Computers on Safehold

USMA74 wrote:The early US M-60 series of tanks used a mechanical computer with a range finder. E-Bay has FM 17-12-2, Tank Gunnery, M60, M60A1, M60A1(AOS), and M48A5 Tanks (1977 edition) for sale for US $12.00 if anyone is interested in learning more about how the US Army found a solution without the use of electricity. Admit that the readouts were lighted so you could read them. I could hit a stationary tank size target reliably from a short halt at 4,400 meters if the 105mm M-68 cannon was zeroed properly. Freely admit that having both the target and shooting platforms moving on their X, Y, and Z axis complicates the gunnery problem significantly over what I accomplished.

I've ordered the FM. Questions. How the hell do you SEE a tank at 4,400 meters!? Do tank rounds come with variable powder charges? My guess is not, cuz you wouldn't want little baggies of powder sliding around the floor/deck/whaddayacallit.
Sounds like pretty good shooting!

EdThomas wrote:

USMA74 wrote:The early US M-60 series of tanks used a mechanical computer with a range finder. E-Bay has FM 17-12-2, Tank Gunnery, M60, M60A1, M60A1(AOS), and M48A5 Tanks (1977 edition) for sale for US $12.00 if anyone is interested in learning more about how the US Army found a solution without the use of electricity. Admit that the readouts were lighted so you could read them. I could hit a stationary tank size target reliably from a short halt at 4,400 meters if the 105mm M-68 cannon was zeroed properly. Freely admit that having both the target and shooting platforms moving on their X, Y, and Z axis complicates the gunnery problem significantly over what I accomplished.

I've ordered the FM. Questions. How the hell do you SEE a tank at 4,400 meters!? Do tank rounds come with variable powder charges? My guess is not, cuz you wouldn't want little baggies of powder sliding around the floor/deck/whaddayacallit.
Sounds like pretty good shooting!

Bionic eyes??!!!

Don

-

That is why tank have rangefinders (mechanically linked into the fire control computer) that are fairly powerful. Also tank ranges tend to be fairly open areas of land. My memory of the Tank Table 8 range at Fort Hood, Tx in the 1970s was that the starting point was somewhat elevated over the course and the long range engagement was one of the first shots a tank crew made. It was a line-of-sight engagement. I welcome anyone in these forums to confirm or refute my almost 45 year old memory.
Hitting a stationary 10 foot by 30 foot target at 4400 meters in daylight from a short halt with a properly zeroed cannon wasn't difficult. The M68 105 cannon designed by our British friends is an accurate gun.

I don't see the problem, mechanical systems can be very strong indeed if properly designed. Remember that mechanical time fuses were used in shells from before the first world war so a mechanical system can survive being fired from a cannon, then the impact of hitting a target before exploding. As for miniaturization it is possible to make a mechanical computer on a silicon chip (it just isn't as small as current silicon equivalent) and you can even include digital displays by using retro reflective mirror elements (these were being used for projection displays in the 90's). The issue with all these is that the development requires sub optical wave length imaging (electron /atomic force microscopes) to get to the really small sizes while the manufacturing is essentially a mechanical process. Making anything much smaller than a woman's wrist watch is a serious technical challenge without electrical something better than an optical microscope. That said, building a perfectly capable ICBM or satellite launcher with mechanical controls (like an DC-3 or V-1/V-2 autopilot) is perfectly possible, just difficult to correct trajectory after launch without a crew.
As for programmablity, punch cards and punched tape work in mechanical systems like the prewar IBM census machines and tabulators (and these date back to 1798 when they were used to control weaving looms - yes Numerically Controlled machines date back to when Napoleon was rattling around (and he may have even seen one given that they are a French invention).

Silverwall wrote:I really don't see a device like this being viable once you start taking hits from enemy shells.

Given how often pneumatic and electrical circuits were blown out in WW1 and WW2 I can't see somthing this fragile surviving concussion or shell hits on the hull. Also I am not sure how quick programable they would be.



All the historical analogue computers from WW1 onwards relied on electricity to power them however and you don't get the same level of control with pneumatics (or the same level of miniaturisation).

AirTech wrote:I don't see the problem, mechanical systems can be very strong indeed if properly designed. Remember that mechanical time fuses were used in shells from before the first world war so a mechanical system can survive being fired from a cannon, then the impact of hitting a target before exploding. As for miniaturization it is possible to make a mechanical computer on a silicon chip (it just isn't as small as current silicon equivalent) and you can even include digital displays by using retro reflective mirror elements (these were being used for projection displays in the 90's). The issue with all these is that the development requires sub optical wave length imaging (electron /atomic force microscopes) to get to the really small sizes while the manufacturing is essentially a mechanical process. Making anything much smaller than a woman's wrist watch is a serious technical challenge without electrical something better than an optical microscope. That said, building a perfectly capable ICBM or satellite launcher with mechanical controls (like an DC-3 or V-1/V-2 autopilot) is perfectly possible, just difficult to correct trajectory after launch without a crew.
As for programmablity, punch cards and punched tape work in mechanical systems like the prewar IBM census machines and tabulators (and these date back to 1798 when they were used to control weaving looms - yes Numerically Controlled machines date back to when Napoleon was rattling around (and he may have even seen one given that they are a French invention).

