Jump to content
Companion

M111 / DM23 - Why was this round such a big deal to Soviets?

Recommended Posts

Well, according to an article published by Armor Magazine, Kubinka tests of M111 "Hetz" convinced Soviets to design a new glacis for T-72A, which was also implemented on T-72M1

But so far in my short SB Pro PE 2.5 experience, the round's performance is somewhat underwhelming - T-72Ms are shrugging off this round just fine (both turret and glacis) even at close ranges unless hit at weak spots.

Any explanation of this contradiction is very appreciated :confused:

Maybe the information that the article has used was somewhat inaccurate?

(just googled it to refresh memory - it's "The Secret Testing of Israeli M111 'Hetz' Ammunition: A Model of Failed Commander's Responsibility,")

Share this post


Link to post
Share on other sites

I googled that article, and it is indeed a good read:

http://www.israelmilitary.net/showthread.php?t=679

As far as the "Hetz" round vs. T-72M in steelbeasts goes, I'll leave that to the brain's here. I will say that with the "M" being a newly added vehicle, it may need tweaking. This is a common thing, as it is near imposable to test every ammo type, at every range, on every tank, in a beta cycle.

Share this post


Link to post
Share on other sites

In general, enemy capabilities were always exaggerated during the cold war. Defense contractors on both sides loved that.

Share this post


Link to post
Share on other sites

I'm not sure about the T72M1 armour quoted... I was under the impression that the face plate was 16mm of soft steel, and the T72A glacis array was initially unchanged apart from this addition - however I also had an impression that the proportion of steel above and below the Stef was slightly different in the T72A than the T72 already. Mind you this is from memory only and subject to error..

T72 80mm RHA/HSS + 105mm STEF + 20mm RHA

T72A 60mm RHA/HSS + 105mm STEF + 50mm RHA

T72A(late) 16mm MS + 60mm RHA/HSS + 105mm STEF + 50mm RHA

T72B got an improved array with multiple layers of steel and a two layer BDD insert:

T72B 60mm RHA + 5mm rubber + 3mm Steel + 19mm Air + 3mm Steel + 5mm rubber + 60mm RHA + 10mm Ballistic Liner + 50mm RHA

ISTR that this also got an applique plate (plus of course the now ubiquitous ERA boxes/plates).

So: I'd argue that - except for the 16mm applique - the 'redesigned Glacis' referred to in the article is the T72B - and this wasn't tested or demonstrated to be flawed in the Gulf... (Not that it would necessarily have turned out much differently given the conditions the battles were fought in).

As for the importance of the M111 - my take - it was performing much closer to the normal penetration than the cosine law would allow - so the 68 degree slope of the glacis - assumed to be near immunity based on the performance of Soviet rounds, was a marginal perforation (modern APFSDS rounds apparently treat this slope as nearer 62 degrees for perforation length)... it wasn't so much the few mm of extra steel required to stop it that was alarming to them, but that suddenly all the non-protected roof, and the whole of the glacis was vulnerable to developed versions of this type of ricochet resistant APFSDS round (nose shape and composition, plus being a mono-bloc tungsten round).

Share this post


Link to post
Share on other sites
Well, according to an article published by Armor Magazine, Kubinka tests of M111 "Hetz" convinced Soviets to design a new glacis for T-72A, which was also implemented on T-72M1

But so far in my short SB Pro PE 2.5 experience, the round's performance is somewhat underwhelming - T-72Ms are shrugging off this round just fine (both turret and glacis) even at close ranges unless hit at weak spots.

Well, if the redesigned glacis is shrugging off mutiple rounds, then maybe their redesign was a good one.

What does that round do against a T-55 or T-62?

Share this post


Link to post
Share on other sites
... Kubinka tests of M111 "Hetz" convinced Soviets to design a new glacis for T-72A, which was also implemented on T-72M1

But so far in my short SB Pro PE 2.5 experience, the round's performance is somewhat underwhelming - T-72Ms are shrugging off this round just fine (both turret and glacis) even at close ranges unless hit at weak spots.

