Seite 1 von 3

RF1000 Which doesn't want to print properly

Verfasst: Sa 9. Mai 2020, 19:48
von speckledpear
Hello everyone,

First of all, I want to apologize for making an English post, I hope this is allowed.

I've got a secondhand RF1000 that I've been trying to fine-tune for quite a while. Initially, I thought the problems were in the Slic3r settings and I didn't touch the hardware or firmware. Recently I've come to the conclusion that the main problem is in the hardware or calibration.

Here's what I know from how I received the printer:
  • The RF1000 came with a used nozzle which I assume is 0.5 mm. I've now also got a new 0.3 mm nozzle.
    The Z-endstop has a modification to lower it.
    The filament is fed from the top, not from the side.
    There is a cooling modification that's made for cooling the nozzle. I don't think this works effectively since I barely feel any air coming out of it
Here's what I've done:
  • I've tightened the belts.
    I've upgraded the firmware to community version 1.45.
    I've calibrated the z-axis and performed the z-axis calibration.
    I've tested around with Slic3r settings.
Important details:
  • I use 2.85 mm PLA that I print between 190°C and 210°C at a bed heated at 65°C.
    I replaced the driving gear with a MK8 drive gear.
    I use Slic3r and have the following as my start G-code
    G28 ; home all axes
    G1 Z5 F5000 ; lift nozzle
    M3001 ; Activate Z-Compensation
    The M3013 P1 results are as follows
    19:45:03.256: M3013: current heat bed z-compensation matrix:
    19:45:03.256: front left ... front right
    19:45:03.256: ... ... ...
    19:45:03.256: back left ... back right
    19:45:03.256: ;7;0;15;35;55;75;95;115;135;155;175;195;215;235;245
    19:45:03.256: ;0;-0.36;-0.36;-0.30;-0.29;-0.28;-0.30;-0.32;-0.29;-0.28;-0.31;-0.30;-0.35;-0.35;-0.35
    19:45:03.257: ;30;-0.36;-0.36;-0.30;-0.29;-0.28;-0.30;-0.32;-0.29;-0.28;-0.31;-0.30;-0.35;-0.35;-0.35
    19:45:03.257: ;50;-0.33;-0.33;-0.26;-0.28;-0.30;-0.29;-0.30;-0.26;-0.28;-0.31;-0.32;-0.33;-0.33;-0.33
    19:45:03.257: ;70;-0.35;-0.35;-0.30;-0.30;-0.32;-0.30;-0.31;-0.29;-0.27;-0.33;-0.33;-0.36;-0.38;-0.38
    19:45:03.257: ;90;-0.36;-0.36;-0.32;-0.31;-0.33;-0.31;-0.31;-0.31;-0.30;-0.34;-0.37;-0.37;-0.41;-0.41
    19:45:03.257: ;110;-0.35;-0.35;-0.31;-0.31;-0.33;-0.31;-0.32;-0.31;-0.31;-0.34;-0.37;-0.38;-0.41;-0.41
    19:45:03.257: ;130;-0.35;-0.35;-0.33;-0.32;-0.34;-0.34;-0.34;-0.32;-0.33;-0.36;-0.39;-0.42;-0.44;-0.44
    19:45:03.257: ;150;-0.36;-0.36;-0.34;-0.32;-0.34;-0.36;-0.36;-0.32;-0.35;-0.37;-0.40;-0.43;-0.45;-0.45
    19:45:03.257: ;170;-0.36;-0.36;-0.36;-0.37;-0.38;-0.38;-0.40;-0.36;-0.38;-0.41;-0.44;-0.49;-0.50;-0.50
    19:45:03.258: ;190;-0.37;-0.37;-0.36;-0.38;-0.41;-0.38;-0.41;-0.40;-0.40;-0.44;-0.47;-0.46;-0.54;-0.54
    19:45:03.258: ;210;-0.34;-0.34;-0.33;-0.33;-0.39;-0.36;-0.41;-0.41;-0.40;-0.43;-0.47;-0.45;-0.54;-0.54
    19:45:03.258: ;230;-0.36;-0.36;-0.36;-0.37;-0.43;-0.41;-0.44;-0.46;-0.46;-0.47;-0.52;-0.52;-0.59;-0.59
    19:45:03.258: ;245;-0.36;-0.36;-0.36;-0.37;-0.43;-0.41;-0.44;-0.46;-0.46;-0.47;-0.52;-0.52;-0.59;-0.59
    19:45:03.258: offset = -334 [steps] (= -0.26 [mm])
    19:45:03.258: warpage = -427 [steps] (= -0.33 [mm])
    19:45:03.258: g_uZMatrixMax[X_AXIS] = 14
    19:45:03.258: g_uZMatrixMax[Y_AXIS] = 13
    19:45:03.258: g_nActiveHeatBed = 1
    Pictures
    Print results. The left cube I made a few days ago when still using the old firmware. The right cube I just printed after installing and calibrating with the community firmware.
    Bild
    Extruder and fan. (fan .STL the old owner used: https://www.thingiverse.com/thing:2239814)
    Bild
    Bed and Z-endstop
    Bild
I've got a few problems. To start, without using a glue stick the filament doesn't stick to the heated bed.

