Workhorse auto level failing

I need some help with my workhorse. I currently have to set the Z-offset to +1.15 in order to print. I noticed that during the auto bed leveling process the nozzle of the extruder doesn’t go right into the middle of the circular bed leveling plate that is in each corner of the bed. It misses the middle and touches the plate. Is this the reason I have to set my z-offset to such a huge value (+1.15) to get my workhorse to print? If so, does anyone know how to adjust the x-axis position so that it hits the leveling circular plate right smack in the middle?
Mini_Leveling_Washer (2)

That’s where it should hit. It should not hit the screw in the center.

Normally the Z-offset is a negative number… mine is currently set to -1.1 (yours may be slightly different … mine was previously a little more negative but I tweaked it a little because my printer was occasionally over-squishing more often than I like.)

You may want to check and verify your profile configuration is correct for your printer.

I scraped the heck out of the nozzle and it seems to be working now at -0.95 offset. I guess I had some filament stuck on the extruder which made it difficult to detect the cicular leveling plates. I guess I should take a metal polisher to the extruder every now and then since it relies on electric conduction for the auto leveling process. Something still seems off because -0.95 seems too high a value as well. Will continue to tinker with different z-offset levels. I have to adjust WHILE it’s printing the first layer for it to work almost every time.

I use a brass wire-brush … roughly the shape of a toothbrush (I got mine at Harbor Freight). I wait for the hot-end to read soften temp (e.g. for PLA that’s about 180°) and then give the nozzle a bit of a scrubbing (if it isn’t at soften temp then the brush wont clean off filament as easily).

Be careful when doing this because the wires that operate the hot-end are just on the side and you don’t want the brass to hit those wires. I am careful to just brush the nozzle (not the whole hot-end).

I’ve noticed that I never really have bed-leveling issues since I started doing that. I used to have problems even when the nozzle appeared to be free of debris. My theory is that a very thin layer will build up (think of the “seasoning” that builds up on cast-iron cookware) and this acts as an insulator. By using the wire-brush, you take that coating off so it gets better conductivity when the probes touch the washers.

Also… make sure none of the washers are loose. I had one sort-of-loose washer in the back right corner of my printer and noticed I was getting frequent errors… but always on that corner. I grabbed the hex-key and tested it and sure enough… it was loose. The washers don’t need to be tight (just snug) … but mine was pretty loose so it wasn’t conducting electricity well enough for the probe to work.

Very useful tips. Got my brass wire brush ordered on Amazon and will check all the washers to make sure they aren’t loose. I’ve had issues with this right out of the box so I still think my workhorse is defective and will never fully function. The very first print I attempted scraped my bed. I tried it again and it worked. It always seems to be hit or miss even when the nozzle was brand new. I hope these tips will help. I got the more expensive workhorse because I was frustrated with all the problems I had with a very cheap 3D printer. Sadly, this printer has taken more troubleshooting time than that one ever did.

1 Like

Possibly there is an issue … but keep in mind how bed leveling works.

The system has “end stops” for each axis… actually a pair of them (e.g. x-min vs. x-max, y-min vs y-max), etc.

Then the printer hits an end-stop, it hits a contact switch that completes a circuit – basically two wires touch so that a circuit is completed. The board detects the completed circuit and realizes the printer has reached the travel limit on that axis.

So the idea of the bed-leveling system is basically the “Z-Minimum” end-stop. When the metal hot-end nozzle touches the metal washer, a circuit is completed.

If you had a broken or disconnected wire, then the system would never detect the washer. If it sometimes detects it… but reliability is poor… it usually means something is acting as an insulator so that you only get a weak connection.

The most obvious place for this to happen… is debris build-up on the extruder nozzle. Even when it “looks” clean, there can be a micro-thin layer that reduces conductivity.

But there are some other areas where problems can creep into the system. I mentioned the washers being snugged down (but not so tight that you crack the glass bed). All washers ground to the plate bed… the plate bed has a single wire that connects back to the RAMBo controller (after going through a couple of wiring harnesses.) The harnesses are areas where you could (but probably don’t) have a weak connection. The more likely place fo the connection to be weak is if the bed-washers aren’t snugged.

The extruder nozzle could be “loose” (but probably isn’t) but it could also have a metallurgical reaction that, over time, reduces efficiency of power flow and starts to act as a resister. Similarly… the heat-sink fan on the print-head has one of the four screws with a single red wire…that’s the magic wire used to complete the circuit back to the board (again… going through a couple of wiring harness connectors).

