Video není dostupné.
Omlouváme se.

3D Print Farm After Hours - Ep 1 - How many Bambu Labs would replace our current fleet?

Sdílet
Vložit
  • čas přidán 15. 08. 2024
  • The Bambu Lab X1 is an amazing machine that is changing the 3D printing landscape for the better with increased printing speed, great reliability, and superior user experience. However, would it be fair to compare them to industrial 3D printing solutions?
    In this video, we break down how many X1's we would need to completely upgrade our fleet using theoretical maximum production rates, with some comparison to our real-life empirical data as well.
    We suspect this video might be a bit controversial, so we look forward to hearing thoughts and opinions in the comments.
    00:00 Introduction
    01:00 Defining 3D Print Farm
    01:20 Compare to Industrial 3D Printing?
    01:55 Bambu X1 Production Benchmark
    02:56 Vs FDM Print Farm
    03:57 Vs MJF (Multi-Jet Fusion)
    05:22 Considering Workflow & Quality
    If you have any high-volume 3D printing needs, you can visit us at i-solids.com.

Komentáře • 23

  • @Sask3D
    @Sask3D Před 5 měsíci +1

    Wow that is an impressive facility! Also what an awesome video getting to see some of the inside of companies and people who are really making it with 3D printing

  • @carlosnvarro8532
    @carlosnvarro8532 Před 3 měsíci +3

    It quite clear that the Hp is a production monster, and as an SLS user I love MJF, however is not fair and true to compare that in one hour the HP produces 5000cm3 of parts while the fdm/bambu produces 57cm3/h, as the HP is only fusing around a 12% of the build chamber, what will give 600gr of fused powder per hour, making the real comparison at around 10 machines.
    There are also other factors to take into account that can be even less beneficial to this comparison, such as that in fdm parts are not solid, what implies also that you will not need as much material and therefore parts are faster to procuces that if they were solid as in your comparison, and also that parts must get cooled for 12-24h.
    Even in that case you are comparing a 500k$ system with a 2k$ and only telling you need 88 of that to make it equal... Again not very wise

  • @ErikBlack
    @ErikBlack Před 6 měsíci +1

    Thank you for this video! Helps a ton to understand just how powerful your HP machines are

    • @i-solids
      @i-solids  Před 6 měsíci

      Thanks @ErikBlack! These machines have a lot more capacity than most people realize but I can see how it might be like comparing a shovel to an excavator.

  • @glennstar2121
    @glennstar2121 Před 6 měsíci +5

    So I understand not replacing your hp machines as they are different and not a fair comparison. However, comparing the FDM printers to the Bambu labs x1 , not only do you get the faster build time but by reducing your fleet by 30 would decrease your total power usage and your employee overhead because you would have 30 less printers to maintain. Not saying you should scrap all of you printers lol. Just saying if you are looking to expand your FDM machines it might make sense to look at getting into their ecosystem. Not to mention the lack of upgrades you would have to do to them right out of the box to match (or get better) results. Love the videos and thanks for sharing!

    • @i-solids
      @i-solids  Před 6 měsíci +7

      Hey glennstar2121! First off, thanks for taking the time to watch the videos and we appreciate the support!
      There is a lot we could unpack on this question but there are essentially four primary drivers as to why we haven't upgraded the FDM machines.... yet:
      1) Firmware / software integration - we have a lot of custom software that has been developed and refined over the years and it integrates directly with our MES. Bambu has some solutions here as well but ours is specific to our workflow and it also meshes with Formlabs, Nexa, and HP equipment so we have a single production ecosystem regardless of the OEM. I think with time we could also get Bambu's to work in this same way - its just a good amount of development time we would have to factor into an ROI.
      2) The peak power is actually quite a bit higher for a an X1 versus what we have. For example, our 130+ machines have between 300 and 350 watt power supplies so we can fit 6 to 8 printers on a single 20 amp circuit. The X1 on the other hand is a monster 1000 watt supply - this is great for heating times but it also means there we can only fit 2 on a 20 amp circuit. This is over 3 times the number of circuits we would need for the same capacity. Of course, at steady state the machines wouldn't draw anywhere near 1000 watts so we could theoretically run a lot more but we would have to intentionally delay or stagger prints which can complicated things but more importantly, it would be a safety concern to overload a circuit.
      3) We currently average just 1.5 people per day dedicated to actually maintaining and running the FDM print farm portion of our business. This does not include a lot of shared support roles like application engineering and quality control which may use portions of their time but these contributions are relative to the number of parts produced and not the number of machines so we don't anticipate any major changes in the manpower requirements.
      4) The main factor in my opinion will always be based on economics. Everything we have now is operational and fully paid off - if we were to instantly switch over the entire fleet it would be $120,000+ investment not including additional manpower costs and even potential lost revenue from downtime. Although there would be improvements in speed and efficiency, I think it would still take around 3+ years just to make make a return on that investment.
      With all of that being said - these opinions are based on using our current operations with a lot of harsh lessons learned over time as the benchmark. If we were building a print farm today from a clean slate I think think our opinions would be much different.

    • @meikgeik
      @meikgeik Před 6 měsíci +1

      You're also assuming that the Bambu printers would be equally easy to maintain as their existing fleet. Nothing could be further from the truth. When a Bambu goes down, it is down for a while due to the way Bambu handles their support and parts distribution. They do not use nearly as many off the shelf parts, and they are kind of a pain to repair. When a Flash Forge Creator Pro (what it looks like they have there) goes down, EVERY part on it is a very cheap off the shelf part that can be fixed... The exception would be the main board, which it sounds like they have changed when they converted to Klipper. I have 7 of the FFCP clones, and they're really good workhorses. They can log thousands of hours between needing maintenance, meanwhile Bambu users have had their warranties voided for not lubing the carbon rails EVERY WEEK. I know people who run Bambu farms, and it's not all roses and rainbows. Like any other printer, they have their upsides and downsides.

