Bachmann Deltic Review.

=>

=>> Would for instance the main exhaust vent on the 44 have been omitted if =>the =>> manufacture had closely looked at one of the preserved examples? =>>

=>> John. =>>

=>

=>Maybe, maybe not. A model by the Australian importer Austrains had no =>exhaust outlet. The reported reason was that detail photos where obtained of =>the prototype maintained by a museum. =>

=>The chinese managed to perfectly reproduce the cover that is kept over the =>exhaust outlet when the loco is not in use. =>

=>TW

Thanks for this -- it illustrates perfectly what I meant when I said that the moldmakers had to interpret the data they received. And, it seems, so do the mnaufcatueres when they vet the moldmakers' work.

HTH

Wolf Kirchmeir ................................. If you didn't want to go to Chicago, why did you get on this train? (Garrison Keillor)

I have tried doing white-metal castings in silicone-rubber moulds (just a beginner, you understand) of bogie frames. On dimensions up to about 70mm, I cannot find any measureable difference between pattern and casting (measuring to the nearest 0.01mm). Small details do come out differently from the patterns (due to reasons other than shrinkage), but due allowance can be made. I don't know the details of injection moulding relating to if/how the finished product differs from the mould dimensions.

Ian

I've checked the dimensions on Bachmann's model corrsponding with Francis' data taken from 55015. I'll first list the 4mm scale equivalents of the above dimensions:

To bufferbeam: 2.67mm Horiz To rear of cab door: 53.67mm Horiz To point of windsceen vee: 21.67mm Horiz, 25.67mm Vert To base of nose 'dome': 1.25mm Horiz, 18.67mm Vert To line extended from nose top slope [1]: 0mm Horiz, 24.67mm Vert Point of windscreen vee to front rail of bonnet hatch: 14.33mm on slope

Now for the dimensions that I took from the model, keeping in mind that the datum point (top centre of cut-away around buffer-beam) may be a little higher or lower than it should be (it's 2.4mm above the buffer axis):

To bufferbeam: 1.3mm Horiz To rear of cab door: 53.5mm Horiz To point of windsceen vee: 22.0mm Horiz, 26.25mm Vert To base of nose 'dome': 1.15mm Horiz, approx. 19.4mm Vert To line extended from nose top slope [1]: 0mm Horiz, approx. 25.1mm Vert Point of windscreen vee to front rail of bonnet hatch: 14.4mm on slope

The large difference with the first of those dimensions is due to Bachmann's buffer-beam being curved (convex) rather than straight as it should be. The curvature is for bogie clearance.

Presumably Bachmann took measurements from either 55019 or D9009 which doubtless differ slightly from 55015.

Ian

"Clearly shows"? The curve is barely discernable at the bottom of the grille and not visible at all at the top of the grille. The area behind the yellow dividing line is essentially flat.

In the same phot referred-to above the yellow section of the nose is much more bulbous than it is on the model. In the opinion of Steve Jones that entirely wrecks the character of the nose ends which are the distinguishing feature of the 44. Steve Jones is a modern image fan who has been waitng a long time for the

1. I am not.

Bachmann's Deltic has a similar problem. Due to the inclusion of of a sloping front plate which does not exist on the prototype except as an illusion of perspective, the blue and yellow version looks as if a large wedge has been removed from the yellow section of the nose.

Immediately above the close up of D1 in British Rail Main Line Diesel Locomotives is a side-on view of D8 which matches the picture of D1 on the front cover of Bachmann's magazine almost exactly. It is clear from that even on the green version there is too much curvature on the cabside grilles. As I've said, you can easily ignore it on the green version. The six inch diameter flag poles either side of the doors are another matter and give a toy-like appearance to an otherwise serious model.

kim

As I said, the same volume includes close-ups of the nose ends of D1 and

40008 as well as a side-on photo of D8 which exactly matches a publicity shot of D1 issued by Bachmann. It is clear from many photos that the nose end of the model 44 is not sufficiently bulbous and that is particularly noticable on the blue & yellow version. Bachmann's Deltic has a similar problem.

As for credibility, I am not in business selling Bachmann.

(kim)

Whereas I had asked myself what's a "couple percent" of the body length, which is where the embarrassment would manifest itself when trying to get the floorpan to fit in the body.

Cheers, Francis K.