One thing you can do is grab a multi-meter and test the electrical resistance through the system.

When I test my printer, I find that as I test the system, I tend to get about 1Ω resistance through the washers. Maybe a couple of ohms through the nozzle. As long as the number of ohms are small single-digit values, you probably wont have a problem. I’m not actually sure what the limit is here… other than saying that when my multi-meter shows resistance around 1Ω I never have a problem… even a couple of ohms seem to be ok, but as the resistance increases… the reliability of bed-leveling becomes a problem.

But when one of my washers was loose, I was measuring resistance greater than 10Ω at that washer (which is how I knew to clean it and re-snug it).

If you have a multi-meter, grab it and start working the connectivity checks back to the direction of the board to see if you can find a weak link somewhere.

For example… touch one probe to the nozzle, and another probe to that red-wire on the corner of your hot-end heat-sink and see what sort of resistance you read. When my nozzle is scrubbed clean, I measure 0.3Ω between the nozzle and the red-wire (front/bottom corner of the heat sink on the left side of the print head). When it’s dirty… sometimes it’s 5+Ω or more.

Before reading these suggestions I went ahead and ordered a new Hardened Steel Tool Head. I got frustrated because it seems to randomly keep scraping the bed or starting too high above the bed. There is no consistency, no matter what my Z-offset is set to. It’s like the accuracy is ridiculously low. I will check those washers to make sure they are snug, of course. If the new tool head doesn’t do the trick I will at least eliminate the tool head as the source of the problem.

1 Like

So I traced the problem to the bed being a different distance from the nozzle in different locations. The bed is closer to the nozzle towards the left and farther away from the nozzle towards the right. This was confirmed when I printed a big X and the left side was smudged and the right side was so high off the bed that it didn’t stick to the bed. I performed a movement x/y/z calibration and finding the z-offset all over again. Here’s hoping that this works! Oh, and cleaning the nozzle has really helped with more consistent leveling and enabled me to discover the bed is at a different distance from the nozzle in different spots. thanks for that

Printer is working better than ever but my big X still doesn’t print “perfect” at .15 mm. It gets close though, and it does span the entire bed. I don’t know if I’m setting the standard too high (it IS a $3,000 3d Printer though!). I’m going to try some more complicated prints that had failed in the past. I have my z-offset at -1.08. A new problem that emerged is that my filament would just stop extruding mid-print. I lowered the printing temperature and tightened the filament knob. My thinking was that there was some heat creep combined with a loose grip on the filament. The same print seems to be succeeding now. I ordered a replacement tool head that I probably didn’t need but will try my big X with that new tool head once I get it at 0.15mm again and see if it succeeds.

What type of filament are you printing and at what temp? Also… who makes it?

For example… one PLA is not necessarily as good as the next.

Also, I often tweak the settings for any given material to get the printer dialed in nicely.

For example… I print PolyMaker PolyLite PLA … and after doing a “temperature tower” and “retraction tower” test, I find that it tends to print best on a Workhorse at 230°C with a 55°C bed and usually use a travel speed of 40mm/sec (but usually slow down for the outer perimeter to around 30mm/sec).

The “cold end” fan (on the left side of the extruder) should be running anytime the printer is powered on (regardless of whether it is actively printer or heating the hot-end). If that fan is not running… you can get “heat creep” (there is nothing to control that fan in software … it simply powers on when you start the printer and runs continuously).

My filament tension knob is typically around 1/3rd to 1/2 of the travel.

For PLA, I typically set retractions to about 1mm (direct-drive printers don’t need high retraction amounts). Sometimes material gets chewed up if there are excessive retractions. If that happens, set the max retraction count to a low number (less than 10 … e.g I often use 5) and then set a retraction window to around 2-3mm. This means that if the job calls for retraction more often than 5x before feeding 3mm of filament… it will ignore extra retractions. This helps by not chewing up the side of the filament and creating a jam.

If your filament stops extruding, do a ‘change filament’ operation to pull out the filament and check the side of the filament to see if was getting chewed up. Cut off that chewed up section, then re-feed the filament.

