Old school design methods

[JSeward]

Hi everyone,

This question has really been bugging me for ages so I may as well try and ask it.

Before the advent of CAD I can understand how the designers/engineers of plane could come up with 3 view sketches (front side and top) but what I don't understand is how they came up with wing rib drawings or fuselage frames, that would fit the design at any point in the wing or fuselage?

http://www.seqair.com/Falco/Kits/Kit302/Kit302.gif

How did they decide the measurements and where frames such as this would be placed?

Thank you very much

[FlightPathOBN]

Come on now, give us a break...look at all of the stuff designed before cad...

If you have the fuselage or wing profile, you can get the measurements for the ribs at any point...all you need is some graph paper...

http://www.outerzone.co.uk/images/thumbs/plans/2414.jpg

[JSeward]

How could I have missed wing profiles! apologies for that, what about the fuselage profiles since the shape is drastically changing?

[Genghis the Engineer]

Basically the drawing board is your tool, you work from initial aerodynamic shapes, and construct the aeroplane on paper, on a series of sheets of paper - the bigger the better. Working in 3-view, you put faint working lines across that allow you to join everything up. It is reliant upon having a good and continuously developing mental picture of the aeroplane or component you're designing.

I worked for a while in the drawing office(s) at RAE where I did a fair bit of this in the latter days of pencil designing when CAD was starting to become new - and happily designed quite a lot of complex shapes.

Genghis Sr. was for rather longer in the Vickers Armstrong drawing office at South Marston where he worked on wing sections for aeroplanes like the Attacker and Scimitar. Comparing notes with him from time to time, his tasks were similar - but those chaps were very good engineers, with an excellent mental picture of what they were designing. Also however they had A LOT of people - think large rooms full of drawing boards, full of intense young men with thick framed glasses, slide rules and collections of data books and drawing instruments on the side table. The amount of training they had, and the level of skill in the task, was arguably rather greater than many CAD operators nowadays.

Mike Whittaker, who's still out there designing using CAD these days, designed his earlier designs just this way - (Plans | MW Club (http://www.mwclub.org/?page_id=82) if you'd like a set to look at) and he'll send you a complete set of drawings for any of his designs. I used to use them for teaching aircraft design when that was my job and very good they were too.

[Flash2001]

Does anyone remember the wonderful Jeffries drawings in Air Trails magazine?

After an excellent landing etc...

[Brian Abraham]

I don't understand is how they came up with wing rib drawings or fuselage frames, that would fit the design at any point in the wing or fuselageIt's done by a process known as "lofting". Originated in ship building. Google "aircraft lofting". A primer

http://www.dept.aoe.vt.edu/~mason/Mason_f/ConfigAeroGeomIssues.pdf

[jxk]

I was around in the days of the Attacker, Swift and Scimitar and remember that a full size wooden mock-up of the later was in the 'sheds' at Hursley (near Winchester). To reach this mock-up, draftsman had to negotiate a lengthy walk across a field of nettles armed with a ruler to take measurements and then go back to their drawing-boards and create the 3 view drawing.

[Intruder]

Lofting classes are still available in many boatbuilding schools.

[rennaps]

Slightly off thread
Most of the conventional instrument approach procedures (ILS, VOR, NDB etc.) that you fly Now were designed using only pencil, ruler, tracing paper and a calculator :8
No CAD

[dixi188]

rennaps.

A calculator????

Slide rule I think!!!!

I never could get the hang of moving the decimal point every time you moved from one end to the other.
I used to do an approximation of the result on paper to get the decimal point right.

[roulishollandais]

teaching aircraft design when that was my job

Welcome coming back from Banningland! Future is behind us. :ouch:

Time is not compressible in Insurance and hazard and Air safety is not solvable in stats..

[stressmerchant]

Thanks for some good info - I remember the drawing offices at my old work filled with lots of boards and hard working men and women. No computers ;-)

My impression is that the CAD tools sometimes separate us from the reality of design. The old designers may not have had the fancy software, but they had a lot of common sense behind them. I battle daily with young engineers who will blindly believe the numbers that the computer spits out.

