Jump to content
This site uses cookies! Learn More

This site uses cookies!

You can find a list of those cookies here: mysite.com/cookies

By continuing to use this site, you agree to allow us to store cookies on your computer. :)

Roy vd M.

McLaren Honda MP4/6 Ayrton Senna 1/12

Recommended Posts

Just caught up with you again Roy, she is coming along well.

It has given me food for thought when I get to build my couple I have in the stash.

 

Simon.

Share this post


Link to post
Share on other sites
21 hours ago, sharknose156 said:

The watchmaker job you did with the copper and soldering iron is amazing. A great learning point, many thanks.

Also the clever way you use different glues. This `Gator glue is amazing, i have to get some like presto.

 

Ps This Honda engine block looks much better than the one on the current McLaren's. 

 

Thanks.. when you buy Gator Glue be sure to get the brown one, rather than the white Gator's Grip Glue. White Gator's Grip glue is better known to modellers but works very differently. 

 

For sure this engine looks way better than those you'll find in modern Formula 1 cars, including the current Honda block. Although I think Honda's 2016 engine sounds nicest and loudest of them all. Let's hope 2017 will show an improvement, although not much change is to be expected engine-wise. The looks of the 2017 cars will probably be way better though.

 

16 hours ago, Dr. Fiat said:

Wow is all I can say!  Great detail and documentation!  This really inspires me for Aventador build coming soon!

 

I assume you mean the Pocher kit? Please give me a notice when you start that topic, I'm looking forward to following one of those builds from the beginning. It's a sensational model. 

 

12 hours ago, Jnkm13 said:

Absolutely amazing skills, patience and eyesight!

 

Thanks Jeremy :)

 

1 hour ago, Spookytooth said:

Just caught up with you again Roy, she is coming along well.

It has given me food for thought when I get to build my couple I have in the stash.

 

You have a couple of these? Meaning the Tamiya MP4/6 kit, or Tamiya 1/12-race cars in general? :o :)

 

 

104. I've been busy playing with Gimp, Inkscape and Blender, because these software programs have come to interest and fascinate me and @NickD has been preoccupied in another project. I can't thank Nick enough for introducing me to these new opportunities for modeling. 

 

First let me explain in simple words what I've been trying to do and why. 

 

It is my goal to determine the size and shape of two suspension parts, so that I can scratch build them to a realistic appearance, to fit in their correct positions. 

 

The suspension parts are support brackets for the rear wheel suspension, as discussed in the previous posts. Not only am I looking to learn their dimensions, I also want to know where exactly they are to be positioned relative to the gearbox housing. 

 

On the next picture you can see these parts, left side of the gearbox. Mirrored they are also, as visible in the picture, on the other side. 

30054685943_a32030f155_b.jpg

The Top Studio parts meant to represent these support brackets are too broad, their shape is incorrect AND they are not positioned correctly. See the next picture (it's all about the white model parts). 

29923034090_9c7cbe2330_b.jpg

I'm currently in the process of trying to fetch a 3D-shape from 2D-pictures. Had there been official technical drawings of the McLaren MP4/6 I could have used those, but unfortunately I have not been able to find them. Therefore I will have to use photographs. I checked if Top Studio's technical drawings are correct but came (as did Nick) to the conclusion that they are not. I'll describe that process hereunder, first I'll finalize this introductory section. 

 

The main problem with photographs is that there is lens distortion, which needs to be corrected. But first, I have to find usable photos.  

 

Suppose I want to make a scale model of a roll of solder wire. As a reference I have a couple of photographs. For example there is this picture:

30600695121_397a0f7037_b.jpg

The problem with this photograph is that I cannot assess how tall the roll is nor how broad the roll is, because the photo wasn't taken frontally or from above. What's more, the sanding stick is blocking the view. Therefore I cannot measure the roll in an accurate way. 

 

Suppose I have pictures taken from above and from the side; then I can use those pictures to replicate the object. 

 

Here's a picture of the roll's side:

30388643160_8fe7dfa0b6_b.jpg

From this picture I could assess the relative hight and width (diameter) of the roll, if the image isn't distorted. 

 

Next there's this picture taken from above:

30688878645_8db2c14889_b.jpg

The outer diameter and inner diameter (=the hole) could be derived from that... again, if the image is not distorted. 