More importantly, I can't seem to get a constant extrusion and the filament tends to slip. Now as can be seen in the pictures, the springs are held quite tight and should hold the filament strongly against the driving gear. However, a little bit tighter and my driving gear clicks and a bit looser and the gear will turn without gripping against the filament (grinding?)

I've had moments where I got lucky and managed to get both these problems out of the way. Then I still get poor print results. I believe the nozzle is held too close against the filament because I can see some marks of the nozzle grinding against the filament in the top layer. Additionally, all corners of the cube slowly get blobs while printing. This gets worse as the print continues.

I've been using Google Translate to read through the forums and due to this have switched to community firmware. Apart from this though, I don't have a good idea of how to continue and fix the printer. Any help is greatly appreciated.

Thank you,
Stefan

Re: RF1000 Which doesn't want to print properly

Verfasst: Sa 9. Mai 2020, 22:50
von Digibike
Hi,

Welcome to our Community. My English is not so good, but here are a lot of people, who want to help you - and the language is ok...
I´ve had a look at you start-sequence. This is very short. No Purge and so on? Do you print on the orginal headbed or have you something else on it? The orginal is not so perfect. How many Digits has you? How much flow has you at Slic3r? How high is your speed at first and other layer?
You has now a 0,3 mm Nozzle, or want to tell us, that you know can use additionally 0,3 mm Nozzle, but print actually with 0,5 Nozzle?
The filament Diameter has you take from the manufacturer information or has you measured yourself it and set in Slic3r?

with regards, Christian

Re: RF1000 Which doesn't want to print properly

Verfasst: Sa 9. Mai 2020, 23:02
von rf1k_mjh11
Hi speckledpear/Stefan,

While it is always difficult to do a remote assessment, I'll give you my first two cents worth of tips and comments.
speckledpear hat geschrieben:I believe the nozzle is held too close against the filament because I can see some marks of the nozzle grinding against the filament in the top layer.
This 'bug' is quite normal in many cases - it can be minimized after you have gathered more experience and succeeded in 'dialing in' your printer better. Don't worry about it too much for now.
speckledpear hat geschrieben:However, a little bit tighter and my driving gear clicks and a bit looser and the gear will turn without gripping against the filament (grinding?)
This sounds interesting ... Normally, increasing the spring preload should not cause 'clicking' in itself. The clicking you hear is presumably the extruder stepper losing steps. If the extruder starts losing steps once you've tightened the springs more, then you are probably grinding filament with the springs not tightened as much. This is an indication that your extruding forces could be on the high side. Now there are several possible causes for high forces. Before we get to those, you should check just what your 'F-Digits' read as you print. The F-digits are shown on the display of the printer, right below the 'Z' value, i.e. on the third line. Two values for the F-Digits are important here: the idle value and the higher (highest) values seen when printing.
The idle value is the number shown when you turn on the printer (preferably without any filament loaded), or when the printer has been at printing temperature for a minute or two, but is not printing. The idle value is a result of build variation and internal stresses within the system. Ideally, this should be close to zero, but don't worry, I've printed for more than 4 years with an idle value of around 1600.
The actual F-Digits encountered during printing can be much higher. On my RF1000, I start hearing the extruder lose steps when the F-Digits start exceeding "idle value + 6500", in other words, when the F-Digits start going above 7000 or 8000, depending on my idle value (and spring pre-load).
Let us know what kind of values you see (idle & printing).
What can cause high extruder forces?
  • Printing speeds too high
  • Temperature too low
  • Clogged nozzle
  • Mis-aligned filament lead-in
  • Sharp edge at filament lead-in
  • Friction in the original filament feed tube (does not apply to you!)
Some of these are easy to check and remedy.
Re: Printing speeds: anything below 60mm/s shouldn't be an issue (provided the temperature is OK).
Re: Temperature: I don't think this is an issue in your case, but it is easy to check - run the same print with a temperature 20° higher and see what you get.
Re: Clogged nozzle: this also doesn't appear to be the main issue for you, as it looks like there is enough filament being extruded. But play it safe and use a filament cleaner (the simplest is just a half-inch cube of sponge that you poke the filament through.
Re: Mis-aligned fliament lead-in: This was an issue on some early RF1000s. If you remove the filament (and your printed filament guide) and look through the hole in the aluminum piece from the top, check to see if the brass filament lead-in lines up nicely with the hole. The slop in the assembly holes allows for a good amount of play - which can result in mis-alignment.
Re: Sharp edge on filament lead-in: This was also an issue, early on. The brass filament lead-in had a fairly sharp edge on some units. With an otherwise acceptable degree of mis-alignment, this sharp edge would cause the filament to continuously get shaved off on one side, increasing friction, with the shavings getting caught in the lead-in, also increasing friction. Check the bore and use a needle file, or other suited tool to remove the sharp edge, should there be one. Be sure to REMOVE any file residue or chips when doing this!