A few recent gems:
"I don't care what the flight test results say, my prediction is correct. I looked up the formula in Roskam and I've double checked my maths!"

"OK, so we designed to the finite element predictions, and the finite element overpredicted the stresses by 50%. At least we know the structure is strong enough".
"Yes, but it's overweight"
"And the problem is??"

"But the diagonal tension field didn't show up on NASTRAN, how am I meant to know about it? "

Children of Magenta?

[john_tullamarine]

I used to do an approximation of the result on paper to get the decimal point right.

which is why all us olde phart engineers have difficulty with the youngsters who believe their computer output garbage without doing a back of a fag packet sanity check rough calc ...

Slide rule I think

I can't recall a single occasion where my slide rule stopped working because of battery problems ... and, for many calculations, its accuracy is just as practical as the 495 irrelevant decimals which the computer spits out.

stressmerchant, I'm with you good sir.

Even in more recent times old techniques die hard. Jess Smith, who was the force behind the short lived MAMBA single in Melbourne started the design in chalk on his hangar floor ..

[Bye]

It's done by a process known as "lofting". Originated in ship building

after my apprenticeship at RR i went into the aerodynamic fairings Design office, where i learned to loft engine fairings and doors and pylons etc.

we used to draw on sheets of nice shiny aluminium with lots of french curves and trammels, with erasing done by emery cloth, these lofts were scale 1:1 and were taken up to Hucknall to be carefully cut out to make egg box molds, from which tooling was made to profile aluminium sheets. being rascals we used to put the shiny aluminium off cuts on the floor when the tea trolley came around we played guess what colour :eek:

also believe it had something to do with parachute packing on long tables which were in a loft as it was the only space long enough for these tables.

we used to draw in ink on finest Irish linen which could have the starch soaked out to make cracking hankeys and dusters. i still shake my pencil when drawing as a throw back to getting the ink flowing in the old pen knibs.

with regards to the bits between the sections, we used to rely on the toolmakers eye to blend it all in nicely, which CAD does for us now.

The problem is nowadays that people forget that the computer is just a tool of the job, as was the drawing board.

and in some cases the limitations of the tools or the operators can influence the design solution rather than looking for another tool, like the one in your head.

CAD is great though and using virtual reality CAD to see if you can close an engine door latch without lying down is a real step forwards in terms of designed in safety ;).

[Genghis the Engineer]

Thanks for some good info - I remember the drawing offices at my old work filled with lots of boards and hard working men and women. No computers ;-)

My impression is that the CAD tools sometimes separate us from the reality of design. The old designers may not have had the fancy software, but they had a lot of common sense behind them. I battle daily with young engineers who will blindly believe the numbers that the computer spits out.

A few recent gems:
"I don't care what the flight test results say, my prediction is correct. I looked up the formula in Roskam and I've double checked my maths!"

"OK, so we designed to the finite element predictions, and the finite element overpredicted the stresses by 50%. At least we know the structure is strong enough".
"Yes, but it's overweight"
"And the problem is??"

"But the diagonal tension field didn't show up on NASTRAN, how am I meant to know about it? "

Children of Magenta?

Anybody else ever come across a dH era structures professor called Dennis Mead at Southampton? I recall him once saying in a lecture "The trouble with aerodynamicists is that the can assume absolutely anything, except for responsibility".

When I ran an airworthiness office I was often presented with elaborate FE predictions built upon sand. If all else fails, my standard response was "thanks, that's given me confidence that we can now go and build and test a sample structure to destruction on your budget".

The other classic quote there was "The difference between theory and practice? In theory it's right, in practice it isn't."

[jxk]

I seem to remember my days in the drawing office where I believe we used design criteria laid down years before by the Royal Aircraft Establishment at Farnborough pre WW1.
That is probably why we here in the UK we ended up using proper nuts and bolts instead of self tapping screws like the USA and of course the increased weight and complexity.

