2D drawing error

[mrehmus]

I have a 30 sheet 2D drawing from a previous model engine design in 2018. I'm developing a new engine using many of the internal parts and I don't want to recreate 30 pages. So I deleted the irrelevant sheets and started inserting the new components.
The cylinder for this engine will lay out cleanly if I start a new independent sheet not part of the 30. But if I insert the cylinder into the old 30 sheet drawing, I get 14 errors as shown in the image. Note the hidden holes showing up on the end views where they should not (but they are deleatable) and the bizarre lines representing I don't know what on the side view. I've opened and laboriously saved the 30 pages in V23 and in a new directory (the laborious bit) so it should have 'updated' whatever it needed to in the database. The View Type and Edge Display were not changed from the default.
A single sheet drawing of the cylinder, however, displays properly.
Any ideas?

 

[HaroldL]

It appears to me that hidden lines are shown the two end views. The side view is the weird one, it looks like cosmetic threads. Were the holes created with the Hole tool and do you have cosmetic threads shown in that view and also in the isometric views?

 

[mrehmus]

It appears to me that hidden lines are shown the two end views. The side view is the weird one, it looks like cosmetic threads. Were the holes created with the Hole tool and do you have cosmetic threads shown in that view and also in the isometric views?

No cosmetic threads. Created with the hole tool. The hidden lines should not be there either. You can see the solid model is clean. Here is the same part but on a single sheet drawing.

 

[DavidJ]

Please send the files to Support. This could be a simple setting, the drawing file could be corrupted, or there could be some other issue (possibly related to the change of Alibre version).

If using your own 2D drawing template, add that to the support ticket.

 

[mrehmus]

Actually, the model and drawing made in V23 has errors when centers are inserted. The centers on the hidden holes are displayed. I think they should only be displayed when hidden lines are invoked.

 

[albie0803]

This is an outstanding issue that I have asked about quite a while ago, to only have centres and threads shown on visible faces.
At this stage, its not a bug, its a feature. Either manually insert the ones you want by right clicking a hole or add them all and delete the ones you don't want.

 

[mrehmus]

I was told that the hidden holes are being displayed as dashed lines and that turning off that layer will remove them. And it did. However it did not solve the unwanted center lines or the solid lines on the side view. They are still investigating that. I am speaking of the drawing as part of the 30 pages. The unwanted centers are present in a new V23 drawing as shown.

 

[mrehmus]

Here is the latest report I received from Support: "I did verify with QA that currently centerlines, as well as cosmetic threads, will show for all holes regardless of the view. So, if you only want the centers and threads of the holes showing in the view, you will need to apply them to the holes individually rather than to the view itself. "
I consider that to be a major bug.
Then, right-clicking on one of those errant lines on the side view and then on Select by Layer, I got these blue shapes on the drawing:

 

[simonb65]

That's definitely a major bug!

 

[HaroldL]

Here is the latest report I received from Support: "I did verify with QA that currently centerlines, as well as cosmetic threads, will show for all holes regardless of the view. So, if you only want the centers and threads of the holes showing in the view, you will need to apply them to the holes individually rather than to the view itself. "
I consider that to be a major bug.
Then, right-clicking on one of those errant lines on the side view and then on Select by Layer, I got these blue shapes on the drawing:
View attachment 34054

So who would have thought that was a good idea when creating the program. It would be nice it the programmers knew something about detailing a drawing.

 

[simonb65]

So who would have thought that was a good idea when creating the program. It would be nice it the programmers knew something about detailing a drawing.

^ This !

Sadly, most developers I've worked with over the years have very little or no knowledge of the product they code, other than it has to meet a requirement or functional specification determined by someone else ... who sometimes had an idea of what they wanted to achieve but no real experience themselves!

I've also worked with engineers in India who design car transmission and powerplant systems, many of whom can't drive, so getting them to determine failure modes, and analyse them in an FMEA, of a transmission was 'challenging' to say the least. One thought that just putting the auto transmission in 1st gear if it had a sensor failure would be the 'best' thing to do. I pointed out that it wasn't a good idea if the cars doing 60mph at the time! Would have tested if the seat belt worked I suppose

Not saying that the Alibre team are like this, but I have to say, there are some things that occasionally make you wonder. Be nice to think they have a select group of 'power' users who use the product day-in day-out that they consult as part of their product feature requirement capture phase of development cycle.

 

[BobSchaefer]

^ This !

Sadly, most developers I've worked with over the years have very little or no knowledge of the product they code, other than it has to meet a requirement or functional specification determined by someone else ... who sometimes had an idea of what they wanted to achieve but no real experience themselves!