    • @Defyant123
      @Defyant123 Před 6 měsíci +2

      @@meikgeik ive got a farm of bambu labs and this is simply untrue.

  • @OldManJimmy1
    @OldManJimmy1 Před 6 měsíci +5

    I think it's hard for people to get their head around the difference between an industrial FDM machines and a consumer level machines.

    • @i-solids
      @i-solids  Před 6 měsíci +3

      We would definitely agree. Sometimes when we say this people may assume that that means we are downplaying consumer printing - it seems silly because we are obviously fans of consumer grade machines as well with our FDM farm. They are just designed for totally different environments and scale.

  • @Defyant123
    @Defyant123 Před 6 měsíci +2

    Trying to swap your HP machines for x1c's is ludicrous .....Crazy comparison... you want say it but i can :) But that old fleet of "upgraded" and "klippered" filament machines is ....well.....an old fleet. 130 of them you say. A delta of 23 per machine x 130 i would think is huge when it comes to output.... Unless! demand is not there? I love your farm your business and the time you take to talk and show us.

  • @edwardmcrichy2985
    @edwardmcrichy2985 Před 6 měsíci

    I think people would be interested to see what you guys have done to the creator Pros you use. Upgrades, changes from stock, any timing you do etc.

  • @slawomirgontarek4213
    @slawomirgontarek4213 Před 6 měsíci +1

    Do you also receive orders from Europe?

  • @slawomirgontarek4213
    @slawomirgontarek4213 Před 6 měsíci +1

    How do you manage energy for such a 3D printer farm?
    What is the energy demand for such a farm?

    • @akselmani
      @akselmani Před 5 měsíci

      ~100 KW average with much higher peaks. If you start all FDM printers at once you can get a huge spike in power draw. The Bambo X1C has 135W avg with 850W peak when starting. I think Kason went through the actual amps over in another video on this channel but you will need three-phase power and a bunch of different circuits for sure.

  • @olafmarzocchi6194
    @olafmarzocchi6194 Před 3 měsíci

    Which FDM machines do you use?
    Bambu is not an option even simply because of the lack of Klipper which gives quite boosts for certain use cases (not talking speed but customisability)

  • @floodingchen
    @floodingchen Před 6 měsíci

    I wonder if infill should be considered when you are comparing X1 with MJF. Since usually you print with sparse infill with FDM while you may need to use lattice structures to do sparse infill for MJF.

    • @i-solids
      @i-solids  Před 6 měsíci +1

      I think this is a really good point to bring up. I considered adding some thoughts on this in the video since we are running several projects right now specifically on FDM because of the ability to easily adjust infill - one application where we needed to have a specific mass / density for calibration purposes and another where we used infill as a design function. However, we can in most cases still create an "infill" or lattice structure so generate the same affect to reduce weight or material - its just not as quick and convenient since this would occur on the design side.
      My only counter argument on infill's ability to increase production using FDM would be that if the infill is sparse then it might not be necessary to the designs strength and could be optimized better. In many cases, we see parts designed for FDM not for functionality but for printability and this can tend to make things bulkier and thicker than needed to avoid support material or layer weakness.
      Overall, we are big proponents that each 3D printing technology has pros and cons and work better for different situations.

  • @NotseenNinja
    @NotseenNinja Před 6 měsíci +1

    Why do the MJF printers not replace the FDM printers? Is it the materials that can be used and the cost of those materials that make it more cost effective to run the FDM printers for certain clients?

    • @i-solids
      @i-solids  Před 6 měsíci +1

      This is a great question! In many cases it does - especially for complex parts with thin walls or a lot of support material. However, the economics are just very different.
      With FDM the machines and materials are relatively low cost so the labor because the biggest cost factor. With MJF it is normally the opposite - material and machines are expensive but we can produce A LOT more parts relative to the amount of labor required. This yields a situation where really big bulky parts will typically be cheaper with FDM because the material costs will be too high with MJF.
      FDM also has a lot more flexibility when it comes to material and color options that can be important in a lot of application.
      Finally, FDM can often provide some creatives solution when you are able to manipulate the slicing for a specific function. For instance, we are currently running a project where parts have to be within a specific mass +/- a few grams. Being able to adjust the infill allows us to do this easily.

    • @NotseenNinja
      @NotseenNinja Před 6 měsíci +1

      @@i-solids Very interesting, thank you! We run a farm of 60x Creality Ender 3 and 10x Bambu Lab P1P and have been looking at the more advanced and larger options. It is easy to find cost breakdowns between FDM and resin because they are similar enough, but finding real numbers for MJF is not quite as easy. Would love to see a video taking a few random projects from small, medium, and large and doing a cost breakdown between the three types of printers. The numbers would be very estimated since costs will vary depending on suppliers, that the materials are not exactly the same, and many other factors; but it would still be quite interesting.

  • @pHuzi0n
    @pHuzi0n Před 6 měsíci

    Totally distracted by the terrible green screen... Is this an 80's TV ad?

    • @i-solids
      @i-solids  Před 6 měsíci +1

      Strangely, this is not a green screen at all so I'm guessing its the lighting. We try to run very lean and keep in mind that our customer pay us to produce parts and not create youtube videos so we are just making these in our spare time without a lot of video editing experience but I think we are getting a little bit better with feedback. I do kind of miss the 80s style ads though - the jingles were a lot more catchy.