[Owain Glyndwr]

Anybody else ever come across a dH era structures professor called Dennis Mead at Southampton? I recall him once saying in a lecture "The trouble with aerodynamicists is that the can assume absolutely anything, except for responsibility".

Never came across the guy, but for many years I was an aerodynamicist regularly treated to that remark. Then I got a position where I had to oversee the work of the stress office as well as aerodynamics, and guess what - the stressmen made even bigger assumptions. Yes I know they had to sign off on the drawings, but so did the aerodynamicists ;)

[djpil]

This pic was taken quite some years before John and I started but very similar to our design office. Loft table was in a separate room. http://www.ctie.monash.edu.au/hargrave/images/CACDESOF_1000.JPG

Actually, I do recall some slide-rule energy failures.

Owain, I experienced the same.

[Bye]

nice photo. our lofting tables were about 50 ft x 20 ft wide.

it was shoes off and get up and on it, with old jumpers taped around knees.

[john_tullamarine]

Actually, I do recall some slide-rule energy failures

.. all mine were after long lunches down at Port Melbourne .. Wriggles used to get quite huffy at my postprandial nodding off.

Strange about one's physical failings ... been a problem all my life .. I had a reputation for nodding off in the cockpit and the boys at work make jokes at my expense on the subject ..

[Bye]

probably not true, but i used to have a stress engineer colleague who claimed to have been on the original comet stress team, and he always told us how some of the numbers they were getting from certain areas, were far above expected, so they disregarded them as spurious.

so here endeth the lesson, don't disregard, disprove.


edit, just been in the loft ( attic ) and found my old slide rule, and a mechanical calculator with instructions, with a certain 2 letter logo embossed in the hammerite painted case, she's in fine fettle, must be worth a few bob on ebay.

[Owain Glyndwr]

Slide rules wouldn't be any use today - you have to be able to add and subtract using pen and paper.....

My old boss once remarked that when he was a very junior stressman and his mistakes cost the company a few hundred, he was given a large circular slide rule, but now he was TD and his mistakes might cost millions he only had a six inch rule.

[Machinbird]

I went to McDonnell Aircraft to pick up a new aircraft for my squadron. A good friend who was working in engineering showed me his work area.

It was in a large room the size of two basketball courts chockablock with narrow aisles and desks shoulder to shoulder in rows.

The theory was to put the engineers close enough together that they could take a few steps and communicate with the all the other engineers in their work group, or take a few more steps and communicate with adjacent work groups as necessary.

This was how the F-4 and F-15 were designed in the days before CAD.

[LeadSled]

I can't recall a single occasion where my slide rule stopped working because of battery problems

Still have my Hemmi 255, works as well as ever. But I must admit I have not used all my old pens for a long time, given the excellent refillable drawings pens available, and where CAD is not worth the effort.
One thing I have noticed recently is how much the price of wide carriage printers has dropped.

[Bye]

god the time i spent washing out ink pens in the sinks was horrendous, i even resorted to taking the pens on holiday for 2 weeks once to avoid the dreaded monday morning dried out pen.

we used to draw on vellum after the linen was replaced, and my old section leader would come around with his pipe alight, take a look at the scheme and blow down the pipe, the following shower of sparks would ignite lots of little holes in the vellum which then had to be designed around, which is why the core fairings on the E4 are the shape they are ;)

[LeadSled]

which then had to be designed around

Folks,
Love it. Only slightly of the thread, but reminds me of being handed up sudden major heading changes in the middle of the night, in the middle of the Indian Ocean. Seems an unthinking F/E had put his teacup on the N/Os chart.
The heading changes were to navigate around the teacup!!

[stressmerchant]

I went to McDonnell Aircraft to pick up a new aircraft for my squadron. A good friend who was working in engineering showed me his work area.

It was in a large room the size of two basketball courts chockablock with narrow aisles and desks shoulder to shoulder in rows.

The theory was to put the engineers close enough together that they could take a few steps and communicate with the all the other engineers in their work group, or take a few more steps and communicate with adjacent work groups as necessary.

