extrusion.
Tried a few different rolls of PLA without change. Did a cold pull on the extruder, no gunk in the hot-end.
Because of the previous 'clicking' problem, I changed the stepper driver for the extruder and adjusted.
The thrust bearing in the extruder (presses against the hobbed gear) seemed a bit loose so replaced that part of the extruder. New bearing just in case. grub screws on extruder stepper OK
Still the same.

Now the extruder is no longer heating. Resistance of the hot-end is around 6 ohms which is correct for a J-Head. No obvious power problem - motors work OK as does heated bed. Next test is the FET that drives the hot-end. More coffee needed!

At least what does print, sticks to the bed.

EDIT: Wire had come unstuck from the extruder heater resistor.

[geoffs]

Bought a new brand of hairspray and now the PLA printer are sticking to the glass. Happy!

After a few disappointing prints (poor finish), I decided to print this thin wall tower http://www.thingiverse.com/thing:337957. There's also a video about it
https://www.youtube.com/watch?v=FSOPsRiiOZk
It involves editing the gcode so the temerpature is reduced every x layers, X=30 in my case. Worked well except on the finished print, I can't see ANY difference in the print quality and picking out the layer where the temperature changes is nearly impossible. Because of this, I dropped my print temperature down to 200 and printed a case for the LCD. Not too bad although I think I could have increased the print speed a bit.

One problem I did have was that about 3/4 of the way through the print, the temperatures started to vary wildly +10 to -25 with resulting messages about cold extrude prevented.
I don't think this is PID settings otherwise I would have had this happen from the start.
More troubleshooting required.

[geoffs]

Well I think I have this thing pretty well sorted now.
There's still the issue with fluctuating temperatures but I've only seen this on one filament so perhaps it's to blame (too much moisture?)

Printing the temperature tower was worth the time and has let me find the correct temperature for the filaments I have. They're all PLA from the same supplier but as the reels are slightly different, perhaps the actual manufacturer is different. Some of the filament I've had for more than two years (bought shortly after I got the Ord Hadron kit) so I'm surprised it's printing as well as it is. I also have a couple of reels of PLA that are the same age (more than 2 years) but have only been opened up recently. The difference in printing is noticeable..
I do have a lot of trouble with the black filament The correct temperature yesterday is not what it wants today and it seems to be over extruding. (I've have measured all of the filaments and plugged the correct values into slic3r.)
I'm still not happy with the level of the heated bed. While I can get the glass level (or pretty close) the PCB has a distinct bow so the edges of the glass don't get hot enough to allow me to do larger prints. I have a Mk2 PCB and a sheet of 2mm aluminium so I think it will be going in as soon as I can track down some suitable spade connectors (and a few blue LEDs!)