 

We know that the hole runs all the way through the roll. The height of the two concentric shapes is therefore equal. 

 

Suppose we also know how wide such a roll is (Googling tells us: 6 centimeters). Now one is almost able to make a 3D-model. I say almost because these photographs are subject to lens distortion. Because a photograph is made using a more or less rounded lens, there's always a slight distortion of the captured image. 

lensvervorming.jpg
(Copyright Eoszine.nl, educational purposes only, picture will be deleted upon first request)

So first of all, this lens distortion has to be corrected. Nick introduced me to certain software (Gimp) that does just that. Not easy, but possible to a great extent. I'm currently in the process of learning a lot of stuff from a book written by one of our forum members, @Witold Jaworski. The book is called 'Virtual Aircraft' and can be found and purchased here. For those who are curious about the possibilities Witold has explored, see this amazing thread of his.

 

Back to the story. Using both distortion-corrected pictures of the soldering wire roll, in the software program Blender you can make a 3D-shaped roll. Simplified it works as follows: import the photograph taken from above into Blender, draw two circles exactly along the outer perimeter resp. inner perimeter; import the photograph taken from the side into Blender; rescale that picture in such way that the width of the roll becomes exactly the same as the perimeter of the largest circle; and in final, 'pile' as many of those concentric circles on top of one another until you reach the correct height of the roll, as can be deducted from the side view.  
 

That way one can draw complete machines or vehicles with great accuracy, using a 3D-program like Blender. 
 

 

104a. First exercise was to check if Top Studio's drawings of the engine block and the transmission house (left side view and top view) are relatively consistent. Nick already told me that the drawings didn't match each other and I soon found out that was very true (I didn't at the least doubt Nick's assessment, just wanted something to practice on).  

 

First I imported the drawings in Gimp (the top drawing was mirrored). The two oil containers were matched: relative to each other they appeared to fit very well. See the blue guide lines hereunder.

 

Right away you see that the engine on the lower drawing is much smaller than the top one, and that the gear box on the top drawing is longer than the one as depicted on the lower drawing. I placed guidelines at the 'H'-logo. Now the question is, which drawing is (more or less) correct, if either is.  
 

30343245100_771b23494b_b.jpg

The next drawing clarifies where the differences are. As said, the oil containers are equally large on both drawings and are positioned in the same spots. That part of the drawings is, as such, coherent. But there are at least three remarkable differences. I marked them on the next screenshot. 

30555960351_ef9f1b30d3_b.jpg

 

 

108. Next I used a photograph, seemingly taken almost straight from the side of the gear box. The photo was taken from the right hight. Lens distortion was not corrected in this photo. 

 

Ad 5. This circle matches the position on the circle seen on the picture of Tamiya's model as well as the 1:1-picture. Top Studio's drawing deviates slightly. That's good news for usability of the Tamiya-part, not so good news for usability of Top Studio's drawing. 

 

Ad 6. Before the photo could be used I checked whether this circle was nice and round. The circle-check-method is one of numerous tips to be found in the E-book Virtual Airplane by Witold Jaworski. As mentioned before I've been using that book as a manual to get to know the software programs of Gimp, Inkscape and Blender. Almost all methods to be described in this thread will be used as found in that book, be it that the book describes modelling of a WWII plane. But the techniques can just as well be applied to other objects such as a F1-car. A big difference is that I'll mostly have to use photographs because accurate MP4/6 drawings or blueprints are not available.

 

Ad 7. While here the rear slanted line above is equal to that on the Tamiya model, the slanted lines below differ. 

 

For the rest it can be concluded that the real measurements (at least) roughly comply with the side drawing and Tamiya's part. That's good news, because the most important thing is that the Tamiya model (literally and figuratively) matches the real thing, grosso modo. 

30039052783_b122ef9a3d_b.jpg

 


Spent time: 25 hours (build) + 34 hours (study)  

Share this post


Link to post
Share on other sites

Sorry Roy, 2 1/12 Tamiya Cars, The Porsche 935 "Martini" and the Yardley MacLaren .

Picked both up quite cheaply a couple of years ago from Modelzone before and as they went tits up.

 

Simon.

Share this post


Link to post
Share on other sites

Wow, thanks for the credits - but I seem to have spawned a monster! Outstanding analysis. You clearly have gone up your learning curve spectacularly quickly and highlighted some tools I knew about but had forgotten to use - doh.