This was how the F-4 and F-15 were designed in the days before CAD.
When I was an undergraduate we had a guest lecture by very senior design engineer who referred to this "everyone together" approach - apparently it was considered to be the best way of ensuring ease of communications. I subsequently worked at a few places where they adopted this approach - large open plan areas, people sitting close to each other. The idea was that you should be able to talk to your immediate team without having to walk out of one office and into another. I was subsequently worked for a firm where seniority came with its own office. I declined the office, and insisted on sitting in the open plan area with my team. Seemed to work well, and some of the other groups commented that they wished their managers were more integrated with their teams.

My current employer places people in different locations - the CAD design draftsmen are in a different building to the senior engineers managing the design, and the junior engineering was outsourced to another country. To say that communications and teamwork have suffered would be an understatement.

[Bye]

sorry, even further off thread, Leadsled, a very good mate of mine known as "the model citizen" in OZ was ATC for many years down there and has told me some stories that make your tea cup avoidance procedure very very real, in fact mild by all accounts.

[Hursleypark]

Hi jxk

I have just found your post from a few years ago about Supermarine at Hursley ...

I was around in the days of the Attacker, Swift and Scimitar and remember that a full size wooden mock-up of the later was in the 'sheds' at Hursley (near Winchester). To reach this mock-up, draftsman had to negotiate a lengthy walk across a field of nettles armed with a ruler to take measurements and then go back to their drawing-boards and create the 3 view drawing.

I am the (volunteer) historian at Hursley Park and would very much like to chat about your time there with Supermarine if that is possible?

[Pugilistic Animus]

When Computers Were Human...excellent read
https://www.google.com/search?q=when+computers+were+human&client=ms-android-metropcs-us&source=android-browser&prmd=ivsn&source=lnms&tbm=isch&sa=X&ved=0ahUKEwjhhp-UxfDSAhWpKsAKHRiFD2UQ_AUICCgB&biw=320&bih=486#imgrc=gYt8vEND5idZUM:

[DaveReidUK]

When Computers Were Human...excellent read

Or. for a somewhat more lowbrow, but nevertheless entertaining, illustration of human computing:

http://www.kalambakal.com/wp-content/uploads/2017/01/hidden_figures.jpg

[Zaphod Beblebrox]

Perhaps some of you old timers can tell me if my thinking is correct in this regard. I have a picture in my mind of the old time design and manufacturing process. It had its limitations and one of these in the matter of stress and strain was estimates.

The old boys built em good! There was always a little bit more than required in each part and collectively those additions made for a very strong airframe. My company operated the BAC-111 until the 90's. We had DC-9's until 2002 and they were all designed "hell for stout" and stood up very well but they were very heavy airframes. The old joke was if you drove an tug into a parked BAC-111, you broke the tug. The old Douglas Direct Cable and Airplane Company came in a close second.

By way of an example, TWA Flight 841,On April 4, 1979, at or around 9:48 p.m. Eastern Standard Time (01:48 UTC), while flying over Saginaw (https://en.wikipedia.org/wiki/Saginaw,_Michigan), Michigan (https://en.wikipedia.org/wiki/Michigan), the Boeing 727-31 (https://en.wikipedia.org/wiki/Boeing_727) airliner began a sharp, uncommanded roll (https://en.wikipedia.org/wiki/Flight_dynamics) to the right, and subsequently went into a spiral dive (https://en.wikipedia.org/wiki/Spiral_dive). Pilots were able to regain control of the aircraft and made a successful emergency landing at Detroit Metropolitan Airpor (https://en.wikipedia.org/wiki/Detroit_Metropolitan_Airport) Despite the best efforts of the flight crew, the aircraft spiraled out of control, diving about 34,000 feet (10,000 m) in just 63 seconds.[1] (https://en.wikipedia.org/wiki/TWA_Flight_841_%281979%29#cite_note-AAR-81-08_Final_Report-1):2 During the course of the dive, the plane rolled through 360 degrees twice and crossed the Mach (https://en.wikipedia.org/wiki/Mach_number) limit for the 727 airframe (https://en.wikipedia.org/wiki/Airframe). Control was regained at about 5,000 feet (1,500 m) after the pilots extended the landing gear (https://en.wikipedia.org/wiki/Landing_gear) in an attempt to slow the aircraft[3] (https://en.wikipedia.org/wiki/TWA_Flight_841_%281979%29#cite_note-Peterson.2C_Iver_1979-3)[4] (https://en.wikipedia.org/wiki/TWA_Flight_841_%281979%29#cite_note-4) The plane suffered substantial structural damage, but made an emergency landing at Detroit Metropolitan Airport (https://en.wikipedia.org/wiki/Detroit_Metropolitan_Airport) in Michigan at 10:31 p. m. EST (https://en.wikipedia.org/wiki/Eastern_Time_Zone) without further trouble. No fatalities occurred among the 82 passengers and seven crew members. Eight passengers reported minor injuries related to high G forces. https://en.wikipedia.org/wiki/TWA_Flight_841_(1979 (https://en.wikipedia.org/wiki/G_force))