I also typically print at .25mm layer height. Remember this is a .5mm nozzle so it’s not great for “fine” detail (although you can use a .15mm layer height for “fine” … you still have a wide nozzle). There is an option “SL” nozzle (.25mm nozzle) which is great for fine detailed prints. For this reason, I seldom use the .15mm layer height. For the face shields I’m printing for medical staff… I use .35mm layer heights because it lets me get through the print jobs faster. (This is where the HS or HS+ nozzles (.8mm and 1.2mm nozzles) would print even faster yet. (By “faster” I mean fewer layers because each layer is thicker).

If you have not already done so… go into the “Move” menu (on the printer control panel) and run the “Auto Calibrate”. This will utilize the calibration cube (that small cube on the front-right corner of your build plate) to work out backlash and other settings. (BTW, doing this does not necessarily enable the backlash compensation… there’s a g-code that enables that and I’ll try to follow up with those instructions).

Also… I do have a tutorial that walks through how to enable “Linear Advance”.

When things like Linear Advance, backlash compensation, etc. are all dialed in… the printer does amazing work (well… given decent filament. Not all filaments are equal. And several types of filaments will absorb moisture from the air and that moisture “boils” as it goes through the hot end … producing really bad looking print quality.)

I’m using Gizmodorks ABS filament ( Do you know of any reputable source to know if a filament is of high quality or not?

I’ll redouble efforts to use filament recommended in the Cura Lulzbot profiles but many seemed like they were no longer on the market. I wanted something tougher than PolyLite PLA as things like screws and nuts were breaking, although I still have that problem with my ABS filament. I tried printing ( and the screw snapped when I put in the nut.

Next time I load up Polymaker PolyLite PLA I will use your recommended settings. The cold end fan is always on. My filament tension knob is just a touch past 1/2 of the travel. I will play with the max retraction settings. I will also remove the chewed up portion from now on whenever it stops extruding, which makes sense in hindsight.

I am ordering the .25mm nozzle for those fine prints. I recently ran the calibration you’re describing in the Move Menu and it’s when it started printing significantly better. The left side of the bed is still a tad closer to the nozzle I think but it still seems to be printing more complicated prints than ever before and my z-offset is now at -1.08.

It seems to be printing better than ever, but I was expecting to have fewer issues when I shelled out this much cash for the Workhorse. I value more fun printing and less learning about all then nuances of what makes a print fail. I suppose 3D printing is still in its infancy. The whole process needs to get easier if 3D printers are ever to become as commonplace as 2D printers. I was just naive in that I expected more money to translate into less troubleshooting/things-to-learn. Either way, now that my prints are working better I’m enjoying the process a lot more.


Like anything… the printer is “just a machine” or a tool. I uses this analogy all the time in photography when I hear “this camera doesn’t take good pictures”. Here’s the story:

A world-famous photographer is invited to a dinner party at the home of a word-famous chef. As the photographer arrives, the chef greets him and says “Your photographs are fantastic. You must own an amazing camera.” The photographer thanks him, but says little else on the topic. After dinner is over, the photographer exclaims to the chef “Your dinner was incredible. You must have an amazing stove.”

Moral of the story… it’s not the tool… it’s you. No matter how good the musical instrument, if you hear a wrong note, it was the musician… not the instrument.

With that aside… can you get a clunky 3D printer? Sure you can. There are loads of them. The LulzBot printers are well thought-out designs and highly durable. There are lots of features I notice about them when I compare them to other printers … the rugged nature of the tower, the dual belt-drives on the z-axis, the double rods on the y-axis table, how the cables are all routed and secured to avoid strain or wear over time. The print-head is 3-screws and 1 wire-harness to detach the entire thing from the X-axis carriage (making it really easy to swap from head to head for different print applications), the fact that it uses direct-drive instead of Bowden drive and the all-metal hot-end of the AeroStruder which let you print with hotter materials and you don’t need a special head to deal with flexible like NinjaFlex, etc.

When I started 3D printing, I made a lot of mistakes. But I followed a lot of articles, and 3D printing experts online and, bit by bit, starting learning how to tweak things. Today, I’ve got my printers dialed in pretty nicely and parts come off looking pretty clean. Every so often I get a part that isn’t a keeper and have to figure out what went wrong. But I get a lot more keepers these days than I used to … even though it’s the same printer. Mostly because over time, I’ve learned which settings make a difference. You will get there and it probably wont take too long. There’s a huge community of folks who are happy to share what they’ve learned.

