#Inconsistent Flow on each print/ day
1455 messages Β· Page 2 of 2 (latest)
These are not post processed, test single cube first to see how it looks, if it looks bad I will post process it and then we compare again, if it looks good then I will post process the rest and we will continue testing
this will narrow down if it's slicer or post processor
if all look bad...
well
I'm leaning towards attempting removing the ptfe tube and trying direct feed one more time
after that...
we go nuclear
reflash and resetup
The bottom left is worse and the top right is good? If there is something loose it would have to be in front/rear direction then. (hold the nozzle and pull/push?).. The bottom left one looks like inconsistent extrusion really but only when toolhead is going in one direction.. Weird..
Just the top right looks good all others have gaps between the lines
So deactivate the post process agin ?
is this linespacing matching the top solid linewidth in slicer? or are you supposed to have 2x more lines here? indicating printer is placing a line more or less right on top of the previous line when alternating direction so that the plastic has nowere to go, nozzle pressure causes undulations/waves around the line etc..
Here should be one line too
yes, this looks like left going lines are 100% overlapping on the previous right going line in some areas since we can see the two narrow wavy lines on each side (plastic forced outside on each side of the previous line)
Yes but I donβt know if these (the red ones) are the move from left to right or from right to left
Here is the same printed part as before just a little bit clearer
The left vs right doesnt matter it was just an example. But you can not tune EM correctly with this going on.. i guess you could print a long narrow cube with only perimeters, so that all "fill" is going in the same direction as the top right cube when tuning flow/em,
Yes you are right till now this was like playing lotto at one moment it was good and the next it was bad but yeah I will test it
sounded like you were stumming around 90hz on Y belts and 80hz on lower belts? If this is a 500 size machine Then isn't that a very high belt tension? idk. i think mine plays around 75 on the lower belt and 60ish on the top belt, (i have a VC 3.1 500 hybrid)
if you have binding somewhere in the motion system it could get worse with high tension..
Yes normaly I have less tension I just tested this yesterday if it does anything
I will set the upper belts on 60Hz and the lower on 75 hz
Here I have 0,71
But it should be 0,84
(One line is 0.42)
I just had 40hz on the upper belts
Where have you measured the hz ? With the x gantry in the middle ? Or in the back ?
Is this normal ?
You have to watch with sound
I measure / compare tension with the gantry pushed to the back.
And then you have these values π± isnβt it way to much ?
maybee you have a belt that is rubbing on a bearing flange, if fame is not alligned then belt will often rise or lower depending on direction.
Yes i set my tension a bit too high myself as well, my gantry is a bit hard to push also, but at least it is the same in both directions. (i have hybrid so all 4 steppers rotate when i push the gantry back/forward. side to side it is a bit easier bc, then i am rotating just 2 steppers.)
If everything fails you have the option to go back to basics, (remove a toolhead to rebuild to corexy non-idex) with or without Y(hybrid) motors. But that is a serious job just to troubleshoot..
with idex it is harder to setup and tune, because you need to use the 2 y motors from the start i think.
Yes this a thing I will do if I have up haha then I will build it back
Okay
So After 4 top layers after the infill the gaps are back again
This was the 5. layer after infill
Just up and down which I just thought have no problems
Have you tried printing a cylinder in vase mode using same mm3/s flow. it does not show any extrusion problems there? Have you confirmed the extruder/hotend is not the issue by running a max flow test in orca using the same temperature?
Yes I reached a max flow of 25 and I am using 15 max flow
Not now
This was the Pa test tower from yesterday and the outer walls looks good
But yeah I will test It later with the cylinder and one wall
you can also go down to 10 flow cap in filament setting. just to remove as many variables as possible. and for simple testprints temperature can also be increased just to lower the viscosity / extrusion force. (it is difficult to know what the nozzle temperature really is as it is very depending on the heat transfer and proper assembled hotend)
Yes testet also with just 5 max but yeah I will test this again
Yes I can print the next prints with 255 degrees instead of 235
don' t trust the temperature in mainsail. I'm even struggling printing PLA under 230 degrees with rapido 2 plus using the uhf adapter and tungsten nozzle. and that is with 20mm3 flow limited.
