28 Sqn AVRO 504K double trouble

[giemme]

Could you print the rigging as well?

 

 

 

  OK, OK, I'll run  

 

Ciao

[hendie]

37 minutes ago, Brandy said:

Would it be feasible to model the struts with a recess down the entire front edge, and add the brass as the front of the struts instead of through the middle?

I use a similar process (adding plastic strip to bent brass) for undercarriage in 1:72.

 

Ian

 

I tried that as an option with the Wapiti struts and it didn't work out too well Ian. That was with the Mar's 1 though and I've no doubt the 3 would handle it much better, however I think the through hole is the preferable option as there will be no seams to take care of afterwards.  From the limited experience I've had with the 3 so far I feel confident I can make the through hole option work.

 

26 minutes ago, giemme said:

Could you print the rigging as well?

 

 

 

  OK, OK, I'll run  

 

Ciao

 

Don't think I hadn't already considered that as an option Giorgio - especially in that cabane area 

[CarLos]

22 hours ago, hendie said:

 

 

 

 

 

 

All valves in the same position? Tsss tsss...

 

😉

 

Fantastic work that I am following with great interest.

 

Carlos

[The Spadgent]

Cor lummie! That’s an impressive selection of updates. I’m wowed by  your skill and determination in this build. It’s making me yearn for the printer. Looking forward to getting back to Dewie after the Moths.

 

 Johnny

 

(edit). Then again why can’t I do both? 🤔

[Fritag]

On 4/29/2022 at 12:38 PM, hendie said:

It's all just yer basic extrudes and cuts.

 

Yeh well, a renaissance sculptor might say that it’s just yer basic hammer and chisel.  But we know a master craftsman when we see one  

[TheBaron]

So much going on here currently Alan that I've no sufficient way to praise the consummate level of problem solving and technique those last two pages alone contain. Absolutely inspiring work in every respect (though the stitching is my favourite...).

 

7 hours ago, Fritag said:

Yeh well, a renaissance sculptor might say that it’s just yer basic hammer and chisel.  

Don't encourage him Steve. It's bad enough him knocking out genius-level gear like this from resin without coming back next week to find he's carved the pilot out of Carrara marble whilst killing time waiting for the next print to finish....

[hendie]

On 4/30/2022 at 6:24 AM, CarLos said:

 

All valves in the same position? Tsss tsss...

 

😉

 

Fantastic work that I am following with great interest.

 

Carlos

 

I did try to model them at different angles (and they are - slightly) but I was limited with the amount of space I had available inside the cowling.  I could have shortened the cylinders to achieve that but...

 

On 4/30/2022 at 8:24 AM, The Spadgent said:

Cor lummie! That’s an impressive selection of updates. I’m wowed by  your skill and determination in this build. It’s making me yearn for the printer. Looking forward to getting back to Dewie after the Moths.

 

 Johnny

 

(edit). Then again why can’t I do both? 🤔

 

I see you're back Dewing again - nice

 

On 4/30/2022 at 8:41 AM, Fritag said:

 

Yeh well, a renaissance sculptor might say that it’s just yer basic hammer and chisel.  But we know a master craftsman when we see one  

 

as much as I'd love to argue with you there Steve... I'll take the compliment

 

On 4/30/2022 at 4:23 PM, TheBaron said:

So much going on here currently Alan that I've no sufficient way to praise the consummate level of problem solving and technique those last two pages alone contain. Absolutely inspiring work in every respect (though the stitching is my favourite...).

 

Don't encourage him Steve. It's bad enough him knocking out genius-level gear like this from resin without coming back next week to find he's carved the pilot out of Carrara marble whilst killing time waiting for the next print to finish....

 

I must show you the writing desk I made out of Carrara and an old Singer sewing machine base some years back.    Genius level gear?   Read on.

 

This week I have been mostly fighting corruption.  I would warn you that this post is going to be picture heavy but if you've followed this thread this far you'll realize that's par for the course.

I'll srt off with some the highlight then.

 

I printed the cabane strut with the wotsit on it and although my hopes were high, my expectations were somewhat muted. Guess wot though?

The cabane strut which was only 0.8mm thick with a 0.5mm through hole printed successfully.  Here it is with a 0.5mm rod which will be used as reinforcement.  I'm still in disbelief that the printer coped with that. That bodes well for the main struts coming later.

