QCAD question

Phil O

Western Thunderer
Hi

I was using Draftsight until it went to a paid for software and I'm now getting my head around QCAD, I have followed the tutorials that Adrian did a few years ago with some success, thank you Adrian. Having mastered it as far as those tutorials go, but now I have hit a brick wall.

My question is how would one go about drawing coach turnunder as non of the drawings give a radius or radii for it and thus no centre(s). The photo below shows where I'm at any help or pointers gratefully received.

Coach turnunder.png
 

JimG

Western Thunderer
If you are tracing over a picture, I would use the three point Arc like Mike suggests. Then, if you want to fine tune, you could use the centre of the arc's circle to use as a guide to adjust. Or you could use trig if you know the height of the turnunder and the offset at the bottom. :)

By the way, I've started using NanoCAD to replace Draftsight. Like Draftsight it is a pretty close copy of AutoCAD LT. There's a free version which is probably limited in some facilities but I haven't come across any problems as yet. I think the paid version is around £180. The download page for the free version is here

nanoCAD

Jim.
 

adrian

Flying Squad
I have followed the tutorials that Adrian did a few years ago with some success, thank you Adrian.
Thanks for the feedback - it's good to know they have been useful.

My question is how would one go about drawing coach turnunder as non of the drawings give a radius or radii for it and thus no centre(s).
I'm afraid I'll have to offer an alternative option as I don't think an arc with 3 points is suitable. The reason being that the top end of the arc needs to be tangential to the vertical side of the coach, the 3 point arc won't guarantee a tangential arc. If you play around with the points then you will get something close but in my experience it never quite works because you are trying to judge very fine angles with the Mk1 eyeball.

To a certain extent it depends on the drawing you are using to trace. If they are old blue prints then sometimes it is possible to work out the radius, the original draughtsman for the drawing would have used a compass to draw the arc, if you look carefully at the drawing there may well be some witness marks, just a small black dot, in the region of the centre of the arc. This is indication of where the draughtsman stuck the point of his compass into the paper to draw the arc. So even though the radius is not labelled it might be possible to work out the centre of the arc used.

If no marks suggest a centre then I would suggest trying the "Tangent, point, radius" option.
Screenshot 2019-08-27 at 23.55.14.png

Start by selecting the vertical side as the line entity, set a smaller than anticipated radius, on first effort the arc might not meet the vertical side so go back a step an increase the radius until they meet as close as you want. This is too small and doesn't meet so I need to increase the radius used. I prefer this method because it allows me to type in a number for the radius which I find easier to fine tune rather than clicking a point on the drawing.
Screenshot 2019-08-28 at 00.13.41.png

Finally extend or trim the 2 entities to each other to ensure the end points are co-incident and the shape can be enclosed for any fill.

This method should ensure that the arc is truly tangential to the vertical coach side.
 

Big Train James

Western Thunderer
Actually the 3 point arc is still suitable even with the requirement for the tangential condition. The trick is to think of the tangent point as the midpoint of the arc, and then mirror the second point (the curve termination at the bottom of the car side in this case) about the horizontal axis through the tangent point.

A video is invariably more successful than a bunch of words in my experience, so here's a quick one I just recorded. I'm using autocad in my case, but I think the programs function enough alike that this will apply to Qcad as well. I also drew a 3 point circle, but an arc could be drawn the same way. I forgot to say at the end to trim off the extra parts that are no longer needed.


As always, feel free to ask questions as needed.
Jim
 
Last edited:

Phil O

Western Thunderer
Thanks again Gents,

I have just got around to having a go and I settled on Jim's solution and it seems to have worked OK.
 

Phil O

Western Thunderer
Hi Gents,

I'm back again with another query that I haven't been able to get an answer to. The roof line is elliptical and I have offset the beading, but I can't get it to put the fillet curves to the rest of the beading, when I click over the straight lines there's no problem, but i cant get it to click on the semi ellipse.
Any ideas are gratefully received and thanks for your time.

End query.png
 

simond

Western Thunderer
Phil,

I’m not a QCAD user, but in TurboCAD, I’d break the semi-ellipse where the beading lines meet it, and then put the filleted corners in.

If you can’t actually select the semi-ellipse, I’d wonder if it’s actually there. Try a re-draw and see if it disappears?

I’m sure someone with direct QCAD experience will be along at some point, but I guess they’re all at Telford...

HTH
Simon
 

Phil O

Western Thunderer
Hi Simon,

I have just tried that and I can do that, but again when I come to fillet the corners. I'm getting the same result. It does prove though that it does exist, otherwise I wouldn't be able to break it down. I'm pondering whether I can redraw it using the three circles method I was taught as an apprentice, but that will mean stirring the grey matter as that was 45 years ago, since I had to do it.Thanks for the idea.

End query 2.png
 

simond

Western Thunderer
Phil,

I now suspect that the fillet tool doesn’t accept an ellipse as a valid “object to operate on”. I suggest you proceed as follows.

Let the radius of your fillet be “R”.
Create a parallel line offset R to one side of your vertical beading.
Offset your ellipse segment downwards by R
Corner these lines, this now provides a centre for a circle radius R, approximately one quarter of which is the desired fillet.
Breaking the circle on the vertical is easy. I suspect you need to find a co-tangent of the circle and the segment of the ellipse (or a line that is a common radius to both) to define the contact point, at which you can break/trim the ellipse. You might have to play with this!

Atb
Simon
 

AndyB

Western Thunderer
In my experience (based on works drawings), so called 'elliptical' roofs are never true ellipses - they are always a series of arcs that run into each other.
I'll see if I can dig out some examples if that would help?
Andy
 

Big Train James

Western Thunderer
I'm in agreement with Andy that this roof is formed by a series of arc segments. To that end, I've done another quick video demonstrating how I would go about creating such a roof. It's mostly done with the 3 point circle command, much as the car sides were, but with a few extra steps. The usual caveats about Autocad versus Qcad apply.


Jim
 

Phil O

Western Thunderer
Hi Jim

Thanks for that, I will give it a go later, sorry about the above post, I thought I had posted it yesterday, but obviously i neglected to do it until I saw it just now.
 

Phil O

Western Thunderer
I have finally got around to giving it a go, but for some reason I can't trim the circles back to give the 3 arcs required, despite the fact that the two circles that link the 2 end circles to the centre one being tangents. Sorry no screen shot at the moment aa I'm on my tablet.

Many thanks to Jim for his time and effort.
 

Phil O

Western Thunderer
Had some success in sorting the problem, by adding a line between the tangential intersection and another point the trim tool worked on the circles, I didn't use the added lines for trimming. I thought I'd add the solution incase anyone else encounters the same problem/
 

Phil O

Western Thunderer
Hi All

I'm slowly winning on the QCAD front, but I have hit another snag, which has left me scratching my head.

I finally got all the loose ends tied up to be able to hatch, but one coach end has been totally hatched, whereas the other has hatched as expected. Any ideas, as to what I have done wrong, as far as I'm aware I produced them in the same way, in fact the right hand one started life as a copy of the left. I have done them both individually with the same result.

Auto end hatch.png

Sorry to be a pain again.

Edit, to add the additional info about individuality.
 
Last edited:
Top