Silverwall wrote:I really don't see a device like this being viable once you start taking hits from enemy shells.

Given how often pneumatic and electrical circuits were blown out in WW1 and WW2 I can't see somthing this fragile surviving concussion or shell hits on the hull. Also I am not sure how quick programable they would be.



All the historical analogue computers from WW1 onwards relied on electricity to power them however and you don't get the same level of control with pneumatics (or the same level of miniaturisation).

Guys, I can come up with pneumatic or wind-up work arounds for most of the problems involved in taking ranges and computing rate of change for them. The biggest problem is how, without electricity, you get centralized fire control to work properly. The whole notion of fire control revolves around the centralized control of the moment of firing, so that every gun is controlled by someone other than a sight setter peering through his own telescope and firing the gun independently. I am not saying you won't see dreadnoughts (not saying you will either, of course) but simply pointing out that the devil is in the details. We tend to think in terms of how they get the necessary data; the real problem is how they can use the data once they have it.

EdThomas wrote:

USMA74 wrote:The early US M-60 series of tanks used a mechanical computer with a range finder. E-Bay has FM 17-12-2, Tank Gunnery, M60, M60A1, M60A1(AOS), and M48A5 Tanks (1977 edition) for sale for US $12.00 if anyone is interested in learning more about how the US Army found a solution without the use of electricity. Admit that the readouts were lighted so you could read them. I could hit a stationary tank size target reliably from a short halt at 4,400 meters if the 105mm M-68 cannon was zeroed properly. Freely admit that having both the target and shooting platforms moving on their X, Y, and Z axis complicates the gunnery problem significantly over what I accomplished.

I've ordered the FM. Questions. How the hell do you SEE a tank at 4,400 meters!? Do tank rounds come with variable powder charges? My guess is not, cuz you wouldn't want little baggies of powder sliding around the floor/deck/whaddayacallit.
Sounds like pretty good shooting!

The T46E1 stereoscopic range finder used in the early M48 models was designed to provide the commander with a system that could accurately range targets out to around 4000 meters. The same sort of stereoscopic target ranging had been used in naval gun turrets for years (being particularly well developed and used by the Germans in WWII) and this type of system was gradually replacing coincidence range finders in many world navies, including the United States and England. The problem with the T46E1 stereoscopic range finder in the M48/M48A1 tank was that aligning all those marks was not easy on your eyes or your brain. You had to allow your eyes to relax and focus differently than normal and your brain had to accept some pretty tough visual assignments. Also, if one of your eyes were strongly dominant, it would be next to impossible to align all the reticle marks anyway. And thirdly, although the system provided very accurate ranges when used properly (more accurate than a coincidence range finder, for instance), the whole thing had a tendency to make you dizzy and give you headaches after prolonged use. In training demonstrations it was documented that it took 2000 practice readings to produce a consistently accurate range reading from the average tank commander. The T46E1 range finder had a baselength of 6ft 7in across the width of the turret. It provided the commander with a 10x sight, which was an improvement over the 7.5x magnification of the M12 unit used in the M47 tank.

So, the range finder was changed from a stereoscopic type to a coincidence type in the M48A2. The coincidence range finder used in the A2 (and later vehicles) was based on a simpler and also well-proven principle. If two optical telescopes are situated some distance apart from each other, and both are aimed directly at a distant target, the angle between the lines of sight of the two telescopes is relative to the distance to the target. That is, if the target is a long way off, the telescopes will be almost parallel when they are aligned to see the target as only one shape. And if the target is sitting close in front of you, the telescopes will be angled in toward each other in order to see the target as one clear image, and the line of sight 'angle' between them will be much greater. The range to target can therefore be determined by that angle between the two lines of sight. So, a mechanical indicator can therefore be made that will automatically provide the range from a scale when the two images are brought into focus by, say, a knob on the range finder. Since the farther apart the two optical telescope lenses are mounted, the more accurate the range data is, the two armor protected ranging lenses on the M48 were mounted on either side of the turret, up near the roof, around six feet apart from each other.

In use, the tank commander first determined the range with his range finder and the gunner then manually entered it into the T25 Range Drive. This process provided the correct superelevation for the gun, and then the gunner only had to aligned his M20 periscope's sights onto the target and pull the firing trigger. In case the M20 periscopic sight was inoperative, the gunner could use the backup telescope, elevating the weapon until the appropriate range line in the reticle overlay the target. This was all pretty simple stuff and was generally common procedure at that time (early 1950's). Don't get confused... this was not an electronic 'computer' in the modern sense of the word; there were no memory chips and nothing in the way of electronic components inside the box. Instead, the ballistic computer was merely a mechanical piece of hardware that provided useful information and allowed quick adjustment of the elevation of the gun. You simply entered the commander's range figure and the ammo type, and then turned the crank until the pointers lined up.