 

 

More tweaks were expended upon the old 504. Originally I was just going to print the tail skid and use brass as the supporting structure, but on further thought I had a bash at modeling the entire shebang with this result

 

 

In theory, this should slot nicely into the opening I've left at the back end.  It works - I know because I printed it and did a dry fit - and forgot to take any photos

 

 

Some more detail was added around the living quarters. Again, just because.

 

 

and now for the fight against Corruption.  No, I put my cape and mask away a long time ago. Corrupt STL files I'm talking about.   I ran a couple of test prints through the week and ran into a few issues.  Blindsided I was.

 

Riddle me this CADman - where did that obscene diagonal bump thing come from?  That bump that runs diagonally opposite the angle on the access panel.  When I washed the print and saw this I was a bit disappointed to say the least. What on earth...?

 

 

It wasn't on the model. I know cos I checked.  During the postmortem I found this unwelcome visitor to be the cause of death

 

 

You can see in this shot how it runs diagonally through the fuselage and how it aligns perfectly with the bump on the exterior.  Confirmed.  The bump/bulge has been caused by resin being pushed out of the way during the curing process and just slowly enough to form that horrible bulge on the exterior. Bummer.  Still don't know where it came from though.

 

 

I went back and checked the model. Nothing. Definitely nothing in the model to cause that.   One thing I had changed in the interim was to customize the STL settings to increase the number of triangles, maybe that was it?

I then spent a ridiculous number of attempts with different STL's trying to get to the root cause.

What I did find was that whatever was causing it was not visible in either the model or in Chitubox.  

 

 

Nope. Definitely nothing there to be concerned about. (this was just a quick auto support during my root cause analysis). By this time I was thoroughly confused and running out of ideas.  I know SolidWorks doesn't have the greatest reputation when it comes to STL output, but despite creating several different STL's, even stepping back to earlier versions of the model, I couldn't find anything.  

 

 

I had been receiving an error message about the STL being damaged when I imported it into Chitubox but I couldn't see anything untoward when I examined the model in there.  I had also received that same error message when I ran the first test prints and they printed fine. In desperation I went back to SolidWorks and found they have a "check model" function, and sure enough when I ran that it found a bunch of errors. 211 errors to be exact. In good old SW fashion however, they don't provide any tools to help you fix those errors .  It seems the only way to get around that is to remove the features and rebuild them and given that the feature tree is now several hundred items long there was no way I was going to delete any of them as it was bound to blow the model up and cause several weeks of rework.  It would have been nice if SW had actually informed me at the time of creating the feature that there was an error. Nope. I now had 210 open faces and 1 invalid line, whatever that is.  I checked them all - and none of them are anywhere near this ghost feature that keeps cropping up - they're all related to those zigzag stitches that took forever to create.  They're staying. (At some point I ran the repair function on the model but it didn't make a blind bit of difference)

 

Futher postmorteming was called for. Then I stumbled on something strange.   It only shows up after the slicing operation.  I can't remember exactly how I stumbled upon it, but as I walked through the slices I found this... the culprit.

Caught in delicto flagrante.   Well, that is without doubt the cause of the bulge but things got even strangererer.

 

 

The elephant in the room was only visible for a few slices and only when stepping through the slices manually. A few slices outside of that region and the darn thing disappeared. It just disappeared - until printing of course.  This was frustrating.

Hours or frustration and trial and error later I made another discovery... The ghost bump only appeared after the slicing operation and it was always parallel to the build plate. I changed the angle of the model a few times and it was always parallel to the build plate, and always around slice 2500 and change.

Right here...

 

 

I should mention that I am currently using Chitubox Pro (mainly because it was free for a year with the printer) and based on all this I feel there is a bug living in the Chitu box somewhere.   Okay, I've spent enough time on this... I downloaded Lychee Sicer and ran the 504 through Lychee. Guess what... I got the same error message about the STL being damaged.

Okay, no problem, I ran the auto repair function in Lychee, set up the supports, sliced the model and stepped through the slice just to be on the safe side.

Great!  Everything looked good so I saved the slice and threw it into the printer.  

9 hours later...

 

 

What a mess.  I didn't even bother rinsing it as I could tell it was beyond saving in any form.  So much for Lychee then.

 

Back to Chitubox and more hours of scrambling about.  What irked me was that the offending bulge thing was only visible under certain conditions. It only showed up after slicing, and even though I could track it down to a few layers, once I found it there was still no way I could edit it to remove the cause - I couldn't cut a hole in it, I couldn't delete it. Sheer frustration at this point.  I had also figured out by this time that the flaw lay in the STL.  Both Chitubox and Lychee recognized there was damage in the STL, but they both (I assume) interpreted and handled it differently.