Oh okay Crazy
maybee i should try to stick a termocouple down into the meltzone of my own rapido later today to see what is the temperature with my setup.. i just last week "upgraded" from rapido 1 HF plus to a rapido 2 UHF plus and have had only issuess. until i increased my temperatures a lot.
I also tested a dragon uhf hotend
But I will test other temperatures
What diameter have the nozzles ? I am thinking about to buy a pt1000 with a round mount so I can mount it between the nozzle and the uhf adapter just to measure the real temp
6mm
m6 threads
could use a m6 threaded sensor as well, just replace the nozzle with the sensor π€
Deactivate your post processor if the single cube looks bad, print again, if it still looks bad idk.
At the moment the 4 cubes are printing and if I come home later I will update you
How did single and double cube come out?
Or did you just go straight for 4 π
Straight 4 π
Understood lol
Because I left the home and I thought I can print the longest print while Iβm gone
Uhhhhh
0.89 but I can double check when I get home
This looks promising however
It's consistent
Donβt know π
there is a very consistent geometric pattern on these prints. could it also be the top fill that is sagging a bit the infill (gyroid?) due to high temp or too little cooling on the top layers ?
The infill is gyroid, so this is definitely likely
Temp 255, im not sure about fan. Maybe 35%
Again, gotta check what I sliced when I got home π
i looked at the file, it was just two top layers. I would be cool to see another attempt but with minimum 5 top layers.
bcause two top layers is in reality just a bridging flow layer + a single top layer
I'll slice and send in a few hours, not a bad idea π just didnt wanna waste a bunch of filament and time 
Perfect thank you
Until we find the problem, then the filament doesnβt matter
π
You are using 0.4mm nozzle, right?
the gcodes i looked at was for 0.4
nevermind, is saw you confirme that the other dayπ
Yes ππ
Haha my rapido 2 wasn't so wrong anyway. I set the temp to 220 and shoved a 0.8mm probe all the way down to the inside tip of the nozzle. (inside the molten plastic). it was actually super stable and only changed max 0.2degree over a minute.
That is really interesting, and do you start the fan ?
no, didnt think of that.. this was just to check if the temperature reported was correct.
Okay even with the max flow of 3 I got holes / gaps
Hmm maybe the motion system is not the problem because if I go down with the speed the result should be better even if i go down from 15 to 3 max flow π€
I also feed the filament by hand directly to the extruder here
There has to be something wrong with commanded values
We have verified correct X, Y, Z, and E travel
So the machine itself is moving the correct distance
Now is it being commanded the correct extrusion amount
Based how with my sliced files you get no gaps
Over-extrusion, sure, but not gaps
Maybe the overextension hides the gaps
I donβt think the slicer have the problem because I use it also for my prusas and bambulabs
Thank you
would you like less flow?
Yes please maybe 2% less
And maybe just 50x50
so I can look at it before I go to bed haha
Thank you
this looks good and normal
other than maybe SLIGHT overextrusion
but your extrusion is consistent
Yes think so too
I know you said you use this on your bambu... but man it's looking like you just need prusaslicer with a default ratrig profile π
I just tried this π
If this next one works I'm just sending you my prusaslicer config file and never ever change anything on it ever 
ok if this single one works, final test is to print 4. because it's done gaps everytime you've done a long print
if the 4 print works... I'd say it's clear and any artifacts after that is just slicer tuning
Oh yeah
Still overextrusion maybe 2% more less
Can you tell a difference between this and the last one
like is flow working lol
I think yes but it is hard to tell but it is very odd normaly the flow between orca and Prusa should be the same
Looks way better can you now make 4 pieces please ππ€―
0.85 flow. 25% infill, 4 cubes
Tank you
And this one is with 85 flow or ?
yessir
Perfect
So I think we have to less top layers because of the longer layertime I think the cooling goes down for the overhang / first layer after infill
Maybe I accidentally uploaded the 2 layer one again... π This can also be solved by dropping the temp, or increasing fan speed.
I'll send you the config file. Looks like it's a consistent extrusion and things are coming out well
Yes but I donβt want to change any values
I just want to see if it can be printed without the gaps
If we prove that it works then I will start to change the values
There's no gaps in the one above?
