Welcome to the SeeMeCNC Community

mhackney wrote:Yes, the pre-made units are pricy and may not be the best size for this application. The Delrin will make a great low friction bearing surface and be easy to machine and light weight. One of my friends called me "Dr. Delrin" last weekend because I constantly promote the virtues of the material! I use it a lot in my reels too.

How are you planning to hold the balls? Are you drilling into them? I'm thinking you are putting the magnet into a delrin thingy with a socket in it machined to fit the ball?

I just feel compelled to add some commentary on my RAMPS excursion...

The RAMPS controller has been around for a while (it is in version 1.4) and thoroughly tested and supported by the main firmwares out there. I was able to use the RAMPS connector that came with the LCD display to hook it up to the RAMPS board with no hassles. Then, configuration for the RAMPS was about as painless as I could imagine. Even the LCD came up and displayed "Rostock MAX" the first time. It seems the RAMBo causes a lot of unnecessary pain because of its newness and unhardened support in the firmwares out there. Hopefully this will stabilize in the near future but it was quite refreshing to work with a board and firmware that have been seasoned!

xnaron wrote:How are you planning to hold the balls?

Decorum prevents me from answering that question!

The magnet will be in a Delrin cylinder. One end will be milled with a 3/8" ball mill (an extra sharp one for the job) and the other will have a 3/8" hole for the magnet. It will be "drilled" with an endmill so the hole is flat bottomed so the magnet can get as close to the ball in the socket as possible - 1mm should be sufficient based on my calculations.

The ball will be drilled and tapped with a stud inserted. I'll likely fix the ball to the carriage and delta head since they have the lowest profile. The magnet cylinder will be attached to the rods. I am currently looking at aluminum arrow shafts. These things are incredibly stiff - much stiffer than carbon fiber tube of similar dimensions.

mhackney wrote:

xnaron wrote:How are you planning to hold the balls?

Decorum prevents me from answering that question!

LOL! I thought that sounded funny as I typed it.

I have built many tri, quad and hexa copters and found this http://www.hobbyking.com/hobbyking/stor ... 750mm.html
to be exceptionally strong and extremely light. Dimensionally it would be similar in size to the rmax arms. It actually looks much nicer than what they picture. It has a nice carbon fiber pattern.

I have a similar idea with the magnets and balls. But I was planning on printing the sockets for the magnets. Hopefully by the end of the weekend I will have a sample to post. My parts are due in tomorrow.

mhackney wrote:Yes, the pre-made units are pricy and may not be the best size for this application. The Delrin will make a great low friction bearing surface and be easy to machine and light weight. One of my friends called me "Dr. Delrin" last weekend because I constantly promote the virtues of the material! I use it a lot in my reels too.

So your intent is to machine the cups in Delrin, and mount the magnets behind them?
That was my first thought, my concern was whether the magnets would keep the ball seated though the delrin, but 10lbs is a lot of holding force.

Exactly. Delrin is a nonpermeable material - meaning it has no significant effect on magnetic fields. So the design consideration comes down to distance of the magnet to the ball. A gap of 1mm at the closest (and then decreasing from there) can be estimated. Here is an online calculator. At a distance of 1mm, assuming a flat plate of steel, the pull force drops to 3.69 pounds. Decrease that to .5mm and you get 5.39 pounds.

Also consider, the ball is nestled in a cradle of Delrin that minimizes the possibility of dislodging sideways. In fact, a cradle at least 25% of the diameter of the bearing is needed to insure accurate centering. With a cradle like this, considerable sideways force is required to dislodge. Even gripping the ball and trying to pull it free requires a lot of force.

mhackney wrote:Exactly. Delrin is a nonpermeable material - meaning it has no significant effect on magnetic fields. So the design consideration comes down to distance of the magnet to the ball. A gap of 1mm at the closest (and then decreasing from there) can be estimated. Here is an online calculator. At a distance of 1mm, assuming a flat plate of steel, the pull force drops to 3.69 pounds. Decrease that to .5mm and you get 5.39 pounds.

Also consider, the ball is nestled in a cradle of Delrin that minimizes the possibility of dislodging sideways. In fact, a cradle at least 25% of the diameter of the bearing is needed to insure accurate centering. With a cradle like this, considerable sideways force is required to dislodge. Even gripping the ball and trying to pull it free requires a lot of force.

I guess you could probably get down into the 0.005in range relatively easily. I'll be interested in how much force you really have, because the percentage of the sphere in close contact is going to be small.
UHMWPE would be a lot cheaper than delrin and probably comparable in this application.

The call me "Dr. Delrin"!

The cost of the material is inconsequential, there is so little of it. Delrin machines like a dream and I have years of experience using it. I also have lots of it in my shop already and no UNHMWPE. AND, it comes in black!

Like I said above, I've already made one of these cups with about .1mm of clearance. I have a difficult time grasping and removing the ball. Not quantitative, I know, but there is a lot of force there.

Steel balls? Arrow shafts? Better hope you don't drop one on your borosilicate glass, or crash into the build plate due to a software glitch. That could get expensive.

- dan

Did you order stainless steel balls or just hardened chrome steel balls? How are you planning to drill into them? Assuming center was found I was thinking of using a vbit to spot the hole and then drill it out with the correct bit. I thought about also doing it on the lathe if I could get the ball chucked up.

