Novak Conversions Jeep Wrangler TJ radiator

Sab-a-dab-a-doo! The back-country LJ build has officially started

Today's entry is an update on the center console insert project. As mentioned in the previous post on the subject, I had some revisions to do in order to clear everything else occupying space inside the console. Those items are the subwoofer, the step slider control module, and the cutting brake handles in the fully-engaged positions. If anyone reading this has made a complicated model in Fusion before, you know the dread you face when you want to make changes and know that you'll be presented with many errors as references break and you have to slowly work through the part history fixing things. Well, it took me all evening to work through all of that, and at about 9pm, I started a test print of this latest design:
View attachment 661849

Since my last test print in PETG HF was a disaster using the PLA/PETG support at the interfaces, I decided to switch back to PLA and the PLA support filament because I've had much better luck with clean, easy separation between the base material and the support material using that combination. I also went to .24mm layer thickness for a quicker print and because @Mike_H discovered that layer thickness affects support removal, and thick layers are easier to remove support. Here's a picture on the print bed just before I removed it:
View attachment 661850

And this is how support with the support filament at the interface is supposed to work (look closely, and you'll see the layer of white support material on the supports on the right below):
View attachment 661851

Nice and clean, and easily done! Sadly, I'm not sure if it's the layer thickness, the change in materials, or something else that did the trick. But I'll take it!

Here are pictures of the top and bottom of the console with everything installed:
View attachment 661852

View attachment 661853

I have a bit more tweaking to do. I fixed all the clearance issues, but the interface with the top opening on the console is too tight. I'll have to do more battle with broken references in Fusion to get more clearance before I print the final piece - hopefully very soon! I also made a little more progress on a model for the slot dust covers. Stay tuned...

Looks great! Glad you had some success. Let us know how the Support for ABS turns out. The TDS says its HIPS and I have a roll of that for supporting my tail light housings. Haven't played around with that too much, since I've revised it to a point that the support is kind of a non-issue now...but cleaner removal is always welcomed.
 
  • Like
Reactions: sab
I worked on three different parts to print on the Bambu today: 1) the center console insert, 2) a dust cover for the slots in the center console that the cutting brake levers will ride in, and 3) plugs for the large holes in the top of the front frame-side spring mounts.

Center Console Insert
After some frustrating work, I was able to remove all the support from the PETG HF test piece and fit it up, and as I suspected, I have some revisions to make in order to clear everything. That is a work in progress, and I'm also making some changes regarding the difficulty removing the supports. I don't have any pictures to post, yet.

Dust Cover for Cutting Brake Lever Slots
I thought of this a few days ago. For months, I've been contemplating what to do to keep things from falling into the console, never to be seen again, through the slots I cut for the cutting brakes. I had thought about brushes, like the parking brake lever, but that'd be a bit cumbersome to make look good. It suddenly occurred to me that I can print a dust seal in TPU and try that. So, that was my first project of the day. Here's the model I ended up with:
View attachment 661724

Here's the printed part:
View attachment 661726

The slot is purposely off-center because the lever is off-center in the slot (hey, nobody's perfect!) I used TPU 95A HF with PLA as the support for the inner lip, and unlike the center console insert, the support easily peeled off with no residue. Yay! Here it is installed:
View attachment 661727

That was a "proof-of-concept" piece, and I think I can make it work. I'm going to have to match the contours on the left one in the picture above, for sure, because it won't fit the goofy top and bottom of the slot (I tried). I'm also going to try TPU 90A, which is a bit softer. The TPU 95A HF seems like it will work fine, but I'd like to try softer, just to see if it works better (or worse).

Plug for Top Front Frame-side Spring Mount
@sierradmax requested a print file for this part, and I saw his request this morning. I wasn't even aware that there was a huge hole up there, and I can see the benefit of having it plugged. So, I went to work. It's actually a pretty easy part to model, and since I was already printing TPU 95A HF for the slot cover above, it was easy to print at the same time as the slot cover. Here's the design:
View attachment 661729

I put four small notches (circled above) in the top so that a screwdriver can be slipped in to remove it, but I don't think it'll be necessary. I was able pop it out with my fingers. There's a small lip to hold it in, too. Here are the two pieces I printed for the LJ:
View attachment 661730

And here's a shot of it installed:
View attachment 661731

If you'd like to download the files for the plug, I put it over here. Merry Christmas!

