32 replies to this topic

[AndrewK]

I was just wondering why the straight eight was used in place of a V8 back in the '50s on cars like the 300SLR Mercedes. I never have found a good place to ask, so I'll put it here.

Andrew

[Chui]

It has the same advantages of an inline 6: inherent balance, pkg space for equal length intake and the classic "bundle of snakes" exhaust manifold. Please refer to the "Why BMW Uses Inline 6 Engines?" thread.

[Kaha]

This question was recently discussed at The Nostalgia Forum http://www.atlasf1.c...&threadid=35021

No conclusion was reached however.

[Catalina Park]

Well there had not been any successful V8 F1 engines yet, but there had been quite a few successful straight 8 Grand Prix engines.

[Wolf]

And there was, as I've mentioned elsewhere, Bugatti T251 from '56 with transversely mounted straight 8 behind the driver, and full width nose... 'Twas raced only once, IIRC.

[marion5drsn]

I have never understood the “real” reason for the adoption of the straight eight over the V-8 as Mercedes had what I consider a better basic car in the old 1939 W165/M165. The only reason I have been able to figure is the desire to show the world that they could build a desmodromic valve operating system. One must remember that Mercedes was at very low ebb at this time and maybe they felt that building an engine of a very different nature would be of a more valuable propaganda/sales than resurrecting an old model that was built under the Nazi’s period of government. There may also have been a conflict in the regulations governing the supercharger as in my estimation the M165 had not even come close to its maximum potential. In some ways one can see a lot of the W165 in the Lancia except for the supercharger. Since this engine was made in Germany it used the DIN firing layout. Same as some Fords 1974/88 # 1-3-7-2-6-5-4-8 this means it had a 90 degree crank which did not need exhaust tuning due to the use of Roots superchargers. Yours M.L. Anderson

[Wolf]

Marion, maybe they put in desmodromic valves because they figured to stay loong in GP racing and trash opposition with development of M196 (revving it up constantly). I think they figured they could rev it at 10k rpm after few seasons... As for firing order of that engine, I think it would be quite easy to deduce- the crank layout is a bit unusual, but I figure one would be safe to assume that consecutive ignitions took place on alternating sides of flywheel (seems logical thing to do ).

[Powersteer]

Post by Catalina Park

"Well there had not been any successful V8 F1 engines yet, but there had been quite a few successful straight 8 Grand Prix engines."


The most successful F1 engine until today happens to be a V8

[david_martin]

Powersteer, reread Catalina Parks post with emphasis on the tense, and then imagine it is 1955

[Catalina Park]

Originally posted by Powersteer
Post by Catalina Park

"Well there had not been any successful V8 F1 engines yet, but there had been quite a few successful straight 8 Grand Prix engines."


The most successful F1 engine until today happens to be a V8

The question was why the straight eight was used in place of a V8 back in the '50s on cars like the 300SLR Mercedes.

The Most successful Grand Prix engine since 1956 is a V8.
The Most successful Grand Prix engine before 1956 is a straight 8.

[marion5drsn]

Wolf; in the other listing at the Nostalgia section you stated that the throws were in a certain position but failed to state the starting position as being on the top vertical or bottom vertical or where.
I am wondering if you have the firing order The only one used in the U.S. by G.M. and Chrysler and others is # 1-6-2-5-8-3-7-4, starting at the front of the car turning left drivers view. This is also listed as one of three firing orders in Bosch’s, “AUTOMOTIVE HANDBOOK”.
The manufacturing of the exhaust is not the simple thing some believe as # 1 & # 8 are 360 degrees from one another, a very long throw for a manifold.
# 2 & # 7
# 3 & # 6
# 5 & # 4
# 1 & # 8 are going to have some very long pipes and # 4 & # 5 are going to be very short. This same thing applies to the intake if one is going to use the two-plane carburetion theory. This may explain the reason why Buick’s dual carburetion did not work very well. If I remember correctly it drank gas like it owned Spindletop. But that’s a long time ago! M.L. Anderson