Donβt see one but not clear because of the uneven top layer
Thank you π
Thank you
So β¦ gaps again
You changed anything in the settings or ?
Oh very minor gaps
But not on all of these 4
I strongly feel like your problem lies in temp/fan/flow rate
I wont be near my computer until sunday but I can maybe manually adjust the hcode for you to do again
Okay at the moment I will test with orca slicer again but other fan speed and temp
Do you know which settings you are used before ?
Before was 15% fan and 255 temp
0.85 flow
25% infill
4 top 2 bottom
2 walls
Alternatively
You can print my gcode again, and when the fan turns on, go into RATOS and turn the fan off and the temp up to 255 and that should be permanent
Yes at the moment I try orca with 10% fan and 255 degrees
Okay the Prusa slicer can not be the solution
Because I now tested Prusa slicer with standard settings (also 15% fan and 255 temp and 0.85 flow)
I only set the top layer to 6 and after the 4th layer (top layer) the gaps come back again
Yes
What do you mean
I think it just means you need to run 4 top layers π
π
If you need more strength, more walls, or more infill
But this canβt be the solution ππ
But 6, may be preventing excess filament from expanding into the infill like you explained before
100% infill has always caused issues
And doing 6 solid top layers is similar
Yes but why I mean this isnβt normal
Tonight u will print a print with minimum speed so acc to 200 for all max flow 1 and jerk 0,5 for all
Oh
Yes I am testing at the moment 80%
Ahh ok
Because im thinking it could be partly dumping so much heat into the top layers
So I can say if its the mechanical or not because if it prints so fucking slowly it should look waaay better
Yes maybe
In 10 min we knew mor
More
KNOWLEDGE JULIAN KNOWLEDGE
Okay the first test looks great
Just look in the middle the other is warping π
I think we are on to something
I test now less temperature (225) and 60% fan
ποΈπποΈ
Really odd
Temp is significant drop but im here for it
That is odd. Still at 0.85 flow?
But we will see the result in a few minutes
Ah I see I used the false flow
Slightly higher temp
Or more flow
But you're nailing it
If it's not curling the fan speed is good
What do you mean with curling ?
Warping :)
More temp
You went more down 
Less temp = less flow
More temp more flow
Compromise: 240c
This looks good though
Especially with ABS
225c is a little on the cold end
Especially when you're trying to push around 30mm^3/s
But even on standard/older printers
250c is the average
Okay now I print with 240 but I go to bed and write tomorrow morning
I have 3 pieces
One very fast
One very slow
And one normal
Look forward to it! Nice work, making progress again :)
Thank you vor your help
I hope so π€ (what I think is β¦ maybe the rapido 2 UHF needs more speed for the printing so maybe without the UHF it is better because also the dragon hotend I tested have the UHF and to compromise the UHF I need less temperature and more fan β¦ but just a thought) and now good night I will update you tomorrow
And thank you again π
Hello againβ¦ so not what we thought β¦
This was very very slow (50mm/s speed for everything and 1 jerk for everything and 500mm/s)
And it looks perfect β¦ this is the result I want
This was fast (5000mm/s)
Looks okay also without gaps
And here is the βnormalβ I am printing with every time this one have gaps again
At the moment I print these again just with the change that it prints by object and not by layer
Another interesting thing is that also the parts from Ratrig have this problem β¦
This is the plate between the x and y Motor just saw this last week while I am rebuilding the xy cube
I wouldn't worry too much about ratrigs parts
I think however, we're changing a little too many variables at once to get good conclusions
(at least im getting confused now lol)
Lets do this:
normal speed
240c
60% fan
0.9 flow
Now the surface quality is good on this
With minor gaps
These gaps are classic underextrusion
So we bump the flow a little bit back up from before
I wanna reiterate though that I think we are proving this is not a printer issue (as your printer has followed commanded values exactly).
We are simply just at a difficult filament tuning stage
And ABS is not the easiest to print with
Yes if I know this for sure than I am a little bit happier because then I have a direction where too look at
Do you have any idea how I can test if the printer is good
Will test this later if I got home
Yes at my Bambu and Prusa I donβt have problems with abs so I thought it is as easy as on these machines
You already have.
You have tested:
Dimensional accuracy β
Extrusion accuracy β
Which are the 2 things important to a printer. You havent had any errors, this is genuinely just filament tuning.