It is important to keep in mind that even though for a computer game the armor/ammunition interaction model in Steel Beasts is pretty refined, we explicitly reject all suggestions by our customers that it will provide an analytically correct prediction of plate perforations.

Our model is overengineered to the extent that many of our assumptions and parameter tweaks aren't based on scientific, quantitative data - but, lacking a reliable basis for any but the most fundamental facts, are based on what appears plausible for us.

Of course we try to get as much right as we possibly can, but even the best understood and most detailed models in our object library have glaring gaps in the data background that would be necessary to actually predict with reasonable confidence the outcome of shot X at impact angle Y and velocity Z in location Q.

SB Pro doesn't claim simulation results that will hold up to scientific scrutiny. These are all ballpark estimates. They certainly are better ballpark estimates than, say, a hitpoint based model. Still, a statement "SB Pro proves that round N cannot perforate tank M at range P" would be a folly; don't make that mistake.

Share this post


Link to post
Share on other sites
Well, if the redesigned glacis is shrugging off mutiple rounds, then maybe their redesign was a good one.

What does that round do against a T-55 or T-62?

Hm, was that mentioned redesigned glacis also applied to the M (not M1) version?

Against T-55 and T-62 the round pierces them mighty fine.

It is important to keep in mind that even though for a computer game the armor/ammunition interaction model in Steel Beasts is pretty refined, we explicitly reject all suggestions by our customers that it will provide an analytically correct prediction of plate perforations.

Our model is overengineered to the extent that many of our assumptions and parameter tweaks aren't based on scientific, quantitative data - but, lacking a reliable basis for any but the most fundamental facts, are based on what appears plausible for us.

Of course we try to get as much right as we possibly can, but even the best understood and most detailed models in our object library have glaring gaps in the data background that would be necessary to actually predict with reasonable confidence the outcome of shot X at impact angle Y and velocity Z in location Q.

SB Pro doesn't claim simulation results that will hold up to scientific scrutiny. These are all ballpark estimates. They certainly are better ballpark estimates than, say, a hitpoint based model. Still, a statement "SB Pro proves that round N cannot perforate tank M at range P" would be a folly; don't make that mistake.

I don't think anyone knows what would truly happen at battlefield if certain round hit certain tank at certain range/angle...

I wasn't trying to prove that either the article or the SB is wrong; rather, knowing that SB has "good enough" estimates, I was asking for possible explanations that links together the alerted soviets mentioned in the article and the results of the SB model. (Well, I certainly didn't conduct a proper armor/ammunition model interaction experiment with SB though. Oh, and I admit that I assumed that Esim people had read and known the contents of the aformentioned armor magazine article years ago, and created SB after their own evaluation of the article.)

Sigh... all these Soviet tank alphabet&number cocktails gives me headache whenever I try to know more about them :(

Share this post


Link to post
Share on other sites
I wasn't trying to prove that either the article or the SB is wrong; rather, knowing that SB has "good enough" estimates, I was asking for possible explanations that links together the alerted soviets mentioned in the article and the results of the SB model.

It's pretty simple, actually: If in doubt, SB Pro is wrong - even if other simulations are "even wronger".

Share this post


Link to post
Share on other sites
It's pretty simple, actually: If in doubt, SB Pro is wrong - even if other simulations are "even wronger".

Heh, am I the only one to think that you're too humble?

Share this post


Link to post
Share on other sites

No, I'm serious here. As far as perforation mechanics of APFSDS and HEAT rounds are concerned, our model is pretty reasonable and generally on a basis of credible data as far as ammo performance and armor thicknesses are concerned. However, especially the case of marginal perforations cannot be simulated with high confidence - for that our data aren't good enough (we rely on what's publicly available, after all) - and also our approximation of armor strength variation by the simple cosine of impact angle is actually a gross simplification which tends to overpredict armor protection especially in the case of shallow impacts.

The biggest question mark however is what will happen AFTER the successful plate perforation. There simply aren't suitable data available. Should the base likelihood of, say, radio damage be 10% or 30%?

Nobody can really say .. but a margin of error of 200% is pretty severe. In the end, we go by gut feeling here, and that can be quite wrong (even if nbobody notices this because nobody knows better (or at least those who do can't say)).