I happened to have a wee look in my copy of "BR Main Line Diesel Locomotives" by Messrs Marsden and Fenn, and compared the nose front profiles of DP2 and Class 55. There appears to be a slight difference between the two, the 55 looks to be "flatter" in the vertical end than DP2, the drawing of DP2 seems to look slightly better in this respect, IMO than the 55. Considering that they shared the same body, (did DP2 not use the 23rd production body?) I was wondering if anybody else has noticed this?

Cheers,

Clearly you see things differently than I do, so others must look at the photos and see what they see rather than taking either of our words for it.

snip

We sure do see different things, in my copy its D7 .

They are at least flat! 6" is exaggerating but certainly twice as wide as they should be and much to thick, just like the 24. Obvious candidates for replacement but not an easy task. Anyone with hints and tips for replacement cab door handrails please provide them, at least that can be fixed, as can the missing exhaust. And on that topic anyone know what size the hole should be?

Keith

Make friends in the hobby. Visit Garratt photos for the big steam lovers.

=>Whereas I had asked myself what's a "couple percent" of the body =>length, which is where the embarrassment would manifest itself when =>trying to get the floorpan to fit in the body.

OK, OK, I was being flippant.

Seriously, most comparisons of prototype and model show a difference of about

1 to 2mm in body length, which is actually less than 1% to slightly over. As for the floorpan fitting the body - I don't understand. The frame is just a hunk of metal for mounting the motor, trucks, etc, and some detail that's visible from the side, so as long as it doesn't fall out, what's the problem?

Wolf Kirchmeir ................................. If you didn't want to go to Chicago, why did you get on this train? (Garrison Keillor)

"Keith Norgrove" wrote

Different editions ?

John.

...

I used to do a lot of white metal casting (mainly 15mm and 25mm figures), straight forward casting as you describe is comparatively easy, however detail is poor and it's quite a slow process. To get better detail you need a casting machine - this is basically a centrifuge that spins very fast, the mould is locked in place on the centrifuge, the metal is poured whilst the machine spins and because it's spinning you get a very high degree of detail - in effect the metal is really forced into the mould. It's also a lot faster in that you can make several castings at a time (obviously dependant on scale/size) because the mould is generally larger than you'd use for casting by hand and also because the machine forces the metal into the mould you get far fewer rejects.

The one drawback is that a certain degree of compression takes place. That is as the metal is introduced and forced into the mould the mould actually stretches/compresses a little - there's no way round this (or at least there wasn't when I was doing it). Whilst I'm sure that there's some form of mathamatical formula that describes this process you tend to guess then as experiance grows get a feel for it. As a result, taking figures as an example you build your masters so that they are all about 6'4" with a build like twiggy so that once cast (depending on just how slim you made them) they'll come out between 5'10" and 6' with a proportionate build.

Now I've never moulded plastics, however from what I've read the mould are hard metal, the plastic is soft and far less dense and whilst injection moulding does rely on pressure being introduced to for the details any mould compression has got to be miniscule compared to metal casting using silicon moulds as the nature of the materials used is reversed. IOW (and I stand to be proved wrong) the tolerances achievable in injection moulding plastics should be much higher than soft metal casting as the materials are that much more predictable.

Just my 2ps worth.

We're definitely getting sidetracked here!

You were suggesting a 'couple percent' shrinkage was a reasonable expectation from injection moulding technology.

I was saying that the technology is demonstrably better than that, because I have many examples in my collection where bodies and floorpans fit snugly together. The implication being that noticeable deviations between the model and prototype proportions are down to lack of rigour at the model design stage, not due to shrinkage. (Unless you're talking about the radiator slats on a Hornby-Dublo Class 08!)

Cheers, Francis K.

Thanks for that casting info' Chris. I had considered investing a few hundred pounds in a centrifugal casting machine (plus an electric melting pot). It might be better to stick to hand-casting for my purposes to avoid the distortion that you described. I have achieved the level of fine detail I required after experimenting with extra vents in the moulds and thickening parts of the patterns. Thin flanges did not cast at first because the metal cools and solidifies too quickly, but after drilling tiny mould-vents and making the flanges thicker on the patterns, they cast okay.

One other option would be to try lost-wax casting in brass. The "Reid Technique" is supposed to be a simplified version for home casting small components. However, I have not read any independent reviews of it so far.