Then somehow I happened on another discovery.  In all my prints up til now I had the longitudinal axis of the fuselage aligned with the X axis in the slicer, and the lateral axis aligned with the Y axis of the slicer, and had it pitched at 60 degrees so it would fit on the build plate. I have no idea how I discovered this but if I rotated the part ( think yaw) by about 5 degrees, the elephant disappeared.  Don't ask me how, why, or wherefore - I have no idea. It just wasn't there any more.

The proof is in the print.  Here's one I made earlier

 

 

Well, if I'm honest there's still something going on.  There's an ever so slight witness mark where that bulge would have been, and the two ports on the access panels are blocked when they should be open.

That offender is mostly gone - I found a small lip on the inside where it would have been.  Good news though - it looks like clean up on the underside won't be too bad at all.

 

 

... and the undercarriage mount features work as well, so it's not all bad.

 

 

I think I stated that I was fairly certain that it was not related to any feature in the model but I may be starting to second guess myself now.  I've noticed in the photos that the ghost always appears in close location to those ports in the access panel.  Those ports were present in my first test prints (and printed fine). One feature I have added since then was to create a chamfer on the rear face to give the appearance of a thin skin around the port.  I wonder if that chamfer has anything to do with this?  That's easy enough to test.

 

I'm also going to raise a ticket with the Chitubox team and see where that leads to.

 

In the meantime I'm running a test print on the engine. Maybe tomorrow.

 

 

What was that about master craftsman again Steve?

 

 

 

 

[Fritag]

58 minutes ago, hendie said:

The cabane strut which was only 0.8mm thick with a 0.5mm through hole printed successfully.

 

Luvverly.

 

1 hour ago, hendie said:

Some more detail was added around the living quarters. Again, just because.

 

Luvverly too.

 

59 minutes ago, hendie said:

What was that about master craftsman again Steve?

 

 

Well seeing as I don’t think I’d have a clue how to go about tracking down that issue and you did.  I’m gonna double down on that.

 

 

Isn’t that wot folks say over in your adopted homeland?

 

 

 

[The Spadgent]

I hate issues with 3D models that you just can’t fathom. Usually In max it’s a corrupt mesh or what not. Maybe a stray vert or a twisted polygon. Is there Any way of flattening the models in solid works to a single mesh? I know if there are troubles in max of this ilk you can select the faces within the mesh and make a copy of them, creating a new model in the process and usually eliminating any CrAzy model corruption. Then you can export a new .STL ?   Just a thought. You seem to have fixed it regardless but it makes me twitch when I don’t know why it went wrong. 🫣

 Nice struts by the way. 😃

 

 Johnny

[TheBaron]

11 hours ago, hendie said:

 

I must show you the writing desk I made out of Carrara and an old Singer sewing machine base some years back.   

I bloody knew there'd be marble in there somewhere!

11 hours ago, hendie said:

I then spent a ridiculous number of attempts with different STL's trying to get to the root cause.

What I did find was that whatever was causing it was not visible in either the model or in Chitubox.  

Sincerest sympathies Alan as I've experienced a slew of what might be termed 'X-Files' incidents over the last couple of years. The closest I can think of from my own situation to your own experience is a similar (but not necessarily identical) problem when printing an early prototype of the engine bays for XN708 over a year ago:

Nothing was visible in the CAD original or sliced versions in Chitubox but the rear of the port intake duct there was completely sealed off by a thick layer of resin about 1.5mm thick, whilst the stbd one was fine. Nothing visible in Fusion or any of the free .stl checking programs I tried. I'm lucky in that I can access Autodesk stuff through work so was able to install the full version of Netfabb and straight away it identified the culprit as some invalid ghost geometry (I think an invisible line left over from drawing which I forgot to delete and it messed up the negative space integrity of the opening) which it automatically fixed.. This was the only software that seemed discriminating enough to recover such errors. I did read somewhere that Fusion has Netfabb access built in on a recent  update but haven't had a chance to check yet.

 

Not surprised to see those issues with Lychee - I do find it superior to any other program I've tried for the layout and designing of supports structures but don't at all trust its slicing and anti-aliasing in conjunction with mars printers so just export the supported models as .stls into an old version of Chitubox for slicing and printing. This combination seems to work fine.

