Why does my SSTO lose control on take off

[Raptor kerman]

https://kerbalx.com/RaptorWorKs/SSTO-mk2

 

Here is the craft file, take off is at 110m/s. Having looked in the hangar for a long time, I can'y find any obvious problems with the SSTO prototype. Its CoG is entered, Fuel is distributed evenly. So why, immediately after take off does this SSTO bank sharp right (or left) before inverting and crashing without any control inputs. This was built in 1.1.3 and I run KSP on a mac.

[Snark]

2 hours ago, Raptor kerman said:

So why, immediately after take off does this SSTO bank sharp right (or left) before inverting and crashing without any control inputs.

Hard to tell for sure from that screenshot-- could you post one from the SPH (preferably with the CoM displayed)?  Or sitting intact on the runway?

In general, an unstable airplane is some combination (one or more) of the following:

• Your CoM is in the back, so it wants to fly backwards.
• Your control surfaces are right next to the CoM, so they have no lever arm to work with and might as well not be there.
• Less of a problem than the above two, but a smaller contributing factor:  roll authority isn't turned off on the vertical stabilizer.

The solution to the above problems:

• make sure the CoM isn't way at the back
• make sure you have plenty of control surface as far as possible from the CoM
• make sure that your vertical stabilizer has only yaw enabled.

That's very general advice, though-- hard to say for sure without seeing a better screenshot of the craft.

 

[OHara]

It flies nicely hands-off, but the ailerons roll controls are reversed so SAS fails to control roll.

The automatic assignment of controls sometimes gets confused, and in this case I don't see an easy way to un-confuse it.  This plane flies nicely, though, if you have each control respond to just one of pitch/yaw/roll, and reverse the control-authority on roll to compensate for the error.

[Raptor kerman]

9 hours ago, OHara said:

It flies nicely hands-off, but the ailerons roll controls are reversed so SAS fails to control roll.

The automatic assignment of controls sometimes gets confused, and in this case I don't see an easy way to un-confuse it.  This plane flies nicely, though, if you have each control respond to just one of pitch/yaw/roll, and reverse the control-authority on roll to compensate for the error.

Thanks!

 

11 hours ago, Snark said:

Hard to tell for sure from that screenshot-- could you post one from the SPH (preferably with the CoM displayed)?  Or sitting intact on the runway?

In general, an unstable airplane is some combination (one or more) of the following:

• Your CoM is in the back, so it wants to fly backwards.
• Your control surfaces are right next to the CoM, so they have no lever arm to work with and might as well not be there.
• Less of a problem than the above two, but a smaller contributing factor:  roll authority isn't turned off on the vertical stabilizer.

The solution to the above problems:

• make sure the CoM isn't way at the back
• make sure you have plenty of control surface as far as possible from the CoM
• make sure that your vertical stabilizer has only yaw enabled.

That's very general advice, though-- hard to say for sure without seeing a better screenshot of the craft.

 

I'll get a screenshot in later today!

[Raptor kerman]

Tell me if there's any problem with this layout

[Jarin]

10 hours ago, Raptor kerman said:

Tell me if there's any problem with this layout

First problem I see would be "lost in Mk2 hell". You're eating drag all over the place with that design. That ship would probably fly 100% better if you completely replaced those outer mk2 nacelles with 2 mk1 nacelles each (half the drag with double the fuel). I know, I know, I love the way the Mk2s look, too, but the numbers are terrible. Nowhere near enough lift to justify the drag, and no bonus fuel at all. You really only want to use them where necessary for things like cargo bays. A central mk2 fuselage is fine, but there's never a good reason to make secondary fuselages mk2 where a mk1 would do the job just as well. 

Other than that, just make sure to restrict the axes of control on those control surfaces. The leading-edge pieces need to be restricted to roll-only and have their control authority value set to negative. The back ones should be fine, though I'd consider restricting them to pitch (not necessary, but possibly helpful on turn stability). 

I can't tell what's going on at the back of the engine fuselages, but it looks draggy. Make sure you're not leaving any attachment nodes open, and don't have any mis-sized matchups (mk1 points attached to mk2 without an adapter, etc)

Unless you're morally opposed to them, use autostruts rather than manually strutting parts together. Struts are draggy as hell for some reason and drag is SSTO death.

