I was firing them up to 100.7km with I think less than 100m variation between repetitions.
If you are interested im my results from this I have typed them in below
The test vehicle used:
Nose assembly
RC-001S
FL-R25 ballast tank to approximately even out the weight between nose cones (ballast figure is amount of RCS fuel loaded)
HammerSRB
2xElevon 1 for stability control
Modus operandi:
Throttle full, SAS on (+ open/close intakes), launch, no time acceleration in the atmosphere, get peak hight from F3 once vehicle starting to descend
Results for 4.2t 1.25m tests
Nose assembly
Ballast Added
Peak altitude with 40% thrust limit
Peak altitude with no thrust limit except for survivability
Fairing 2 section
0
122km
69.5% thrust => 145km
Fairing 1 section
25
116km
N/C
Fairing shortest possible
50
55km
N/C
FL-A10 NC
50
98km
100% => 102km
Aero NC
75
79km
100% => 79km
Shielded clamp-o-tron
50
66km
100% => 59km
Heat shield
0
52km
100% => 42km
Advanced NC
50
105km
71.5% => 121km
Closed shock cone
50
101km
93% => 117km
Open shock cone
50
101km
93% => 117km
Tail cone A
25
108km
71% => 128km
The accuracy of this comparison is limited by the VAB mass readout in steps of 0.1t. Adding ~0.1t to the Fairing 2 section reduced the 40% altitude from 122km to 105km so it is hard to measure the difference between the following with confidence
* Fairing 1 or 2 sections
* Advanced Nose cone
* Tail cone A
If you want to be more accurate then you need to control the test vehicle mass more closely. Or do something like the suggestions in http://forum.kerbalspaceprogram.com/index.php?/topic/129655-what-options-are-there-in-ksp-for-quantitatively-determining-the-drag-produced-by-parts/
From the above experiment I have gone with a 1 section fairing over a shielded clamp-o-tron for it's much lower drag and higher temperature limit.
Regards,
Richard