xTool Pro 20W Review: Mandala Art using a laser cutter and engraver of 20W power

Karl. This is Karl here. The xTool D1 Pro 20W is a laser cutter/engraver that has a work area measuring 340x390mm. The product has received a lot of attention and I’ll comment on some points that I didn’t see mentioned. For these reviews, I participate in a number of Facebook groups to receive real-world feedback. As with almost all types of digital fabrication I frequently read about the frustrations and misconceptions caused by unknowingly buying a machine which requires patience and a significant amount of time to master.

This review was based on a daily log that I created and which will give a summary of the time taken to finish one project. I got sidetracked in a couple of areas and will share those details. It is exciting to share my results from the laser power tests I’ve been conducting.

The logs will be brief and straight to the point. Some minor editing was performed. I didn’t mean to alarm anyone. My goal is to educate and inspire. It’s a challenge for me to learn new processes and techniques. I find it very entertaining. I believe there is a list that almost everyone needs and must be included in budgeting. The cost of extra items adds up fast.

I am not a beginner as I have invested a lot of time into different CNC machines.

Prior to starting, I need to set up my workspace for the project. It is winter in my workplace so I have to review it in there. Particulates are produced by lasers and need to be eliminated from indoor areas. I have purchased an enclosure softened with exhaust, and would like to share my experience.

My project required me to create a present for my wife. Although I knew the design elements I wanted, I didn’t have a detailed picture. These were the elements I was looking for.

• Turning your back towards the box
• Modular Mandala Art
• Make thin strips of metal, and then artificially rust/weather them
• Laser cut 45deg corners
• Grave art 
• Stretch goal to ingrave shading
• Construction mainly from plywood

This was a major mistake.

This project took 3 weeks for me to set up. After the kids are asleep, it takes me up to 3 hours to complete this project before I’m ready to put my hands up.

Summary

It’s a great product overall. Very strong 4-diode laser module. It is amazing that it’s possible. This is amazing. It’s fascinating to me how they assembled it.

You might like

• You are completely contained. You won’t find any moving parts in the vicinity.
• Use offline mode. Although not well documented, it is an advantage.

Do not like

• Mesh loom to manage wires Simple to use
• Concentrating is basic
• Firmware cannot be edited or changed.

Unknown

• Motion system metal on metal. It is unclear if this should be a problem. These last two images show round metal rails and metal rollers. Would love your opinions. Share your experiences with the subject.

My longest task was connecting all the small cables.

• 1/26 at 9:21 AM ET Test/Setup

Setup. The enclosure is now complete. I determined I cannot do this on my 24” deep desk. The hose is attached to the computer desk by using my CNC bed. This is the final part of my exhaust. Design and printing. After printing I discovered that I had missed some geometry. I will need to print another piece. Lightburn was updated and I renewed my $30 license. I downloaded the Lightburn profile from the support forum, and then installed it. Lightburn doesn’t recognize it automatically. It took me the longest to update the firmware. I used the standalone program. The instructions weren’t good. xTool Creative Space ended up downloading an older version of the tool that was not available for download. It was not possible to properly install the cables that I had complained about while assembling them. Limit switches could not be connected. Limit switch issues could be fixed by seating fully.

I had the weekend off. I still printed and put in 2 pieces: leg and wedge. I’ll add about 15 minutes. Today is the day to perform power tests using a DIY meter, and complete the exhaust. My Dad and I discussed the project. I am concerned that a single fan won’t be enough. He suggested that additional 120mm fans be added. This is something I could try if the one you have doesn’t work.

Many mistakes were made. Although not bad, the power tests have been inconsistent. It was difficult to get the timing correct for the power test. 17.88W…..but I had 2 failed tests and the aluminum is already heated. I will move on to fusion, and complete the fan. Once it cools down, I’ll retest the fan at the end.

I have done some research on combining fans. I think it’ll be fun to test. The fan, which measures 80mm in diameter, moved air well when I tried it. It will need static pressure to work. The air will need to be pulled in through cracks. A test was done inside, and the device could take in some smoke. Amazon offers fans with promising features that are just a click away. I also “overclocked” the fan to 19V for a few seconds and then it was moving some air. Some bitminer fans claimed speeds of 5000-6000 rpm. The price is right and I am hoping that I don’t have to go all out and get a $100 exhaust fan. A exhaust vent seems to be a necessary accessory for any maker. A resin printer review was stopped by me. I couldn’t handle the fumes. I probably could press through but I was getting headaches and didn’t want to change it. Anyway….a hood would have been great for that. This would also be great for printing. A removable curtain could be added to an industrial hood.