I'd like to start another test print of the center console insert this evening, but I have a fair bit of modeling to finish in order to do so. It may have to wait for tomorrow. Stay tuned!

I was looking at that stupid hole without a plug on top of the front spring perch again today. Receptacle for dirt and moisture. Formula for rust city. Been there and fixed them. Very cool plug. Now I need a Bambu. I can’t avoid all the discussion about them. My 21yr old son is working on his engineering degree and taking some solid works courses on the side. He wants to go in halfsies with me on a printer. I told him we should go for the ~$2k version and he was like, “No, like $400-500.” 😂
 
I was looking at that stupid hole without a plug on top of the front spring perch again today. Receptacle for dirt and moisture. Formula for rust city. Been there and fixed them. Very cool plug. Now I need a Bambu. I can’t avoid all the discussion about them. My 21yr old son is working on his engineering degree and taking some solid works courses on the side. He wants to go in halfsies with me on a printer. I told him we should go for the ~$2k version and he was like, “No, like $400-500.” 😂

The college he’s attending doesn’t have one?
 
The college he’s attending doesn’t have one?

Yes, it does, but he can get it free directly through Solidworks, get the certifications he needs for credit, keep his course load where he wants it, and graduate in Spring 2027.
 
  • Like
Reactions: FarFire70
Very cool plug. Now I need a Bambu
Or a friend with a Bambu. I have a mis-matched pair of plugs left over. One is the prototype and the other is the final design. There are only two differences between them, and they don’t affect function. The differences are: 1) the prototype has a solid center, and 2) the screwdriver notches are deeper on the prototype. Once installed, you can’t tell the difference. If you want them, PM me your address, and I’ll send them over.

My 21yr old son is working on his engineering degree
An excellent career choice, if I don’t mind sayin’!

Edited to add a picture of the two leftovers (the prototype is dirty from installation on a dirty Jeep):
1766409344863.png
 
Last edited:
Overnight, I did a test print with ASA using Support for ABS filament at the interfaces (3 layers). The Support for ABS filament has been in my AMS for about a month (pretty much since I bought it), and I dried it when I put it in there. However, I believe that it had absorbed some water because the ASA printed beautifully, but the Support for ABS was stringy and crumbly (note all the "crumbs" on the build plate):
IMG_8470.JPG


The print still worked out. Support was very easy to remove, but with the crumbly support material, the surfaces touching the support were rough. However, it was a successful test. And I learned something that surprised me. This printer was drawing a lot of current when printing the ABS because the bed and chamber were heated. My wife is a glass artist and fired up her propane/oxygen torch, which gets its oxygen from two oxygen concentrators running in parallel that each draw over 10A (they're on separate circuit breakers). One of the concentrators was on the same circuit as the printer, and not long into her work, that breaker popped - about 80% through my print. It was actually far enough along that I would have been able to check fit as-is. However, when I reset the breaker, the printer remembered where it was in the print and let me resume without a hiccup. Cool! I thought the print was done at that point. Bravo, Bambu!

I dried the support material after the last layer using it was printed - while the printer completed the test print using only ASA filament. I'm hopeful the drying will solve the crumbly/stringy issues. As for the part, I tweaked the opening length and width for the remote pocket because it fit tighter in the ASA part, and I adjusted the mounting hole locations for better fitment. The final part (fingers crossed) is printing now. It's a very long print, even using variable layer thickness. It'll be done Friday after lunch. :oops:

And here are all the test prints done to get this thing designed:
IMG_8471.JPG


What are not shown are the countless paper templates I printed out to ensure my design matched the odd surfaces correctly. I must have close to 100 hours of design into this silly part! It was well worth the effort - the part fits darn near perfectly, my "design itch" was satisfied, and I've learned a lot more tools in Fusion during this process.

