Daggy bridges

So I have had a bad few days for dags. Nothing I did was working, then I made a tiny change to retraction settings and got this.

0.2 layers 40mm/s outer perimeter speed. Stoked!!
So I thought I would try in 0.1 it went really well. I'm not sure I want to mess with it, but there are a couple of stand out bits that could in theory be improved.



The points of interest are
the dag in the corner. I think this may be because I have wipe while retract on.
The small pillar is over extruded and has a lot of webbing. Similar on the 0.2 print too. -Maybe dial back extra length on restart 0.05 at a time?
One of 4 bridge attempts didn't stay stuck on the origin side causing a dag. Bridging at 80mm/s
Also not visible in these pics. The 0.1 is a bit wider in the middle. Looks a bit like heat slump?

Printing ABS at around 265-270.

If I did want to fix those small issues. what would everyone suggest? Am I doing it right?

It'll take some time to reply to the other thread, been up through the night, long prints what a ball ache, i've no experience with abs, so can't offer much advice, but was wondering about a few points of your setup
I'm not sure what your nozzle size is, is it 3mm filament?, whats your z pitch, or if you have enclosure, do you have anything ontop of the heatplate glass? I dont see anything, I would have chosen the alu side for printing, but all that aside,
I wouldnt try for .1 layers or .4 until you get .2 (or .24 or .28) right, get some fixed or flexible couplers for the z,
hopefully by the time I wake, you'll have made further progress, and i'll finish my reply to the other thread.

Why are you printing ABS that hot? Try 240c and see. I only print that hot when im trying to print ABS real fast ~80mm sec (yea thats fast for ABS on an i3...)

I'm printing that hot because of splitting and peeling. Any cooler and I can't get an inch off the bed without the part destroying itself. Some parts have lost their head halfway through (which can be a little traumatic for young children watching their favourite cartoon hero get printed, no more toys till I get it right more)... I googled a lot and the sugested fix helped a lot. Print hotter and faster. Makes sense too when you think about the bondage, thermal transfer and cooling rates etc.
I am still looking into improvements here re: enclosure, heat lamps etc. Still need to dummy load my ATX so looking at 5v constant current solutions...
It may also be bad filament, it was nearly half the price of the cheapest ebay filament, so I got 8 kg... May as well work out how to use it.

It wont come out consistently at 240. 250 is the min temp it will extrude smoothly at. 260+ for printing. Thats heat block temp. I have no idea what the actual temp in the melt chamber is, but likely a lot lower. The thermistor is out buy quite a margin, nearly 10%. so programmed temp is around 240-250. 250 via thermistor 250-285 via k probe. Depending on where and how you measure.

No enclosure as such, but efforts have been made to minimise drafts. Room temp is 20-30c most of the time. Got down to 18 in a storm last night. Had some poor layer adhesion on a print that was set to 270 as a result. No warping of splitting, but a few bits broke coming off the bed.

I don't think its actually extruding that hot. Thermal transfer into the filiment seems fairly slow. I'd guess its under 0.2 w/m2k for abs? It passes heat even slower than epoxy... plus I have measured huge variations around the extruder. ~ +/- 15c

Its a cheap Chinese hotend. I'm printing fairly fast. 80-100 perimeter, 50%outer perimeter. 120 infil, 50 solid infil 70-80 bridges.
I can go slower and do need to drop temps if I do, but it prints badly due to inconsistent extrusion and warps and splits. High speed and high temps seem to be the go. The speeds I have have been found from the last 2 weeks of full time calibration printing following various online guides.

If I run over 285c the nozzle boils the surface on slow top infill (less than 30mm/s) and it gets blocked a lot if it stops extruding. So 260-280 seems to be the range Under 260 is terrible. Heat tower is on my too print list.
ie I know this is not recomended temp, but trial and error (lots of error) led me here...

3mm filament
0.4 labelled nozzle extrudes into air at exactly 0.5 at 270c.
Z pitch is claimed m6, 1mm per rev, measured 1.0565. Although that was with missed steps so need to re-check it I think.
I might still have a 3m length, should check it over that length turning by hand.... Will be no worse than the last Tesla Coil I hand wound...

The bed is probably up the wrong way.
I was confused as to how it was supposed to work when it arrived...
Seemed to make sense to put the hot bit where the heat needs to be... I only recently found out about the aluminum, which solves the mystery of the changing hole spacing with thermal expansion... I knew fibreglass PCB didn't normally grow that much!
Not a big fan of PCB heatbeds. Been looking out for a good ceramic tile for ages...
It has 2 layers of kapton on it. Because I am worried the traces will get ripped off by the abs juice, or that I will slice it in half with my Nakiri knife scraper (basicly a large straight razor for peeling veggies [s-media-cache-ak0.pinimg.com] I can shave well with my one, its very sharp, I cut tomato by hand at 0.5) which I need to overcome the ABS juice that I need to use to overcome the warping.... Its a whole chain of stuff waiting to go wrong lol.
I'm good at laying the tape with no bubbles so its hard to see. I'll try it with board the other way next time I need to change tape.

I also have a 5v supplementary heater which I documented for peer review in the developers section of this forum [forums.reprap.org] So I get some nice warm air drifting up and the bed never cools below 40c if the ATX is on. Its a great mod for those with ATX!

I have a fan. I think I need to change it. It blows way too hard. Enough to stop the hotend getting over 150 at 100%. No idea what CFM it is. Probably out of a Plasma TV. I have a funny setup there. Pics will explain better...