Edited March 29, 2017 by Jarin

[Wanderfound]

On 29 March 2017 at 9:35 AM, OHara said:

The automatic assignment of controls sometimes gets confused, and in this case I don't see an easy way to un-confuse it.

My first thought is to make sure that the wings and nacelles are attached with lateral rather than radial symmetry. They may be working in opposite directions.

 

5 hours ago, Jarin said:

First problem I see would be "lost in Mk2 hell".

Agreed in general, but:

If you like Mk2 fuselages, there is a non-cheaty way to make them non-draggy: wing incidence.

Use the rotation tool to angle the leading edge of your wing up a few degrees. Keep an eye on the CoL marker as you do this; it will move, so you'll probably need to shift the wings a bit to compensate.

The idea is to set them so that you can maintain level flight at speed and altitude while keeping the nose pointed prograde. It doesn't take much; about 3-5° is usually plenty. It should be just barely perceptible.

I'd still dump the lateral fuselages, though. Mk2s are best as single stack or one Mk2 plus a pair of 1.25m side tanks. If you need anything bigger, go Mk3.

Edited March 30, 2017 by Wanderfound

[Raptor kerman]

12 hours ago, Jarin said:

First problem I see would be "lost in Mk2 hell". You're eating drag all over the place with that design. That ship would probably fly 100% better if you completely replaced those outer mk2 nacelles with 2 mk1 nacelles each (half the drag with double the fuel). I know, I know, I love the way the Mk2s look, too, but the numbers are terrible. Nowhere near enough lift to justify the drag, and no bonus fuel at all. You really only want to use them where necessary for things like cargo bays. A central mk2 fuselage is fine, but there's never a good reason to make secondary fuselages mk2 where a mk1 would do the job just as well. 

Other than that, just make sure to restrict the axes of control on those control surfaces. The leading-edge pieces need to be restricted to roll-only and have their control authority value set to negative. The back ones should be fine, though I'd consider restricting them to pitch (not necessary, but possibly helpful on turn stability). 

I can't tell what's going on at the back of the engine fuselages, but it looks draggy. Make sure you're not leaving any attachment nodes open, and don't have any mis-sized matchups (mk1 points attached to mk2 without an adapter, etc)

Unless you're morally opposed to them, use autostruts rather than manually strutting parts together. Struts are draggy as hell for some reason and drag is SSTO death.

 

6 hours ago, Wanderfound said:

My first thought is to make sure that the wings and nacelles are attached with lateral rather than radial symmetry. They may be working in opposite directions.

 

Agreed in general, but:

If you like Mk2 fuselages, there is a non-cheaty way to make them non-draggy: wing incidence.

Use the rotation tool to angle the leading edge of your wing up a few degrees. Keep an eye on the CoL marker as you do this; it will move, so you'll probably need to shift the wings a bit to compensate.

The idea is to set them so that you can maintain level flight at speed and altitude while keeping the nose pointed prograde. It doesn't take much; about 3-5° is usually plenty. It should be just barely perceptible.

I'd still dump the lateral fuselages, though. Mk2s are best as single stack or one Mk2 plus a pair of 1.25m side tanks. If you need anything bigger, go Mk3.

Thanks for the tips! But will it stop randomly losing control on take off despite everything being aligned?

[Wanderfound]

2 hours ago, Raptor kerman said:

But will it stop randomly losing control on take off despite everything being aligned?

If you fix your control surfaces and make sure CoL is behind CoM it should.

[Jarin]

10 hours ago, Wanderfound said:

 

If you like Mk2 fuselages, there is a non-cheaty way to make them non-draggy: wing incidence.

Oh trust me, wing incidence and I are old friends. But even dead-on to the airflow, Mk2s are strictly worse than a mk1 of the same length, in all respects other than "ability to fit a cargo bay" and "looking pretty".

[Wanderfound]

31 minutes ago, Jarin said:

But even dead-on to the airflow, Mk2s are strictly worse than a mk1 of the same length, in all respects other than "ability to fit a cargo bay" and "looking pretty".

Crew capacity and  heat tolerance.

Edited March 30, 2017 by Wanderfound

[Jarin]

1 hour ago, Wanderfound said:

Crew capacity and  heat tolerance.

Two mk1s do the same job better for crew (the mk1 crew cabin is the most efficient crew transport part in the game). And I keep hearing people talk about heat tolerance issues, but I had an entire line of mk1 SSTOs that could do direct-drop re-entry onto KSC without burning. Shed your ascent heat in orbit, people.

