Insight Calbox FAILURE!

I want to try and get my calibration better for the support and I thought I’d give the Insight calibration a try. You get to it from the main menu – modeler – properties, then click on the micrometer (hey I know what they are) icon. That opens a dialog with three different tabs. I shot some screen images of them, but blush, they are in an another computer. One of the nice things, is it would querey the FDM and get what values are in there.

Anyway I clicked the FDM Send icon to send the Calbox to the FDM and run it. Jog away from corner in X&Y, Set the Z, and un-pause. The FDM zips around making two Calboxes, but they aren’t there. Just a goey mess on the tip — nothing stuck to the foam,

Calbox on the tip
Calbox on the tip

I did this several times, lower the Z each time thinking I was too far away from the foam. The last try, I could see the tip plowing thru the foam (damn… I just turned it over for a FRESH side)….

Plowing thru foam
Plowing thru foam
Still plowing - not sticking
Still plowing - not sticking
Scored foam -- no Calboxes!
Scored foam -- no Calboxes!

This is really weird! I did calboxes before with the MG-47 with FDM Send……

I sliced and ran a Mendel part to see if I futz something up — It ran okay

Front View
Front View
Left View
Left View

Another Stratasys mystery….

—-John

A try fixing the other head

While I had tools out, I thought I’d take a crack on the FDM 2000 head. It had a big puddle of melted model material over the inlet buffer that need to get cleanned up.

I started like I did with the FDM 1650 head

A little different
A little different

The head is a little different, with only one pair of drive rollers for each filament. The guide tube has been extended to reach the eatra distance (and seems to collect crud inside too). There’s the two outside capscrews that hold the block down. One screw is long, and the other short.

BIG Puddle
BIG Puddle

I knew I was going to need room so I pulled the other out too.

Making Room
Making Room

It was only half the work as one screw wasn’t there!

One short one missing
One short one missing

The material is a silver-grey, and not as nice as the tan polyimide. Also, when I started using the hot-knife to clear material I found it had a lower melting point too.

A little cleared away
A little cleared away

 

Getting closer
Getting closer

I think they were using the ABS+ model material, as this was a lot tougher to remove than the ABS Breakaway Support.

As I got further along, I found I didn’t have some Mini-Torx drivers that are needed for this job, and I ended up putting it back together ’til I get some. Here’s some photo’s of the guts…

>1
>1
>2
>2
>3
>3
>4
>4
>5
>5
>6
>6
>7
>7
>8
>8
>9
>9
>10
>10
>11
>11
>12
>12
>13
>13
>14
>14

 

—-John

1st run after head repair

I ran a tray of parts I’d run before to check out the head repair I did. When I was starting the job & I tried to ‘LOAD’ the model material, I kept on getting torque errors. That made me power down & back up to clear the error. In the end I just gave up on trying to load the model material — some was squirting out until the error, so I knew stuff was in there. I just proceeded with the run. There was a glich though, I had powered off-on too quickly and the keypad & display weren’t working. I couldn’t adjust the flow down to the 90% that has been working okay. Another thing, I couldn’t set the ‘Auto Shutoff’…. but I let it go:

Front View
Front View
Left View
Left View

The parts ran okay, and I didn’t get any torque errors when the model material was running. There were some burnt globs and hairs from mushing so much material around. Here’s a few photos showing the problems:

Hairs & Burnt Globs - 1
Hairs & Burnt Globs - 1
Hairs & Burnt Globs - 2
Hairs & Burnt Globs - 2

Still quite usuable parts…. and I’m glad everthing worked okay!

—-John

Removing rat’s nest & cleaning up…

Things aren’t getting done without the printhead, so I tore into it today. It seemed a bit too packed in there to just cut filament:

Before Starting
Before Starting

Moving the stepper wires, uncovers the capscrews holding the drive to the mounting plate:

Uncovering Capscrews
Uncovering Capscrews

Only the two outside screws need to be removed:

Two Screws backed out
Two Screws backed out
What was underneath...
What was underneath...
Melted mess around inlet buffer
Melted mess around inlet buffer
Floor View
Floor View

 

This mess was real stuck down
This mess was real stuck down

 

Heating knife with mini-torch
Heating knife with mini-torch

 

Big stuff removed
Big stuff removed

 

Turning big gear got filament cleared
Turning big gear got filament cleared

 

A little father
A little father

 

Saw some plastic under model drive
Saw some plastic under model drive

 

Small drill to clear some support out
Small drill to clear some support out

 

Larger got the whole plug
Larger got the whole plug

 

0.053 Pin checking alignment
0.053 Pin checking alignment
Cleaned with scale for measurement
Cleaned with scale for measurement
Floor view with scale
Floor view with scale
Back together
Back together
What came out
What came out
And it’s back running. I was getting some torque limit errors loading the model material, but we’ll see how it goes.
—-John

A try to catch up Posts….

I’ve been falling behind in my posts and this is an attempt to catch up.