2. Second test, mostly ambient: 18.37W; another at 18.37W an hr or so later

An epiphany occurred to me regarding the DIY laser powermeter. This is a project that I’ve been working on for some time. The power measurement results from all my tests so far have been very similar to the laser’s rated value. I have seen very few reviewers actually measure laser power so I didn’t have anything to compare to. The safety features on these lasers and the software prevent them from sitting still while firing. This is what I did: set up a laser job for a one-mm diameter circle at 1mm/s with a variety of passes. I had found that LightBurn has been pretty good at approximations and didn’t really question it. It knows every motion parameter. This, my fatal mistake I believe. The xTool showed me that my times were not correct. I performed a test run, and then used a stopwatch for the time. In hindsight, this isn’t accurate enough. A single second can make a big difference when it comes to calculations. What I’m going to do now is record the 1mm code and run it. Measure in video editor software.

This is unacceptable. Two tests are exactly 11 seconds….Two are 11.5 seconds.

Hmmm… 18.37W again… at least it is consistent. It had to have been close. We can’t read too much into this. Other variables have also been changed.

So I removed some of the settings that I had made to the software. Now, I get 15-16W. It needs to cool down completely and then I can try it again.

Tonight, no real progress. I will return to power later. The safety devices are in place so that the laser does not fire while it is still moving.

The problem is likely caused by the DS18B20 thermometer I’m using. It comes with a shell made from stainless steel. To check the math, I will see if the laser can be fired directly at the shell. Okay, it doesn’t work out. That would cause the temperature to rise too high. I’m going to find a different probe.

After looking at a lot of options, I finally settled on the sensor with a slightly different package. The swapping out is easier. You will also need to replace an inactive 3D printer heater. You should place the sensor where you would find the heat cartridge. It works well and can measure 80 Watts in a five second burst. I’ll look into finding a more durable piece of aluminum.

Finish the exhaust. It was much easier than I expected. It’s done. The print is now underway. I’m calling it quits for the night. It was the right thing to do. Close to ready for the cutting. Also, I ordered the temperature sensor which will be delivered on Thursday.

It was important to me that I could find an easier and more repeatable method to test fire. To find these commands, I had to watch the serial port. This took a lot longer than I expected. To find the right formatting, it took some effort.

The exhaust was running. Pitiful. Ordered a 6” duct fan on eBay for $67.

I won’t be able to make much progress tonight. FedEx informed me that the fan should be delivered by 2/6. It is necessary to create a mockup of the exhaust pipe setup. Ok, it’s time to regroup. My CNC bed doesn’t have enough space. My benchtop will be used. This is not enough. I’m thinking I will cut a hole in the top of the exhaust chamber depending on how big the fan is installed directly in the top. It is only 30cm wide. Perhaps I can print it instead of buying a pipe.

I saw that xTool came out with a 40W diode version…This video shows cutting a 2×4 pine board as an intro. This is quite impressive. The 8 diodes are combined. It is very close to the CO2 zone. Tonight is a short night.

Everything is now working much better. I will need to hire a laser meter. To cool the exhaust, I’m using the fan provided with it. The block is not insulation. You will need to glue the temperature sensor in place with thermal compound. Next, glue it into place. The next step is to test the cheap thermal epoxy. This will let you know how repeatable it is.

Have passed several tests

• After a bit of code manipulation, 20.34W.
• 20.85W It was closer to the room temp at this point.
• 18.8W outliner. It was impossible to see the surface perpendicular.
• 20.53W
• 50% Power Approx 10 deg – 10.36W
• 25% power – 4.48W

The initial thought was that the block must be much smaller to allow for a lower-power testing. It only raises 3 degrees Celsius.

The batch file will be modified to allow for length, power, and com port. A few batch files are all I’ve created right now.

• 2/4 10:10 12:07 – Power meter software
  

Tonight was strictly for software to power the power meter. The key to success is repeatability. In my testing I discovered that power calculations were not accurate if the aluminum was at room temperature. To find out the function, I used an online function equation finder site to perform several measurements. The code was then incorporated and the offset calculated. It was not linear, but I wasn’t expecting it. But after thinking about it….the rate of heat loss should be linear. My room temperature was 23 degrees. Next, I will need to determine if real-time temperature measurement is required and integrate them into my equation. Then, I’ll test the existing code in a cooler or warmer environment. At the moment, I’m very happy. These calculations assume that this laser can output 20W. A delta of 1.25 Watts was obtained by combining the results from 1 and 0. If I need to consider the temperature, I’m going to have problems. Tonight, I’m going to turn off my heat in my office. We will see what it does tomorrow morning. Maybe I’ll need to hire some math help.