I'm still working on the dust seals for the cutting brake slots. I've spent two days modeling the front of the console to make form-fitting dust seals. Here's what I have so far (the one on the right doesn't have a slit in it, yet):
1766619665362.png


Designing those was very challenging. First, I had to model the console because the parts need to fit on it. And in the process, I had to learn a LOT of new Fusion tools to make these due to the compound curves on the front of the console. Lots of card stock templates were printed to make sure I had the compound curves modeled correctly. Early on in the process, I did a few test prints, and think I ran into a moisture issue with the TPU, too, because the last print was a disaster. Filament loops hanging everywhere! Before I try a print of the full seals, I'll need to do a few tests to get the dimensions dialed-in. If they don't fit perfectly, I fear they'll fall off when moving the cutting brake levers. Once I get the basic dimensions down, I can try a full print.

Although it may seem like it, I promise this hasn't turned into a 3D printing build! I have plenty of other tasks to go for this build - once the 3D printing binge ends. Stay tuned...
 
Overnight, I did a test print with ASA using Support for ABS filament at the interfaces (3 layers). The Support for ABS filament has been in my AMS for about a month (pretty much since I bought it), and I dried it when I put it in there. However, I believe that it had absorbed some water because the ASA printed beautifully, but the Support for ABS was stringy and crumbly (note all the "crumbs" on the build plate):
View attachment 662450

The print still worked out. Support was very easy to remove, but with the crumbly support material, the surfaces touching the support were rough. However, it was a successful test. And I learned something that surprised me. This printer was drawing a lot of current when printing the ABS because the bed and chamber were heated. My wife is a glass artist and fired up her propane/oxygen torch, which gets its oxygen from two oxygen concentrators running in parallel that each draw over 10A (they're on separate circuit breakers). One of the concentrators was on the same circuit as the printer, and not long into her work, that breaker popped - about 80% through my print. It was actually far enough along that I would have been able to check fit as-is. However, when I reset the breaker, the printer remembered where it was in the print and let me resume without a hiccup. Cool! I thought the print was done at that point. Bravo, Bambu!

I dried the support material after the last layer using it was printed - while the printer completed the test print using only ASA filament. I'm hopeful the drying will solve the crumbly/stringy issues. As for the part, I tweaked the opening length and width for the remote pocket because it fit tighter in the ASA part, and I adjusted the mounting hole locations for better fitment. The final part (fingers crossed) is printing now. It's a very long print, even using variable layer thickness. It'll be done Friday after lunch. :oops:

And here are all the test prints done to get this thing designed:
View attachment 662457

What are not shown are the countless paper templates I printed out to ensure my design matched the odd surfaces correctly. I must have close to 100 hours of design into this silly part! It was well worth the effort - the part fits darn near perfectly, my "design itch" was satisfied, and I've learned a lot more tools in Fusion during this process.

I'm still working on the dust seals for the cutting brake slots. I've spent two days modeling the front of the console to make form-fitting dust seals. Here's what I have so far (the one on the right doesn't have a slit in it, yet):
View attachment 662458

Designing those was very challenging. First, I had to model the console because the parts need to fit on it. And in the process, I had to learn a LOT of new Fusion tools to make these due to the compound curves on the front of the console. Lots of card stock templates were printed to make sure I had the compound curves modeled correctly. Early on in the process, I did a few test prints, and think I ran into a moisture issue with the TPU, too, because the last print was a disaster. Filament loops hanging everywhere! Before I try a print of the full seals, I'll need to do a few tests to get the dimensions dialed-in. If they don't fit perfectly, I fear they'll fall off when moving the cutting brake levers. Once I get the basic dimensions down, I can try a full print.

Although it may seem like it, I promise this hasn't turned into a 3D printing build! I have plenty of other tasks to go for this build - once the 3D printing binge ends. Stay tuned...

I’m enjoying your progress on this. Both learning printing and your design process.
 
