Replacement hot end on HS 0.8mm tool head

General Hardware
1

I have the HS 0.8 tool head on a TAZ 5 (with the adapter plate), but I’m thinking about changing the hot end.

The reasons are: 1. The 0.8mm nozzle isn’t as useful as I’d hoped. 2. The nozzle temperature behaves erratically. I’ve gotten “thermal runaway” errors, but it also seems to run way too hot at times, boiling the filament. I can manually tune the temperature but I still get inconsistent results with it.

Anyways, I was thinking that I could make some use of the tool head if I replaced the hot end, but I’m not sure what a good pick would be. I’m looking for:

0.4 mm nozzle
for 3mm filament
wide range of materials including TPU and nylon/carbon
something that’s an easy swap

The one I was looking at is CLEARANCE - E3D All-metal v6 HotEnd Fully Assembled 3.00mm Direct (24v) | MatterHackers

Thanks in advance

2

The HS 0.8 and HS+ 1.2 nozzles both use the Volcano hot-end to improve the rate at which they can heat filament. The HS 0.8 nozzle has an area roughly 2.5x larger than a standard 0.5mm nozzle … and that means it can move material through 2.5x faster.

As for erratic behavior, have you PID-tuned the printer for the hot-end?

See: Two easy fixes for 3D printer temperature swings - YouTube

In the above video, he uses OctoPrint to access the terminal, but you can do this in Cura LulzBot edition by going to the “Monitor” screen, then click on “Console” to access a terminal interface that will let you send commands directly to the printer (your computer must be connected to the printer via USB).

You’d send it a command like this:

M303 C5 S230 U1

Where:

M303 is the Merlin G-code command to initiate PID tuning
C5 is the number of cycles you want it to run (5 is the recommended default)
S230 is the TEMPERATURE that you normally use when printing (change this to whatever temperature you usually use when printing)
U1 is the flag to indicate that you want to send the results to the board.

Then:

M500

This saves the results to eeprom so that they will be saved even after powering off. If you don’t do this, the printer will revert to previous behavior after you power it off.

That may help you have much more stable temperature when using the head and save you from replacing it. Anytime you change a head, you may want to redo this procedure.

2 Likes
3

Thanks Tim, this is useful to me too.

4

Thanks for the quick reply.

I have not PID-tuned. That does make some sense though because when I swap back to the original toolhead everything’s fine.

I will give that a try (next time I swap tool heads again) and post the results here.

5

I got around to trying this @TheVirtualTim

I have an enclosure, and I noticed the pid values change based on ambient. I did not get the ambient but I know from experience it varies based on the bed temperature.

From cold

Recv: #define DEFAULT_Kp 21.93
Recv: #define DEFAULT_Ki 2.06
Recv: #define DEFAULT_Kd 58.39

From bed at 107.9

Recv: #define DEFAULT_Kp 21.38
Recv: #define DEFAULT_Ki 1.96
Recv: #define DEFAULT_Kd 58.25

with bed at 90.9

Recv: #define DEFAULT_Kp 16.53
Recv: #define DEFAULT_Ki 1.51
Recv: #define DEFAULT_Kd 45.37

So what do you recommend for a procedure? Get the enclosure up to its temperature when printing before doing the M303 commands?

6

It is ideally set from the circumstances closest to what you use when printing.

In thermodynamics, you can lose heat through conduction or convection or radiation (or a combination of any or all of them). Heat will conduct from the heat block up the heat-brake to the cold-end of the extruder, but the extruder’s cooling fan will use convection to dump that heat and keep the cold end cool. That will be less efficient at dumping heat when the enclosure is being used and the bed is hot because the fan only has access to the warm air inside the enclosure rather than the cooler air outside the enclosure.

7

Ok, so I put the HS 0.8 toolhead back on, did a firmware update, then the PID autotune with 8 cycles. Unfortunately, there’s no noticeable change.