[Wolf]

Marion5drsn- well, I made the crude sketch of the crank in AutoCAD. BTW, my assumption it would fire at alternate sides of the flywheel is, unfortunately, false... And, I don't have the firing order of M196, just ordinary straight 8 (1-4-3-2-8-5-6-7), which is not applicable. :



Possible orders are (assuming CCW rotation) :

1-5-2-6-8-4-7-3
1-5-7-6-8-4-2-3
1-5-2-3-8-4-7-6
1-5-7-3-8-4-2-6
1-4-2-6-8-5-7-3
1-4-7-6-8-5-2-3
1-4-2-3-8-5-7-6
1-4-7-3-8-5-2-6

The first one seems OK to me...

[marion5drsn]

Wolf; You have just discovered the same thing I discovered about 5 years ago. When I was drawing basic cranks and stuff for the old 720 cubic inch Hispano-Suiza WW-1 aircraft engine I found out that a lot of others and my theories about 90-degree and 180-degree V-8 cranks were not possible or not practical. A basic little sketch destroys a lot of romantic theories.
While thinking about what you said I think I figured out how many firing orders are possible for a straight eight engine and it seems to be 0+7+6+5+4+3+2+0=27 possible ways to fire the cylinders. My guess is that it fired as two four cylinders on each side of the output gear. But then again I don’t know how that would affect the crank as to vibration and etc. If this were possible this would allow the use of a very effective exhaust tuning. However the U.S. firing order is just ridiculous as to exhaust tuning.
I don’t know how many old Buick 320 cubic inch engines you have seen but my Dads 1937 Century 4 door 320 seemed to fire # 1 before the flywheel knew what was going on! (Slight exaggeration.)
What kind of drawing program do you use?
Yours M.L. Anderson

[Wolf]

Marion, since I don't have scanner, I had to use AutoCAD r13... BTW, I think I have listed all possible firing orders for that layout- the rest is obtained by switching cyls 1 and 8, but those layouts are symmetrical (e.g. switching 1 and 8 in first sequence would result in last given sequence shifted by a revolution). That would give 16 possible firing sequences not 28 (You forgot the one, and have two zeros in that eqn). I think You got it wrong: assuming first cylinder fired You can choose only second, third and fourth cylinder that can be fired (since there are pairs of throws at the same position)- the rest is then fixed... So the no. of permutations is 1x2x2x2x1x1x1x1=8.

Speaking of firing orders and ignition in real ( ) cars, I think I have rare solution in mine... It has no distributor, fires in 'all' cylinders simultaneously but firing order is 1-2, a revolution apart...

[marion5drsn]

Wolf; I have just made a circle to duplicate the crankshaft as in the book of Taylor’s and rotated it as the layout for the Buick firing order and it works perfectly. This makes the throws # 1 & 8 at 0 deg., # 4 & 5 at 90 deg., # 2 & 3 at 180 deg. and # 5 & 6 at 270 deg. This is from the front of the engine rotating to the right (CW), hand crankers view. After doing this, my findings are much like yours. If this is true there is no way the engine can fire as a double four cylinder way. That is a four cyl. in back and four cyl. in front.
According to DIN standard #73 021 an engine rotates from the front in a Clockwise direction. This is from, “Bosch Automotive Handbook”. However this doesn’t make any difference in the firing order, as it still won’t work. The M 196 must have had some weird exhaust manifolding. Certainly not as good as a V-10 or a 180-degree crank V-8.
I am very glad you became involved as I made the mistake of substituting the possible crank layouts with the firing orders. There should be only 6 crankshaft layouts. But in actual practice forging techniques hold this down to two.
I am now wondering if someone knows of a picture of the exhaust manifold, as I am very curious about what it looks like. My guess is that all cylinders had very equally long pipes of fairly small diameters. Eight into one. Yours, M.L. Anderson

[Wolf]

You won't belive this, Marion... Seeing the exhausts were to the starboard of the engine, the side engine was tilted to, I thought this kind of photo would be quite hard to get (I have only engine crossection w/o exhaust and intake manifolds, and detail drawing of desmo). A quick search with Google proved me wrong- there are photos of the die-cast model showing it here...

