Cant get prints to correct size

So I have had a TAZ 5 for about a month now and love everything about the printer except when it comes to fitting pieces together. I have spent hours looking through forums and you tube for videos but I am hoping someone out there can help me.

Every thing I print the interlocks is so tight and never fits. Never Never Never. So I made a basic test cube in tinkercad. 30mmL x30mmD x20mmH I also made a 20x20x20 hollow cube and centered it inside. This way I could measure outside X & Y and inside tolerances. What I found was my printer is printing bigger then 30mm on the x & y by about .3mm and I was also printing to much material on the inside. my print is coming out as follows

30.32x30.28x20.02mm and the inside is measuring 19.81mm x 19.83mm x 20.002mm.

After this I did some research and looked into setting Estep properly. I adjusted this to 853.1mm in the firmware which is as close to perfect as I can get when dispensing 100mm of filament. Reprinted the test cube and now it is closer but still not close enough. I can post the exact measurements this afternoon but I don’t know what I’m doing wrong and how to fix it. I have been trying to print this for my 6 year old

I had a buddy print it and the pieces snap together and twist lock no problem on his ultimaker. I for the life of me can not get these things to print and fit. I break them every time by trying to force them into each other. How can I as a noob calibrate this printer to print a 20x20x20 cube? Do I need to calibrate x/y? I assume those are set correctly. I have never changed anything besides the Esteps.

If your e-steps are realy calibrated very well (853 sounds very high to me, but might be due to different hobbed bolt), then there is basicaly only one variable left: do you measured your filament diameter with a caliper on at minimum 3 different points?

You can also post a close up picture of the top surface of your cubes, it should be possible to see if you are over extruding.

Edit: At which speed do you run your extruder for the 100mm test?

So the default was 858 and I lowered it to 853.1 due to over extruding. I don’t know what the default speed in CURA was and don’t see where it says but its whatever the default speed is from the Lulzbot Cura normal profile.

Next I reprinted again this time with same thing and a 20mm cube on the highest quality setting lulzbot cura profile.

So my finding were as followed. X axis was 30.33mm on the outside, Y was 30.26mm, and Z was 20.08. Inside hole was X 19.92mm, Y 19.87, Z was 20.09mm.

A image of the top layers attached.

Could be slight over-extrusion. Try setting the flow % to 95, keep the filament diameter at 2.85.

Another way to gauge the flow is checking for roughness on the top layer where the infill overlaps with the perimeter.

This may need to be adjusted based on the color of the filament… for instance eSUN ABS Black needs to be reduced by 1-2% compared to White.

I was wondering what the dial on the front of the LCD screen did. I know it says FR 100% and it lowers it it or raises it but that seems to increase the overall speed of everything. Is that the flow rate or do I need to do it in CURA under flow percentage? Thank you for the help.

You can change the flow rate in Cura or using the LCD controller with the tune menu during the print. Changes made with the LCD controller will be lost when you turn the printer off unless you save settings.

Yep… what he said. :slight_smile:

Try it in Cura… its under the expert (full) settings. If switching from quickprint to full, make sure to transfer the profile.

So I changed the filament size to 2.89 which was the average and printed like 5 different test cubes. It is most accurate at 85% flow rate. It’s almost perfect outside and inside dimensions. It seems like that is low doesn’t it? I’m going to try printing some pieces that interlock and see how they turn out. Thanks for the help so far I’ll post a picture tomorrow when it’s done printing.

So this is the cubes I printed last night at 85% flow rate which are as close to perfect as it is going to get for dimensional purposes. I tried to take close ups so you could see layer detail and surface detail.

So last night before bed I printed a few pieces for this light saber my kid wants and BAM fits great now. There were some small imperfections in the print but that could be from multiple things not necessarily the flow rate. What are your thoughts about the flow rate being set that low? I guess i’m just surprised at how low the flow rate is. Does the layer and surface look OK to you guys?

Hard to say on the pictures… If there are no tiny gaps between the lines on a top surface, you are not under extruding. Watch at the parts with a bright light source at different angles is the best you can do.