Although I do my best to not work weekends, it was still a lazy weekend. I am beyond thrilled with the outcome. The results are even more precise this time with an approx.5W. The ambient temperature doesn’t seem to have an effect. The results came out as I expected when tested at 16C. However, I found that the time factor can affect results. The results of testing below 5W were inconsistent. Because the temperature rises only by a few degrees, I tried increasing the laser’s timer to 20 seconds. The variable was adjusted accordingly. Again, it was not what I had hoped for. But I’m not certain if it is possible to modify the current formula. Do I need to calculate each duration? This is enough for this side-project. This xTool will help you. This side project would allow me to test your laser at 40W.

Although I expected the inline exhaust to arrive tonight, Fedex has proven wrong with its estimation. It takes almost always an extra day. My wife finally agreed to sit down with me and I began explaining my plans in order to gain her feedback.

• A box is what you should be leaning toward
• Modular Mandala Art
• Make thin strips of metal, and then artificially rust/weather them
• Laser cut 45deg corners
• Engrave
• Stretch goal to ingrave shading
• Construction mainly from plywood

She liked some of the examples so she is now working on her Mandala wall art design. It is not clear what program I will be using. Inkscape, Fusion 360 and Fusion 360 are my options. Fusion 360 is what I plan to use for full rendering. While I can buy premade designs, I would prefer to make my own. A beach theme would be great. The beach is her favorite place.

Ok, no progress tonight. This is the first time I have tried to create this myself. A Mandala Art Generator was what I came across. In the end, I had a few minutes to experiment with it. If I make it work, it will be easy.

After looking at some photos, I decided to change my strategy. I was trying to go from the back to the front…I am going to try the opposite direction tonight.

Today, I was gifted a new fan inline. Although it is of good quality, the fan shakes at high speed. It is quieter, and moves a lot of air, even when there are obstructions. It is quite weighty. The enclosure is heavy. I wanted to create a loop and then connect the two together.

I am feeling defeated. Perhaps I will just have to buy and find. I don’t want to spend much more time in this phase. Very little progress.

• 2/8 9:21 11:15 Setup/Project

Finally, I found the right guide to help me make a 3D mandala. The test was a success. Also if you read my last review….I am now back to bare metal with the 3900X and letting it stretch its legs. It was then that Fusion rendered it. This gives you an idea of what the end result will look like.

I also designed and printed a 6” flange for the enclosure. Super glue will be used to attach the enclosure to the vent hole. It is a problem to vent pipe. It doesn’t matter where I look….I see negative reviews saying it is extremely flimsy. I’m going to the hardware store, and will likely go rigid. This is a very short job and I have only enough rigid duct to get it done.

It’s a great feeling to create. We just need to find better applications. Fusion didn’t like the complexity of the geometry. It isn’t optimized for mirroring or circular patterns. Although it was great to be able update one pattern, all of the others were updated as well, Fusion is not optimized.

The pattern was created using a new method that I found on the website. It worked well so I tested it. Everything went smoothly. A couple more tests and then get my wife’s input on the different styles. Then I’ll see her reaction and decide whether or not to go with the customized design. I like the idea of custom designs but dang it takes so much time…..She is in bed already and will see what she thinks tomorrow and make a decision. This generator is only for reference. The vectors were not closing properly for me. Variables were also tried. I am able to modify offsets in masse.

• 2/13 9:04 10:11 Setup/Test

There was not much to be done tonight. Super glue didn’t get picked up by me. For the blower to be hung, I made some command strips hooks. Printing off a transition from 6” pipe to square hole. Modeling this took just a few moments. I tried vase mode, hoping the .6 nozzle would be rigid enough but it wasn’t. Printing with 3mm thick walls again. I got feedback from my wife over paid designs vs custom….I will find one to purchase. I liked the thought of a custom one but don’t really want to put all the time in. I’m going to sit down with her tomorrow and get one picked. If all goes according to plan, I’ll be setting my sights on tomorrow.

• 2/14 9:16 10:00 Setup/Test

I attached the flange to my enclosure. It should dry over night. One final piece was designed to be used as a transitional piece. Fusion has just allowed me to make an extra piece. Instead of creating a completely new piece, I will glue the piece together. I really don’t like this final setup but will move forward to get this project done. It’s time to cut tomorrow.

• 2/15 9:19 11:15 Setup/Test

So I took a while to glue and secure the last pieces for my exhaust. I also found some coasters made of wood in the Dollar General’s arts and crafts department in my hometown while I was applying glue. It looks like balsa. It is a method I’ve been meaning to attempt for some time. I etch the shape, then fill it with UV resin and sand. My wife hasn’t finalized on her design so picking up a few side projects.