True but the bambu is self tuning and the prusa is slower... Different beasts!
At the moment I print these settings but with the same flow because the print I did today (object by object) was perfect
I just speed up a little bit the walls and something
We will see in 8 min π
So this was the result of the normal speed from the object by object thing
And this is same temp same fan but just a little bit faster in the settings
Now I try the exact same with 260 degrees
Same settings just 260 degrees
Soooo
240 degrees 1% more flow 60% fan
And I have a usable result
Maybe 1% more flow and then
I will start the big test over night π
Fingers crossed
I donβt know why but the 5th top layer looks perfect and the 6th have gaps
I changed the Acceleration for the toppest layer to 3000 like the 5th layer in the hope it looks the same but no I got more gaps then before ππ
And I also set 1% more flow
But yeah more gaps
What happened between this test
And this
I only changed the acc of the top layer to the same value as the massive infill acc
So this one was again with the same acc as massive infill but I change the line with to 0.4 for the top layer instead of 0,42
I canβt explain this to me why does the fan changes this result
Okay donβt think we have it
I printed these with the same settings as this one
The walls are good so I think we can really say that the mechanic of the printer is good
I mean the layertime is much longer on these prints
I agree, I think the mechanics are good
(also you look very close on your first layer, this wont affect top tho lol)
Yes the first layer is rubbish at the moment but I want at first to get the top layer good and as you say this affect not the top so yeah
I think I will give it a try to reset the complete firmware
As I know 0.4 is the best because we also have a 0.4 mm nozzle the other values are possible too but the best should be 0.4 is what I know
This.... This may be something
Look what I found on this CNC kitchen video
From most of what I've seen, (including default settings) 120-150% is common for extrusion width.
Start your Squarespace free trial today at http://www.squarespace.com/cnckitchen and use code CNCKITCHEN to get 10% off your first purchase.
Extrusion width defines how wide the lines of extruded material are. Higher values require more pressure to squeeze the molten filament out of the nozzle that also presses the layers together. I investigat...
Watch if you feel so inclined :)
Could we try this:
240c
60% fan
0.9 flow
2 walls
6 top layers
Line width for ALL layers 0.45
Now I am printing different models
- 0.48 layer lines
- 0.45 layer lines
-0.42 layer lines - 0.4 layer lines
I have a similar issue: https://discord.com/channels/582187371529764864/1315555504558182410
but not yet found a solution. I suspect it is related to the mesh (although I have only .15 deviation) as this usually only ocures >7mm in z
Yes looks like the same
To exclude this, an object must have at least 1 layer infill then it is not mesh related
I am printing 50x50x2,3mm cubes to test this
the prints have all multiple layers of infill
Then it have nothing to do with the mesh
my though is atm that there is maybe some sort of either phase out of the mesh or maybe bi-metal effect after certain z
I donβt think so because I got perfect first layers and also I have an titanium tube
And also I got perfect 1 layers oder 400x400
But the top is trash
that is interesting
is it just the top layer or also top -1?
on mine top-1 is perfect and top shit
I bring now my child to bed and then I will write again and start testprints again
Sweet, would love to see the results! Thanks :)
So
Yes top -1 looks good but sometimes I see just a little gap but I would say 90% the top -1 looks good
And if you reduce the thickness to I.e 5mm the top surface is not overextended?
What do you mean ? Sorry donβt understand
In my case I observe the overextrusion pattern usually if z > 7mm
No these prints are just 3.4mm height
RR tech support claims this is warping but for me I can exclude that
Yes this is not warping 100% not
I use orca slicer and Prusa tested both
Oh yes tested multiple times haha
Very interesting that we have different setups but experience the same
Oh yes what machine do you have
But no ti tube
Hmm have you already re installed the firmware ?
No just updated to current
Okay
I will try it now
I canβt flash the board π€ it says it falsh the board but then it is again on the false firmware
At first I will go on
So with everything default just the flow of the Ratrig abs profile to 0.93 (which I used in orca before)
Minimalistic gaps
Now I make IS and put the fan to 30-50%
Okay it gets worse
Now I just try the settings from Bambu
Tomorrow I will see the result
Can you print 50x50x2,3mm cubes so we can see how does these look
Does this have pressure advance or no?