When you do such cube fit tests, don’t forget the imperfect cube edges. It would be better to print the inner cube with a small radius on the 4 edges to eleminate this effect. You might recognise that you can increase your flow rate a little bit again with this…

The thing to keep in mind about the flow rates is print speed and adhesion to the previous layer.

Slow down the print in general for better precision. That should cure the blister effect on the outer most perimeter. A dremel is the surefire remedy. :slight_smile:

To increase adhesion, I extrude a degree or two higher… and if using ABS minimize cooling from the fan. In some cases, you’ll know you’re printing too hot if vertical circles look like ovals.

Great thread Josh. I’ve had my Taz 5 since May 2015 and have struggled getting a firm handle on the extrusion rate as it relates to getting perfectly fitting printed objects as well. I also printed the same light saber you mentioned and it’s one of the better prints i’ve done. You can see some pics here.

As it relates to getting perfectly fitting prints, you’re findings are very interesting that you’ve had to go as low as 85% on the ‘Flow %’. I’ve never tried that low as i always assumed this would adjust the flow rate directly and I would never have imagined that 85% would extrude enough. Moreover, all the reading i’ve encountered suggests that if you get your e-steps right (which i’ve spent a lot of time doing, like you) then the ‘flow %’ shouldn’t need to be adjusted in theory.

Here’s the snippet out of the Cura manual relate do the ‘Flow %’.

Filament Flow %
This controls how much filament your printer is extruding in relation to
speed. This setting is mainly used to adjust for filament density variations.
Leave this value at 100% as changing it can lead to surface quality issues.

Now i’m completely confused as I’m not really sure what this means. …extruding in relation to speed??? and filament density variations?? And if the suggestion is to leave it at 100%, why is it a feature at all… Still scratching my head on this one. What do you make of it?

Thanks again

I’m no expert, but I’ll give you my take on your questions.

You’re on the right track, but probably need to take a step back and stop over-analyzing. :slight_smile:

There are a lot of ways to adjust the proper amount of filament deposited during the extrusion process. Proper e-steps is the basis, so make sure when requesting 10mm, the stepper motor advances/feeds 10mm of filament. Make sure that you have a good average for filament diameter. The flow rate falls in line with adjusting the proper amount of deposited filament, but in a much quicker fashion.

Filaments come in various formulas but in general still fall into the general family of ABS, PLA, PET…etc. Again, I’m not an expert on filaments, but I suspect different brands use more or less filler which could dissipated during the extrusion process. So the flow rate allows adapting to these differences quickly without wasting a quarter of the spool figuring out the new esteps… only to do it again for the next spool.

I believe there is some relationship between flow rates and speed. The slicing software should be smart enough to make compensate, but in some case you may want to increase the flow rate when speeding up prints.

Regardless, the easiest way to gauge proper flow rate is visually inspecting the top layer. Measuring and fitting is good too, but generally if your top layer is a bit rough where the layer infill overlaps with the perimeter… best seen when using a rectilinear pattern for the top solid layer infill.

If your machine, filament… everything is dialed in correctly, leave flow rate at 100%. For any other circumstance, vary flow rate to suit the project or your liking. :slight_smile:

Hope that helps… and again, I’m no expert… so maybe others can chime in.

When I did the Esteps calibration I ended up with a number like 860 or so. But even with carefully measuring the filament diameter, I find that the calibrated Esteps number is too high and way over extrudes in practice on the Taz. I had to set it to 830-ish.

Instead of the Esteps calibration, If I need to recalibrate (changed stepper or different extruder), I run a print of a thin part and adjust Esteps until I get a good top surface finish without any overextrusion or underextrusion. Of course making sure the input filament diameter is exactly correct and all the other relevant slicer settings are perfect, and 100% flow.

After doing this I get much better parts out of the TAZ. For whatever reason the method for calibrating esteps is flawed. The printer overextrudes with the calibrated number, even when all the other setttings are correct (nozzle dia, filament diameter, etc.). If the calibration is “correct” the flow percentage should be set to 100% for perfect parts. Anyway this is how my machine behaves. And I have tried different extruders. They always over extrude after calibration.

Perhaps the nozzles that are supplied with the Taz are out of spec? That would be one explanation. I have no idea.