Overnight, I did a test print with ASA using Support for ABS filament at the interfaces (3 layers). The Support for ABS filament has been in my AMS for about a month (pretty much since I bought it), and I dried it when I put it in there. However, I believe that it had absorbed some water because the ASA printed beautifully, but the Support for ABS was stringy and crumbly (note all the "crumbs" on the build plate):
View attachment 662450

The print still worked out. Support was very easy to remove, but with the crumbly support material, the surfaces touching the support were rough. However, it was a successful test. And I learned something that surprised me. This printer was drawing a lot of current when printing the ABS because the bed and chamber were heated. My wife is a glass artist and fired up her propane/oxygen torch, which gets its oxygen from two oxygen concentrators running in parallel that each draw over 10A (they're on separate circuit breakers). One of the concentrators was on the same circuit as the printer, and not long into her work, that breaker popped - about 80% through my print. It was actually far enough along that I would have been able to check fit as-is. However, when I reset the breaker, the printer remembered where it was in the print and let me resume without a hiccup. Cool! I thought the print was done at that point. Bravo, Bambu!

I dried the support material after the last layer using it was printed - while the printer completed the test print using only ASA filament. I'm hopeful the drying will solve the crumbly/stringy issues. As for the part, I tweaked the opening length and width for the remote pocket because it fit tighter in the ASA part, and I adjusted the mounting hole locations for better fitment. The final part (fingers crossed) is printing now. It's a very long print, even using variable layer thickness. It'll be done Friday after lunch. :oops:

And here are all the test prints done to get this thing designed:
View attachment 662457

What are not shown are the countless paper templates I printed out to ensure my design matched the odd surfaces correctly. I must have close to 100 hours of design into this silly part! It was well worth the effort - the part fits darn near perfectly, my "design itch" was satisfied, and I've learned a lot more tools in Fusion during this process.

I'm still working on the dust seals for the cutting brake slots. I've spent two days modeling the front of the console to make form-fitting dust seals. Here's what I have so far (the one on the right doesn't have a slit in it, yet):
View attachment 662458

Designing those was very challenging. First, I had to model the console because the parts need to fit on it. And in the process, I had to learn a LOT of new Fusion tools to make these due to the compound curves on the front of the console. Lots of card stock templates were printed to make sure I had the compound curves modeled correctly. Early on in the process, I did a few test prints, and think I ran into a moisture issue with the TPU, too, because the last print was a disaster. Filament loops hanging everywhere! Before I try a print of the full seals, I'll need to do a few tests to get the dimensions dialed-in. If they don't fit perfectly, I fear they'll fall off when moving the cutting brake levers. Once I get the basic dimensions down, I can try a full print.

Although it may seem like it, I promise this hasn't turned into a 3D printing build! I have plenty of other tasks to go for this build - once the 3D printing binge ends. Stay tuned...

Sounds like you need to do some filament calibration vs changing your models to fit what the printer is giving you. Either works, but once you have a filament "calibrated" you can model to nominal and let the slicer take care of the scaling / Shrink Factor.

Click one of the little "triple Dots" by the filament you want to use. Click Edit.
1766719826831.png


That will bring up another Dialog box. In that box look for "shrinkage"

1766719891917.png


That is where you enter the correction. If your prints are too big, you want to enter your percentage as a number over 100. If they are too small, enter it as a number less than 100 (the tool tip tells you how to enter the number).

Once you get that dialed in, the slicer will calculate your shrinkage for you, instead of having to incorporate it into the model. You'll save the new material profile and you select it when selecting your filament. The ones here that say calibrated are the ones I entered and saved after running some calibration prints.

1766720139652.png
 
...once you have a filament "calibrated" you can model to nominal and let the slicer take care of the scaling / Shrink Factor.

