Interesting new Diamond hotend 3x filament

[626Pilot]

AdaFruit sells PermaProto breadboards. They are like the usual perforated breadboards, but electrically, it's set up more like a solderless breadboard. There are power rails on the top and bottom, and the solder pads are connected to their neighbors in the same column. There are no connections between the middle two rows of solder pads, so the top and bottom prototyping areas are disconnected from each other, making it ideal for soldering up standard ICs. This way, you don't have to solder a bunch of jumper wires to the underside. They're really useful for small projects.

https://www.adafruit.com/search?q=permaproto&b=1

[JJPowelly]

Yeah, I need to pick up more PCB components, I was just working with what I had at the moment... But I definitely don't want that motor powered with 24/26 gauge jumper wires permanently...

[JJPowelly]

My setup now... I completely changed what I did in terms of setup since making the video I posted above... I bought a little stepper expansion board from reprap.me... I bought it to play around with it - not expecting it to work - however, when I hooked it up to the Rambo and played with the firmware, it ended up working out for me... So there's that...

Right now I'm designing my own EZ(er)-Struder (I really don't like the EzStruder it's not really that ez, it's bulky, heavy, hard to feed, and the filament likes to get hung up in the assembly... I really think my redesign is 100x better, IMO)... and new 'Flyin Struder' to support the three stepper motors that drive the filament into the Diamond... I have some pretty heavy nema 17s in the (240 grams each) so 240 x 3 = 720 grams... In total, my assembly weighs 850 grams which is way too much weight on the effector, belts, and motor... I'm waiting on some new motors I ordered that weigh only 180 grams yet still have ~2100 gram/cm of holding torque (pretty strong)... With the lighter motors and weight reduction in my assembly, I'm hoping to get the total weight down to about 600 grams in total... This will work since the stock extruder motor with the printer is relatively heavy (~400 grams w/ezstruder - each)... I had a decent single extrusion system with a "flyin struder" that worked well and in total that was about 500 grams... So adding 100 grams for two more motors seems like a good trade-off to me...

But everything is working for me, I'm just trying to figure out how the setup should work now... I'm also dealing with all the extra wiring that comes with adding these extra motors and how to route that up to the top of the machine... If I get around to it, I would also like to change the thermistor to the PT100 coupler sensor for more accurate readings and the potential of reaching 400C...

Overall, I'm about 1-2 days away from having this 100% set-up and printing...

Here are a few pictures....

[Eric]

It sucks the rambo doesn't have very much in terms of technical documentation (especially v1.3, pretty much nothing)... To find the best pins and voltages, I literally had to unravel the schematic because the only schematic that gives (accurate) digital pin outs was for the v1.1... So I literally had to solve a riddle to figure out which pins (that weren't already used) go where... That's what took the longest for me... I'm a computer science guy by trade, coding is one of my strengths... So I had the firmware good before I even started with the electrical... However, The digipot is still a riddle to me right now...

I found the nomenclature for the digipot and looked up the manufactures specifications and data sheets. It's a 6 channel digipot, but only 5 of those channels are declared in the code... The main channel line is declared in pins.h

#define DIGIPOT_CHANNELS {4,5,3,0,1} // X Y Z E0 E1

To me, this is a very odd array of channels, especially when you look at the board - it makes no sense why the would number it that way... clearly there are 6 overall channels and channel 2 (or 3 if you're starting with 1) is not used...So there is an open channel on the digipot, it's just a matter of finding where it leads to... I would imagine, since the digipot's primary concern is current control and micro stepping, that the vmot aux that I showed in the video (the one seeme uses as a board fan) is actually that 6th channels voltage out, or can at least be mapped to it... Even more proof is that, the 12v (vmot) out is also under the same fuse as the rest of the motors... and then the microstepping pins (MS1 and MS2 for 4982) route out to the analog auxiliary due to the fact that the majority of the MS pins are in the pin 60-69 range - which is also the auxiliary out ports' range... But like I said, the documentation to verify all this is nonexistent, so I'm literally solving this problem like a detective...
I would love to have at least one accurate pinout/schematic... Their github doesn't even have anything of great use... Even an eagle file would be great... I mean, the board exists, it had to be made somehow, this shit does exist somewhere.

