Latest comments

  • It's alive!
    Lee
    Hi, any chance of you sharing your config code for this project.

    Read more...

     
  • It's alive!
    Lee
    Hi Danny, great piece on converting an Overlord to use an MKS Sbase. I'm replacing my Overlord MB ...

    Read more...

     
  • iHobby Expo 2008
    Alison
    Sweet blog! I found it while surfing around on Yahoo News. Do you have any tips on how to get listed ...

    Read more...

     
  • Getting back into it
    online word Counter
    Everything is very open with a very cleaar description of the challenges. It waas really informative.

    Read more...

     
  • TiVo HD upgrade
    BHW
    I blog often and I really thank you for your information. This article has truly peaked my interest.

    Read more...

Danny's Corner

Designing and printing

As of late, I've been designing and printing all sorts of things. No theme between them. The only tie that binds them, is that I'm making an effort to design and make more things. It somehow calms me, oddly enough. Is this what it feels like to have a mid life crisis?  =)

This one has got to be oddest of the bunch. A small clamp that fixes itself against BaT's roof rail and acts as a perch for an antenna.  And well, the antenna's only purpose in life is to hold a Jack in the Box antenna topper...

Small projects though they may be, they do bring a smile to my face.

Giving credit, sort of

Small printer upgrades are still happening. Amazing just how much I can polish a turd. :)

I ordered a handful of 8 diode filters specifically made for TI DRV8825 stepper drivers that are configured for micro stepping.  These random, and rather highly specialized little PCBs, fix a rather problematic behavior of the TI chips under circumstances that are common when 3D printing.  The original solution was published by a friendly sleuth named cabristor.  Which then got copied and sold by the thousands with not even a whiff of who came up with the fix. That sucks. The images on his site were even being direct linked in ads to sell the darn things.  He's since taken down his site, which is a shame, but I get it.  Me being me, I took a PDF of it for safe keeping and is linked here.  Hopefully cabristor comes back online with his website.

I designed and printed up a small bracket that holds the diode filters in place on one of my printers. These diode filters do consume some power and convert it into heat. Luckily, I've got a good sized case fan sending quite a bit of air over them.  With these installed, my print quality noticeably improved. There are plenty before and after examples on the internet so I won't dogpile. Safe to say though, if you're running stepper drivers based on TI DRV8825 ICs, are using microstepping, and making slow movements, these are definitely worth looking into.

Who knew, Pyropen Jr can do SMD

A long while back, I was gifted a set of Trinamic TMC2100 stepper drivers. This was long enough ago, that a batch were created with mislabeled pins. But, the internet being what it is, someone else figured it out and left bread crumbs on how to fix the issue.

And the fix if you have Anycubic TMC2100 drivers that look like this, is purportedly to bridge CFG3 and get the MS1 pin to ground.  This will get you the much beloved combo of 16 micro steps, interpolation, and Spreadcycle.  I haven't quite tested it yet, but will soon.

The biggest issue I had, was bridging the dang pads of CFG3 though. I got lucky on my first attempt. I managed to drag the solder to bridge the two dots. It didn't happen again. No matter how much flux I added, the dang solder simply didn't want to do that again. So, plan B. Solder paste + a tiny piece of flattened wire + my butane soldering pen.

All things considered, I think I did pretty well for my very first try. I was pretty skeptical that my Pyropen Jr would be able to get the job done with the hot air tip. But it sure did. I wouldn't want to start a big SMD project with it, but in a pinch it worked a treat.

 

Printer update

I mentioned how my printers had been more "fun" to mess with than my Shapeoko in the last post. A lot of that fun was me simply learning and gaining experience in various bits of CNC motion assemblies, wiring, and other nuances. However, I think I've gone as far as I can with my two printers without more or less starting from scratch.

Still, I tinker on. I've managed to attach an E3D Titan Aero replica to my blue printer. It wasn't a particular hard bracket to design, but it sure did require a lot of rewiring. After several prints, it's pretty clear that my print quality has gone way up at the expense of print speed. Not too bad a trade off to me.

The blue printer is going to get some diode filters to "fix" the issues with DRV8825 drivers and microstepping.

The black printer will eventually get Trinamic drivers. It will also get a new cold end extruder, but I haven't made up my mind if I'm going to go with a Titan Aero setup like the blue printer.  I'd like to keep one of them nice and speedy.  I may look into getting some Capricorn tube to get the retraction issues under control.

It's alive!

That right there is the result of a lot of tweaking and represents my first truly successful print with the new MKS Sbase microcontroller. Been a long road, but we're up and running. More tweaks to be made, but I'm really close now. Time to recap.

The MKS Sbase lacks a way to switch off power to itself. The commands are available, but there's no physical hardware like a relay or what not to do it. So, I tried to put one together with mixed results. Here, I'm trying to use an LM78xx 5v voltage regulator. In my tests with a 9v battery it did great. The temps creeped up, but topped off at around 104F or so. However, once i switched over to 24v, the temps went through the roof. I didn't bother to figure out how hot, it simply got way too hot to touch.

I then went and designed around a DC step down converter. Much better. But this too ultimately failed because my relay is blown. It seems to be stuck in a NC state on the NO side. Really weird. I'm going to try a few different things to it, but in the meantime I've got more 5v low relays inbound. The rest of the wiring was pretty uneventful and not worth writing about. Plugs were cut off and new ones crimped on.

Another hurdle was my hotend setup. I inadvertently snapped the all metal heatbreak I was going to use. The part wasn't made to spec and the shoulder of the transition zone was too thin. But, I do have quite a few (5) 4.1mm throughout heatbreaks and I'm going to be printing almost exclusively in PLA formulas, so I'm good right?  Turns out I was wrong.

A E3D style hotend seals itself when the heatbreak and the nozzle clamp against each other in a heater block. One of the two differing metals is going to yield and result in a nice seal. Typically it's the brass nozzle that yields against the stainless steel heatbreak. When the mating doesn't happen, you end up with leaks in two places. It didn't matter if I heated up the nozzle, or wrenched down on it as hard as I could, it kept on leaking. So, I dug a little deeper.

A 4.1mm throughout heatbreak leaves very little surface for the brass nozzle to mate against. That's a small problem but something to keep in mind. But since the heatbreaks I bought haven't gone through any meaningful quality control, I ended up with a bit of a mess. Only two of the heatbreaks had what I thought was a decent enough mating surface.

It also turned out that out of the 5 heatbreaks I had, only 3 had mating surfaces that were perpendicular to the vertical axis. One of these off balance heatbreaks was in the hotend assembly. All of the heatbreaks also weren't quite threaded properly either. In fact, really only one was up to par. The one in the middle above. It had the best 6mm threading, a nicely sized mating surface, and was perpendicular.

I can really see now why the argument exists about buying a real E3D v6. Their quality control saves quite a few headaches.

While I was diagnosing my hotend issues, I was using 3D Benchy as my test print. One of the small issues I also had to figure out was layer cooling. The print on the left had no layer cooling and suffered for it. The two others did have layer cooling, but suffered from the unpredictable nature of the leaky hotend. 

Things are looking up now for my Black Devil Overlord Pro. Configuring Smoothieware wasn't too bad. I got the simple bed leveling and the mesh leveling routines to work without much fuss with my microswitch Zprobe. I did have to cut off the roller wheel, but after that it was smooth sailing. I'm contemplating getting another Smoothieware based microcontroller for the Gem Blue Overlord Pro. We'll see.

Edit: link to what my blue printers SDcard contents look like.