Ahh, that was great engine- desmodromic valves, GDI, cylinder barrels with integrated heads...

[Ray Bell]

I have a query with the Hudson straight 8.. but I have to get on my way now and will address it when I get back.

In the meantime, Marion, do you know the firing order for the Hudson 8 and are you aware of the cylinder starvation it caused in 1 & 8?

[marion5drsn]

Ray; the Hudson had the same firing order as the Buick. I use Buick as an example as they probably made more straight eights as all the rest of the carmakers combined between 1932 to 1952. This Buick engine was made in two sizes 232 to 344 cubic inches altho most engines were of either 248 to 320 cubic inches after 1937. Besides it was the first car I ever drove. And yes I can understand the lean/rich problem in any straight eight engine.
Wolf; from the pictures in the drawing I seem to have made a correct guess on the manifold layout i.e. being an eight into one.
Since this crankshaft was made up out of pieces of Hirth jointed pieces I wonder if they made experiments of different crank patterns. Then testing them for Shake and Vibrations of various types. This should have been “easy” if the joints were divisible by four and eight. If I get my things together I will send you a QuickCAD sketch of my work to see if it is similar to what you did. It is a sketch of what the crank would have to look like if it were two four-cylinder engines back to back. However it will be several days before I get it finished. The firing order is much different from what either you or Bosch have come up with. This is due to the rotating of the crank throws to allow the use of the two four cylinder 180 degrees firing principle. What we need now is for Karlcars to find out the firing order of the M196 engine. There are two firing order that might work, but one of them will have to do for now. Yours M.L. Anderson

[Catalina Park]

In L.J.K. Setright's book The Grand Prix Car 1954/1966 he says that:
".......the power take-off point was not at one end of the crankshaft but at its middle, so that the shaft's susceptibility to torsional flutter was halved. In fact, some problems were still experienced in the Mercedes-Benz, leading to some experimental variations in the firing order, before eventually settling upon the sequence 1,4,7,6,8,5,2,3.

[Wolf]

Thanks, Catalina Park, I guess that settles it.

[marion5drsn]

To make the straight eight fire as two four cylinders back to back would require a firing order # 1-8-2-7-4-5-3-6, which obviously the M196 did not. Either Mercedes did not appreciate the four-cylinder firing sequence power increase due to exhaust tuning or it might have shaken the engine to pieces. But we will never know! But it does explain the long exhaust pipes. Yours M.L. Anderson

[Powersteer]

Sorry Catalina Park because I saw the 'F1' that made me think otherwise.

[marion5drsn]

After much time I have finally got around to finishing the firing order on M196 engine and the only thing we missed was stating that the firing order that L.J.K. Setright gave it had to be viewed from the drivers seat or the engine turned the wrong way. If anyone can see an engineer at M.B doing the thing wrong. But there is always a possibility. After all I missed the fact that if you tried to hand crank a Corvair engine from the back you would have a tough time of it. If it was a front engine car it turns the wrong way for a right handed person.

To View the crank in the M196 see Karl Ludvigsens book ,"CLASSIC RACING ENGINES" on page #96. Be very carefull as the picture is small. I will make a sketch of it but Wolf's is very close.

http://home.earthlin...NKSHAFTM196.JPG

M.L. Anderson

[Ben]

I remember once seeing an interview with Rudolf Ulenhaut the chief engineer (and bloody good test driver) at Mercedes. He said that the decision to run a straight 8 in the W196 was political due to the road cars having them.