You might find this of some use for the 1.3 version info: https://ultimachine.com/sites/default/f ... al-1-3.pdf

For the digipot channel layout, it makes perfect sense if you look at it from the hardware viewpoint. The AD5206 chip has two outputs on the left side and four on the right. Look at where it is on the board. X and Y are on the left, Z is below, and E0 and E1 are on the right. So when laying out the board, the left outputs which are pins 11 (output #5) and 2 (output #6) go to X and Y, and the right outputs (pins 14(#3),17(#1),20(#2),23(#4)) serve the rest. Just as the chip itself uses whatever pin is most convenient for outputs, the board layout guy picks the digipots that allow the most direct routes and the fewest vias. The fact that the software guy then has to put things in a logical order is usually both irrelevant and the lowest-cost solution to the hardware guy.

Anyway, odds are if you want to make use of the unused digipot, you'll have to solder a wire directly to the AD5206 chip, pin 14, to get access to that signal. If you've never tacked small wires to surface mount chips before, I'd suggest getting some practice on some junk board first.

[JJPowelly]

Eric, good info... I also used the link that you provided me, however, it's still not "exactly" what I needed - but it helped... The best resource that I found was an excel spreadsheet on github of all the pins and their traces back to the 2560... It was good enough, but it did have a few errors...

As for the digipot... I don't want to run the risk of soldering a trace on the board directly... I managed to get the third extruder to work (somehow) so I'm at the point now where I really don't want to fuck up the magic... "if it aint broke"...

Right now, my biggest issue has nothing to do with motors or extrusion... My biggest issue is heating the hotend... I'm currently maxing out at around 170C in the hotend... I've nailed it down to two things, either my cooling fan is working too well, or I'm not getting enough heat to the hotend... I really don't believe I'll be able to run two heat cartridges to the hotend, so I may end up having to resist the fan or put a potentiometer on it somewhere to control its speed. I should have it figured out in the next couple of days...

[Eric]

Right now, my biggest issue has nothing to do with motors or extrusion... My biggest issue is heating the hotend... I'm currently maxing out at around 170C in the hotend... I've nailed it down to two things, either my cooling fan is working too well, or I'm not getting enough heat to the hotend... I really don't believe I'll be able to run two heat cartridges to the hotend, so I may end up having to resist the fan or put a potentiometer on it somewhere to control its speed. I should have it figured out in the next couple of days...

http://reprap.org/wiki/Diamond_Hotend#B ... his_design

In your pictures, I see no sign of the thermal isolation sheet between the hot end and the cold ends. It also prevents any fan airflow from hitting the hot end, which is probably its most important functon in this case, and why it's cut larger than the hot end itself.

If that's not the whole of the problem, measure the resistance of the heater cartridge and do the math (P = V**2/R) to make sure you're getting the wattage you expect. You could have a defective one with an internal short.

[JJPowelly]

I figured out what it was... I'm using XT60 connectors on the heater cartridge so I can do quick disconnects of the entire diamond assembly without having to take it apart... Well, one of my solder points into the XT60 connector was quite literally hanging on by a thread... I re-soldered and it is working perfectly now... It's always something stupid....

Also, today, I decided to redo to the firmware (yet again)... This time I upgraded my old firmware from Rep 0.91 to Rep 0.92... By doing this I was able to enable the "mixing extruder" feature that didn't exist in 0.91... I couldn't find any 0.92 firmware written for the Max v2 on rambo... So I think I may have the only working upgrade right now (I couldn't find any on google or on here)... It wasn't too difficult, however, the Z-probing and distortion features gave me a TON of grief... Everything else configured pretty easily. Although the UI on the controller is a little funky now... But still usable...

[mhackney]

Most of us who have upgraded to .92 use the config tool. Since my config is unique, I don't typically publish my updates but will provide it to folks on request and IF they understand what they need to do to change to support their config. The mixing virtual extruder support in .92 is very cool and works great.

[JJPowelly]

The mixing virtual extruder support in .92 is very cool and works great.

Yeah, I'm still trying to get it configured right... As of right now, when I enable it, all three extruders start extruding and cause a traffic jam in the nozzle... I'm trying to play around with the mixing python post script to see if that does anything... But I haven't quite figured it out yet.

Also, Mr Hackney, I do believe you recently sent me a friend request on FB, is that right? My pseudo-name on there is Fletcher Biceps.

[mhackney]

Yup, you accepted the friend request.