Call me doubtful regarding this statement, Mike. I aborted my first full and final print yesterday after about 30 hours of run-time because one of the support towers warped and the print head knocked it off the build plate. I should have taken a photo, but was pretty disgusted and didn't think about it. That was with automatic tree supports with the style set to "Tree Strong". After some research on warping, I changed to automatic normal supports with the style set to Grid and slowed down the support speed to try to prevent the same thing. Look at the warping that's now occurring in the grid supports:
IMG_8474.JPG


If ASA is this prone to warping, I don't think calibrating will solve the problem because warping in a part is very hard to model or predict. Like welding sheet metal, the temperature gradients causing the part to warp vary with the smallest change in conditions (like small drafts from environmental influences like a person walking by while you are welding). I suspect that the calibration only works with the print you used to calibrate. Repeat the part, and the calibration is no longer valid. That said, the same problem occurs when you change the model like I did.

I have to say that I'm getting very frustrated with 3D printing this part. I'll probably be aborting this print today, too. I need more rigid supports that won't warp as much (a lower surface area to volume ratio or more connections), but I can't find a way to do that. I thought the Grid option would tie the support walls together in a rectangular grid (from top view), but they are just zig-zagging back and forth with no ties at 90° to the walls shown. The part itself is much more rigid and isn't visibly warping, so I think if I can get the right support, it'll work. Do you know of a way to accomplish that?

Stay tuned for more frustrating sab 3D printing adventures!
 
One other observation: This printer must use machine vision to adjust the nozzle position while printing a layer because it is laying the next layer accurately on those warped supports, even though they've moved. That's impressive! If not for that, I'm sure this print would have already failed.
 
Call me doubtful regarding this statement, Mike. I aborted my first full and final print yesterday after about 30 hours of run-time because one of the support towers warped and the print head knocked it off the build plate. I should have taken a photo, but was pretty disgusted and didn't think about it. That was with automatic tree supports with the style set to "Tree Strong". After some research on warping, I changed to automatic normal supports with the style set to Grid and slowed down the support speed to try to prevent the same thing. Look at the warping that's now occurring in the grid supports:
View attachment 662609

If ASA is this prone to warping, I don't think calibrating will solve the problem because warping in a part is very hard to model or predict. Like welding sheet metal, the temperature gradients causing the part to warp vary with the smallest change in conditions (like small drafts from environmental influences like a person walking by while you are welding). I suspect that the calibration only works with the print you used to calibrate. Repeat the part, and the calibration is no longer valid. That said, the same problem occurs when you change the model like I did.

I have to say that I'm getting very frustrated with 3D printing this part. I'll probably be aborting this print today, too. I need more rigid supports that won't warp as much (a lower surface area to volume ratio or more connections), but I can't find a way to do that. I thought the Grid option would tie the support walls together in a rectangular grid (from top view), but they are just zig-zagging back and forth with no ties at 90° to the walls shown. The part itself is much more rigid and isn't visibly warping, so I think if I can get the right support, it'll work. Do you know of a way to accomplish that?

Stay tuned for more frustrating sab 3D printing adventures!

What is your chamber temp set to? I think I'd be running it to max, trying to keep the differential cooling down. Your build plate is probably 100° right? Printing with the nylons and glass filled ABS, I actually wrapped my printer in a blanket to keep the temp higher (was running around 53-54° C). I was having trouble with the models warping and pulling up in spots.

This stuff definitely isn't CNC machining.
 
  • Like
Reactions: Tob and sab
What is your chamber temp set to? I think I'd be running it to max, trying to keep the differential cooling down. Your build plate is probably 100° right? Printing with the nylons and glass filled ABS, I actually wrapped my printer in a blanket to keep the temp higher (was running around 53-54° C). I was having trouble with the models warping and pulling up in spots.

This stuff definitely isn't CNC machining.
Thanks, Mike! My bed temp is set to 65°C, which is the maximum. The print is still hanging in there. The top surface of the supports, with each layer printed, seems to be fairly planar (unlike the sides which have warped even more since that last photo), so maybe this print will survive?
 