The Printer

As for LulzBot printers vs. other machines…

I have friends with other 3D printers and they can’t use materials that need more than 250°C to print (they don’t have an all-metal hot-end). They can’t easily print with TPU (e.g. NinjaFlex) because they use Bowden-drive extruders instead of direct-drive like the LulzBot (if they use TPU … it’s a mess for them.)

There’s a lot more to point out … but I digress a bit… so back to the topic.

One last point (before getting back) is that the machine is basically a robot that does what it’s told. It’s the filaments that have “personality”. You’ve got a hard plastic material … but you heat it until it turns into goo. In that goo state… it has non-linear properties that compress, expand, stretch, etc. and this happens with a bit of a delayed reaction. When you “retract” it doesn’t just back up the goo… it relieves pressure. TPU is VERY stringy. PETG is mildly stringy. PLA is less stringy. So you learn to use settings to deal with these “personalities” of the filament and that’s where your experience will factor in. The printer will just do as it’s told.


ABS is a slightly more difficult material to work with (it’s considered “intermediate” difficulty). The issues with ABS are that it is highly prone to “warping” and is very sensitive to environmental temperature and drafts. For this reason, it is recommended that anyone using ABS should be using a build enclosure.

E.g. here’s the one on the LulzBot store (by PrintedSolid):

This both helps block any drafts … but also helps to reduce how quickly the heat is lost (there are open areas so it’s not sealed up like an oven … but it blocks most of the sides so heat is lost a little more slowly).

Warping occurs because nearly all filaments (probably all filaments) shrink as they cool. The bottom is heated via the build plate … but the top isn’t. This means as the layers cool, the layers at the top can cool more … causing them to contract more. This pulls them in and causes the part to warp … pulling the corners off the build plate.

One question would be: Why ABS?

Once upon a time, the two common materials were PLA and ABS. PLA is VERY easy to use. It’s a “beginner” material. The downside is it is a bit brittle so parts have less impact resistance (they can crack). ABS is a bit better with impact resistance. With that aside… depending on what you want to do with the part, even ABS may not be strong enough.

Another downside to ABS is that the gasses it gives off while printing are not healthy to breathe. Some people (who live in suitable environments) will put the printer in their garage and not their house so they don’t have to worry about the family breathing the gasses. Some people modify the enclosures to vent them to the outside, etc.


Other materials have since come on the market that are more popular. While there’s a list of them, the MOST popular of these is PETG.

ABS was popular because it was the only alternative once upon a time. Then some co-polymers came out as alternatives to ABS … but the earlier alternatives were expensive. PETG did not yet exist.

PETG came along and has many of the properties of ABS in terms of strength and impact resistance (it isn’t brittle) BUT… it is a much easier material to work with … I’d say it’s as easy as PLA but that’s not quite true… although it is pretty easy to use. It doesn’t need a build enclosure.

I follow a lot of 3D printing YouTube channels and among that crowd, the general consensus seems to be that there’s probably not much reason for people to be using ABS anymore (certainly not as much reason as their used to be). If you have an application that needs ABS… sure, go ahead and do it. PETG seems to be taking over.
The nuances to know about PETG are that you really need to use a release agent on the build plate (PVA such as “Elmer’s Glue Stick” works well) … there are stories of it sticking so well that it can damage the glass (there are stories of flakes of glass pull away when trying to release it if you didn’t use a release agent.) I use the glass side of my build plate and use the PVA stick (Elmer’s Glue Stick … I get the “purple” stuff that turns clear when it tries because the purple stuff is easier to see how much you’ve applied. The white stuff goes on nearly invisible and I’m never quite sure if I missed a spot.) PVA glue dissolves completely in water (That plastic-like material used in dishwasher pods … that dissolves in water … that’s PVA).

I set the Z-offset to get the nozzle a little higher when using PETG. Most materials need the first layer to have a tiny amount of “squish” on the build plate. PETG shouldn’t be “squished” on because it sticks to everything … including the nozzle. If you over-squish it then it will start to cling to the nozzle and make a mess. Instead the technique I was taught is to “lay” it on the build plate… don’t “squish” it onto the build plate.

I use IC3D brand PETG. There is a Cura LulzBot Edition profile for this filament. I tweak it just a little. I print at 260°C and use a 75°C bed. It’s also sensitive to print speed… I print at 40mm/sec (but I think that’s the default). If it prints too slow it can goop… if it prints too fast it will be stringy. So there’s a happy zone in the middle and that seems to be around 35-40mm/sec.