Considering that CAD/CAM now makes airframes much lighter and that computer aided design makes limits a better known quantity, would an A320 hold up as well in the above scenario? Is there as much buffer in the newer airframes.

[Peter H]

This might be of interest:
https://en.wikipedia.org/wiki/Slide_Rule:_Autobiography_of_an_Engineer

Cheap new & s/h copies available (re-search with your location & currency).
https://www.bookfinder.com/search/?ac=sl&st=sl&ref=bf_s2_a1_t1_1&qi=kEVCH1mblSuWRfjAhgaUHSj3Vf8_1490486193_1:4:19&bq=author%3Dnevil%2520shute%26title%3Dslide%2520rule

[john_tullamarine]

It had its limitations and one of these in the matter of stress and strain was estimates.

Not really .. the numbers come out of equations although some of the set up conditions may require a bit of artistic licence at times. The bigger problem was the time and labour resource limitation. Hence one needed to be a tad more conservative to make up for not being able to look at a widget in sufficient detail.

With the advent of electonic computers which are great at doing LOTS of sums very quickly, the stressman can drill down a lot further into smaller and smaller parts of the structure. At the end of the process you can realise some weight savings.

Another consideration, if I recall correctly, is that some of the Airbus Types negotiated a reduction in the proof to ultimate factor which really gave a weight benefit.

[Owain Glyndwr]

John,
I had a lot to do with Airbus loads and stressing,and I have never heard of any such negotiated reduction.
The big difference between now and then is that where in the old days resources limited one to studying say half a dozen cases on whatever depth ones budget and skill allowed, nowadays it is common to calculate the detailed loads for many more cases and many more locations and then to scan the total data set to get the worst case at each critical location. It is a different world from adding a bit of strength in case.
As a matter of fact, quite a lot of the airframe these days (or at least in the last days of metallic construction) is/was designed not by static strength but by damage tolerance, so there is a margin over the static strength requirements which I think was not there at the time of Zaphod's 727 incident.
Zaphod
Difficult to say how an A320 would have come out of that because the "Gs" are not known. A spiral dive of itself doesn't generate much g so it would depend on the severity of the pull up. All aircraft are designed to withstand 3.75g ultimate load at MTOW, so at lower weights they can stand much more. That 727 I think was at the end of the flight so it might well be capable of (say) 5g. In similar circumstances I would think that about right for an A320 also.
The overload due to high speed gear extension is another matter which would depend not only on the overspeed but also the detailed design of the gear.

[john_tullamarine]

OG,

The advent of computers for everyone certainly has meant that one can spend the effort drilling down to that which is, rather than using that which appears to be, critically limiting.

Re factors, perhaps I dreamed it all.

When the A320 first was introduced into Oz by AN in the 80s, I had a number of discussions regarding the aircraft with whatever name CASA was that week (particularly in respect of flight test matters - which would have been with one of our PPRuNe posters who was the then local regulatory test pilot). My quite distinct recollection was that Airbus had negotiated a reduction from the usual 1.5 factor on the basis of all the electronic gee-whizz protection capability.