 

Your tenacious approach to problem solving continues to inspire Alan and demonstrates that there's no 'magic plateau' with this kind of technology, just a constant process of discovery.

 

A landscape unto itself. Wonderful:

11 hours ago, hendie said:

 

 

 

[giemme]

2 hours ago, TheBaron said:

and demonstrates that there's no 'magic plateau'

I beg to differ: this is all sorcery of the highest grade, and you all practicing it will be doomed forever.....

 

 

Well, no, I'm just envious and this whole report shows that Alan is a 3D master

 

Ciao 

 

 

[heloman1]

More fantastic work going on here Alan. I doo like the brass reinforced strut.

I'm bending wire fr the outside of my Wessex 3 build, slow progress.

 

Colin

[Brandy]

I've no idea what all that was about, but it does worry me what you and Tony are saying about Lychee as that is what I use exclusively since Chitubox flat out refused to work for me. I may have to revisit Chitubox....

BTW have you seen this...

https://www.facebook.com/groups/523218208401203/permalink/1038394710216881/?sfnsn=mo&ref=share

 

Ian

 

 

[AdrianMF]

Blimey! I'd be off sacrificing a goat and a chicken at midnight on a full moon in a field in front of the printer! Good work.

 

Regards,

Adrian

[hendie]

On 5/6/2022 at 6:27 PM, Fritag said:

 

Luvverly.

 

 

Luvverly too.

 

 

Well seeing as I don’t think I’d have a clue how to go about tracking down that issue and you did.  I’m gonna double down on that.

 

 

Isn’t that wot folks say over in your adopted homeland?

 

 

 

 

maybe in the big cities Steve - round here most have trouble stringing a full sentence together without the chewing baccy falling out.  (this county (Cecil County) is often referred to a Ceciltucky)

 

On 5/6/2022 at 6:41 PM, The Spadgent said:

I hate issues with 3D models that you just can’t fathom. Usually In max it’s a corrupt mesh or what not. Maybe a stray vert or a twisted polygon. Is there Any way of flattening the models in solid works to a single mesh? I know if there are troubles in max of this ilk you can select the faces within the mesh and make a copy of them, creating a new model in the process and usually eliminating any CrAzy model corruption. Then you can export a new .STL ?   Just a thought. You seem to have fixed it regardless but it makes me twitch when I don’t know why it went wrong. 🫣

 Nice struts by the way. 😃

 

 Johnny

 

Oooh, can't be doing with stray vertices, and twisted polygons are just downright painful. Unfortunately SW doesn't provide you a lot of tools to investigate - but there is progress if you care (dare?) to read on

 

15 hours ago, TheBaron said:

I bloody knew there'd be marble in there somewhere!

 

In a former life I worked in marble and granite for about 5 years. I designed and made everything from kitchens, bathrooms, tables, illuminated columns, and anything that anyone was willing to pay for, including a lectern stand and matching chair for a Chapel in Glasgow.

WHen I lived in Edinburgh I made myself an Adams Style fireplace and with the scraps left over I cobbled together a writing desk. Sheesh, that would be about 35 years ago now.  The wife has nabbed it now as her working from home space - you have no idea how much junk i had to remove before I could take this shot, and missed the sweetie wrapper on the floor.  I wish I had some photos of the furniture I made but that was back in the day before digital even became a word.

 

 

15 hours ago, TheBaron said:

Sincerest sympathies Alan as I've experienced a slew of what might be termed 'X-Files' incidents over the last couple of years. The closest I can think of from my own situation to your own experience is a similar (but not necessarily identical) problem when printing an early prototype of the engine bays for XN708 over a year ago:

>SNIP>

Nothing was visible in the CAD original or sliced versions in Chitubox but the rear of the port intake duct there was completely sealed off by a thick layer of resin about 1.5mm thick, whilst the stbd one was fine. Nothing visible in Fusion or any of the free .stl checking programs I tried. I'm lucky in that I can access Autodesk stuff through work so was able to install the full version of Netfabb and straight away it identified the culprit as some invalid ghost geometry (I think an invisible line left over from drawing which I forgot to delete and it messed up the negative space integrity of the opening) which it automatically fixed.. This was the only software that seemed discriminating enough to recover such errors. I did read somewhere that Fusion has Netfabb access built in on a recent  update but haven't had a chance to check yet.

 

Not surprised to see those issues with Lychee - I do find it superior to any other program I've tried for the layout and designing of supports structures but don't at all trust its slicing and anti-aliasing in conjunction with mars printers so just export the supported models as .stls into an old version of Chitubox for slicing and printing. This combination seems to work fine.