I’m using NinjaFlex Armadillo filament, and when I start my print at 220C the extrusion bubbles and makes a sizzling sound. If I tune the nozzle temp down to about 175C the bubbling subsides, and at 170C it prints pretty normal. As the print continues though, it seems I need to raise the temperature more. Basically, I have to keep adjusting the nozzle temp during the print to keep it about 10 degrees below where it bubbles – my only reliable indicator of actual temperature.

So this brings me back to my original post, what to do? I haven’t seen a known issue that fits, and I’m thinking a hot end replacement is the easiest option. I see some online in the $70 range (much less than a $400 toolhead!), but unsure what will be an easy/compatible swap.

Thanks again

Update: had another test going, manually tuning the temperature and flow rate, and received “E1 Heating failed” error.

Update 2: Repeated last test, got E1 error again.

8

I have good news and bad news.

First the good news: As soon as you mentioned “sizzle” I immediately know what the problem is. There is but one explanation for sizzling filaments.

Now the bad news: You need to dry the filament and that can be … tricky.

Sizzling means you need to dry the filament. It has absorbed moisture out of the air. It sizzles and pops because moisture is absorbed into the filament. Water boils and converts to gas at 100°C … and you print at much hotter temps. So when the filament moves into the hot-end, the moisture immediately “boils” … and that’s the sizzling and popping sound that you hear (the boiled water converted to gas popping out of the filament as it is extruded.)

Storing your filament in sealed bags or containers with desiccant packs will reduce how much moisture it can absorb… but desiccant packs can’t “dry” filament that has already absorbed moisture. You have to dry it using a filament drying or oven.

This gets tricky.

Unfortunately I was not able to find drying settings for Armadillo. A LOT of filaments can be dried at 65°C … for various lengths – anywhere from 4 hours to 12 hours. If the filament is dried too hot then it can melt together on the spool and be ruined. But often you can dry it at a cooler temp … but for many many more hours to compensate.

You can try to contact NinjaTek for recommendations. I’ve used NinaFlex and Cheetah … just never Armadillo. Most TPU can safely be dried at 55°C. While I love the “quality” of the filaments … as a vendor, NinjaTek is not good about providing care instructions. The “better” filament vendors include information on how to properly dry the moisture out of their filaments.

When you don’t have proper drying instructions … you can sort of hedge your bets by using a lower temperature (just to be safe) but keeping it in the oven for many more hours (e.g. more like 12 hours rather than 4-6 hours).

To make things more tricky … it’s hard to control the temperature in a convention oven. Conventional ovens can have temps swing by as much as 20° as it cycles on and off. This means the filament can be much hotter than it can handle.

I destroyed several spools of expensive filament trying to dry them in my kitchen oven. Consider yourself warned.

I bought a couple of filament dryers and was disappointed because while the heater at the base of the dryer warmed up, the chamber isn’t insulated. When I put a thermometer in the chamber … it was nowhere even remotely close to the temps needed to dry it (and it was at the max setting possible). I wasn’t happy with either of these dryers designed to dry 3D printing filaments.

Then (this is going to sound like a sales pitch) I watched (of all things) America’s Test Kitchen talking about the quality of toaster ovens. They confirmed that ovens have massive temperature swings that can ruin food. But mentioned their “winning” oven made by Breville … which has a PID sensor that keeps the temp within 1°C of the temperature you set.

I immediately bought one … one of the smartest things I’ve ever done. The oven has dried countless rolls of filament and print quality is MUCH better. I was NOT disappointed. I should further caution… the model I bought has a “dehydrate” mode … this allows the oven to operate at lower minimum temps than most toaster ovens. CHECK the MINIMUM temp of any oven you consider buying.

I basically never print with a new spool that hasn’t been through that oven.

2 Likes
9

@TheVirtualTim

I appreciate your detailed response, but I think it’s the plastic itself boiling, not the moisture within it. This toolhead has been nothing but trouble, with filaments that work fine on the other toolhead.