Check these and report back.

Now to a different topic altogether - The MK8 drive gear:
You need to re-calibrate the extruder in the firmware if you use a different drive gear (the STEPS/MM value). There is a good thread on the topic, here, plus a revised one, here.

Bed adhesion:
This is always a topic. This forum probably has a hundred posts on the issue. Aside from the obvious, it is also important to have the correct distance between nozzle and bed for the first layer. This can be a challenge for a beginner. Your Heat Bed Matrix states "offset = -334 [steps] (= -0.26 [mm])", which tells you where your bed is with respect to your nozzle. Normally you can compensate for this distance in your start gcode. I'm just not sure how the Community Firmware handles this at the moment, it may be automatic. Nibbles or AtlonXP could help out with that. The non-Community version of firmware would benefit from a slight Z offset in the start gcode, since it does not automatically compensate.
Once you have the nozzle-to-bed distance down pat, we can discuss other adhesion enhancing options.
speckledpear hat geschrieben:There is a cooling modification that's made for cooling the nozzle. I don't think this works effectively since I barely feel any air coming out of it
The fan, provided you are using the stock hot end, is used to cool the printed part and should not cool the nozzle, if possible. The stock fan was not the greatest engineering solution, many in the forum have come up with alternatives. You photo seems to show an alternative solution, also. There is little air flow on many of these alternative designs - increasing the cooling power can lead to other issues.

... Enough for now, I'm hitting the sack...

Good night and Health to All!

mjh11

Re: RF1000 Which doesn't want to print properly

Verfasst: Sa 9. Mai 2020, 23:14
von Nibbels
There are many tips already.

About the StartCode:
You maybe want to activate the feature "SenseOffset" of the community mod.
This works as overpressure relieve control when you print the first layer. That means if the hotends inner pressure gets too high (you set a limit for that) then the printer will automatically adjust the distance from nozzle to bed.
wiki/index.php/SenseOffset
wiki/index.php/Gcodes#M3909_-_SenseOffs ... aktivieren

I cannot tell much about the picture you made from the cube which looks concave.
Without thinking mutch I would say the part cooling fan was not working, the speed to high - without PCF. The bed too hot. (?) Weird.
Maybe Advance could help a bit, but this should not be the actual reason that it looks crappy.

Re: RF1000 Which doesn't want to print properly

Verfasst: So 10. Mai 2020, 00:34
von Digibike
Your Fan is nice, but I´m not sure, that this is a good Idea... There are some Problems about: 2 Fan is critical for the RF1000 board. Look at this thread... Your Fan-channel is for only small power output. 2 fan´s with much more power will kill them. Perhaps not yet, but it will kill them...
The next is, air flows opposite will cancel each other out.
The next is, your construction see me, that it will bring many cool air to the top of the Nozzle... The heater must work very hard to hold the temperature, I think so... Silicon socks are perhaps a good Idea to hold the cool Air at a distance to the Nozzle.

Re: RF1000 Which doesn't want to print properly

Verfasst: So 10. Mai 2020, 11:28
von speckledpear
Thank you for your responses everyone!