My recollection may well be flawed, of course, .. or the brain, as the seconds tick, slowly is following the body into decreptitude.

The poster to whom I referred probably will see this thread and correct my thoughts in due course ?

[Pilot DAR]

When I worked for deHavilland in the early '80's, during the infancy of the Dash 8, it was designed with the aid of rather basic computer assistance. I think that even though computers were involved, there was a lot of old school slide rule calculations which really made the plane.

Following the fatigue test program of the "iron bird" airframe, the wing was to be pulled to failure. I recall that they installed the longest stroke hydraulic pull cylinder they could find - 8 feet of stroke. They pulled up on the wingtip with it (with the wing root very firmly held in the frame). The cylinder used up all 8 feet of stroke, yet the wing did not break. I don't recall if it "yielded", but had that been done in flight, the aircraft would still have been whole. Imagine a Dash 8 wingtip displaced 8 feet up, and not broken!

[Allan Lupton]

With the advent of electonic computers which are great at doing LOTS of sums very quickly, the stressman can drill down a lot further into smaller and smaller parts of the structure. At the end of the process you can realise some weight savings.

I remember the stress office at Hatfield found that when you got the computer to calculate LOTS of sums you still needed LOTS of staff to plough through the bales of paper output to approve (or otherwise) the MANY results.

When CAD started to appear in drawing offices some of our older friends caused a drawing sticker to be produced with a DBH logo - "Drawn By Hand".
If we Future Projects numbers men defined an aeroplane, one of those DBH drawings of a three-view GA could be ready by lunchtime, whereas coding it for CAD could take a week. The latter was better if you were going to try a multitude of small variations, but as we rarely did that DBH won.

[john_tullamarine]

DBH.

For sure.

A good mate of mine was (is) a superb on-the-board designer and draughtsman.

Utterly refused to be seduced by the new computer God .. and, likewise, could run off an original drawing far quicker than CAD output .. not to mention they were far more pleasing to the eye.

CAD, like WP .. great if you can develop and exploit a library of drawing elements and use previous drawings as the basis for new.

[Volume]

Allthough modern CAD software seems to do magic, the geometry formulae behind are quite simple and can typically be solved geometrically. One of the first lessons engineers took in the past was how to dertime the intersection of two cylinders just be use of a ruler, pen and sheet of paper. It really is not black magic.

the numbers come out of equations although Or out of diagrams or Nomograms some clever scientist were coming up with.

In old style manufacturing you anyway required simple 2D drawings, for example to do a sheet metal cutout. You needed some simple basic geometry data to build your rigs, jigs and templates. So were used to already design that way, you defined your geometry in a way that you can determine the intersections and shapes you needed.

Today you design however you like (and however Nastran gives you some nice results) and then you hae to mill the part from a huge block of metal or (in the very near future) 3D print it. Old school engineers always designed with production and strength calculation in mind. They designed shapes for which reliable stress/strain/stiffness/buckling formula were existing and which were easy (and precise) to manufacturer. Many Details on the drawings were actually just "artists impression", as they did not define actual dimensions. Especially rounded sahpes (e.g. of a fuselage nose) were somehow accidental and high tolerance between frames / stringers, as the exact shape simply does not matter. Today in CAD you have to define all shapes exactly, as the computer can not deal with "some arc between these two points", it needs to know everything exactly. Much more exactly than it can finally be manufactured...

There are great benefits of modern technology CAD/CAM/FEM tools, but when it comes to initial design, to get quick estimates, to decide whether something can be realistically done or is practically impossible, old style scratch and simple formula technique work still great. If you want to optimize the last kg of dead weight out of an aircraft, modern tools are perfect. If you want to have a quick estimate, they fail.

[Derfred]

Interesting.

The nose of a B737/B747/B767/B777/A320/A330 and indeed a DC9/MDx/717 is physically "attractive" to me, whereas the nose of an EMB1x0/B787/A350/A380 is not.

Perhaps because computers have been put in charge of aircraft design instead of humans.

This could also explain why the Toyota Prius is one of the most unattractive cars ever developed.