Qidi X-Max 3: First impressions

[mcdanlj (Michael K Johnson)]

@mcdanlj

https://www.amazon.com/Noctua-NF-A4x20-FLX-Premium-40x20mm/dp/B072JK9GX6

I may cut out the noise-creating grille from the back plate and put the fan on the outside. No need to worry about static ratings that way. And I’d rather use an adjustable buck converter than the resistor to slow the fan down.

I ordered that Noctua fan. It’s possible to install it on the inside even though it’s twice as thick as the original. I didn’t modify the panel at all.

I used one of the $1 LM2596 buck converters to convert from 24V to 12V.

I used a JST connector to connect wires to the buck converter, which I then soldered to the “omni” connector that came with the fan. I used liquid electrical tape as strain relief for the wires at the solder points, and put kaptan tape across the bottom of the LM2596 board and around the wires to isolate it.

Because I installed the original screws instead of substituting 25mm M3 screws, I had to install them with the heads inside the frame, and they protrude out the back. I could have either used this mounting technique with shorter screws or used longer screws through the hole fan case and avoided the problem, but now I’ll just occasionally scrape my hand and mutter when I am working on the back of the printer. I didn’t modify the printer at all, and could trivially replace the original fan if I wanted to.

I tucked the LM2596 board in among the wires at the bottom of the electronics bay and there’s plenty of room now.

I can still hear the fan if I’m listening for it, but it’s so much quieter. And it probably cools the board even better, because the 20mm deep fan has higher static pressure than the 10mm deep one it replaced.

[mcdanlj (Michael K Johnson)]

I now have two working cameras in the printer.

@mcdanlj

Haven’t had a chance to test it yet. [Edit: It works, but I’m not sure whether I’ll end up using the official one or the one I installed first. It’s not obviously better, in any case.)

The official camera flickers when I soft-PWM dim the case light. The one I set up doesn’t. So I’m back to full-power case light.

Because I’m using wired ethernet now, I removed the USB wifi dongle that the printer ships with, which gave me an additional port in order to be able to plug in both cameras.

The way I plugged it in, /home/mks/klipper_config/webcam.txt is the camera I first added, and has

camera_usb_options="-d /dev/video4 -r 1280x720 -f 10"

Then for the official camera, I’m using this device spec in /home/mks/klipper_config/webcam1.txt:

camera_usb_options="-d /dev/v4l/by-id/usb-SYX-230524-J_HD_Camera-video-index0 -r 1280x720 -f 10"

To make sure they are found, as root, systemctl restart webcamd and then less /var/log/webcamd.log

Then I have configured two cameras in Fluidd. The front camera in my setup uses /webcam2?action=stream (which, confusingly, comes from webcam1.txt), and the rear camera uses /webcam/?action=stream (which is from webcam.txt) I’d rather things made more sense, but I don’t really want to customize this printer more than I have to.

[mcdanlj (Michael K Johnson)]

@mcdanlj

I’ve been getting this error repeatedly lately

…

They responded in less than 30 minutes to let me know that it was a hardware issue and to ask for information to ship me replacement parts.

They shipped me three really separate things.

A new front and back cover, which have more air slots and each have a tiny 20mm 5V fan. This includes a breakout board with a regulator to power them from the same circuit as the 24V heat sink cooling fan. There were already unpopulated holes for adding this breakout board. I guess they had planned ahead to have room for a fix for something?

This I installed, and re-ran input calibration since I clearly changed the mass of the print head.

It was nice to get a new front cover, since one of the four tangs on the original cover had already broken.

A new USB-C cable (over which they run CAN) and installation hardware for it. Interestingly, they must have tested and bought some off-the-shelf cable, because both ends are clearly hand-trimmed to fit.

For this, I am waiting to see whether I keep getting the error before taking the effort to install it. If I do keep seeing it, my USB-C cable is probably defective. But it’s rather a pain to route through and install, and since I normally got that error message with the printer at rest rather than during a print, I doubt that the cable is the problem. For this, I’ll wait and see.

An inductive bed leveling sensor and a soft magnetic bed since the 9 hard high-temperature magnets would make the inductive sensor not work. They said that it’s more robust than the bltouch. However, I already bought a spare bltouch just in case this one breaks, I want to be able to swap out surfaces without having to have different profiles for different surfaces, and I don’t want to add a soft magnetic surface when they already ship high-temperature magnets. In fact, I’ve been pondering changing my other printer over from an inductive print head to a bltouch after experiencing failure of an inductive sensor, and in doing so I would also like to change from soft magnetic sheet to hi-temp rare earth magnets.

Fortunately, the new cover with the new fans still fits around the bltouch, so I’ll keep the soft magnetic sheet and inductive sensor aside for now. If I go through two bltouch sensors, I can always reconsider.

This does make me assume that they have switched to shipping new units with the inductive sensor.

[Daniel_Quites]

Hi Michael, I Had this issue for the past two months… they sent me 3 different usb-c cables. Its works until it doesn’t. The erro occurs mostly when printing ABS with chamber heater ON. Bought a high quality usb-c(with ground) and so far so good.

[Daniel_Quites]

I replaced the magnets with the soft bed but Im still using the bl-touch. Bed leveling improved a lot. The downside is, when printing larger/longer prints, the bed plate tends to lift/pull from the soft magnet. Im countering this by printing larger stuff diagonally. Im wondering if the inductive sensor will be even better for auto leveling. Let me know if you try it.