Honestly, the only difference I've ever gotten any real benefit from is impact tolerance. Mk2 fuselages are a lot likely to go kaboom if you flub something.

 

- I feel the need to clarify here, I'm not actually arguing against using Mk2s. I'm willing enough to make minor performance sacrifices for appearance that every light SSTO in my current fleet has a Mk2 core; the mk1 lines have been almost completely retired. I'm just grumpy about game balance anyway.

Edited March 30, 2017 by Jarin

[AeroGav]

@Raptor kerman I had a play with your craft file, and made several stages of modifications.

First,  was a minimal changed version to resolve the control issue you were complaining of.  If waggle the stick before starting engines (real pilots do that too!) you'll see that the inboard elevons are moving in the opposite direction to the outboards.  So, i set their authority to -66% and the outboards at +66%.  Now they move in the same direction in response to roll input, but they try to work against each other again when you go nose up/down.   I resolved this by just disabling the pitch function of the inboard elevon.   Also , I disabled yaw on everything except the rudders , which i set to yaw only.

For the leading edge "slats", I gave them -80% authority at set them to pitch only.   The outboard ones i disabled completely,  they are aft of CG so it's hard to work out what they're gonna do.        The canards were moved forward and set to 150% and pitch only.   The advanced canard is pretty small and only deflects to 10 degrees (the big elevons do 25) so giving them 150 doesn't stall them.

For the landing gears, I disabled the steering on the rear tyres and used friction override on the nose gear to lower friction and make it "understeer" on the ground.

Finally, with regard to those triple engine mounts - you have one pair of rapiers on the centreline, two above centreline, and the rocket engines on the bottom.  This means most of your jet power is above centre of mass and creates a pitch down force, that makes it hard to take off.     However since your rocket engines are underslung you'll get a big pitch up effect when you start the swivels.

I put the swivels in the middle with rapiers top and bottom, that way all is balanced.

Craft file of this version , here 

https://www.dropbox.com/s/wdz13qqjykp2ese/SSTO mk2.craft?dl=0

 

Having taken off, i was unable to get through the sound barrier on my first three attempts.  I am not sure if it can be done.  Mk2 hell indeed !

However, your method of engine mounting adds to the drag.     You have a mk2 fuselage joined straight onto a 1.25m tri adapter.  The size mismatch creates drag.   You should have used a mk2 to mk1 (1.25m) adapter before attaching the triple engine mount, assuming that's what you want to go with.        But  since you have 3 mk2 fuselages in parallel,  you'd probably just be better off with a mk2 bicoupler at the back of each fuselage.

Here's a minimally changed version, with bicouplers on the two outer fuselages, giving 4 engines total.  2 Whiplash and 2 Rapier.    Just barely makes it through the sound barrier.   Now, I was about to ask why the hell you fitted swivels to the original.  I normally reckon on 60kn of close cycle power per 15 tons of weight to get to orbit,  so 2 RAPIERS in close cycle should easily have been able to push this to orbit.  Wrong !  There is still so much drag from all the mk2 fuselage, that it stops accelerating at about 1100 m/s in level flight at 21km - both RAPIERs still kicking out over 220kn plus whatever the Whiplashes still have.   In close cycle mode, the RAPIERs drop to "only" 180kn,  so we start slowing down !

https://www.dropbox.com/s/dhv14ewz6sd8kz9/SSTO mk2 CLEANUP.craft?dl=0

So,  I  made a more radically modified version, I hope I've preserved enough of the essence of the original.

The major change was to rotate the two side fuselages 90 degrees.  This may reduce the drag that comes when the airplane is pitched up to a positive AoA,  but it also allows me to apply an incidence angle to the wings without it looking so awful (mk2 fuselages are meant to have flat wings, otherwise the wing/fuselage blending gets messed up at the leading and trailing edge).

I also changed the engine config, one whiplash, two rapier, two nukes.   It accelerates through mach 1 much easier than the other versions, despite having fewer engines.

The shielded clamp o tron was swapped for an inline clamp o tron (less drag, and i need the engine mount point for something else).   The cockpit was changed for a more heat resistant inline one.   