 

One thought - if the core tamiya part (engine and gearbox) is close enough I wondered about (and actually started) measuring it in some detail. It would provide "truth" for (and perhaps help simplify) aligning photos to allow the simplified brackets to be scaled effectively. Of course, as I keep finding, all of this distracts from actually building.

 

Finally, photo alignment is challenging - sensitive to very small misalignments both in angle and focal length. Usually it is highly iterative. Again good truth geometry woudl be a godsend

 

Regards

 

Nick

Share this post


Link to post
Share on other sites

 Really many thanks for these detailed info and taking the time to explain it very well.  Thank you also to Nick !

 

Are you using a Mac for this ?  i plan to order the book of @Witold Jaworski and perhaps enter into this 3d imaging blender stuff during x mas.

 

On Tamiya's 1/12;  i plan to start one of the old 70's F1's after the Mef. but i am not sure which yet, possibly the Yardley McLaren M23 or the Tyrell 003.

i have a few of these 1/12 in my basement :)  .

 

 thank you again :cheers: 

Share this post


Link to post
Share on other sites

I am absolutely not into F1 cars, but I will follow closely. Read this whole thread and I am very impressed. Much to learn for a beginner. Thank you for taking your time for this detailed write up.

Share this post


Link to post
Share on other sites
On 10/31/2016 at 10:52 PM, Spookytooth said:

The Porsche 935 "Martini" and the Yardley MacLaren .

 

Great stuff, I'd love to see the Yardley McLaren being built.

 

On 10/31/2016 at 11:17 PM, NickD said:

One thought - if the core tamiya part (engine and gearbox) is close enough I wondered about (and actually started) measuring it in some detail. It would provide "truth" for (and perhaps help simplify) aligning photos to allow the simplified brackets to be scaled effectively. Of course, as I keep finding, all of this distracts from actually building.

 

According to my findings the Tamiya gearbox reflects the real thing very well (at least from above). I'll definitely use Tamiya's part as a starting point. By the way, as expected I'll have to slightly amend the wishbones... (see hereunder update).

 

On 11/1/2016 at 1:27 AM, sharknose156 said:

Are you using a Mac for this ?  i plan to order the book of @Witold Jaworski and perhaps enter into this 3d imaging blender stuff during x mas.

 

On Tamiya's 1/12;  i plan to start one of the old 70's F1's after the Mef. but i am not sure which yet, possibly the Yardley McLaren M23 or the Tyrell 003.

i have a few of these 1/12 in my basement :)  .

 

Yes I'm using a Macbook Pro, works perfectly for all three programs. I feel Witold's book is a must-buy because it saves a -lot- of time and effort as compared to gathering all information from Youtube, forums et cetera. The book (1300+ pages) comprises everything you need to know, specifically for our purposes. It's simply fabulous.

 

On 11/1/2016 at 1:52 PM, Schwarz-Brot said:

I am absolutely not into F1 cars, but I will follow closely. Read this whole thread and I am very impressed. Much to learn for a beginner. Thank you for taking your time for this detailed write up.

 

I'd kindly like to ask you what part of the thread appeals to you most. Maybe I could elaborate a bit more on one subject or another. Thanks for your interest, especially seen that you're not a F1 car fan.

 

 

 

109. I've been reading through plenty of texts about soldering. My most vivid memory of this will probably be how many different opinions (=truths) there are and how fiercely people will defend theirs. It's absolutely astonishing and even a bit disturbing. 

 

A number of things I found out:

 

1) There's a significant difference between soldering electronics and soldering non-electronics.

 

2) Regarding the second subject, little was written online. 

 

3) In electronics soldering, people almost always use solder with flux core (60% tin / 40% lead), whereas in soldering brass and photo etched metals people tend to use solder without flux core (50% tin / 50% lead).


4) Regarding the latter case, the flux will have to be applied separately. The reason modelers choose that method is that you can control more precisely what route the solder will flow along. 

 

5) There are several types of flux. Home depot-stocked flux is (at least the one in Dutch home depots) not a good idea. Flux for electronics should be used for that purpose only. Tip activator is hard so you can't use it. For that reason I purchased pasta-flux and it works excellently. 

 

The materials I've gathered (and which seem to work)... mind that several alternative good project can be used:

 

* A Toolcraft soldering station (Digital Solder Station St-100D). Crucial is a usable wattage. 100 watt is sufficient for all reasonable purposes. My soldering station contains a cleaning box for the solder tip. 

* Soldering tips chisel-shaped. 

* 50/50 solder from the home depot.

* Nokorode regular paste flux.

* Weller tip activator (to clean the tip).

* Desoldering wire.

* Two 'third hands'.

* Fireproof tile.

 

- Using these tools it's possible to solder photo etched metal or brass sheets / tube. 

 

110. After a lot of testing, at one point I found out how to do what Paul Budzik does in his video (2:20), click here

 

The following piece was soldered in only 10 seconds (well... after 3 hours of repeated attempts), precisely in the way Paul Budzik demonstrates. No corrections have been made to the part.

30717430005_9f4ea6c786_b.jpg

30600590182_19b3b058fe_b.jpg

111. Back to the photogrammetry. 

 

Many pictures of the brackets seem to suggest that the brackets are not positioned perpendicular to the gear box. I checked this using a photograph that has little lens distortion. In Inkscape it's easy to draw help lines. The result: the brackets are truly perpendicular! Amazing what lens distortion can do.

30702832316_e3cc4d4e5f_b.jpg

112. The most important picture taken from above I tried to correct as well as possible (lens distortion). If anyone is interested in how I did this I'll describe it a bit more in detail. That remark applies to every step.

30622955472_5e1e07ed0f_b.jpg

113. On top of the picture of the gear box topside I put a picture of the side. Each picture is on a separate layer, so you can amend one at a time. And it's easy to make one picture (half) transparent, to name an example. The side photo hardly had any distortion. It was scaled and moved parallel to the photograph from the top, using guide lines. After that I did the same using another picture from the side. The very good news: the length lines and positions match almost exactly. This means the lens distortion was successfully corrected. 



30739603795_63f2c34ba1_b.jpg

114. In Inkscape I had drawn a red line indicating the outer lines of the gearbox.  

30651492131_3309d3c0a7_b.jpg

115. Back to the software program called Gimp. The picture from the side comes to rest parallel to the picture of the top. The top side is correct up and including the suspension support brackets. Beyond that point there's still lens distortion. Those positions are not relevant here. 

30651491981_e7eda0d509_b.jpg

116. Next I imported and moved a (scaled) photograph of the Tamiya-parts to fit between the guide lines. I was very relieved and happy to see that width and length are exactly correct (the hereunder picture has the vertical help lines too far to the right). 

 

Excellent work Tamiya!!!

30703080036_b0ff6db9c7_b.jpg

117. If we do the same with Top Studio's drawing we can come to the conclusion that the drawing wasn't based on the real model nor was it inspired by the Tamiya model NOR was it based on Top Studio's parts. In short, the brackets on the drawing don't match any of my sources. 

30651491171_9c61a13060_b.jpg

118. Because the ends of the brackets are subject to lens distortion I overlapped them with better images of the brackets from another (non-distorted) photograph. It makes for a visible difference. Picture 1: the old, distorted image. Picture 2: the new image on top of the old one. By the way the new image needs to be horizontalized. I'll do that in step 121. 

30703505686_b3f3c2abf8.jpg


30703505246_6a4413a33e.jpg


119. Then finally the end result is in sight. To the left is the bracket with (almost) correct positions and measurements, to the right is the Top Studio-part. As can be seen, four important things don't match:

 

1: Top Studio's part is (as we already knew) too wide. The exact difference was unknown until now. After my manual drawings I thought the part needed to be made 2,25mm. wide but I was wrong; it will have to be 2,00mm. That's quite a difference (12,5%... again, thanks Nick for guiding me toward this insight!) but the really important difference is, of course, with Top Studio's brackets. They are 2,7mm. wide. That's 35% too wide!

 

2: Top Studio's part has its 'bend' in the wrong place. We can now measure exactly where the bend has to be. (By the way, everywhere I say 'exactly' I mean 'almost exactly' of course. There's always a fraction of distortion). 

 

3: Top Studio's part is situated quite a bit too far to the front of the gear box (on the picture: it has to be moved a bit down).

 

4: The wishbone will be connected to the bracket in another position. Consequence is that the wishbone will also have to be amended, very slightly. This is also something I had anticipated because I had noticed that the rod between the two brackets on one side (=the short bar of the wishbone triangle) should run slightly angled rather than parallel to the gearbox. Tamiya and Top Studio have incorrectly made that bar run parallel to the gear box. 

30622954432_47e85fc221_b.jpg

120. Eventually this will be a great visual difference, small as these parts may be. Here five screenshots that document the difference: 'From Tamiya to scratch in five steps'.

30440079330_6cdb125013_b.jpg

30623885762_a5d344f633_b.jpg

30623885702_3a372cd55e_b.jpg

30623885572_8bb383a49e_b.jpg

30623885472_724046f733_b.jpg

I now feel as if I solved and finished a jigsaw puzzle: very satisfied! It's a great pastime pursuit (to me) because it feels like a quest for the truth. Sounds a bit dramatic maybe but that's how I have experienced it. Side advantage is that I now know that Tamiya performed their measurements very well as regards the large shapes. Only the brackets have been given a different shape but I still think that has to do with the required strength of the bracket parts. 

 

121. To finalize the above story... correction of the bracket ends. 

 

How it was...

30624493852_3e073b8893_b.jpg

... and how it became:

30108972054_fc4df12986_b.jpg

In this 9 second video you can spot the difference more easily.

Spent time: 25 hours (build) + 40 hours (study)

Edited by Roy vd M.

Share this post


Link to post
Share on other sites

Amazingly, while reading through your great post i was thinking of sending you or posting a link to Paul Budzik's video ! then it appears a few lines later....by the way he is accessible and replies to emails. i am sure he would greatly appreciate this thread. 

Share this post


Link to post
Share on other sites

Roy,

 

Mega-post. I have never managed to get a grip with Inkscapes interface. Combination of it being quirky and me being very lazy about being methodical. Another fine bit of analysis.

 

Couple of observations - again, I might not have read the post closely enough but a couple of things struck me (in no particular order).

 

  1. None of the pictures is truely vertical - difficult to do without a platform or actually standing on the car (which might not go down too well). There is therefore some residual pitch angle which will shorten the fore/aft dimenstions. i.e. if the width is OK the length is too short. Could fix with a linear stretch but it starts to feel a bit arbitrary. I'd hope it did not affect things too much but have been bitten before.
  2. Warning - This is definitely "I may be wrong but..." territory - The undistorted image will suffer from 2 different effects - parallax/perspective (due to camera focal length) and barrel distortion. The method you describe fixes the barrel distortion but obviously not the parallax. The approach raised two slight worries for me. In the original picture, barrell distortion will curve straight edges. If you look at the orange bodywork, they look to the naked eye quite straight. Only Inkscape will tell. But it would be worth a check. In the "corrected" picture the orange bodywork is definititely a bit curved to say the least. I think if it was me. I would have corrected by refering to the long straight edges (bodywork, driveshafts and transmission edges) and then attributed the rest to parallax and camera angle. The other really usefull straight edge is the anti-roll bar which appears very curved. Might be worth a bit of further study.
  3. If the Tamiya G'box is generally accurate, it seems strange that the suspension pickups are so far out in terms of Gearbox position. If they had access to better information than us, and they are certainly better modellers than me, why would they get that wrong. If it is not wrong, what drove them to simplify the model in that way. Thought of in this way can you think of a motivation that might have caused them to put the longitudinal position of the bracket in the wrong place?

Hope that makes sense/is constructive/ and ultimately is wrong!

 

ATB

 

Nick

 

PS the only way to fix the perspective distortion properly is with a tool like blender.

I

Share this post


Link to post
Share on other sites
17 hours ago, Roy vd M. said:

I'd kindly like to ask you what part of the thread appeals to you most. Maybe I could elaborate a bit more on one subject or another. Thanks for your interest, especially seen that you're not a F1 car fan.

 

Hi Roy,

your thread is generally appealing to me. Especially superdetailing and scratching practices are very interesting to me, as this is the stuff that's coming up with my project right now.

The analysis of photographs is another huge topic. I do not have the time to dig too dip in there yet, but that may very well happen in the future and therefore I bookmarked this thread.

 

Don't worry about going more into detail. You already do a superb job, describing what you're doing. If there are open questions I will speak up.

 

Thank you!

Share this post


Link to post
Share on other sites
9 hours ago, NickD said:

Mega-post. I have never managed to get a grip with Inkscapes interface. I

 

Personally thus far I haven't experienced much difference in ease of use between Inkscape or Gimp, although it's clear that Inkscape wasn't programmed for the Mac. Nice thing about Inkscape as far as I've been able to find out is that you can draw diagonal (and even multi-angled) guide lines whereas you can't in Gimp. And drawing of contours is quite easy. It is all described and explained in Witord's book on pages 1193 and further, and pages 694 and further. 

 

Thanks for your ongoing support and help in the matter of getting the measurements right. I feel I have an answer to most of your issues and still stand, largely, behind the derived measurements. However you are certainly right on some aspects (actually you're correct on all accounts but I think I thought of most things you described) and I will correct the images to that extent. Your help has been invaluable!

 

I'd like to comment on some of the issues you discussed, including aspects that require improvement.  

 

1. None of the pictures is truely vertical - difficult to do without a platform or actually standing on the car (which might not go down too well). There is therefore some residual pitch angle which will shorten the fore/aft dimenstions. i.e. if the width is OK the length is too short. Could fix with a linear stretch but it starts to feel a bit arbitrary. I'd hope it did not affect things too much but have been bitten before.

 

The main picture (starting point) is almost vertical to the central gear box section. In my estimate it was taken approximately above and toward the drive shafts. Now the only part of the picture I have effectively used for measurement purposes is this:

 

30758619385_c3e692a769_b.jpg

 

That's only 1/11 of the surface of this picture. The selection also includes the parallax-neutral point somewhere around the drive shaft, so the surrounding distortion might be expected to be minimal. When I used 'correct lens distortion' in Gimp I was only looking for getting the gear box length lines straight, as well as getting the bracket lines straight. As I reckoned the differences between the pre-correction and post-correction images is minimal:

 

30758619385_c3e692a769_b.jpg

 

30126544024_cf8487dd23_b.jpg

 

The difference is a few pixels at most. The remainder of the picture is of no use for me at this point, so I didn't amend it further.

 

Room for improvement #1: distortion correction needs to be originated from another parallax focus point. 

 

I used Witord's method of photo lens distortion correction, to be found in his book on pages 1189 and further. 

 

After correction, the length (from front to aft) was checked alongside two side view photographs. They matched very well.. From these pictures I only used these measurements: 

 

30123199363_2dbc0bf522_z.jpg

 

30123199443_0843f45ddf_b.jpg

 

The latter picture was mirrored. 

 

Room for improvement #2: as the pictures key measurements fitted the top view so well and they looked so non-distorted, I didn't correct them. I'll check that, just to be sure.

 

Key positions in these pictures proved to match the corrected top view remarkably well. As these pictures from the side are almost non-distorted and the several parts along the above-shown lines are approximately at the same distance from the viewer, I trust the findings to be correct.. be it that I will perform the above-mentioned check anyway just to be completely sure. 

 

To position the brackets as correctly shaped as possible I copied them (using the bolt positions as seen in the main picture) from a picture taken almost from above a bracket. See paragraph 118 above. 

 

2. Warning - This is definitely "I may be wrong but..." territory - The undistorted image will suffer from 2 different effects - parallax/perspective (due to camera focal length) and barrel distortion. The method you describe fixes the barrel distortion but obviously not the parallax. The approach raised two slight worries for me. In the original picture, barrell distortion will curve straight edges. If you look at the orange bodywork, they look to the naked eye quite straight. Only Inkscape will tell. But it would be worth a check. In the "corrected" picture the orange bodywork is definititely a bit curved to say the least. I think if it was me. I would have corrected by refering to the long straight edges (bodywork, driveshafts and transmission edges) and then attributed the rest to parallax and camera angle. The other really usefull straight edge is the anti-roll bar which appears very curved. Might be worth a bit of further study.

 

Don't mind the orange bodywork, as already mentioned above. I don't use that part of the picture right now and I don't think I will need it later, because I have better photos of those parts. 

 

Barrel distortion was corrected until the bracket lines AND the length gearbox lines were straightened. Parallax effect is minimal because of the very small photo surface used. Plus, as said, parallex effect was successfully checked against two side views. I used guidelines in Gimp to get the lines dead straight. 

 

Room for improvement #3: you're right about the anti-roll bar. I'll check whether using another distortion correction origin will have an influence on that. I think that for the end result it won't matter more than 1 or 2 pixels but still it's worth a try as every step towards higher accuracy is welcome. 

 

3. If the Tamiya G'box is generally accurate, it seems strange that the suspension pickups are so far out in terms of Gearbox position. If they had access to better information than us, and they are certainly better modellers than me, why would they get that wrong. If it is not wrong, what drove them to simplify the model in that way. Thought of in this way can you think of a motivation that might have caused them to put the longitudinal position of the bracket in the wrong place?

 

As mentioned in paragraph 120 and somewhere earlier in this thread I am convinced that Tamiya's engineers thought it preferable to have a strong suspension bracket part, as the weight of the car (partly) rests on it.

 

Some aspects:

- Not only does the weight of the car rest on it, Tamiya has also intended this car to feature functional suspension (who uses or tests that??). The stresses on the plastic material are enormous. 

- That (required strength) is most probably why the bracket was bent centrally rather than at the end. That's also why it is so wide and that's why the plastic is so thick. 

- The reason why the bracket was moved forward a bit is (educated guess again) for visual attractiveness. If the thicker part were to be in its original position, the suspension part would appear over-scaled, clunky and deformed. It's the left end of the suspension, so it was not opportune to have that left end move toward the right end as to prevent the wishbone from getting too narrow.

- Top Studio followed Tamiya's take on the bracket part as they had to work with resin, not the strongest of materials. Furthermore they probably didn't want to change any suspension pivot point, to prevent further corrections. In final, Top Studio was bound to Tamiya's bracket position. The reason is that the whole suspension arm assembly goes through the gear box housing - again, for strength purposes!

 

If these assumptions are correct I truly understand Tamiya's worries. I've been thinking for many hours how to solve the strength issue. Ultimately I have not been able to find another solution than to solder the suspension construction. 

 

On to correcting previous calculations and measurements!

 

46 minutes ago, Schwarz-Brot said:

your thread is generally appealing to me. Especially superdetailing and scratching practices are very interesting to me, as this is the stuff that's coming up with my project right now. (...) Don't worry about going more into detail. You already do a superb job, describing what you're doing. If there are open questions I will speak up.

 

Thanks, I will just continue as usual then :)

Share this post


Link to post
Share on other sites

I just showed the above workflow to my girlfriend who is an art teacher with experience in photography.

 

She is usually the one to point out flaws in my modeling (telling me "what? You told me to be fully honest" upon seeing my annoyed face while I'm taking the sanding stick to redo the part). Confronted with the above-described schedule and workflow, she deems the measurements to be highly accurate. 

 

Of course it will never be perfect unless I'll have access to the real deal. I don't think McLaren will be happy with me measuring their priceless car. One scratch alone would be unacceptable and unforgivable... 

 

I'll be very happy if the final part looks like the real deal. I'm not looking for an exact scaled-down replica. Suppose in the end there's a 3% deviation. That's still way better than the 35% difference between the real brackets and those provided by Top Studio. And then there is the difference in bracket position... 

 

So don't worry too much about it, I think I'm going as far as I reasonably can with the help of these pictures. It's enough to make me excited about these findings and the way this project has turned to. 

Share this post


Link to post
Share on other sites

if i were you i would contact the Mclaren factory and ask to measure it,  the worst you could get is a 'no'. they would send you the measurements most probably.

and if not the factory, if i were you i would try to contact Neil Otley directly, or Steve Nichols.

 

it s a pity this is the first car Gordon Murray was not involved in, am pretty sure he would reply for such a request.

Share this post


Link to post
Share on other sites

@sharknose156 That being a good idea in itself, I'd prefer to reserve a bit of credit with McLaren for questions that may arise in the near future. I could e-mail Mr Neil Trundle requesting what you suggest and, friendly and courteous as he seems to be through the short e-mail correspondence we had, it would not even surprise me if he were kind enough to take these measurements... but I'd rather contact him when it is really, really necessary.

 

Plus, I must confess that this drawing business is rather exciting stuff and I much enjoy it. It's also a good practice for a future scratch build... :)

 

Hope you understand. 