I use a new 0,3 mm nozzle but also have a used nozzle available which I assume is 0,5 mm. I've shared my Slic3r settings in the pictures below but I don't think these should gain too much of a focus now considering I can't get a constant extrusion.

The filament diameter comes from the manufacturer. I've got callipers with a 0,1mm accuracy that I use for measuring 50 mm to extrude. However, I can't get this consistent because of the extruder clicking or slipping.

I've added a sponge and tube for the filament lead-in. I didn't find the lead-in to be particularly sharp but now that I've got filament loaded and look from the bottom I can see a bit of residue that could be grinded filament. I'll take apart the lead-in section and will file it down.

I assume the F-Digits are shown with the P: value on the third line?
When starting up the printer the value was -867, after heating up the nozzle this value is 457 and while printing this value is usually around 4500 but reached up to 6000.

About the nozzle-distance, isn't a bit of distance between the nozzle and the bed normal? I thought a gap of ~5 pages of paper was the goal because otherwise the filament would get squished when extruded.

I've put the SenseOffset AutoStart on. However, how do I configure the Level value? When sending M3909 to the printer I get the following error: "M3909: INFO SensiblePressure is currently disabled."

That fan topic looks problematic. Would you recommend I cut of the power supply before it breaks anything? I considered replacing it anyways. Any recommendations? I'm currently looking at this:
https://www.thingiverse.com/thing:4023550
https://www.thingiverse.com/thing:3813963
https://www.123-3d.nl/123-3D-Ventilator ... 15051.html (or would 24V be better?)
This option would however only be viable once I get the extrusion working and can print with ABS.

Bild
Bild

Re: RF1000 Which doesn't want to print properly

Verfasst: So 10. Mai 2020, 12:21
von AtlonXP
Hello speckledpear,
you definitely print too quickly.

The Hot End V2 can usually not be that fast.
An E3D V6 can be faster.

Your P values should stay below 5000 digits.
5000 digits and more are harmful to the printed image.
Therefore try it with max. 30 mm / s to print only.
I would turn off your fans.
Lower the bed temperature as much as possible.

Here is an example of SenceOffset:
M3909 P5500 S300
See also:
wiki/index.php/SenseOffset

Here is an example of DigitFlowCompensation:
M3911 S4900 P5500 F-30 E-30
See also:
wiki/index.php/DigitFlowCompensation

These G code commands should be inserted in the start code.

To your fans:
If these are connected to 12V on board (X24), then this output is already broken!
The 24 V (X20) output can withstand this load.
viewtopic.php?p=25943#p25943


LG AtlonXP

Re: RF1000 Which doesn't want to print properly

Verfasst: So 10. Mai 2020, 12:23
von speckledpear
For getting the F-Digits I printed the same cube as I did yesterday. This one came out much nicer. I however barely changed anything, says something about the consistency? I've added the filament loading, removed and replaced the filament and turned SenseOffset Autostart on. I also removed the fan attachment as it wasn't cooling the printed part.
Pictures
Bild
Bild

Re: RF1000 Which doesn't want to print properly

Verfasst: So 10. Mai 2020, 12:33
von AtlonXP
speckledpear hat geschrieben:I also removed the fan attachment as it wasn't cooling the printed part.
But determines the nozzle of the V2 hot end tremendously!

LG AtlonXP

Re: RF1000 Which doesn't want to print properly

Verfasst: So 10. Mai 2020, 14:07
von Nibbels
Nice :D

That cube already looks better.

The M3909 is not needed if set the "Autostart: on". Autostart means you only use M3001 and with it the M3909 is activated. That is how I like it personally.
(You cannot use M3909 when M3001 did not successfully start or was not started. You can however use M3909 after M3001 within the startcode to set different limits for different filaments.)

The default way to tune in SenseOffset is to watch your hotends force value (F or P value in digits, 2000/3000/4000) when you print normal upper layers. That is your default hotends inner pressure value [for the given filament and speed and temperature].
In case your hotend shows some good value like 1500... 3000 then this is the lowest value you should set as SenseOffset upper force limit.

Rule of thumb:
Add +1000 or 20% to your default pressure then Senseoffset works well. And play with it to get more or less perfect first layers.

(If you have default 1500, set it to 1800 ... 2200
default 2000 -> limit 2500 to 3000
default 3000 -> limit it to 4000)

If you set a lower value than your default hotend inner pressure then your printer will space and space and space and space and it will not reach the goal pressure. So every first layer is crap.