My usual stock SSTO are designed for max delta V and take a while to get to orbit.  This one has more power than my usual designs and gets to orbit quite smartly,  but still has plenty leftover fuel.  I would have a lot more delta V if it still wasn't carrying so much oxidizer.  With all those mk2 fuselage adapters you're always going to have loads of oxidizer whether you want it or not.

 Well,  you can always tweak that down in the VAB, but i suppose all this oxidizer might be needed if it was flying up with a heavy load instead of empty.

https://www.dropbox.com/s/845dvk1zt7lhj2m/SSTO mk2 CLEANUP2.craft?dl=0

original -

mine -

(i made a copy of my KSP directory to  do this because i have installed Ferram Aerospace on my main game, unfortunately i forgot this meant F12 no longer takes screenies because it's no longer running as a Steam game, so none of the pictures from orbit survived)

Action groups 1 - toggle nukes

Action group 3 - rapier switch mode

Edited March 30, 2017 by AeroGav

[Raptor kerman]

2 hours ago, AeroGav said:

@Raptor kerman I had a play with your craft file, and made several stages of modifications.

First,  was a minimal changed version to resolve the control issue you were complaining of.  If waggle the stick before starting engines (real pilots do that too!) you'll see that the inboard elevons are moving in the opposite direction to the outboards.  So, i set their authority to -66% and the outboards at +66%.  Now they move in the same direction in response to roll input, but they try to work against each other again when you go nose up/down.   I resolved this by just disabling the pitch function of the inboard elevon.   Also , I disabled yaw on everything except the rudders , which i set to yaw only.

For the leading edge "slats", I gave them -80% authority at set them to pitch only.   The outboard ones i disabled completely,  they are aft of CG so it's hard to work out what they're gonna do.        The canards were moved forward and set to 150% and pitch only.   The advanced canard is pretty small and only deflects to 10 degrees (the big elevons do 25) so giving them 150 doesn't stall them.

For the landing gears, I disabled the steering on the rear tyres and used friction override on the nose gear to lower friction and make it "understeer" on the ground.

Finally, with regard to those triple engine mounts - you have one pair of rapiers on the centreline, two above centreline, and the rocket engines on the bottom.  This means most of your jet power is above centre of mass and creates a pitch down force, that makes it hard to take off.     However since your rocket engines are underslung you'll get a big pitch up effect when you start the swivels.

I put the swivels in the middle with rapiers top and bottom, that way all is balanced.

Craft file of this version , here 

https://www.dropbox.com/s/wdz13qqjykp2ese/SSTO mk2.craft?dl=0

 

Having taken off, i was unable to get through the sound barrier on my first three attempts.  I am not sure if it can be done.  Mk2 hell indeed !

However, your method of engine mounting adds to the drag.     You have a mk2 fuselage joined straight onto a 1.25m tri adapter.  The size mismatch creates drag.   You should have used a mk2 to mk1 (1.25m) adapter before attaching the triple engine mount, assuming that's what you want to go with.        But  since you have 3 mk2 fuselages in parallel,  you'd probably just be better off with a mk2 bicoupler at the back of each fuselage.

Here's a minimally changed version, with bicouplers on the two outer fuselages, giving 4 engines total.  2 Whiplash and 2 Rapier.    Just barely makes it through the sound barrier.   Now, I was about to ask why the hell you fitted swivels to the original.  I normally reckon on 60kn of close cycle power per 15 tons of weight to get to orbit,  so 2 RAPIERS in close cycle should easily have been able to push this to orbit.  Wrong !  There is still so much drag from all the mk2 fuselage, that it stops accelerating at about 1100 m/s in level flight at 21km - both RAPIERs still kicking out over 220kn plus whatever the Whiplashes still have.   In close cycle mode, the RAPIERs drop to "only" 180kn,  so we start slowing down !

https://www.dropbox.com/s/dhv14ewz6sd8kz9/SSTO mk2 CLEANUP.craft?dl=0

So,  I  made a more radically modified version, I hope I've preserved enough of the essence of the original.

The major change was to rotate the two side fuselages 90 degrees.  This may reduce the drag that comes when the airplane is pitched up to a positive AoA,  but it also allows me to apply an incidence angle to the wings without it looking so awful (mk2 fuselages are meant to have flat wings, otherwise the wing/fuselage blending gets messed up at the leading and trailing edge).

I also changed the engine config, one whiplash, two rapier, two nukes.   It accelerates through mach 1 much easier than the other versions, despite having fewer engines.