 

Your tenacious approach to problem solving continues to inspire Alan and demonstrates that there's no 'magic plateau' with this kind of technology, just a constant process of discovery.

 

A landscape unto itself. Wonderful:

 

 

 

I have to thank you Tony - your comments spurred me on to don my deerstalker, little tweed shoulder cape and light up my calabash in anticipation of a bit of sleuthing 

 

13 hours ago, giemme said:

I beg to differ: this is all sorcery of the highest grade, and you all practicing it will be doomed forever.....

 

 

Well, no, I'm just envious and this whole report shows that Alan is a 3D master

 

Ciao 

 

 

 

Sorcery is removing the masking tape to find the paint still intact.  This is just digital lego Giorgio 

 

12 hours ago, heloman1 said:

More fantastic work going on here Alan. I doo like the brass reinforced strut.

I'm bending wire fr the outside of my Wessex 3 build, slow progress.

 

Colin

 

italeri don't make it go any faster does it Colin?

 

11 hours ago, Brandy said:

I've no idea what all that was about, but it does worry me what you and Tony are saying about Lychee as that is what I use exclusively since Chitubox flat out refused to work for me. I may have to revisit Chitubox....

BTW have you seen this...

https://www.facebook.com/groups/523218208401203/permalink/1038394710216881/?sfnsn=mo&ref=share

 

Ian

 

 

 

Have you tried contacting their helpdesk Ian?  They were very responsive when I used them

 

3 hours ago, AdrianMF said:

Blimey! I'd be off sacrificing a goat and a chicken at midnight on a full moon in a field in front of the printer! Good work.

 

Regards,

Adrian

 

Being seen with farm animals in the dark around here is a bit risky Adrian -  see Ceciltucky, abiove. 

 

After reading the comments above (thanks all), I felt I couldn't ignore the corruption issue and had to attack it head on, and now rather than later as it wasn't going away on it's own.  Based on Tony's & Johnny's comments above I went a-hunting down the feature tree looking for wayward sketches or likely candidates for getting it all vertice about polygon. 

I couldn't find any.     Drats.

I did remove some unnecessary elements and tidied things up as I went along so it wasn't a wasted journey.

The feature tree at this point consists of a few hundred features and there was no way I could go through them all one by one but I had to figure something out - then I had one of those idea things that I hear people talk about sometimes...  I couldn't go through several hundred features one by one... well I could, but I wasn't about to.

What I could do though was to step through the feature tree in batches and create a separate STL for each batch, starting from a very basic model incrementing up to the full build sequence.

I ended up with 11 STL files - the first one containing about 20 features, and each subsequent STL contained an additional 15 - 20 features from the STL previous.  That meant I only had 11 STL's to go through in Chitubox.

The plan was it iterate through those STL's checking each in turn in Chitubox until the problem appeared.  That would narrow the issue down to only those additional features added since the previous STL.  The I'd have only 15 to 20 features to examine in order to find the culprit.  Nice theory.

 

It worked!

 

STL Troubleshoot Number 2 had the issue. To be honest that was way earlier in the feature tree than I had anticipated but there it was staring me in the face.

 

 

Some more detective work was required and after some back and forth and running a few additional STL's to verify the issue I can now present to you the culprit in all it's shameful glory...  The access panel.

For this panel I used an Extruded Surface, much as I had done for all the reinforced linen etc.   Why this one failed I still don't know - based on what I see on screen it appears to be extruding successfully but the sliced STL tells me otherwise.

 

 

Once I had the culprit nailed down I suppressed the features, which then proceeded to blow up about about eleventy six features further down the tree.  That wasn't too bad and all I had to do was edit each one in turn and fix the broken references. In total I probably spent 4, maybe 5 hours with the Sherlock cap on but I feel a lot better about things now.

I should mention that I tried recreating the access panels using the extruded surface option again, but this time as the last item in the feature tree - it still didn't work. In the end I created the panels using an extruded boss and had to clean up afterwards with a revolved cut or two. Not as elegant a solution as I'd have liked but it was the solution this time around.

 

As a bonus to myself I allowed the relieved hend to make a main strut for the wings.

 

 

Tweaks (as always) required but the struts are almost complete. I can just about begin work on the interior now

 

 

[TheBaron]

Superb sleuthing Holmes.