Cheers
Michael

[JJPowelly]

Michael, did you manage to get Z-probe and distortion working in your 0.92 FW? If so, can you send me your config.h? I can't get it to compile with it enabled at all... as soon as I go 1/true, I get more compiling errors than I have the fortitude to try to debug.

[JJPowelly]

In your pictures, I see no sign of the thermal isolation sheet between the hot end and the cold ends. It also prevents any fan airflow from hitting the hot end, which is probably its most important functon in this case, and why it's cut larger than the hot end itself.

I actually cut a little piece of cork and used that as insulation... That fabric was extremely annoying how it just frays out so easily, I got tired of it quickly and decided to come up with my own alternative... I had some extra cork pads laying around from when I used them to dampen the vibration in my x,y,z motors... So I cut it up to insulate the nozzle.

[mhackney]

No, I run Duet and RepRapFirmware on my Rostock Max. I have the Cyclops and Repetier on my Taz 4 cartesian printer so it won't help you.

[mhackney]

Got a Diamond in the mail this week. Now I need a mount. I got three of the E3D V6 Lite heatsinks. Eegads are they heavy. I might have to replace them with aluminum and maybe a close fitting water cooling jacket. Anyone running 3 Lites on a Diamond on a Max?

[teoman]

Get the chinese copy and chop the heatsinks to your liking.

If they are not garbage.

[mhackney]

I'll fabricate my own from scratch since I would be doing water cooled jackets anyway.

[Polygonhell]

Got a Diamond in the mail this week. Now I need a mount. I got three of the E3D V6 Lite heatsinks. Eegads are they heavy. I might have to replace them with aluminum and maybe a close fitting water cooling jacket. Anyone running 3 Lites on a Diamond on a Max?

Yes, that's what I'm running, and yes the entire assembly is on the heavy side.

[mhackney]

It's hard to believe that E3D can machine that hunk of SS for the Lite hot end for such a low price. It's really well done.

[mhackney]

I've decided to go with 3 water cooled jackets for the Diamond setup to keep things compact. Here's a drawing.

Here's a comparison with 3 E3D heat breaks:

This will be quite a bit smaller and lighter. I'll use the press in fittings like E3D uses on their hot ends. I'll press fit the 4mm OD aluminum tubing into the top for the hose connections. Short lengths of plastic tubing will connect each jacket with one serving as the inlet and one the outlet.

I'm thinking about mounting options now. I like the 3 post designs that a few folks have printed. I could use something like that which snaps over the jackets like a big C with the opening facing up and to the middle. The entire assembly could be snapped in and out easily.

[Polygonhell]

I'm using a version of the 3 post design, just makes sure you model it, the heatsinks and the platform. Or do what I did and print it 3 times.

I ended up modifying the original one, I got from somewhere because there was too little clearance between the bottom of the platform and the bottom of the nozzle, I think I removed 10mm from the legs, and that modification resulted in the heatsinks not clearing the platform, and I ended up 1/2 way between the two.
Depending on how the water blocks compare with the E3d V6 Lite heatsinks, there might be some messing about to get a mount that will work.

[626Pilot]

I really don't like the EzStruder it's not really that ez, it's bulky, heavy, hard to feed, and the filament likes to get hung up in the assembly

They should have called it the EZierStruder. It really is way easier than the gigantic monstrosity they were shipping before. It took me half an hour to assemble that thing.

Maybe this year they'll revisit the EZStruder design and make it less annoying, and more compatible with flex filament. They should really think about going with a geared extruder. They skip FAR less, and anyone who wants to run a Cyclops is gonna have zero luck with a non-geared NEMA17.

[JJPowelly]

My new design has been pretty solid so far... Only problem I have is that I put too much tension on the springs and it grips a little too tight causing the filament to get ground down by the gear... I made a few adjustments in the plunger and found the sweet spot for tension though... The hardware used in my mod is the same hardware as an mk9.

[Renha]

I have updated my file on Thingiverse, added older version without EL hoop support (but with better LED ring support because outer holes are free there), and added link to EL hoop. Link is same http://www.thingiverse.com/thing:1128037

[mhackney]

I'm building a water cooling system. This will greatly decrease the overall size of the hot end and remove a lot of mass (compared to the V6 lite).Just wrapping up the machining this week.

[Renha]

mhackney, that's best news EVER! well, after release of Diamond, but right after it!