I've also worked with engineers in India who design car transmission and powerplant systems, many of whom can't drive, so getting them to determine failure modes, and analyse them in an FMEA, of a transmission was 'challenging' to say the least. One thought that just putting the auto transmission in 1st gear if it had a sensor failure would be the 'best' thing to do. I pointed out that it wasn't a good idea if the cars doing 60mph at the time! Would have tested if the seat belt worked I suppose

Not saying that the Alibre team are like this, but I have to say, there are some things that occasionally make you wonder. Be nice to think they have a select group of 'power' users who use the product day-in day-out that they consult as part of their product feature requirement capture phase of development cycle.

Simon, you hit on a pet peeve of mine, speaking as a developer, I whole heartedly agree with you. I've also seen too many times where something is developed to fit exactly what was asked, but is completely unusable. I am a firm believer in attempting to actually use an application I'm working on or developing. Now, you're also correct that a lot of the time the developers don't necessarily know what they're coding is meant to do, other than what the spec tells them, but you might also have stumbled across something that regression testing would normally catch. Keep in mind, testing costs money, and time and they have lots of users screaming for things to be fixed and updated, so they have to balance this. So, complain that it exists, but please don't blame the developers or the testers, they're doing their best to get all the requests balanced and released in a timely manner.

Oh, and I don't work for Alibre in any shape, I'm a user just like the rest of you!

Bob Schaefer

 

[mrehmus]

I was just told that the errors on the side view are from turning on Cosmetic Threads. Certainly I can erase them and clicking on insert cosmetic threads does indeed generate them. Not only did I not click on insert cosmetic threads, if I did so, it is still a bug.

I worked in Silicon Valley for many years as the Director of Marketing for the first digital cameras and the first desktop scanners. The engineers didn't start writing code until I had designed the user interface and described how each item would function. I also worked with AutoDesk before there was a public release of AutoCad and we made certain the interface was as a draftsman would expect.

 

[simonb65]

Keep in mind, testing costs money, and time

Testing should be part of the coding process, and if it isn't, it should be. It's the best investment (apart from marketing) that any software product could have. I can't say what level of testing takes place, I can only judge on the types of bugs that pop up just modelling everyday objects and the 'simple to see' bugs that appear in beta releases.

So, complain that it exists, but please don't blame the developers or the testers, they're doing their best to get all the requests balanced and released in a timely manner.

It wasn't a complaint, it was an observation. Most issues that we currently see are from changing the underlying framework (i.e. Legacy to HOOPS, New ribbon, etc), its issues that have been introduced to what was already good code ... but I understand the reasons why they changed (we've all been there with 3rd party dependencies). If you're going to change something so fundamental, then investment in testing is a no brainer if you want to meet short timescales.

I know the team are working on issues that I've personally raised, that have been introduced in recent releases, and that's extremely welcome. But, the question to ask should be ... should that be done now (so far down the line), and not at the time of the changes, as it is currently constantly eating into time that could/should be for moving AD forward (functionality wise).

Out of curiosity, I'd personally be interested to know how big the dev team is at Alibre actually is. For the size of the AD user base, changes, work that needs to be implemented, testing, etc ... it suspect it may not be quite big enough to fulfil all our needs in the timescales we would like.

However, overall, I think they are doing a great job with whatever resources they do have.

 

[mrehmus]

Testing is the most efficient way to get a product to market. Having been in the position of releasing a fax board for the PC and when it got out there, it essentially did not work. We had to bring all the inventory back and fix the problem.
Since support is frequently the most expensive part of selling product, good testing is mandatory.
I get SolidWorks for $20 a year and still think Alibre is good enough for my magazine production.

 

[HaroldL]

I get SolidWorks for $20 a year

How did you manage that or is that a special deal for your magazine? And what version/level is it?

 

[simonb65]

How did you manage that or is that a special deal for your magazine? And what version/level is it?

You can get SW for Makers for $49.50/yr by being an active member of the Experimental Aircraft Association (EAA).

 

[simonb65]

This is what the EAA offers it's members ...

 

[HaroldL]

Sooo -- then everything you create with it is on the cloud.

 

[mrehmus]

How did you manage that or is that a special deal for your magazine? And what version/level is it?

They offer the deal to any U.S. or Canadian military veteran. You get absolutely everything they have as far as I know. It is just for personal use and it puts a title on every 2D drawing stating it is the Student Edition. It has to be renewed every year for another $20.
It is always the most current version.