Most stainless is not magnetic so these are steel. I will spot them and drill. I can do this on either the lathe (with a 5C chuck) or mill (also a 5C chuck). Balls this size can be drilled easily so I don't anticipate a challenge.

mhackney wrote:Just wanted to capture this in my build thread:

KISSlicer 1.1 RC2 (released today on Mac)
.5mm skin
2 loops
50% infill
.55mm extrusion width and infill extrusion width
.25mm layer thickness
1.0 / 45° Jitter (new feature as of today, just learning how to use it!)
195°C nozzle
60°C heated bed on blue tape
10mm prime/suck
5mm wipe
20mm/s speed
1.5mm min jump
5mm trigger

On the printer side you can get the accel and extruder speeds on my EEPROM screen shot:

Screen Shot 2013-03-26 at 7.47.42 PM.png

stepper digipot 175

Let me know if I missed anything!

Oh yeah, while it was printing I was dancing up and down with a counter clock wise spin!

Do you KISSlicer to just generate the G-code and then export to repetier or also print?

I don't understand the question aeroula.

Flateric wrote:http://www.ebay.ca/itm/221168698861?ssP ... 757wt_1398

Have you seen these yet? Looks like a nicer cleaner way of heating the hotend. Less chance of shorting out too.

Can 2 of them be used to replace both factory resistors sent by Seemecnc? Or only 1 is allowed to be used at a time? I'd like to put 2 of them on my hot end. Take out the 2 resistors and replace them with these clean ones. Does any one know if 2 of them can be put on the hot end at the same time just like with the 2 factory resistors?

Not sure why you would want to do that but there is no problem with doing it. They each draw about 3.33Amps at 12V. One is more than sufficient to heat the nozzle.

mhackney wrote:Not sure why you would want to do that but there is no problem with doing it. They each draw about 3.33Amps at 12V. One is more than sufficient to heat the nozzle.

Thanks mhackey. I wasn't sure if we had to install 2 of them. I will go with 1 and a platinum Thermistor on the other side.

Also, are you still printing using the KISSlicer ini setting files you posted on this thread? Or have you changed from those settings, if so could you provide us with your recent KISSlicer settings?

I posted my current KISS settings in the KISSliser topic a few days ago. Those are what I use but I do tweak things for specific prints, etc.

mhackney wrote:I posted my current KISS settings in the KISSliser topic a few days ago. Those are what I use but I do tweak things for specific prints, etc.

Thank you. Downloaded and installed into kisslicer. Only change I made was 100% infill. Am working on some prints requiring 100% infill. I was trying KISSlicer for the last 3 days and could not get it to print properly for me. It would always start the prints on me pressing down on my glass bed with painters tape, even though my nozzle was perfectly leveled and calibrated. Hopefully with your settings I won't get that anymore. Will let you know.

I'm printing defcad's V5 lower, used Slic3r and enabled honeycomb support. Boy is it sucking up a lot of filament, but looks so cool cause it looks like a scaffolding structure built around the part itself. Have you ever used the honeycomb support feature?

I wanted to switch to KISS though cause you told me they produce better prints and better support structures. Is that correct?

Well, it was easy enough to get the RAMPS board set up and firmware configured. Jogging manually is smooth and indistinguishable from RAMBo control. After verifying steps-per, temp calibration (new thermistor), and extruder movement, I attempted to do some simple .5mm calibration cube prints. That's when it all came apart!

There is a very noticeable stutter or jerkiness to all 4 steppers. The machine just does not sound good at all and the delta head actually vibrates. This seems like a typical over-amping the stepper problem but that is not the case. I've run RAMPS on 2 other machines and know how to set the current limit pots - which I did to as close to the firmware setting from RAMBo as I could (.9A). The motors do not get hot when just sitting while energized. And, a G0 Z0 F2500 and subsequent homing are as smooth as silk - as are jogs on the X and Y.

The one theory I have is that maybe I am not getting enough Amps to the RAMPS from my 12V supply. The RAMPs has 2 power connectors - one for the heaters and the other for motors, etc. I only have 1 set of leads from the supply to each. The hot end and heated bed heat up fast with no noticeable differences from when I had the RAMBo.

I have checked my EEPROM settings against a known good set from RAMBo (I did a screen capture of the EEPROM window) and checked and rechecked the Repetier firmware Configuration.h. It will likely turn out to be something simple. But, I can't afford not to have this printer running so I did order a replacement RAMBo "just in case".

I am doing some more testing and even manual jogging has a very discrete stutter when I slow the jog rate down. I wrote a quick gcode exercise program so I can test. I have 16 microsteps configured on the RAMPS (3 jumpers per driver) and in firmware and the movement is correct - tell it to move 10mm and it does, just with obvious discrete quick steps.

FWIW I ran my MAX on a RAMPS board for about a week while I was waiting for fuses to arrive for the RAMBO and it does work.
I was running with Marlin at the time.

Thanks. I know the homebrew Rostock and Kestral guys run with RAMPS. This is a known good RAMPS board although I do have another new one to try if I run into a dead end. I suspect it will be some subtle setting. I might try Marlin to rule out issues with firmware.

Well, chalk this problem up to "operator error". Basically, I thought this RAMPS had Pololu stepper drivers like my first RAMPS. It doesn't, it has StepSticks. The current limiting pot / circuit is not the same as the Pololu! I thought I had set the stepper current to .9A but it was actually set to something like .2A. Doh!

Anyway, it is fully operational on RAMPS now. And my new replacement RAMBo arrives tomorrow. And, the parts I ordered to try to fix my broken RAMBo arrived today! Too many controllers, too few printers.