Bug? Mammoth overheats without fair warning

[ShadowZone]

So basically I have this problem: I build a huge launcher (if you know me, you know I can't do small) but my usual designs fail miserably due to mammoth engines suddenly overheating without warning:

http://gfycat.com/DisastrousInnocentDuckling

This is consistent behavior every time I group stacks of 3.75m tanks with mammoths, however it's always a different engine that decides to pop.

Any ideas?

[sal_vager]

I'm actually a bit surprised you have not run into this before ShadowZone, engines will get hotter when clustered and this has been the case well before the recent thermal changes, and this is supposed to happen.

You need to space them apart a bit, you could also use radiators to absorb the heat and use struts to conduct heat from the engines to cooler parts with more thermal mass, such as your fuel tanks.

It's an engineering challenge, you can beat it!

[ShadowZone]

24 minutes ago, sal_vager said:

I'm actually a bit surprised you have not run into this before ShadowZone, engines will get hotter when clustered and this has been the case well before the recent thermal changes, and this is supposed to happen.

You need to space them apart a bit, you could also use radiators to absorb the heat and use struts to conduct heat from the engines to cooler parts with more thermal mass, such as your fuel tanks.

It's an engineering challenge, you can beat it!

I had clustered Mammoths getting hot before, that is not the issue. The issue is them going from "hey, I'm fine" to "OMGXPLODE!" in 2 seconds (I could provide you a screenshot of the Flight log. It literally was 2 seconds) even though the vehicle is just at 50% throttle. I would also expect sort of an even heat distribution (visible on the heat gauge on the staging UI) and not on singular engines at different times. Just look at how they behave like popcorn in the gif.

I put my Ozymandias in orbit with a similar rocket in 1.0.4 and had no issues. Not even one mammoth overheating.

I tried radiators on a different design, but that had no effect. It didn't slow down the heat buildup either. The only thing that worked was removing the center mammoth from a 7 tank/engine stack cluster (1 center, 6 in symmetry around it). But that wouldn't work with this 800 ton payload beast

 

My point is: It has worked before. It doesn't since 1.0.5. And it behaves unpredictably and way too fast.

[sal_vager]

Try moving that mammoth elsewhere, the thing is, heat was tweaked a lot and many bugs were fixed so not everything from before 1.0.4 is going to act the same way.

It's predictable as it's continually happening to you on this craft.

[ShadowZone]

1 minute ago, sal_vager said:

Try moving that mammoth elsewhere, the thing is, heat was tweaked a lot and many bugs were fixed so not everything from before 1.0.4 is going to act the same way.

It's predictable as it's continually happening to you on this craft.

if by predictable you mean it in the sense of "this is never ever going to reach orbit" then I have to agree

However, I'd prefer the kind of predictability where you could anticipate the threshold of your technical equipment: "if put at throttle level x for time y then event z occurs". This is not the case here. The way it is now I can only work around a new problem that was introduced with the latest update. And trusted designs that have worked for months are now completely obsolete.

Color me unhappy...

 

[sal_vager]

Can you upload the craft?  It may have too many parts for me to run smoothly but I can have a go at fixing it, so can others.

[ShadowZone]

4 hours ago, sal_vager said:

Can you upload the craft?  It may have too many parts for me to run smoothly but I can have a go at fixing it, so can others.

No need, I think I found the problem. Apparently the part model for the mammoth is now really strict. Meaning that if those 4 exhausts that do not really keep in line with the 3.75m diameter clip into each other, the aforementioned effect takes place. So after some minor rotating and offsetting (you can't really tell the difference when you look at the vehicle) the Mammoth exhausts no longer touch each other and I managed to get out of the atmosphere!

So yeah, question anwered, it seems

 

That the payload started flopping around inside the fairing and exploded since there is no way to stabilize/spacetape tall payloads to the fairing walls is an entirely different matter...

[sal_vager]

Oh man, payloads, I have had some success by closing the fairing around the very top of the payload then running some struts from the cargo outwards, it's tricky but if you get it right you strut the payload to the fairing and the whole thing is a lot stronger.

[ShadowZone]

I try to use sort of an exoskeleton if there is enough space. Like this here: https://youtu.be/qWZjiuNC4Yg?t=6m38s

But the entire fairing thing still is not really thought through

[sal_vager]

They don't occlude from physics, so you still need lots of struts for excessive loads, if your payload wobbles without a fairing then it'll do it with one.

[ShadowZone]

True - and that's a challenge I am willing to accept

But the problem is the fairing base. It is the thing that breaks most of the time. And you can't attach struts from it only to it. I usually have 6 struts coming from the bottom tank up to the fairing base and from some part inside of the fairing to the fairing base in hopes of it staying on top of the stack. That thing breaks like a twig.

 

[sal_vager]

It is wobbly, worse than decouplers, but thankfully struts attach properly at both ends (have done since 1.0.0) so they help.

One thing to try is putting the fairing on a decoupler on the top of the payload, then building the fairing downwards, you can use a smaller fairing to join to larger rockets that way, you avoid the fairing body lift bug and you don't have your payload on a wobbly fairing base.