So the fan mostly stops the hotend getting too hot and blocking above the heatblock/nozzle. The bit of soda can air flow divider is a version0.1 design. I have a v1 planned. It will use the Coanda effect to get air to the leeward side of the print as well as the blower side. as well as better insulating the hotblock. I'm only diverting a tiny bit of air below the nozzle. If I wet the back of my hand I can just feel it 5-10mm below the nozzle. A dry hand can barely detect airflow at 1mm below nozzle. More than 10mm below the nozzle and there is airflow, but its warm air from the bed.
Obviously this is tunable at the deflector and the fan pivot. (that fan mount was printed with a hand made 1mm nozzle and overstepping z, not bad if I do say so myself) So I can print with no air on the print surface. Its helps with the bridges though.
Normal fan speeds are from 15% to 30%pwm. Over 30% for long cools the hotend and a kill command is issued. I have been bridging at 40% with no issues.

Bed normally runs at 130. Its thermistor is better. 130=130... I was running lower but the heat helped with the peeling etc. I just have to let it cool below 110 to get parts off or they can warp a bit. The big fan is helpful there. I have a Cool bed button in pronterface....


And I've done it again. A quick reply turned into a small novel... Hope I got all the relevant info in.

Where to start? a fan mounted on the hotend with proper duct might be better, cheap as chips, (as long as you dont shop in McD's)
I just wrapped some tape on mine where the air exits so as to point the air up away from print, do people printing abs have any kind of part cooling? your psu will also be blowing cold/warm air around the place its close to printer(i think even with the shield), is your hotend PID tuned? maybe if the room air was 18 that would cause problems, I would be worried about bed traces getting gouged and something becoming live, but I dont know maybe it would just blow something, did you print the parts for the printer?

Edited 2 time(s). Last edit at 09/10/2016 09:25PM by MechaBits.

I print ABS like 90% of my time. NAd believe me: there is no way or anything you tell me to make me believe that the radial fan you have is not afecting your ABS prints.
First, I have to say that your print temp is way too high, I think your temp reading are not accurate or if you really have thant temperature, you should have runid the thermistor by now. "common" thermistor tolerated till 250 C, unless you have a special one.
My advice would be: first, print a fan shroud. Second, you need an enclosure if you really want to print in abs securily. And for you bridging: for ABS you are quite reaaly good there, since useing cooling is not an option, you need to fine tune speed and extrusion rate and temperature.

Thanks again for the help guys.

I have played with retraction setting some more. Reducing extra length down to 0.05 fixed the over extrusion on tiny deposits. Still experimenting here, I may have only needed extra length to compensate for other issues (Insert joke about big cars here).

The printer parts where purchased as a kit from ebay 2-3 years back. I built it then but we had a stack of PEBKAC/didn't RTFM issues (like slicing for 1.75 filament and running 3mm lol facepalm) and it got shelved until 6 weeks ago when I picked it up from my partners house and brought it here (my house) where I can work on it full time. As I have a fairly well equipped electronics lab and metal fab setup.. While the print quality of the kit is great I'm not thrilled by many design aspects and am keen to find or design better parts to print for a new stiffer frame. It all works surprisingly well considering the design of the parts though. I've done a number of prints that have come out a lot better than I ever dreamed of.

I have adjusted the fan so its not blowing on the print at all. Not sure if its made much difference. Its turned down really low and was only just pointed low enough so that most of the air was cooling the heatsink and barely any was going below the nozzle.

I have PID auto tuned a few times. Its reasonable. maintains hotend within +/- 2 or 3c. Not really expecting better from a thermistor.

I agree its not the right fan. I have a clip on fan that came with the hotend. I'll be swapping them over today, likely using the fan always on connection on the RAMPs board (the one between stepper drivers and mosfets). As well as flipping my bed, tidying wires etc up a lot and tightening up whatever loose tolerances I can.
An enclosure is high on the to do list. For now I have improved the shielding and there is no drafts in the build area. While room tamp varies from 18 to low 30's the temps in the build area are in the 30's and 40's. Depending on where reading is taken. Its pretty stable.

The themistor/hotend is claimed to be rated for 300c and nylon. I'm dubious here. Pretty hard to have confidence in cheap ebay stuff. The 30w (measured) heat cartridge cant maintain 300c... Further testing has revealed thermistor reads very close up to the mid 240's then starts to under read as temps rise. Its a 100k glass mode andl at higher temps I suspect its got so low a resistance it can't drop more ohms... I have a K-probe and Maxis 6675 amp. But cant get Marlin to compile with it turned on... Apparently its a known problem. I think I can run it without I2c directly off the ATX and feed its signal to the thermistor input with a little re-coding. I have a mate that can help with the code when he has time off work. Any extra info on that would be very helpful. I'm also open to I2c, but we have a screen and sd unit we want to use and I'm seeing more problems than benefits running k-probe on i2c. (I'll get round to googling this eventually)

What is an ideal build area temp? (I'll get round to googling this eventually)

As to the extrusion temp. I'm thinking more and more I may have bad filament. The seller claimed it was cheap do to closing business. But I'm thinking there is more to that story.... Had some major dramas this morning after changing colour yesterday. It all seems extrusion temp related.... Running so high has not been sitting well with me. Its clearly indicating I'm doing something wrong or have bad filament. I think.

About the take pics of my last few prints and start a thread on the temps, as all my other problems keep leading me back to the temps... Its clearly the root of all the problems I am having. That should be the thread to end all my "I did something stupid and need help" threads.

Edited 1 time(s). Last edit at 09/13/2016 02:24AM by Mongrel_Shark.