I couldn’t be happier about the first test. To see how useful a mask is, I tested it with and without one. This looks like it is time to order some masking tape. Tomorrow will be the actual UV resin experiment. This is a great start. Beautiful and detailed engravings.

There is still a wood-burning scent. It isn’t a perfect solution for exhaust. It is something that I am going to continue working on.

• 2/16 9:10 11:00 Setup/Test

Today I spent 5 minutes to apply a clear coat to the Pepsi Coaster. Wood is very porous. Clearcoat has two functions. Seal the wood so the resin doesn’t seep into the wood, and protect the wood when sanding the black resin and grinding into the wood.

It is now set. I don’t feel like my UV light is strong enough to penetrate all the way. I’m going to put it outside to fully cure and sand. It seems promising.

Steel business cards were something I attempted. It took several tries but it’s a fun little project

Steel business cards 200mm/s 20 power.15 Darkened images to better show Jarvis 4.5min under the right lighting it looks amazing

Also, I made a wooden coaster. Wood 150mm/s 85 1 passes 4.5 min

• 2/20 9:05 10:28 Setup/Test

I really don’t like the focus adjustment. There must be a simpler way. Focus will also be tested. A couple of lasers back, I discovered my focus jig.

It’s over for tonight. The plastic cost about three cents and the result is a small improvement on stock. This is fine for the moment, but I would like to make this right sooner.

A focus jig was also tested, but it wasn’t conclusive. It seems that the Focus arm works well.

• 2/21 9:00 10:30 Setup/Test

I didn’t like the first revision. The Z-lifter was quickly rewritten. It is much stronger and better. The model was easy to create and printed in just a few minutes. 9:45 now. My wife was compelled to choose one of the 20 files I had found. It seems I’m going to do several things. The compass I saw is amazing.

We received more today. 12” masking tape

First cuts tonight… lots to experiment with. It was kerf offset for friction fit parts. This is approximately half the beam width of 0.07, which allows for precise dimensions and friction fit.

Some cuts were tested by me. The design was finalized by my wife via Etsy. She now has to pick the colors.

These parts are very precise and detail-oriented.

Starting cuts. Some hold-downs were printed for the honeycomb to ensure friction.

Masking tape was not necessary for first cuts. I had some problems with my memory, which led to me having to stop cutting. There were 4 layers. I find it amazing.

I’m getting too many spots that aren’t cutting through and it’s taking too long to clean up. Accuracy – 1mm/s running at 90%

Need to improve ventilation, as I’m getting some smoke smell in the workshop. 15-minute cuts. The last cut is much better after slowing down…going to lower the laser a mm or so to focus the laser just below the surface. That mm/s could be possible. A few extra minutes were added to the cutting. The entire process will be finished over several nights.

Tonight, you can go back to it with five more cuts.

Last 5 layers tonight. The only problem was that I failed to place the focus arm again. This snag caused the cut to be ruined. I am almost ready to glue. It’s much easier to do this tonight.

• 3/2 9:14 9:45 Setup/Testing – Off-line mode
  

Tonight’s testing took only a few moments. Just before I went to sleep, last night, I posted a question on xTool’s forum regarding offline laser cutting/engraving. The product is described, but there was no documentation about how it works. I find this very important. It is difficult to stream Gcode from Windows to any CNC machines. Linux worked fine. It is important that $I be the first command in the gcode. The file on your SD card is named tmp.gcode. Just added $I Lightburn Start gcode below Edit > Device Settings > Gcode > Start Gcode. The xTool D1 Pro20W machine cannot be configured automatically by probing it, like other laser engravers. This would be a good idea for xTool.

So that’s it. I’ll wrap up my experiences.

Summaries of Time and Total Cost

It took 33.75 hours to complete, and this does not include any writing or light research reading.

• Installation and testing take 11.5 hours
• 10-hours of working on the power meter
• The project took 12.25 hours.

This is a summary of the message: Be prepared to invest time and money. Nearly everyone should have an enclosure that has an exhaust system. This will increase safety and provide additional benefits to your quality of life.

I admit that the project is not complete. However, this will allow me to continue gluing up as well as framing. The weather was brutal and it is now cold outside.

I love these types of tools and projects. The challenge is what I enjoy and the investment time it takes. While the xTool D1 Pro20W may be a costly investment, it can also lead to some amazing projects. You can order one at Their official site currently sells for $1,103.99 plus an extra $50 discount for St Patrick’s Day, or Amazon Prime for $1,119. Both of these items are currently on sale.

Karl is an avid technology fan who contributes reviews to TV boxes, 3D Printers and other gadgets that can be used for making.