I therefore emphasize that the damage model in SB Pro isn't made for analytical purposes. The results simply aren't reliable enough to decide whether a specific round at a specific range and impact location should ...just... perforate the target successfully or not. SB Pro simulates component failures, but only a fool would use it to predict the number of spare parts at the end of a battle to restore a vehicle fleet to operational readiness.

And I dare to make the prediction that the damage model in SB Pro will NEVER be good enough for this. We have to acknowledge that there are limits of what one can do in a simulation, this being one of them.

Share this post


Link to post
Share on other sites

Well, put some (high quality)FEM software under SB's hood und run a complete simulation of the impacts physics for each hit. This may lower the error at least for the penetration.

It may also lower the framerates a little ;-)

Share this post


Link to post
Share on other sites
Well, put some (high quality)FEM software under SB's hood und run a complete simulation of the impacts physics for each hit. This may lower the error at least for the penetration.

It may also lower the framerates a little ;-)

Huh, more like freeze the entire sim for 2 hours while it crunches data.

Also this would not provide real life results as FEM makes the assumption that all objects are homogenous.

And explosions are pretty random in their characteristics.

(I refer you to Ssnake's mention of Radio damage)

All these effects listed by Ssnake make a simulation creators life hell.

(Especially if they are perfectionists.)

Share this post


Link to post
Share on other sites
How serious a problem are marginal perforations to the victim tank?

According to the Odermatt group of equations, the penetrator residual velocity (and the peak fragment velocities (small fragments on the front of the fragment ellipsoid)) will be around 50% of the impact velocity with armour 95% the limiting thickness.

So IMO, a pretty major 'event' still.

The projectile length remaining will be much reduced though - only around 5% of the original length left as large nose fragments or the unbroken body, the rest eroded during the perforation or liberated as small fragment particles...

Share this post


Link to post
Share on other sites

I hate to nibble on the closed case but to share further SB "discovery;"

The part of T-72M glacis that is "inside" the V-shaped splash guard is in fact vulnerable to M111 / DM23 rounds @ 1000m ~ 1500m.

But because of either my poor gunnery skill or native dispersion of the round/gun, it is quite hard to sharpshoot that vulnerable part of the glacis, hence my nigh-invulnerable glacis image.

P.S: where can I find tutorials for Leopard AS1?

Share this post


Link to post
Share on other sites

Yes, the area between the fuel tanks, and near the driver's hatch is very vulnerable - also the nose armour in the lower central area, and the turret roof armour, above the main frontal array...

These 3 areas are vulnerable on all developed versions of the T72 also - maybe not to quite the same extent, but it is surprising how little it takes when you can hit the right spots..

Share this post


Link to post
Share on other sites

The vulnerable part was bit too large to be considered near the hatch

And surprisingly nose held well

...I'm hoping for SB Wiki armor data updates to see if what I'm conjecturing is right

Share this post


Link to post
Share on other sites
The vulnerable part was bit too large to be considered near the hatch

And surprisingly nose held well

...I'm hoping for SB Wiki armor data updates to see if what I'm conjecturing is right

Yes, you are right. It is substantially weaker in those places (just above the "V" and below the driver's hatch). This is by design as the armor is thinner there as the drivers compartment area rises to the top of the hull. That is at least one compliment to SB's armor system, with inner walls represented, armor is at least realistic in the sense that it is naturally thicker or thinner in places where it would be so in real life. It would be a mistake to assume that all of one continuous flat surface on a vehicle in SB is one thickness level, it varies greatly depending on what is under it.

Right, maybe one day we can get around to adding more LOS thickness images to SBwiki -- just so much to do lately...

Share this post


Link to post
Share on other sites
Also this would not provide real life results as FEM makes the assumption that all objects are homogenous.

Well, as in all simulations the quality of the FEM is defined by how much resources you put into the model. If you know all the armour layers (and penetrator components for that matter)you can model them in CAD and give each layer the right parameters for the FEM calculation you get results that are close to reality.

Key word: resources. the same goes for SB where resources would include development time but also CPU/Graphics power(now use having a perfect model, that only runs on high power workstations)

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

×