Back on 3/2/11, I did some special parts to test the backlash that I think I have in the FDM 1650 X-Y Stage. There were three blocks – two were identical, square with a rib running partway up each side to cause the machine to approach the vertical side from the opposite rotation; one of the blocks was rotated 45 degrees to X&Y — the third block was round. The photos:

Front View
Front View
Left View
Left View
Hanging Edges
Hanging Edges
Tips After Run
Tips After Run

Anyway, Block #1 had the step on the front side but not on the back (the backside approach doesn’t change direction). There was no step on the left or right sides to speak of. Block #2 didn’t have any steps on any of the sides, but wasn’t square. Block #3 was out-of-round. I wanted to measure these on the CMM, but never found the time to measure them.

I think this shows I need to tighten up the Y-Axis wires to as tight as the X Axis wires are. I looked at the X-Y Stage that is out of the machine, and there are two adjustments for the Y-Axis and only one for the X Axis — I wonder why? I think I would like to do the adjustment out of the machine, so it will have to wait until summer time. Some photos of the X-Y Satge on the bench:

Cable Drum On Stepper
Cable Drum On Stepper
Left Side Adjusters (On Bench)
Left Side Adjusters (On Bench)
Right Side Adjusters (On Bench)
Right Side Adjusters (On Bench)

The cabling thru the different pulley rollers is immposible to grasp in photos, so I won’t stick them into the post. I did upload them into the gallery though (not sure how you get to it though?).

I did a number of different runs of Mendel parts. All looked pretty good considering my backlash.

Earlier in the month HaveBlue suggested lowering the flow rate to reduce some of the excess material I was getting. I did a set of calboxes at differnt flows with the MG-47 material at 255 degrees. You can read the results in the photos:

50% Flow - Slow to start
50% Flow - Slow to start
50% Flow - 0.016 Road
50% Flow - 0.016 Road
60% Flow - 0.020 Road
60% Flow - 0.020 Road
70% Flow - 0.023 Road
70% Flow - 0.023 Road
80% Flow - 0.028 Road
80% Flow - 0.028 Road
90% Flow - 0.030 Road
90% Flow - 0.030 Road

Definately shows a linear relationship, but below 60% Flow it seems to take to long to get the squirt started.

I’ve been getting less burnt globs on the parts running at 90% flow. When I run less than 90% the edges are unattached too often.

Some vertex parts with feet (both styles) were printed for a respooling thingy:

Front View
Front View
Left View
Left View

Also, got the printer relocated (but still on a dolly). The new setup:

Window View
Window View
Door View
Door View
Other Side Window View
Other Side Window View

I didn’t get the machine off the dolly yet, as I’m not entirely certian this placement is going to be to efficent. Since the photo’s I’ve added a ‘Photo Frame’ to the part’s table with a slide show, and a Laptop & Dock to the top of the printer for the ‘CAM’ upload & preview. The camera is located on the bottom window sill looking into the chamber. The serial upload computer also has a ‘Desktop’ CAM that uploads seperately so I can view where the build is. The only shortcoming to the CAMs is they are generating a lot of picture files (one pic every 5 seconds in a 21 hour build) …. at least they are small files!

This post doesn’t bring me up-to-date, just a bit closer. I’ll try to get another post later this week.

—-John

Run Finished 3/1/11 & bACKLASH

The last tray of parts run at 90% Flow, 255 Degree model tip temperature, Insight sparse build were done when I got in today. The pictures:

Front View
Front View
Left View
Left View

Still have the unattached outside edge, tried to get a photo below:

Unattached Edge
Unattached Edge

And the tips were gummed up like the previous run:

Gummed Up Tips
Gummed Up Tips

I was looking at one of my one-off parts from the weekend — there is a large (Tall & fairly wide) angled surface that had a rib 3/4 of the way up. The one side of the rib (X- side) the angle was a nice even slope, but the other side (X+ side) had a 0.01″ step at the top of the rib. This seems to indicate to me that the FDM1650 has a serious backlash problem.

Step in angled surface
Step in angled surface

This backlash could also possibly cause the unattached edges I have been seeing. I’m going to model up a couple of three backlash STL test solids to stick in the next run. (I’ll post my STL when I post the results in case anyone else wants to test their machine)

I’m not looking forward to tightening up the Automove, as that’s going to take a fair amount of time (which I don’t have) and stop the flow of parts & tests. I think this is one thing I’ll limp along with until summer…..

{A late thought after I posted this — this could also be STL model has reversed normals on the surface, and Insight going to outside of surface rather than staying inside — now – how to check that?}

–John

Run finished 2/28/11

This run finished yesterday morning (90% Flow, 255 Degree model temp, Insight sparse build) looked better than the lower modeling temperature. The pictures:

Front View
Front View
Left View
Left View

The tips seemed a bit gummier but not as bad as the solid builds at 100% Flow:

Gummier Tips
Gummier Tips

The outside roads seemed a little better attached at the higher tem, but it just may have been the parts I was running.

I sliced up another platter of parts to run at the same settings, and got it started — should be ready when I get in today.

–John