BTW, currently PETG is a little more difficult to get becuase with this COVID-19 stuff going around, there are a lot of people in the 3D printing community who are printing face-shields for medical staff (myself included) and their preferred material is PETG. So much so… that filament makers are running out of the stuff.

The good news for you is that you own a printer that uses 3mm filament. The vast majority of 3D printers on the market use 1.75mm filament. This means that mostly everyone is out of 1.75mm PETG … but you can often get 3mm.

I recently spoke to the folks at IC3D and they told me that they have 3mm in stock … as long as you aren’t fussy about the color (I told them I don’t think the medical staff care so much … they just need the protection of the face-shield.)

Selecting Filament

I suggest that whenever trying a new kind of filament, you stick with the filaments you find in the Cura LulzBot Edition filament profiles list. These are filaments that have been tested on your printer.

You’d think that one PLA is as good as the next, and one ABS is as good as the next, and so on. But the reality is that the filament manufacturers buy the raw material then introduce additives to create color or other properties and this changes them. Also there are issues such as: How well was it dried before making the material? Was it properly vacuumed to release gas bubbles? How accurate is the filament diameter? Basically there are property and quality control differences from vendor to vendor. In the community you can find people who love one vendors material X, but hate another vendors very same material X … because of quality differences in the material.

If you start with a material (including the brand) from the Cura LE menu, then you’re at a good starting point and you can see how that material should behave. I went off the board a few times and regretted it. But now I’ve learned how to make a “temperature tower” and “retraction tower” and other calibration tests that help me dial in the settings for specific materials if I want to venture out trying a new brand where no profile exists in Cura LE.


Most filaments will eventually absorb moisture from the air. Certain filaments are notoriously bad (Nylon has a particularly bad reputation for this and can absorb moisture in a matter of a few hours). If moisture gets into the filament, it heats and boils while printing and the off-gassing of the steam destroys the texture as it prints and you get awful looking parts. Keep your filaments dry (in sealed bags with desiccant). Desiccant wont “dry” filament once it gets moisture… it just tries to prevent the filament from getting wet. I eventually bought a filament dryer (basically a food dehydrator for filament). I mention it because it’s a reason why a printer could produce bad quality even though all your settings are otherwise perfect.

Also, don’t assume that “new” filament is “dry”. My friends and I have all experienced brand new filament that started printing poorly… toss the spool in the dryer for 6 hours and try again… and then it prints nicely.

Stay healthy & safe!

I want you to know all of your advice is taken to heart. You obviously have figured out this stuff more than 99.9% of people and thank you for dispensing your wisdom in this very fun hobby. I’ll be an aged Daniel Son to your Mr. Miagi any day, just don’t make me wax your collection of cars.

I do want to say one more thing here. I work as a computer scientist. Much of my job involves looking for things to automate. If a human is doing some tedious thing that a computer could be doing it then that’s where I come in. It sounds like there are things here that the software can be doing to further automate troubleshooting and I hope developers take it to that level. All of that knowledge you have through undergoing rigorous trial & error sessions with your 3D printer(s) should be coded into Lulzbot’s version of Cura, at least the parts that lend themselves to automation. Cameras have “auto” settings for ISO, focus, and a bunch of other traits to get “decent” pictures. These settings are never going to match what an expert can do with a camera, but they do dynamically change based on the picture being taken. If settings in Cura could change based on the 3D object being printed to maximize the chance of a successful print, I’d be pretty impressed.

When the auto levelling is taking place on my Mini I listen for the little ‘ticks’ at each corner of the bed, if I hear more than 3 ‘ticks’ I abort the print and clean the nozzle. Usually there is some plastic stuck on the backsiide of the nozzle. I printed myself a little cleaning stick which is about 1" by 6" by ¼" with a very rough print surface which I use like a file to clean the nozzle and it works great for me.

I had same issue had to do recalibrate and it took care of the issue.

Check out this post, as well as the entire thread:

May not fix all of your problems, but it may help a few of them.

Before I retired I was a control systems designer and I agree with you that many of the things that people struggle with their 3D printers should have been designed out and taken care of before they ever hit the market. It’s like the manufacturers and many of the users prefer to feel like they are the still in the ‘wild west’ of technology. Most of the technology required in 3D printers has been in use in industry for 50 years or more so people really shouldn’t think they they are on the ‘cutting edge’ of technology.