The combination temperatures being way off, Thermal runaway errors, and now the E1 Heating failed leads me to believe there’s a fault in the nozzle heating element or the temperature sensor.

Again, any thoughts on the hot end replacement?

10

I ABSOLUTELY ASSURE you … if you hear sizzling noises … you’ve got “wet” filament. Hence my assurance that “there is but one explanation for sizzling”. This is based on experience from years of printing. Plastic melts … it doesn’t “sizzle”.

1 Like
11

Tim, let me start by saying I really appreciate the time and thought you’ve put into this.

The thing that doesn’t make sense though is that I can swap the tool head or switch to a different printer and have no issues with the same exact spool of filament.

It’s really only this one toolhead giving me trouble in this way. I can go back to one of my Lulzbot Minis or the standard Taz 5 toolhead, and the only issue is that the flexible filament jams up from time to time. No bubbling or sizzling.

And I’m getting error messages from Merlin about temperature control.

I think we need to consider a hardware failure in the hot end

12

Could the “sizzling” actually be “grinding” noises? When filament “sizzles” you can usually see tiny bits of steam exiting the nozzle and also the texture of the filament will often be rough with signs of bubbles in it.

Thermal run-away is triggered when the board knows it’s been calling for heat for too long … and yet the thermistor says the nozzle isn’t actually hot. Logic on the board recognizes that it shouldn’t be possible to be heating the hot-end for that amount of time and still not have reached operating temperature. The error shuts down heat to the hot-end so that it doesn’t result in a fire. (A real thing with printers that lacked thermal-runaway detection.)

This can be one of two things… the thermistor is correct and the nozzle isn’t heating (so not a real run-away but the head still needs service) or the heater cartridge in the hot-end is working fine … but the thermistor needs to be replaced (so the head is actually much hotter than the thermistor reports).

If the hot-end is not reaching the correct temperature then it may be struggling to force filament through that isn’t quite warm enough … and that could result in grinding. I’ve never seen filament get so hot that it boils unless it has water inside.

The heat-block has a heater cartridge inserted into it (a cylinder slide into it with two thicker wires come out the side … be careful not to bend them. They are usually solid-core wire – not braid. Excessive bending can brake a wire. The thermistor is also a cylinder slide into the side of the heat block – but this one is a little smaller.

Check with LulzBot and they should be able to give you the correct part numbers for these. It might be these:

LulzBot makes an adapter that lets you mount the new toolheads to TAZ 5/6 printers. It gives you the correct mounting bolt pattern. The old toolheads no longer appear to be offered.

LulzBot might be able to service your toolhead … but check with them. Since they no longer build them, they might not offer the repair service.

There is also IT-Works 3D … and they do both build and sell toolheads and parts for LulzBot TAZ 5 printers.

As for the hot-end at MatterHackers … that’s just the hot-end (not a full print-head) – and it looks like it’s a V6 hot-end – not the Hexagon hot-end that is likely in your current print-head. This means it wont be a drop-in replacement for what you currently have. I’m not sure which hot-end is used on your 0.8 print head (on my TAZ Workhorse the default 0.5mm head is a Titan Aerostruder with a V6 hot-end, but in the 0.8 tool-head it’s an Aerostruder but with a bigger/stronger motor and the hot-end is an E3D Volcano rather than the E3D V6. I’m not sure which hot-end is used on your 0.8mm tool-head. The point of the Volcano is that it has a bigger/longer heat-block so it can heat filament faster (filament is moving roughly 2.5x faster through a 0.8 toolhead). You can find the documentation for building the default 0.5mm print-head here:

https://ohai.lulzbot.com/project/2b36d92e-a984-4e61-af92-245ea7fcaeaf/a16b1f2d-5c1f-47fe-a83d-9d68cca5d62c/

When swapping to a completely different design … be aware that the printer firmware has an idea of where the extruder nozzle is located relative to the x-axis carriage. If a new tool-head has the nozzle lower or in a different position (relative to x/y axis) from a factory head, then it can create problems unless adjustments are made to compensate.