The shielded clamp o tron was swapped for an inline clamp o tron (less drag, and i need the engine mount point for something else).   The cockpit was changed for a more heat resistant inline one.   

My usual stock SSTO are designed for max delta V and take a while to get to orbit.  This one has more power than my usual designs and gets to orbit quite smartly,  but still has plenty leftover fuel.  I would have a lot more delta V if it still wasn't carrying so much oxidizer.  With all those mk2 fuselage adapters you're always going to have loads of oxidizer whether you want it or not.

 Well,  you can always tweak that down in the VAB, but i suppose all this oxidizer might be needed if it was flying up with a heavy load instead of empty.

https://www.dropbox.com/s/845dvk1zt7lhj2m/SSTO mk2 CLEANUP2.craft?dl=0

original -

mine -

(i made a copy of my KSP directory to  do this because i have installed Ferram Aerospace on my main game, unfortunately i forgot this meant F12 no longer takes screenies because it's no longer running as a Steam game, so none of the pictures from orbit survived)

Action groups 1 - toggle nukes

Action group 3 - rapier switch mode

Thanks a TONNE. Also do the wing strakes actually give an AoA advantage like on the Typhoon testbeds or does the KSP aerodynamic model not account for this.

[AeroGav]

31 minutes ago, Raptor kerman said:

Thanks a TONNE. Also do the wing strakes actually give an AoA advantage like on the Typhoon testbeds or does the KSP aerodynamic model not account for this.

No the stock aero model does not care about wing geometry at all.     If you build an airplane with modular wing segments and make a WW1, straight winged Fokker Triplane (like the Red Baron used to fly) and then rearrange the same wing pieces to make a highly swept delta whilst keeping the fuselage the same weight, performance is the same.

All you have influence over with the stock physics model is wing loading and trim drag.

For fuselage pieces a lot more things affect drag, but many of them don't match up to real life aerodynamics.    Using angle of incidence on the wings so the fuselage stays prograde lowers fuselage drag.  So does more wing area since that allows you to fly higher or stay prograde more.     Choosing non draggy parts.   Making sure stuff attaches correctly in the part tree.   Every stack starts and ends with something pointy.   Making sure that when you join two parts, the attach node of both parts is the same size .

If you got a chance to fly it, let me know what you think of my version of your ship.    I tuned the handling to the way i prefer to fly stuff, but obviously that's no good if no-one else likes it that way.

[AeroGav]

10 hours ago, Raptor kerman said:

Thanks a TONNE. Also do the wing strakes actually give an AoA advantage like on the Typhoon testbeds or does the KSP aerodynamic model not account for this.

Also, to put this in context of your original design -

The strakes, and the leading edge devices, will not increase the stalling angle like they would in real aerodynamics.    The trailing edge flaps/elevons don't increase the lift of the main wing,  only their own lift increases when deflected due to having a higher aoa than the main wing, but being a small proportion of the total area the effect is small.   Also, because they have increased AoA, they will tend to stall and stop making lift altogether at low speeds.

The leading edge devices need to move in the opposite way to real life.   In real life,  fighters like the F-16, eurofighter etc . droop their leading edge flaps downwarrds (along with the trailing edge flaps) when they want more lift, to increase the curvature of the wing.    KSP does not see overall wing curvature, only the AoA of the individual part.   So the leading edge flaps actually need to deflect upwards, like spoilers, if you want them producing increased lift.   That is why i set their authority to a negative number.

Finally, you speak of an "AoA advantage".    It's not a fighter plane.  High AoA is draggy, you want it to have sufficient lift all the way to space while keeping AoA low.

[Hodo]

On 3/30/2017 at 11:47 AM, Jarin said:

Two mk1s do the same job better for crew (the mk1 crew cabin is the most efficient crew transport part in the game). And I keep hearing people talk about heat tolerance issues, but I had an entire line of mk1 SSTOs that could do direct-drop re-entry onto KSC without burning. Shed your ascent heat in orbit, people.

Honestly, the only difference I've ever gotten any real benefit from is impact tolerance. Mk2 fuselages are a lot likely to go kaboom if you flub something.

 

- I feel the need to clarify here, I'm not actually arguing against using Mk2s. I'm willing enough to make minor performance sacrifices for appearance that every light SSTO in my current fleet has a Mk2 core; the mk1 lines have been almost completely retired. I'm just grumpy about game balance anyway.