 

3 hours ago, hendie said:

WHen I lived in Edinburgh I made myself an Adams Style fireplace and with the scraps left over I cobbled together a writing desk.

Going on that desk, yon fireplace must indeed have been a thing of beauty Alan.

[perdu]

On 29/04/2022 at 14:45, giemme said:

Could you print the rigging as well?

 

 

 

  OK, OK, I'll run  

 

Ciao

I would not advise running too far Giorgio, did you mean something like this?

 

I thought I would give this a try

 

Coming to a Hawker Assortment build near you soon...

 

This ridiculously tiny set even has the ring and bead sight and control surface horns.

 

Detective work of the highest order Alan, utterly enthralling stuff.

 

And I love the sewing machine marble computer desk, my mom used to own one of those Singers with the treadle pedal, instant 'take me back fifty-six years' to our old council house in Moseley.

[The Spadgent]

Nice one Alan. Yup, that’s pretty much how I try and solve the issues at work. It’s an iterative process socutting those processes into chunks and testing is a great way to corner the blighter. 👏👏👏 we can all sleep well again. 😃🤓

 

 until next time. don, Don, donnnn. 🫣😆

[hendie]

On 5/7/2022 at 8:17 AM, Brandy said:

BTW have you seen this...

https://www.facebook.com/groups/523218208401203/permalink/1038394710216881/?sfnsn=mo&ref=share

 

 

I did Ian, yes.  Crisp  brought it to my attention earlier in the week.

I won't be buying one as it's not my scale. It looks a nice kit though it does seem to have some inaccuracies- but I'm only going on some poor quality photos and the box art, so I could well be wrong. 

 

Does that mean there'll be a new 504 kit available around December then? 

[Fritag]

On 5/8/2022 at 2:34 AM, hendie said:

STL Troubleshoot Number 2 had the issue. To be honest that was way earlier in the feature tree than I had anticipated but there it was staring me in the face.

 

 

On 5/8/2022 at 2:34 AM, hendie said:

Some more detective work was required and after some back and forth and running a few additional STL's to verify the issue I can now present to you the culprit in all it's shameful glory...  The access panel.

 

It’s early here in the UK; and I’m only half a cup of coffee into wakefulness…But I’m not sure from the screen prints what was staring you in the face    No doubt it’s bl**din obvious to anyone with half a brain….

[mark.au]

30 minutes ago, Fritag said:

No doubt it’s bl**din obvious to anyone with half a brain….

Clearly, I have <0.5 of a brain… 🤦‍♂️

[perdu]

Just now, mark.au said:

Clearly, I have <0.5 of a brain… 🤦‍♂️

Me?

 

Way less than that.

[Terry1954]

14 minutes ago, perdu said:

Me?

 

Way less than that.

 

Mine tends to 0!

 

Terry

[giemme]

23 hours ago, perdu said:

I would not advise running too far Giorgio, did you mean something like this?

 

 

 

1 hour ago, Fritag said:

No doubt it’s bl**din obvious to anyone with half a brain….

I'm not even trying....  

 

Glad you fixed it, though, Alan  

 

Ciao

[hendie]

 

5 hours ago, Fritag said:

But I’m not sure from the screen prints what was staring you in the face

 

Steve et al, the clue was in this pic/slice.  The access panels have been added since the previous STL, and there is/should be a vent hole in the center of the panel, but when I stepped through the slice I could see that there even though there is a vent hole visible, the slice shows a bridge layer over the gap when it should clearly be just a gap. On the port side you can just see that there is a through hole for the vent yet there is still a bridge layer showing up.  The bridge layer only shows up in two or three of the slices (sliced at 0.05mm) but prints at around 1mm thickness

 

 

While that small "bridge" wasn't as obvious as the full width defect seen in this earlier shot it pointed me in the direction of the access panels somehow being the possible cause

 

 

To test that theory all I had to do was create one STL with the access panels and one without (easy enough to suppress them).  When I ran those two STL's though the slicer the one with the access panels had the defect but in the STL with the panels suppressed, everything looked good.  

I didn't step through each and every slice since there are over 3000 of them but when focusing on the trouble area I could not find anything untoward.

I can't be 100% certain I have eliminated the issue completely until I run another print but at this point I am feeling quietly confident...

 

Now as to why that was happening?  I still have no explanation.  I have used extruded surfaces in numerous locations on this model and all of the others work as expected. I have no idea why this particular surface didn't want to play ball. I'm sure there must be some combination of factors playing into the end result, but what those factors may be is anyone's guess.