[mcdanlj (Michael K Johnson)]

@Daniel_Quites

Bought a high quality usb-c(with ground) and so far so good.

Can you provide a link to the good cable you bought?

@Daniel_Quites

The downside is, when printing larger/longer prints, the bed plate tends to lift/pull from the soft magnet.

That confirms to me that I don’t want to switch. I’m not finding bed leveling to be a problem and I like the strong hold of the high temp rare earth magnets.

[Daniel_Quites]

Hi Michael, I bought the cable locally. It’s Ugreen usb-c cable. I purchase two of them, one rated as 60W and the other 100W. The 100W one had the same problem. So I stick with the 60W version. Be aware that some triming around the connector is necessary to fit the cable properly.

UGREEN 60W Right Angle USB C Cable

[mcdanlj (Michael K Johnson)]

Kingroon is now making an assortment of “three eyes” nozzles in the right length for the X-Smart/Plus/Max 3 series. I haven’t ordered or tested them, but there’s a possibility they will help take more advantage of the fast speeds possible on this printer, since extrusion rate is the biggest limiting factor as far as I can tell.

[mcdanlj (Michael K Johnson)]

@mcdanlj

Qidi brought the price for the X-Max 3 down another $50 to $899 for black friday, effective today. I’m curious whether that will come back up afterward

Now the same price is a Christmas sale…

[mcdanlj (Michael K Johnson)]

@mcdanlj

Instead, they offered me a coupon for a free “smooth” print sheet (PEI without the manganese additive in their default “HF” sheet) which doesn’t have any logos painted on it, and so there’s no graphics to adhere to the bottom of a print.

I did take them up on this offer and paid a few dollars in shipping to get the smooth bed sheet. It has a fairly thick soft gloss surface on the smooth side; another reason to stick with bl-touch and not have to have a separate profile for each different kind of build plate. The other side of the plate is textured PEI (goldish, not the black-ish “HF” of the default plate), and that textured side is still labeled.

I’ve only printed ABS on it, and adhesion has been fine so far.

[mcdanlj (Michael K Johnson)]

Price keeps falling. $899 on the site, and $50 off with coupon code MERRYXMAS right now.

[mcdanlj (Michael K Johnson)]

@Daniel_Quites

I purchase two of them, one rated as 60W and the other 100W. The 100W one had the same problem. So I stick with the 60W version.

I finally read this thoughtfully — the hot end has a 115W heater, I think. That makes me worry about a 60W cord.

[Daniel_Quites]

I was worried about that too, but so far it’s the only cable that works. The new cable they sent me had the same issue after some time, only when using the chamber heater. I read in a different forum, that the real problem is the adapter plate chipset overheating. That’s why the add those little fans to the printer head.
Do you think this 60w cable could cause some damage or even catch on fire?

[mcdanlj (Michael K Johnson)]

@Daniel_Quites

Do you think this 60w cable could cause some damage or even catch on fire?

That’s my concern. But if it is honoring the USB e-tag on the cable and restricting power to the head, maybe it is helping by reducing power to the head so that it doesn’t get as hot?

If I used a different cable than they provide, I’d think about measuring power consumption preheating while I was present, and feeling the cable to make sure it wasn’t overheating.

[Daniel_Quites]

well, Im using this cable for over a month now with no issues. I will further exame it next print, search for warm spots. I email Qidi asking about this 60W cable, I’ll let you know about their response.

[Daniel_Quites]

Qidi replied today, the 60W is no Good. I’m
Replacing it with the cable they sent.

In my opinion, I believe the cable is not the problem, the adapter plate getting too hot is.

I only have this MCU error when printing with heated chamber on.

[mcdanlj (Michael K Johnson)]

I think you are right.

I basically only ever print with chamber heat on. I did only a test print with PLA; the rest has been mostly ABS and a little PETG.

[laicos]

Thanks for the thorough review!

I am considering getting this printer and the toolhead issue is kind of concerning. This is a bit of a stretch but I wonder if the design of the hotend + chip thermals + chamber heat = accidental reflow oven under the toolhead cover or at least just enough heat to soften the solder under rapid accel/decel.

[mcdanlj (Michael K Johnson)]

Oh, it’s clearly not that hot.

And in my case, I’ve had the MKS_THR error it has been after a print has completed, while the printer is cooling down. It might not happen as much if I’d leave the printer open to cool; I like to leave it closed so that I can start the next print without first walking over to the printer.

It might work to change the heater_fan hotend_fan’s heater_temp to a lower temperature, because the fans in the hot end shell piggy-back off the hotend cooling fan.

It ships like this:

## FAN0  喉管风扇
[heater_fan hotend_fan]
pin:MKS_THR:gpio1
max_power: 1.0
kick_start_time: 0.5
heater: extruder
heater_temp: 50.0
fan_speed: 1.0
off_below: 0

I’m thinking of running the fan all the way down to 30° as the printer cools after I finish a print.

## FAN0  喉管风扇
[heater_fan hotend_fan]
pin:MKS_THR:gpio1
max_power: 1.0
kick_start_time: 0.5
heater: extruder
heater_temp: 30.0
fan_speed: 1.0
off_below: 0

Because it became much less frequent when I put the new shell with the fans in place, I might need to go several weeks before having any confidence that it was an effective change.