Hotend for new printer which one?

[DerStig]

So question for everyone here. I'm building up a MetalMax (Finally) and I'm using a Duet controller. Would you use a E3d v6 print head and FSR's for calibration or would you use a HE280 and the accelerometer for calibration?

The HE260 does simplify wiring and build plate mounting but if the E3d and FSR's will give better results I'll use that.

Thanks for the help.

[Tincho85]

I haven't tried neither the FSR, the accel or the IR sensor, I'm still using the manual calibration.
But If I were in your pants, with a DuetWifi, I would give the differential IR height sensing board a try.
https://miscsolutions.wordpress.com/min ... sor-board/

If stuck with Rambo, the accelerometer is very tempting as it has a nice step by step guide.

and regarding the hotend, I can only speak for the E3Dv6 and V5.
The v5 used to have some troubles with PLA jamming, but the v6 performs like a champ.

[Xenocrates]

Personally I would use (and am installing) FSR's. I like it because it lets me use pretty much any hotend I like without having multiple sensors or wiring them, and helps me keep effector mass down, which given all the other bits I've been adding to the effector, is important to me. It's also easy to use FSR's with controllers other than the Rambo, while the HE280 is supportable by others but is not yet implemented in other firmwares(I think. folks like Arthur wolf for Smoothieboard and DC42 for duet would have to chime in). It should also be noted that the HE280's PCB can be used on other hotends too it mounted properly, so either calibration method is usable with any hotend, although it might be slightly odd to use the 280 without using the onboard autocalibration. Another factor is that the accelerometer doesn't work while heating, while FSR's will function all the way to print temperature, so you can probe at full temp.

So I would make sure you know how to configure the HE280 for the duet before going much farther, as I don't know what protocols the Duet supports for probes, and would need to look it up.

[mhackney]

I use and have FSRs, Davids, IR sensor and the new HE280 hot end. I'm probably one of a handful (maybe the only) that has real world experience with all three of these (as well as the old school mechanical switch probe). I have a lot more work to do with the HE280 as I've only had my V3 with HE280 running for full day but I've done a lot of printing in that time and starting to get familiar with the system. I have 1 Rostock Max setup with both FSR and IR Probe and I intend to add the accelerometer to it as well, I got a spare when I ordered my V3. Then I'll be able to do head to head comparisons.

There is no support for the accelerometer in the dc42 firmware for Duet. FSRs are, I assert, an excellent probing system for delta printers. The amount of force it takes to activate is much less than the accelerometer tap. Adding them to a Metal Max is about as easy as it gets. I'm working on finishing off my MM next so I'll have some designs soon(ish).

The IR probe is also a very good solution but a little tricker to mount than FSRs in my experience. If you go with a mainstream hot end like the E3D V6, there are printable mounting systems. If not, you are on your own. If you print on PEI, it requires special preparation too. You need to paint the back black before adhering it to the glass. But that is a small matter and actually creates a nice looking surface that makes it easy to observe your first layer. I also like the non-contact of the IR. In practice, the slightly offset probe point (from the nozzle) has not proven to be a significant detrement to calibration. But, the primary reason I don't use the IR Probe is that the technique I developed to repair and keep my PEI surface pristine (for the persnickety fly fishing reels I manufacture) is that CA adhesive is incompatible with the IR Probe, resulting in "random" triggering.

The E3D V6 hot end is the gold standard for hot ends with a very large user base. Again, I have all of the primary hot ends out there with one exception and they all perform pretty well. But for long term reliability, the V6 wins hands down. A lot of folks tout their hot end du joir and show a print or two but they don't talk about long term reliability. I print many hours every day and reliability is a critical requirement. Only one other hot end has demonstrated reliability in my shop and that's the J Head - but it is only suitable for PLA.

[mhackney]

Xenocrates also brings up an important point. The HE280 is a complete system - hot end, integrated wiring harness, accelerometer, part cooling fans. I liken it to the early days of the PC industry. The PCs had an open bus architecture. You bought a motherboard and whatever "cards" you needed - like display drivers, hard disk controllers, etc. Then the Mac comes along, a completely integrated system designed for the masses. It's success was that it simplified the complexity for the common man and eliminated any decisions. Not to push the analogy too far since the we are now talking about Open Source printers but in reality, the HE280 is more like a Mac with it's integration. The good news is, I've been very impressed with the print quality and ease of setup so far. Long term reliability experience will come with time.

And to reiterate, since you have made the (wise) choice to use Duet and dc42 firmware, don't expect accelerometer support in the near future. David and others have already expressed concern about the length of the cable run for the SPI interface to the accelerometer probe and susceptibility to noise (the primary reason you have to probe "cold" with the accelerometer). How it plays out in practice is yet to be seen.

The other point is that mounting the accelerometer to other hot ends like the V6 is going to require some head scratching. It is designed explicitly for the new SeeMeCNC hotend that has a threaded insert on top that clamps the probe through it's center hole. That hole is too small to fit any other hot end I have.

[KAS]

The PCs had an open bus architecture. You bought a motherboard and whatever "cards" you needed - like display drivers, hard disk controllers, etc. Then the Mac comes along, a completely integrated system designed for the masses. It's success was that it simplified the complexity for the common man and eliminated any decisions.

What's this Apple drug you're pushing? Mac didn't come along with an integrated system. They had a history of open and close architects and numerous expansion boards, not to mention the failed NuBus design.. /PCFORLIFE

[dc42]

Both I and T3P3 have been considering making a PCB effector for delta printers using an accelerometer or strain gauge to detect nozzle contact. I have had an accelerometer module on my bench for a month but no time to try it out. If we do decide to make an accelerometer-based probe then we will include a microcontroller on board to avoid the long I2C run, and hot probing will be possible. But I'd like to hear other people's experience of accelerometer-based probing first, to help me decide whether to put effort into that or the strain-gauge based probe.

[Xenocrates]

The PCs had an open bus architecture. You bought a motherboard and whatever "cards" you needed - like display drivers, hard disk controllers, etc. Then the Mac comes along, a completely integrated system designed for the masses. It's success was that it simplified the complexity for the common man and eliminated any decisions.

What's this Apple drug you're pushing? Mac didn't come along with an integrated system. They had a history of open and close architects and numerous expansion boards, not to mention the failed NuBus design.. /PCFORLIFE

Apple didn't come along with an integrated system, however the original Macintosh was largely integrated, it was the Mac 2 which was one of the first expandable mac's, although machines like the original Apple II were more hackable and expandable. The modern products of Apple are very much integrated systems (Which is quite annoying to me, as I like to tinker and upgrade, but to each his own).

I think it's a matter of pedantics as to which is correct, as the Mac as an actual Macintosh is integrated, however Apple computers were originally much more open, IE, the Apple II, and yet are now generically referred to as Mac's.

DC, if you do end up making those PCB effectors, are there any plans for the onboard micro to be able to switch hotend or fan power? It would be more complex in all likelyhood, but it would allow for cleaner wiring down to the effector itself, which in my opinion would be nice, and would also potentially let you have different supply voltages for the fans Vs hotend Vs bed if you wanted to. it definitely has some downsides in needing to cool the PCB more if you do that, and coming up with a way for the controller to signal it to heat. It's up to you how you make it though.

[DerStig]

Thank you all for your input.

My only concern with the FSR's is build plate constraint. how do I do that?

Thanks again

[mhackney]

Research how others have done it. It isn't difficult if you think about it, especially on the MM.