1 Like
13

Sorry to jump in @TheVirtualTim , but the Breville ovens have quite a range of prices.

Would the BOV845BSS model be overkill for dehydrating filament.

14

Breville’s website lists two different models… the BOV800XL, and BOV845BSS as having a temperature range of 120-450°F … and 120°F works out to 48.9°C (so my guess is it’s really a minimum of 50°C). That should be cool enough to safely dry most filaments.

Their top of the line ($400 USD) Smart Oven Air Fryer Pro BOV900BSSUSC includes a Proof setting as well as a Dehydrate setting that the other models lack … and this lets you set the temp as low 86°F/30°C.

Most filaments dry at temps of about 65°C or warmer. Some TPU’s dry at 55°C … so you’d still be safe.

Only PVA (support filament) has a lower temp (45°C). That’s cooler than the oven allows, but if you don’t have a dual-head printer that uses PVA support filament then you wouldn’t need to worry about that.

15

@TheVirtualTim

So, I got one of those digital laser thermometers. With the nozzle set to 210C I was getting readings as high as 255C, and with the nozzle at 230C it was reading 275C.

Somehow this thread became about drying filament (some good info–thanks), but I’d sill like to know about hot ends that are compatible with the HS 0.8 tool head. Again, I’m looking for:

0.4 mm nozzle (or around there)
for 3mm filament
wide range of materials including TPU and nylon/carbon
something that’s an easy swap

I found this Volcano HotEnd – E3D Online but they’re out of stock of the direct drive 24V 2.85mm. I think that’s what is needed, but I’m not sure.

Any other leads would be appreciated.

P.S. I did contact Lulzbot and they just referred me to a repair shop.

16

The current LulzBot HS 0.8mm toolhead uses a Titan Aerostruder with a Volcano hot-end. But with a 0.4mm extruder you wouldn’t need a Volcano … a V6 heater block would be sufficient.

Nylon prints very hot … and requires an all-metal hot-end (but LulzBot hot-ends are all-metal hot-ends).

The nozzle is changed by heating it up fairly hot (e.g. 280-290) then removing it. I put a damp towel on the bed and wear appropriate gloves … just in case the (hot) nozzle drops you don’t want to get burned and you don’t want the bed to get damaged.

Do the same when inserting the new nozzle … you can start it (turn it in a few threads) but heat it up to snug it in. Aluminum threads are soft … so don’t go crazy tightening it.

The “cold” side of the print-head and “hot” side of the print-head are separated by a gap called the “heat brake” … this is a threaded tube to minimize the contact between the hot and cold side.

The heat-block has a threaded hole that runs from top to bottom… the heat brake is threaded down into the top of the block and the extruder nozzle threads up into the bottom of the block … and the two threaded parts meet in the middle. They must seal … no gap. This means the threads of the print nozzle will “bottom out” against the threads of the heat brake.

Also… when you remove the nozzle … the heat block will be loose … you’ll be able to twist it. So you’ll want to find a way to hold it while you thread in the nozzle.

It’s not too difficult to swap a nozzle … but you want to think through the steps because you’ll be doing this with while it is hot and you don’t want to burn yourself.

Usually the reason for moving to a smaller diameter is to print parts that have finer detail. e.g. LulzBot has a 0.25mm print-head but this is meant for printing small parts with high-detail. Printing with larger diameter nozzles speeds up the print job and moving to a smaller diameter makes the jobs take longer … but can print in finer detail. 0.4 is only slightly better than 0.5. Also… TPU is very difficult to print with fine detail (it tends to be very stringy … especially softer TPUs).

1 Like
17

Interestingly, after doing my last calibration the bed, it overshoots to 107 but settles back to 105 by the time the print starts. Strange. I should not have messed with it. It was not oveshooting before (but maybe it wasn’t heating up as fast as it could have? I did not measure the time).

18

You can always run that PID tune again, generally running it twice can get you within the range you need unless that PID value is all over the place.

1 Like