Remember the 1.5 litre V8 that Mercedes dominated the Tripoli GP with (can't remember the year - sorry), Ulenhaut essentially wanted to do a 3-litre version but wasn't allowed by the bosses.

Ben

[scheivlak]

Originally posted by Ben
Remember the 1.5 litre V8 that Mercedes dominated the Tripoli GP with (can't remember the year - sorry), Ulenhaut essentially wanted to do a 3-litre version but wasn't allowed by the bosses.

Ben

That was 1939.
"Wasn't allowed by the bosses"- Well, the Italian organisers announced late in 1938 that it would be a voiturette race, 1500cc max.
Mercedes took them by surprise and built two entirely new 1500cc V8 compressor cars just for this one race!

[Ray Bell]

Originally posted by Ben
I remember once seeing an interview with Rudolf Ulenhaut the chief engineer (and bloody good test driver) at Mercedes. He said that the decision to run a straight 8 in the W196 was political due to the road cars having them.

Strange... all M-B production cars had either fours or sixes... at least from 1946 to 1963 or so (the 600 'Grosser'), with a heavy leaning in production numbers to fours, I should think through the fifties.

As the last straight eights had already disappeared from production lines by then (or at least I can't think of any that survived beyond 1953 - maybe Packard?), it's unlikely that they were thinking this way at all, as I see it.

[Ben]

The 300 SL had a straight-8

Ben

[Engineguy]

Originally posted by Ray Bell


Strange... all M-B production cars had either fours or sixes... at least from 1946 to 1963 or so (the 600 'Grosser'), with a heavy leaning in production numbers to fours, I should think through the fifties.

As the last straight eights had already disappeared from production lines by then (or at least I can't think of any that survived beyond 1953 - maybe Packard?), it's unlikely that they were thinking this way at all, as I see it.

He probably meant they went with a straight8 because their road cars had inline engines, not necessarilay they same number of cylinders.

[Henri Greuter]

The 300 SL had a straight-8

Ben

==

No: the 300SL had a straight six!
The 300SLR had a straight eight.

Henri

[Amnios]

Originally posted by Catalina Park
Well there had not been any successful V8 F1 engines yet, but there had been quite a few successful straight 8 Grand Prix engines.

IIRC The Ford engine in the 94 Benetton was a V8.

Amnios.

[scheivlak]

Originally posted by Amnios


IIRC The Ford engine in the 94 Benetton was a V8.

Amnios.

Well Amnios, it's verrrry easy to list a page full of succesful V8s (I only have to remember the 1.5 litre F1 that attracted me way back in the 60s, and remember some Cosworth block later on....) - but the original question was: why did Mercedes choose the straight 8 in mid-fifties ...

[marion5drsn]

DESMODROMIC CAM LOBE 288 DEG, OPENING 1954 MERCEDES-BENZ

http://home.earthlin...OPENINGLOBE.JPG

CAM LOBE CLOSING DESMODRMIC 288 DEG.

http://home.earthlin...G288DEGREES.JPG

M.L. Anderson

[jgm]

Michael Riedner wrote a considerable book about the W196. On the question of cylinder layout he says- 'With a 12-cylinder, valve area could be maximised relative to cylinder capacity, enabling the highest possible revs to be attained. On the other hand the car's centre of gravity would be pushed relatively high, unless a 'flat' angle was chosen for the V (about 120 degrees). A very short front section could have been designed for the car had the V layout been chosen, but this was not deemed a relevant factor. Besides this, the 12-cylinder would have been a highly complex and very heavy piece of equipment.

Finally the engineers decided on an 8-cylinder as a compromise. A V-6, V-8 and straight-eight engines were investigated and calculations were carried out (especially on vibration characteristics), the choice finally falling on the straight-eight.'

On firing order he says - 'Theoretically there were three possible firing orders, the choice being the one guaranteeing constant high output in prolonged running at highest possible revs: 1,4,7,6,8,5,2,3,.'