Well, I aborted this latest print, too. Look at this ugly mess:
IMG_8479.JPG


I guess the surface wasn't as flat as it looked because just before it started laying down the second deck (and about the time the layer height dropped from .24mm to .08mm), the machine stopped neatly laying the bead on top the last layer accurately. I'm going to make a few setting changes and try again. What's that old adage about trying again? :)
 
Well, I aborted this latest print, too. Look at this ugly mess:
View attachment 662698

I guess the surface wasn't as flat as it looked because just before it started laying down the second deck (and about the time the layer height dropped from .24mm to .08mm), the machine stopped neatly laying the bead on top the last layer accurately. I'm going to make a few setting changes and try again. What's that old adage about trying again? :)

Not necessarily suggesting it will correct your issues, but have you considered changing the angle of the supports? I have corrected some printing problems previously by doing this.

Changing the support pattern might help, though at the cost of filament and print time.
 
  • Like
Reactions: Tob and FarFire70
Not necessarily suggesting it will correct your issues, but have you considered changing the angle of the supports? I have corrected some printing problems previously by doing this.
Yes, for this last print I increased the threshold from 15° to 60°. However, I did that as a "hail Mary." I didn't think it would work because I'm seeing the support itself fail due to warping - before the part above it starts getting printed, so I don't think that will help. It certainly didn't on the last print.

I wish I could "thicken" the support, or use a true grid pattern, but I can't find any settings for that. All the support options are flimsy (compared to options for the part), and warping is my issue. I'm about to try tree (auto) and Tree Hybrid for the support settings because it seems to be the most rigid support network. If that doesn't work, I guess I'll design it as a two-piece (top and bottom) print, but I really don't want to do that. It feels like defeat, but I can minimize warping because the prints won't take 2 days.

It's frustrating, but I am learning, so all is not lost. Knowledge gained though experience is never cheap. So far on this part, I've burned through about 8 rolls of various types of filament (PLA, PETG HF, and ASA), and the engineering build plate has taken a beating with these ASA prints. This last print was very difficult to remove from it, and I gouged it with the scraper.

Changing the support pattern might help, though at the cost of filament and print time.
As I said above, this is what I've been concentrating on, but I'm not happy with the options. I don't mind burning through filament if I can get this to print. I just don't like burning through filament only to see the print fail after 24 hours.

Stay tuned!
 
  • Like
Reactions: Tob and ColoJeep
Yes, for this last print I increased the threshold from 15° to 60°. However, I did that as a "hail Mary." I didn't think it would work because I'm seeing the support itself fail due to warping - before the part above it starts getting printed, so I don't think that will help. It certainly didn't on the last print.

I wish I could "thicken" the support, or use a true grid pattern, but I can't find any settings for that. All the support options are flimsy (compared to options for the part), and warping is my issue. I'm about to try tree (auto) and Tree Hybrid for the support settings because it seems to be the most rigid support network. If that doesn't work, I guess I'll design it as a two-piece (top and bottom) print, but I really don't want to do that. It feels like defeat, but I can minimize warping because the prints won't take 2 days.

It's frustrating, but I am learning, so all is not lost. Knowledge gained though experience is never cheap. So far on this part, I've burned through about 8 rolls of various types of filament (PLA, PETG HF, and ASA), and the engineering build plate has taken a beating with these ASA prints. This last print was very difficult to remove from it, and I gouged it with the scraper.


As I said above, this is what I've been concentrating on, but I'm not happy with the options. I don't mind burning through filament if I can get this to print. I just don't like burning through filament only to see the print fail after 24 hours.

Stay tuned!

Tree hybrid is my 'go to' support on any build needing supports. Strong, easily comes off final print, minimal material - usually a trunk that will then branch out when it gets to a large flat area.
 
  • Like
Reactions: sab
Yes, for this last print I increased the threshold from 15° to 60°

I meant the angle of the actual support pattern, not the threshold for support generation. You can also change the support pattern and spacing here as well. I frequently play with the angle and spacing, coupled with part orientation. Just for the record, I've never used the tree support type, I think everything has been "normal (auto)" to date, though I suspect there's additional power in the manual settings.

studio_capture.jpg
 
  • Love
Reactions: sab
Novak Conversions Jeep Wrangler TJ radiator