I’m glad to hear I’m not the only one who gets over-extrusion with a carefully set e steps value (and carefully measured filament diameter w/flow rate set at 100%).

I tested and set the e steps on my mini and got 869. That seemed higher than what others were getting, so I checked it two more times at 869 and kept getting 100mm when measured with my digital calipers. It does seem that it is over-extruding a bit. When I print a 20mm X 20mm X 5mm test cube, I get 20.25 in the X dimension, 20.13 in the Y, and 4.95 on the Z. (Part is fully cooled off - was printed 2 days ago). I keep thinking if I have e steps and filament diameter set as close to exact as I can measure, I should be able to leave flow rate at 100% 9or very close to it, and only vary that for fine tuning).

I’m also getting a wider first layer, which leaves a bit of brim around the bottom edge, so it measures 20.55 in the X dimension and 20.51 in the Y. So I guess I’ve got some first layer work to do yet as well.

I get the same thing on a Mini. The factory esteps value was 833. I did the calibration and got 1096. I triple checked with the new value and I get 100mm extruded. I checked with a steel ruler and with digital calipers as I didn’t believe the number, but it’s correct. With both values I get a lot of overextrusion. I’ve been adjusting the flow%, sometimes as low as 60%. The ideal amount varies a bit with the part I’m printing, sometimes it seems to need a little more or less. I find I get acceptable prints within 10% or so of the ideal value for the part. I generally only tune further if the surface finish is important to me or I need more than one anyway.

I get the same results in Cura and Simplify3D, so I don’t think it’s a slicer problem. Though it’s possible they both have the same problem I suppose. I have considered adjusting esteps to get a good base value as billyd says he does, but I wasn’t sure if that would cause other problems. I also thought that the printer should be doing what it’s told. If commanded to extrude 100mm, it should extrude 100mm.

I have similar results from a Flexystruder, so I’m a bit at a loss. The Flexystruder comes with a 0.6mm nozzle and I did change the slicer configuration to adjust for that. The stock extruder has a 0.5mm.

The first layer is nearly always a bit wide by design to help with adhesion. Most slicers have a setting to adjust that though. I find with that PETG and HIPS I get sufficient adhesion and don’t need help, so I turn it down to get more accurate first layer dimensions.

Are you sure your idler tension was set correctly? In one of my early attempts, I got a very high number. I tried again and noticed that the gear was turning, but the filament was not moving smoothly.

Which reminds me: I printed off the idler tension jig and used that to set the tension. If I slip it between the two washers on the idler tension springs, the tension seems low (that’s where I was getting the slipping). If I positioned the jig to miss the washers and pinched it between the plastic of the idler arm and the frame, it seemed to work better. (I also bumped my extrusion temp up 5˚C).

So which is the proper way to use the tension jig? between the silver washers around the springs, or between the black plastic parts?

For those having problems with the estep-calibration: At which extrusion speed are you measuring the 100mm? You have to keep it in a reasonable low area, or you measure other problems and not only esteps…

Yes, the tension is set properly, at least at a reasonable level. The filament moves throughout the whole extrusion run. I didn’t use the jig, but my calipers measured 5mm between the washers.

Speed is another issue. I’m not 100% sure what I was using there, it took maybe 10 seconds for the extrusion to run. I’ll try setting the speed to a low value to make sure that’s not contributing. I don’t remember seeing a speed mentioned in the lulzbot guide. At worst, I can type in a gcode command. Is there a recommended speed setting to use?

10s for 100mm would be 600mm/min, thats way to much! The best speed would be an average print (extrusion) speed. That can be calculated with simple math. Let’s say we want to print 0.2mm layer height with the TAZ5 default 0.5mm nozzle. That means your line with should be around 0.6mm. Final, the print speed: Infill is usualy printed faster then perimeters, lets assume 40mm/s as an average value.
We know that an average 3mm filament has around 2.9mm in diameter, thats 6.6mm². So 4.8/6.6 = 0.73mm/s for the extruder, or about 44mm/min.

I recommend you to do the comparisson on yourself: Make the measurement at 20mm/min (be patient), 50mm/min and 100min/min. There will be a nice difference :wink: