Mercedes-Benz Kfz.384 Refuelling truck (full scratch) 1/72

[Roy vd M.]

Introduction

For a diorama I intend to build this refueller. The Kfz. 384 was produced by Mercedes-Benz and used by the Luftwaffe. It was succeeded in 1941 by an Opel Blitz refuelling truck. No fully intact vehicles have survived but there are one or two specimens that may be restorable. This topic is a translated summary of a build report (in Dutch) elsewhere. As translating takes some time, I'll post this in sections. 

 

The kits

Original idea was to make this a kitbashing project (and maybe I'll be able to use parts of the two kits I have, although I doubt it; if not it will be full scratch). They are kits of Mac Distribution and HP Models. During another build I made 12 attempts at making louvres; the 13th attempt would involve photo etching. Now that I have a milling machine, I decided to give that a try. 

 

The louvres, part 1

The Mercedes-Benz (hereinafter also 'MB') has a very long row of very narrow louvres. In total there are no less than 43. 

 

11. Attempt 13, milling machine at 45 degrees angle.  

 

13. The result was truly the best yet achieved, but the problem was that the louvres looked unrealistic seen from certain angles.  

 

15. Having given this some thought, I decided to sand the piece. The more I sanded, the better the result. On the next picture you see this very clearly, the right side was sanded and the left side originally milled. There are now twice as many lines visible!  

 

But this too wasn't ideal. After an hour of testing: 

 

This way you can also see how the louvres compare to the model. Don't mind the bottom or top, those would have been detailed better. I'm happy with the 'grooves' but perhaps the top and bottom could be optimized; hence the photo etch attempt. 

[Roy vd M.]

To follow are posts summarizing the build progress so far. For those who can't wait (lol) here are some 'teasers': 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Updates to follow shortly. 

[Roy vd M.]

The wheels, part 1

 

20. When making a comparison box art and parts of both kits, I noted the wheels of the Mac kit. These seemed to be far too small on the model. A quick check of the drawing confirmed this thought. So I grabbed the HP Models and, surely, these were much larger. 

 

To be precise, the HP wheels are 12,89mm. in diameter while the Mac Models wheels are 11,94mm. A full 1 millimeter... which, on 1/72, is a lot. See the above pic. 

 

21. Here a comparison:  

 

22. I loaded three images into the drawing software Gimp, to determine the real diameter of the wheel. Literature (among which Nuts & Bolts) is in structural disagreement. I'll spare you the details.

 

27. Using a 3mm. drill. 

 

30. Turning takes one minute.

 

The positive: a bit of tyre profile is still visible. Nice, because now I still have guidelines for cutting the profile a bit deeper. The hub can be turned a bit further. But the bigger issue wasn't clear yet: thickness of the tyre. It should be 2,56mm. whereas the HP tyre was 3,4mm. That's 32% too thick. And it really looks like the tyre ate too much. 

 

33. Install on lathe and go... resin is easily parted.

 

35. The last half millimeter is manually cut using another micro saw. 

 

38. Here the wheel before and after. The two halves (with the desired thickness) are provisionally glued together using a drop of CA glue. 

 

39. Here the difference in thickness. 

[Roy vd M.]

41. I had made up my mind to consider the anonymous blueprint as leading, now that drawing appeared to be the only one without any apparent errors and it also (more or less) seemed to resemble photos of the refueller. Again, some discrepancies (which I'll again spare you). 

 

42. Obviously I was very curious how the HP kit would stand against drawing and model. I must say that it looked better than Mac's kit. 

 

On the above photo you can see that Mac Distribution's kit contains more detail than HP's, in these areas. 

 

45. I had another look at the wheels and concluded that the HP kit specimens cannot be used, because of a wrong number of grooves in the tyre among other things. In the end I decided to scratchbuild the tyres and rims. Using Nuts & Bolts' dimensions and the drawing made in Gimp I could draw the tyre and wheel. Face drawing was more difficult, I had to use photos as references.

 

Turning...  

 

60. After turning, milling. This is three quarters of the groove-work. 

 

61. Comparison with the too large tyre (yellow, HP Models) and the too small tyre (grey, Mac Distribution).

 

64. I used several sources (photos and drawings) to determine the measures of tyre and rim. Here the result seen from the side. 

 

71. After making some setup and method changes: 

 

101. After several attempts of pressing a brass sheet, I decided to vacuform. Making a new mould...  

 

104. The mould, ready for use..

[Mick4350]

14 minutes ago, Roy vd M. said:

 

 

45. I had another look at the wheels and concluded that the HP kit specimens cannot be used, because

that's incredible

[Roy vd M.]

@Mick4350 What is incredible? Thanks for following by the way, as well as to the others who do

 

 

105. In this video I show the vacuum process I utilise (3 minutes). Better to post this video than to use 60 lines to write things down... 

 

 

107. Mould and shape next to one another.

 

108. After vacuforming first a hole was drilled (diameter 2mm. so that the M2 bolt slides right trough). 

 

112. Eventually this is the result.

 

113. The part is mounted onto the dividing table, in its turn mounted on the milling machine bed. 

 

116. Rear side, before the real wheels: 

 

119. Ready.

 

122. To maintain and protect the shape achieved by the vacuforming, I'll use a metal ring for fastening.

 

130. Ring fits fine.

 

133. Result is the first ready inner rim, four holes on the desired level: 

 

153. The rivets are only 8 (0,6mm. diameter). Much smaller are the rivets along the inner rim and, later, at the end of the axle. These are 20 per wheel and 8 per axle end. My daughter (then 5 years old) was perfectly able to make those. The 0,6mm. rivets are created by using an RP Toolz tool + brass sheet, whereas the 0,25mm. rivets are created by using a tool called 'the nutter' + lead foil. My daughter is welcome to make as many as she likes  

 

 

[Roy vd M.]

156. Using the template, attaching the rivets is a piece of cake: dipping cocktail stick into Future (or the like), next using the still moist cocktail stick to push the rivet to its designated location. That's all. After a minute it's quite steady, after some minutes it's firmly attached. In one minute 5 out of 8 rivets are in situ. 

 

157. Template and glue were removed at the rear end. 

 

162. From vacuformed to ready for paining... 

 

165. After turning the new mould, both halves of the wheel were to be joined.

 

171. When I make a drawing, chances are big that I'll misinterpret it even a few hours later. That's pretty dumb. So I had to restart twice. Also, the drawing was incorrect in one aspect. 

 

172. In the end it looks like this. 

 

178. The thing I dreaded most was to turn the extremely thin and long axle from differential toward wheel (in real life: the propulsion axle housing). Miraculously and probably by mistake, this worked just fine, but I'm sure the axle wasn't far from breakage. 

 

181. Dryfit of the outer rim. 

 

182. Here the inner rim in situ. 

[Roy vd M.]

186. Vacuforming 10 specimens cost an hour and 15 minutes. 

 

187. Drilling of 1 (lathe) + 4 (milling machine) holes cost only an hour. This has to b done for 10 wheels + 1 reserve + 1 test specimen that I'll use to determine the position of the 0,4mm. holes. 

 

188. The 0,4mm. holes are drilled into a scrap piece... it works! 

 

191. Then the bolts are mounted. It's crucial to straighten them. Again I used brown Gator Glue. 

 

195. After drilling, they are cut to shape using the lathe. 

 

197. Status quo. It all looks a bit messy, because there are still burrs caused by the drilling. 

 

300. In Qcad I draw the row of required rivets, then I print it. 

 

302. Cutting a strip.

 

206. Glueing using CA and then attaching the rivets with Future. The transparency of the vacuformed parts can be put to use here. 

 

213. This rim is approved and will be one of four rear wheel rims that will be visible on the model. In the photo you see both rims attached to one another. 

[Roy vd M.]

There seems to be little interest for this build.. too slow or detailed perhaps? 

 

Any suggestions to improve, are welcome.

[Roy vd M.]

220. Differential is turned (one half).

 

223. Drive axle only just fits into the differential half. The arrow shows where the axle is wrenched into it.  

 

224. Photo with a coin, lovely cliche  

 

225. I thought, let's see to what extend the frame needs scratching. Which part (Mac or HP Models) is 'more correct'?

Well the results are saddening. 

 

226. Again, many things are incorrect.

 

It's slowly becoming clear why Mac's model does not look right. It's not only the size of the wheels... but once more I'll spare you the long list of details... 

 

This is how milling has to be done after the turning work will have been finalized:

 

238. Filing a bit, sanding and straightening; then the part is ready in its basic shape. 

 

242. Cleaning up and then... ready for priming. 

 

247. Compared to the kit specimens it's an improvement. Don't mind the paint damage, that will be corrected. 

[Robert Jan Scholte v. Mast]

Hello Roy. Like it a lot what you're doing in such a small scale. From your question above I understand a little frustration came up, nobody reacts? I'm simply in awe when people make things with a lathe, beyond my skills. I learned to look over the tech. aspects and enjoy the skills presented? I think a lot of Brit Modellers don't see the modeling, they see a lathe? Which I fully understand. However, keep up the good work in this "Gentleman Scale", that's what they call this scale, or not?

 

Mijn 2 centen (toen er nog Guldens waren).

 

Vriendelijke groeten uit Thailand,

 

Robert Jan 

Edited November 25, 2019 by Robert Jan Scholte v. Mast
additional text and spelling

[Roy vd M.]

Hi Robert Jan,

 

Much appreciate your opinion! Well it's not as much a frustration, as well as my wondering why there were no reactions (as there's some aspects of this topic that could be new and perhaps interesting to certain modellers). I'm always happy to learn and see what I could do to improve my topic so that it becomes more attractive for forum members. For example I removed the most technical bits (e.g. explanation of how drawing software works). Perhaps it's the lack of swift progress, too. Now we're looking at a couple of wheels... great. But what about the rest of the vehicle? I understand these sentiments and I see what you're saying too. You may be right that once a lathe or milling machine is in view, the modelling implied could go ignored  

 

Happy modelling over there

 

Best regards, Roy

[Graham Boak]

I think that you do deserve more response for this excellent work, but I'm not sure that mine will be helpful. 

1. I am interested in WW2 softskins, so was drawn by the heading, but there are a number of things which push me away.  I am interested in the faults you can find with the overall shape of the model, such as the tyre size and short chassis.  However I would not be too concerned about the hub detail or the correct layout of the chassis cross-pieces.  After all, this is a wartime subject and these areas would rapidly be covered in mud and dust.  I would be very interesting to see how this short chassis has affected the body shape and dimensions.

2. I do not own a lathe nor a milling machine.  Anything done on these has therefore no relevance to me and cannot help my build.  (I don't have this one, I do have the truck which I presume has much the same flaws.) 

3. The sheer number of photographs is daunting, and it is difficult to find the ones that would read across buried in the ones I find irrelevant. 

4.  The sheer time and effort spent on one single truck does not appeal - I have dozens waiting to be made.  Were I to reject a kit because the tyres had the wrong grooves, I think that I would have much much fewer kits both made and waiting.

 

All modellers differ, of course.  I would not recommend that you approached your model in any other way than you have, and the work is admirable.  However I do suspect that many modellers on this board are closer to my "style" (if it justifies that name) than yours, and that this explains the lack of response.

 

[Heather Kay]

I think Graham has a point. What you are doing, Roy, is very much miniature, indeed micro, engineering. As such, I think many of us will shy away as it seems like a lot of hard work for something so relatively small. While I appreciate the skills you have, and the work you’re doing, it’s beyond me to get anywhere near to your level!

 

Please continue showing your work, because there must be others watching progress with interest. I will just sit quietly at the back, bodging away at my own models, wishing I had even a tiny part of your skills to use. 
 

By the way, it’s nice to see your daughter getting involved. Please keep encouraging her! 

[Pete in Lincs]

Roy, Some incredible modelling going on here. Way beyond my skills. I've only just found this, otherwise I'd have put 'likes' in earlier.

I don't normally look in the AFV section. Love the word, Geshuurd. I will try to remember and use it in the future.

I'll be back later to see how you're getting on. And, a five year old pair of eyes as a modelling aid, genius!

Pete

[Roy vd M.]

Your comments are much appreciated and are enlightening.

 

@Graham Boak 

 

At 1: I'm sure you'll appreciate the vast differences in appearance. These kit makes clearly didn't have the shop manual(lette) available. You have the Mac Distribution kit available I guess? 

At 2: perhaps I should include some images / vids of how one could use a multitool (Dremel / Proxxon) to act as 'lathe' in a future post. Your remark does make sense. 

At 3: future updates will have less photographs; what you've seen till now is a few months of work summarized. Still I could have picked less photos, I agree. Still, beside the update in this post there are two updates yet to be posted, before I'll be up-to-date.

Thank you for the effort of replying to this topic!

 

@Heather Kay

 

I agree that this micro building is a bit slow and takes lots of energy. I hope things will speed up a bit once the chassis and wheels are ready

 

Yes my daughter was truly involved in the (much faster) build of the 1/12 Honda RA273 by Tamiya. Now almost 8, she has learned to handle the airbrush among other things. I cherish these times, realising that it's rare that a child of this era (let alone a girl) has an interest in this hobby. Last Sunday at Euro Scale Modelling our model, a 50/50 effort by us both, was awarded 'highly commended' and I'm very proud! 

 

@Pete in Lincs

 

Indeed I hadn't expected to see you over here in this section of the forum. I've not visited this much either... but probably will Thanks for your comments and I'll send Karmijn your approvals haha

 

 

251. I got a lesson in soldering (thanks forum member Silenoz!): 

 

271. Rim 2.0.

 

178. Tyre mould.

 

180. Four ready axles.  

[Roy vd M.]

Meanwhile I had purchased this book, an instruction manual for this type of vehicle. 

 

187.  It's impressive to touch, hold and study something used during WWII. Moreso, for a lorry I'm building in scale. Very useful booklet (100 pages) containing blueprint-like overviews. Overlaying these drawings with the Nuts & Bolts version proofs that the latter is quite incorrect. That's why the ICM kit is incorrect. Only the wheels match, fortunately. 

 

188. Blueprint overlayed on scan/photo of a real surviving LG3000. This is a very comforting sight!

 

192. The slightly twisted drawing can be corrected using Gimp. 

 

195. Finally making a true drawing of the frame. 

[Bullbasket]

You are way above my payscale on this, and all I can do is watch this build in amazement and admiration. And well done for getting your daughter involved. Hopefully a future modeller.

 

John.

[Roy vd M.]

Thanks John, that's a huge compliment to me! 

 

My daughter is currently building a Mitsubishi Zero 1/72 by Airfix. Not moving fast by any means, but it's a start Also she's painting a Frozen's Anna figure. 

 

 

 

PHOTO ETCHING

 

So, let's start the photo etching preparations. First we'll have to have a drawing consisting of vectors rather than pixels. I'll use Qcad (vectors) rather than Gimp (pixels). I'll not translate all background info, howto's et cetera, and will just keep to an overview of the drawing results achieved. 

 

197. Here you see pixels versus vectors, in an enlarged overlay. 

 

 

In Qcad you can insert a bitmap image, so that vectors can be traced (=the pink line seen above).

 

First the 'blueprint' from the Mercedes-Benz manual needs to be resized to scale. 

 

200. Here some preliminary test prints. Number D) was achieved by the (enormous) laser printer employed by a copyshop, using the corresponding overhead sheet. 

 

 

[Silenoz]

nice to see this one here. Incredible work on this beast, which I'll suppose hopefully will end with unfortunately no kit parts used... maybe a bit comprehensive, but that is every man for himself to decide. It is interesting on how you approach the problems and address them. And knowing you don't have a mechanical background sure deserves some merits.

 

Hopefully you haven't forgotten about the soldering by the time you will need it 😉

Edited December 4, 2019 by Silenoz

[Roy vd M.]

@Silenoz Thanks for checking in! I think indeed no kit parts will be used, but I've not completely given up on the tanker body. Well perhaps that's not fair, in fact I have

 

I'm quite certain I'll need your help once more on the soldering, once the photo etch can be joined! By the way that beam you soldered had the wrong height so it had to be redone anyway even if I hadn't prepared that part for photo etching

 

221. The bonnet (left hand side) is ready. 

 

228. Now one bonnet, one chassis beam and two (identical) crossbeams are ready. The folding lines have been included. Don't mention pixels or distortion on the drawings, printing will be way more accurate. 

 

230. Now the crossbeam for the rear suspension subassembly must be drawn. This is more complex. I'll have to establish what can be made on lathe and/or mill and what should be photo etched. As I'll need dimensions anyway I decided to 3D-draw this in Fusion 360. On this forum I won't go in-depth as to drawing techniques etc. Below you see the end results. For the 3D drawing I use the blueprint from the vintage Mercedes-Benz manual. 

 

238. The suspension of the rear wheels can 'swing'. I have drawn the swing-connector on one side of the assembly. Here's the top view. 

 

239. The rotatable piece is pressed between the two leaf spring pockets.

 

240. Hereunder you see the rotatable piece (1) which can freely rotate around its arm (2). 

 

241. The arm is fixed to the chassis beam. 

 

So much for 3D-drawing of this 'swing'. It won't be printed, I'll just use the old-fashioned milling machine, lathe and photo etch-bowls, to achieve as nice a result as possible and because I enjoy it

 

It seems that I'm finally up-to-date now! This is how things stand right now.

[Roy vd M.]

242. The reason to draw a separate bit of the chassis is shown here: there would be an 'impossible overlap'. 

 

 

243. But Silenoz advised me to remove part of the front end. It will be almost invisible anyway and can in fact be filled up with solder. The amended drawing: 

 

 

248. Some other aspects of the drawing needed correcting as well. 

 

I decided to etch the leaf springs. They will now be of perfectly equal width (nicer than I'd be able to cut) and will have rounded ends. 

 

 

256. Now it's time to draw the encircled crossbeam. You can see that the drive axle touches the crossbeam. 

 

 

257. Therefore we'll need to make a recess at each end of the foldable crossbeam part. 

 

258. As the drive axle isn't horizontal, one recess needs to be higher than the other. 

 

259. Seen from below, as well as in overview.

 

 

260. So what does this look like in 2D? Here's the basic drawing. 

 

261. Next, the bottom side of the brass sheet. Everything black will be etched away. 

 

262. Top side of that same piece of brass sheet: 

 

As can be seen, there's a large black area around the part, bottom side as well as top side. The etching acid will cut trough the brass from both sides of the brass. This way, the acid will meet in the middle and thus the brass is fully cut through.  

 

The fold lines are created as well, but in a different way. On the bottom side, four fold lines are etched, whereas on the top side two fold lines are etched. They don't overlap, so bottom and top of the same piece of etched brass will look different. These fold lines are called 'half etches', because the brass of each of 6 lines is etched only halfway through. Then, the etching process is stopped manually. 

 

So those six lines are fold lines, but some are on top while others are below, because the folding direction differs. 

[Roy vd M.]

 

275. To draw the wording 'DIESEL', multiple guidelines are set... 

279. After removal of the unnecessary parts and curving of the letter S (not finalised yet) it looks like this. 

 

280. Test print by the copy shop. 

 

I am confident to start the etching process with this. Not everything worked out, as the small Mercedes star proved to be too thin and will be enlarged. 

[Robert Jan Scholte v. Mast]

Nice to follow your processes. Like the tankers. I ordered a German Ford Fuel Tanker / Sd.Kfz.384 in 1/35 at Dnepro Models (full resin build)

Hope to see that one arrive soon.

Waiting for your next update, I say goodbye.

 

Robert Jan

[Roy vd M.]

@Robert Jan Scholte v. Mast I must say I'm quite amazed by this kit you refer to and purchased. As far as I'm aware the Sd.Kfz. 384 refuelling tanker was exclusively built for the Mercedes-Benz LG3000. Starting to doubt, I took a look in the Nuts & Bolts #32 about the LG-trucks. There I found a few clues that in fact there never was a Ford tanker like the one depicted. First the development of these LG (rough terrain) trucks... there were only four companies that participated at first (Büssing, Henschel, Krupp and Mannesmann-Mulag. The latter's chassis proved unsatisfactory during the testing period and was withdrawn from the project. Later, Mercedes-Benz joined the project. So, no mention at all of Ford having participated. 

 

Second clue is found on page 122: "The (...) aircraft fuel tanker motor vehicle (Kfz. 384) was developed by the Schelmer Eisenwerk Müller & Co. AG. It was manufactured commencing in the mid-1930s, and based exclusively on the Mercedes-Benz LG 3000 chassis. 

 

The box art of your kit may have been photo shopped (bonnet does not seem to be in alignment with the tanker). If anyone has proof of this Ford Sd.Kfz 384 having existed I'd be very interested to learn about it. Robert Jan, I'm sharing this to prevent you from initiating a build being under the impression that the vehicle has in fact existed. As a what if-project it's quite appealing. 

 

Another possibility, it's real but not a Sd.Kfz 384. Looking at the tanker it's not exactly like the MB one. 

 

285. Small logos on radiator and grille have a 'thickness' of approximately 0,5mm. The letter 'D' has grown a bit, after a tip by a fellow modeller. 

 

290. Bit of chassis beam where the rear wheel suspension will be inserted, was removed. 

 

291a. Photo etching the radiator housing won't work. the shape is very complex. This is the top of the part seen from upfront (!):

 

 

At the arrow there's a difficult-to-establish folding line. Qcad nor Fusion 360 could aid so I decided to mill this piece. 

 

 

301. The curvature is the hardest part of the radiator to produce.

 

309. Here you can see the interesting shape: the drawing is the front of the radiator housing, the brass part the rear. Both will be 'loftedly joined'. 

 

310. Satisfied thus far.