Regards,

Ian

Can't help you there, never tried brass ... just lots and lots of little soldiers, tanks, ships and aeroplanes :-)

=>You were suggesting a 'couple percent' shrinkage was a =>reasonable expectation from injection moulding technology. =>

=>I was saying that the technology is demonstrably better =>than that, because I have many examples in my collection =>where bodies and floorpans fit snugly together.

That's 'cuz they both shrink almost exactly the same amount. It's when parts are of martedly different sizes that shrinkage becomes rather difficult to adjust so that the parts will fit. And "technical difficulties" is a synomym for "increased costs." BTW, shrinkage also causes warpage - it's possible to control the warp so that the part is the right shape when cooled, but that's not so easy, either.

Just how much a mold is reworked depends on the required tolerance of the ejected part and how much the customer is willing to spend. I had a very interesting chat with an expert in moldmaking a few years ago - he was under contract to a Mexican plastics molding company, teaching their diesinkers how to refine cooling passages (for example) in the molds to reduce cooling time, so that a few dozen more parts per hour could be made - a couple percent difference in output was the difference between profit and loss. That tells you something about costs, and what costs are worth accepting. He told me that the mold for a simple plastic record-player or CD player cover could cost upwards of $100,000 - in Mexico! - and would cost more if the customer wanted a more precise fit. Sure, the technology is capable of all sorts of precision - at a price. No mold works right the first time, he said -- that's why his many years of experience in refining the mold was in such high demand (he was earning about $10K a month plus expenses.) It was quite clear from his answers to my questions that moldmaking is as much black art as science.

Model railway makers just don't have the market to be able to afford repeated refinement of the molds, even assuming they have sufficiently reliable information in the first place. After listening to and learning from the above mentioned gentleman, I had a great deal more respect for the price-quality ratio that we do get in our rather finciky hobby. It's no accident BTW that US models are generally morre accurate - there were many more of them made for many more different railroads, which means there is a much greater potentials ale of any one diesl model, which in turn means that the maker can afford a more refined mold.

Another factor is the recent advances in using CAD/CAM for mold making, which have reduced the cost of making molds, but have also increased the demand for accuracy and precision, simply because the CAD/CAM makes higher accuracy and precision possible. But higher accuracy and precision costs more, so the savings are largely offset by those costs. It's the existence of high quality models in larger markets that has roused the desire for equally accurate models in the small UK market. That demand will be met over time; but expect prices to rise, too.

Wolf Kirchmeir ................................. If you didn't want to go to Chicago, why did you get on this train? (Garrison Keillor)

For the benefit of anybody not bored with the great Deltic-nose-slope-debate :) I can shed some light on the apparent discrepancy between the measurements that appear to tally with those taken from Tulyar, and the visual appearance of the model.

On the Bachmann model, the central part of the nose top (largely covered by the doors) is slightly concave (viewed in profile) towards the front end of nose. That is, a shade too steep for about three-quarters of the way, and then more level near the front. Towards the side edges of the nose top, the slope is constant and hence slightly too steep along the whole length of the nose. This is the area that tends to catch highlights and shows up the slope error from a variety of viewpoints, while the measured drop of the nose along the centre is spot on.

The above errors imply that the front top corners of the nose are slightly too low and the curvature of the nose top (viewed from the front) is too tight. Sure enough, this can be seen from the front view thanks to the edge of the yellow paint highlighting the effect - otherwise it would be very hard to spot. From the front view, the nose top curve and the corner curve do not meet at their respective tangents, giving a slightly humped appearance. Maybe a drawing error?

From Francis' measurements on Tulyar, the nose slope should be 4.0 degrees to the horizontal. This is accurately reproduced from the overall nose top measurements on the model, but the steeper part of the nose is around 4.2 to 4.4 degrees. Very small in absolute terms, but perceptible as a percentage of a small angle.

Ian (still awaiting a job offer from Bachmann :)

"Ian" wrote

Don't push your luck, you might get one! :-/

John.

Not sure why that this is an acceptable excuse. The "gentleman" referred to has quite a reputation for acidic postings and however one tries to dress it up, it is unacceptable. When will people like him learn. The term he uses on so many others - "plonker" is particularly relevant in his case

The man is a prick. What more can one say.