Share this post


Link to post
Share on other sites

Yeah haha CrazyCrank we share some experience about asking questions to a information monopolist who has not been friendly or courteous... :mellow:

Share this post


Link to post
Share on other sites
6 minutes ago, Roy vd M. said:

Yeah haha CrazyCrank we share some experience about asking questions to a information monopolist who has not been friendly or courteous... :mellow:

 

Unfortunately for me, and him ! (contempt has often a boomerang effect) 

Share this post


Link to post
Share on other sites

122. After @NickD's suggestions I completely redid the photo aligning procedure, being much more precise and elaborate than last time and distortion-correcting all used photos. In total I now used six photos. Most lens distortions only needed minor corrections. My findings (partly relative to the previous drawing):

 

- After distortion correction the pictures match very closely (relative to each other), some being very slightly off. 

- The brackets appear to be a bit wider after all.

- The brackets turn out to be positioned slightly more toward the back of the gearbox (the original position on the Tamiya part is, therefore, incorrect to a lesser extent). 

- The large oil container needs to be repositioned toward the back.

- The measurements of the Tamiya part are (still) mainly correct. For example the two oil containers have the correct sizes. To give an example, the size of the small container differs less than 1% from my photographic findings. 

 

Based on the six pictures I made an overview drawing (not finalized yet, two more sets of suspension arms need to be included). Along with drawings of the side and front, this drawing will be used in Blender. 


The guidelines were used to align segments of the six photos with the drawing. Later on they will also act as leads to create the side and front drawings. 


30797247575_f3eab80ced_b.jpg

Without guidelines:

30797247735_80bd297e48_b.jpg

Spent time: 25 hours (build) + 48 hours (study)  

Share this post


Link to post
Share on other sites

@NickD that's for sure! I'll refrain from name checking this time :D 

 

 

123. I dryfitted one side of the rear wheel suspension, because I wanted to compare the preliminary drawing outcome to another photo. For that I needed another picture taken from above the kit parts, this time dryfitted as shown hereunder. The photo used for the drawing is in itself not so interesting, I prefer this for the topic:

30867437825_f1d7f0021b_b.jpg

 

What a great fat tyre haha

124. The bracket ends on the drawing were amended based on my new conclusions and the front arm for the springs was drawn. The current state of affairs:

30566186200_7890a2eb67_b.jpg


Spent time: 25 hours (build) + 51 hours (study)

Share this post


Link to post
Share on other sites

Roy

 

I trust you are well.

 

I am new to the Britmodeller forum but I have been bowled over by the general support for everyone who posts their work. I, like some of the other contributors to your very comprehensive updates of your progress, have the MP4/6 on the 'to do list'. Your dedication and detail bring to mind Andy Mathews's work but I would like to thank you for taking the time to share your thoughts and observations and will await the developments with considerable interest.

 

On the AM front, do you have any views on the Thunder Valley (from Paul and his team) offerings? I have the brake and suspension set which are of very high quality.

 

Many thanks and best wishes.

Share this post


Link to post
Share on other sites

@SlowQ A bit ashamed to admit, but this beautiful kit and its superb aftermarket detail part have become the victim of my new love for lathe and mill. 

 

As a side project I became interested in researching the history, looks and dimensions of the Fiat 806. I wasn't even going to build that. Then, I was going to build it. Then, I gave up on building it, to start work on another legendary Grand Prix car, the Delage 806. As soon as I have the dimensions of that care (hopefully this week) I will resume working on the Delage. The MP4/6, beautiful as it is in real life and as a kit, will remain a future project... 

 

Regarding the Thunder Valley aftermarket parts, I only have:

 

- the 'Calsonic' photo etched sheets which, on second thought, I was not going to use because as it turns out the text wasn't featured on the car during the race season; as well as:

- the cast metal parts. Quality seems to be nice but not perfect.

 

Here are some of the parts:

 

34689146571_db0be27681_b.jpg 

 

Share this post


Link to post
Share on other sites

Roy

 

Many thanks for your reply. I am glad you are still very active as your MP4/6 posts just seem to stop late last year and I feared the worse. 

 

I agree that the TV castings are interesting rather than exceptional but I fear that if you use them all the weight of the finished model would be considerable.

 

You are undoubted a highly skilled modeller and your lathe work seems to be in the same vain. I hope this new passion will serve you well and I will keep an eye out for when you return to the Mac. As you say it is a great model but a mighty project.

 

Best wishes and I will look out for your posts.

 

Regards

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...