To end the debate on Mk2 vs Mk1 is this.  MK2 works better in FAR because they are a lifting surface, unlike mk1s, which are basically lawn darts.  

But the old fashioned saying always comes to mind, "If anything goes fast enough it will generate lift."

 

[Nothalogh]

So I took the craft, made sure my control surfaces were properly set.

She took of and handled well, so I leveled out and attempted to go supersonic.

That thing is a drag monster, wouldn't budge past .98 Mach.

 

So dump the tricouplers, two shock cones can easily feed four Rapiers.

Good rule of thumb is 100m/s as you pass the Control Tower, and ability to break the sound barrier in at least a 10 deg climb.

DV on orbit at 80 to 90 KM should be around 500.

By rejiggering your design to hell and back I was able to do almost all of that easily.

 

One other thing, that Mk2 end cockpit is useless on SSTO's now, it's just a fancy barbecue for kerbals, use a 1.25m fairing as a nosecone on the end of a long Mk 2 adapter, and either an inline cockpit or a crew tank and a drone core.

And yes, I basically consider it cheating, but clip passive radiators into your fuselage parts.

Atmo heating at the optimum speed and altitude for the Rapier is lethal, and Squad gives us no plausible solution to it, so prepare for lots of crispy kerbals, or spam radiators in ways they won't kill you with drag.

Edited April 3, 2017 by Nothalogh

[Jarin]

2 hours ago, Hodo said:

To end the debate on Mk2 vs Mk1 is this.  MK2 works better in FAR because they are a lifting surface, unlike mk1s, which are basically lawn darts.  

But the old fashioned saying always comes to mind, "If anything goes fast enough it will generate lift."

They're a lifting surface in stock, too. They're just way draggier than they should be compared to any other lifting surface. 

2 hours ago, Nothalogh said:

Atmo heating at the optimum speed and altitude for the Rapier is lethal, and Squad gives us no plausible solution to it, so prepare for lots of crispy kerbals, or spam radiators in ways they won't kill you with drag.

Posted on this a few days ago:

I really don't know what people are doing to have such heat troubles. Unless you're trying to hit 1600m/s+ on air-breathing thrust, anyway.

Edited April 3, 2017 by Jarin

[Nothalogh]

1 hour ago, Jarin said:

I really don't know what people are doing to have such heat troubles. Unless you're trying to hit 1600m/s+ on air-breathing thrust, anyway

1. The cockpit at issue here IS the Mk2
2. The Rapier has a power band at a particular speed range, at a particular altitude range, if you want 1450m/s on airbreathing you WILL melt at Mk2 Standard cockpit

The workaround is to use an inline Mk2 with a long adaptor an a 1.25m fairing as a nose, looks like a polished turd, but it's all very low drag and handles heat like a champ.

The problem here is that Mk2 parts are really only built for ~Mach 4 thermal loads.

[bewing]

My MK2 cockpits never melt. I agree with Jarin.

[Jarin]

2 hours ago, Nothalogh said:

• The cockpit at issue here IS the Mk2
• The Rapier has a power band at a particular speed range, at a particular altitude range, if you want 1450m/s on airbreathing you WILL melt at Mk2 Standard cockpit

Okay, firstly, the Mk2 has a significantly higher heat threshold than the Mk1. And from the album linked in the above post: 1450+ on ascent. 

Secondly: are you sure you aren't running 150% heat or something?

Those are both at roughly the same altitude, a bit under 20km.

 

Edited April 4, 2017 by Jarin

[Wanderfound]

5 hours ago, bewing said:

My MK2 cockpits never melt. I agree with Jarin.

They do require a bit more care than an equivalent ship with an inline cockpit, but they're certainly usable:

[Jarin]

8 hours ago, Wanderfound said:

They do require a bit more care than an equivalent ship with an inline cockpit

Oh yeah, absolutely. But if you're not trying to push the theoretical RAPIER maximum velocity, the faring-nose trick really isn't necessary. 

[Wanderfound]

14 hours ago, Jarin said:

Oh yeah, absolutely. But if you're not trying to push the theoretical RAPIER maximum velocity, the faring-nose trick really isn't necessary. 

Even at maximum, it isn't compulsory. A shock cone is a good enough heatshield to hold 1,650m/s for a full circumnavigation of Kerbin if you do it high enough.