These are 100x100x2,9 that look great - just a bit too much overlap
Yesterday exactly same settings but 8.8mm height of top layer and no warping at all - was perfectly flat on bed
Today same settings and part just 6.8mm height of top layer and moved it 2cm to the back as the bed mesh (.15deviation in total) has on the front the high point. Cannot show this one better as it is still printing. Top layer is perfect
This is why I say ai have absolutely no clue π
For my two VC4s I can tell it is somehow correlated to height of the top solid infill layer and maybe to positioning of the part
But I am far away of narrowing down the cause
BTW all PLA
@helge yes
Ah okay then I think you have a different problem
Another part same settings but random underextrusion in one corner (also no warping)
Yes could be. Anyway, Iβll follow this thread. Sorry that I am not really a help
Yes but my problem is not in one corner my problem is random in the mid of a print and just gaps not completely underextrusion
No problem we are here to write and look if we can help each other π
Do you have the same experience with a different material/brand and if it is idex also with T1?
Yes tested Petg and ASA and tool 1 also
And same gcode on different printer comes out perfect?
And with slightly adjusted settings on another brand?
You can send it to me I can print it
But the other one is really misaligned π
But please PLA settings
Okay I will make later one with pla and send it to you
Thank you very much this shows 100% if it is printer related or software
This was the settings from Bambu and till now no gaps I will print now the longer print which always have this problem π€
I think despite all this that the problem is mechanical because it is so irregular. And if I print very slowly itβs gone.
Possibly the belt tension is not correct or the motors have a problem
After the tests with pla later I think I would go back to hybrid and try then
Okay now I have a print with pla with the same result
Donβt look at the edge I had a too high travel speed and thatβs why I have a few layer offset
This is 1h and 16 minutes
43g of filament
Can you print this ?
Please
There is a "phase out" from the mesh. Its referred to as fade_start and fade_end
[bed_mesh]
horizontal_move_z: 2
mesh_min: 30,30
mesh_max:470,450
probe_count: 40,40
fade_start: 1.0
fade_end: 10.0
mesh_pps: 2,2
algorithm: bicubic
bicubic_tension: .2
What was the result?
I print now with 1% less flow
But looks really good I think
This was with 94% and the other (the picture before) was 97% now I print with 96%
And I got a huge better speed
100% fan this thing is screaming π
And on hybrid gone?
So this is the result not as good as my bambu but usable
But yeah it is not as good as my bambu or xl
Can you send me the stl?
Till now I got good results maybe just a settings thing now for the top layer but yeah I can use this machine for orders finally. π€
And my IS is MzV and ZV so yeah I think I donβt will go back to Idex for a time π
Next successful print
Now I print 10 of these in one print
Think its really working now π₯°
beautiful prints!
So the final test if this prints without an error then I would say it is solved
But if I go back to Idex I donβt know if it comes back π©π€
Nice
i also am going to print the Chonk parts as well soon. But the big parts are scary. What material u used? Did it end up without warp?
This one was ABS but it was just to test the printer the real part I will use, I will print out of PC Fibr20 or ASA CF15
Amazing to print this big in ABS wihout warp. When you slice the big bottom plate with some sort of CF filament, doses the slicer calcualte less than 1kg will be used? when i sliced for PET-CF, 25% gyroid and 4 walls it came out scary close to 1kg. dont remember bu tit was like 996g or something. What are the strength settings you use or recommend?
Yeah you are right with PC Fibr20 I need 1051g so I need a spoolchange π© or I just glue the end with the new roll together with the sunlu tool
Yes I am using the same 25% infill gyroid and 4 walls
most exiting filament change after emptying a whole expensive roll on a part. Be carefull with the splicer method on such brittle filament, the splice might snap in bowden before it even reaches the hotend. Perhaps better is to stay close and monitor the filament transition, then you could just push the next filament behind the other until it is grabbed by the extrudergears.
Ah yes this is also a method ππ
Okay I print it out of ASA CF because I have 2,5kg rolls π
How is your printer working ? Do you found anything ?
So what was the problem in the end?
I donβt know
After Build it back to one head it works perfectly
I donβt change anything else