Jump to content

IBob

Members
  • Posts

    3,116
  • Joined

  • Last visited

  • Days Won

    27

Everything posted by IBob

  1. Them's pretty big vanes at 4" x 16". Be interesting to see the detail of how they reduce drag, which the article says is the result of the upswept loading door/s shape.
  2. KJ, if you watch the briefing, they are very specific about what they have so far in terms of altitudes. In fact they go through it about 3 times: for the Blackhawk at time of briefing, they only have ATC radar derived altitude. These choppers are part of a continuity of government arrangement, for moving government personnel out if the seat of government is threatened in some way. In those circumstances, they definitely wouldn't want to be transmitting their location by ADS-B or any other means. So maybe they run dark on these exercises???
  3. Yes, I just saw their NTSB press briefing. While no doubt they have a long way to go, they sound fairly confident of the 325' figure at this point, derived from corrected ADS-B and another source that I didn't catch. For the helo, so far they have only the ATC radar readout, which they say shows 200', albeit a less granular (precise) source. This is understandably causing some press confusion, despite them going over it several times. I also watched commentary from another pilot who pointed out that in the necessary low level banked left turn for a landing on the shorter 33 strip, the airline pilots would have been tightly focused to the left then ahead............with the helo meanwhile coming from the right.
  4. Deano, the Blancolirio summary I watched had the Helo at 200'..........then 300' just before the collision. I don't know how those numbers were derived. I guess we'll find out in due course. Or not.......( One of the reports right after the event contained the assertion that the airliner 'suffered a sudden loss of altitude passing over the river', or words to that effect. Where that came from, and whether it was just uninformed conjecture, I don't know. Either...
  5. How was the Helo altitude derived? And does/did that value require correction for QNH?
  6. You mean the gentleman's aerial carriage, Nev?........)
  7. This is how ICP present it for the Savannah, facthunter: a graph with 2 lines showing the forward and rear limits. Using the 'arms' provided and the weights (for fuel, people and baggage) you combine the results with the original aircraft w & b figures, then plot the result on the graph to see where you are on the allowable range for that all-up weight. That part is very straightforward (and it's also near impossible to load outside the limits unless you were to grossly overload the baggage, which has a limit of 20Kg.)
  8. Thanks all for your input.....I'm learning! I see now that there is no fixed centre of lift, and that a forward datum avoids possible errors with negative values. What brought this up is the question of weight in the baggage area (immediately behind me) vs weight in a tail locker. In a steady flight situation, a much smaller weight in the tail locker will have a similar balance effect as a larger weight in the baggage area. But how much smaller? Measuring from the LE as the POH does, the tail locker 'arm' would be approx 3.35M. As previously stated, the baggage area 'arm' is 1.32M. This suggests that 10Kg in the baggage area has the same w & b effect as almost 4Kg in the tail locker. That just doesn't feel right, somehow.............(
  9. My Savannah (no sweep on the wings) came with a set of standard 'arms' in metres for use when calculating W & B for a given load. Eg for baggage area: Kg baggage X 1.32M, 1.32 being the given 'arm' for baggage. These arms all appear to be measured from the leading edge of the wing. I would have thought they should be measured from the centre of lift. As any weight forward of that would push the nose down and vice versa. I'm interested as I've been thinking about the difference of weight in baggage vs weight in tail, and with the given arm for baggage, the (theoretical) difference does not seem as great as I would have expected. Can anyone explain this (why the leading edge?) to me.....or point me to a thread that will do that?
  10. My tank seepage was a bit scary: No signs of it during initial ground runs or test flights. But shortly after someone said I had a damp patch under, and when I popped the belly hatch the entire outer walls of the tank were damp and shiny: fuel making it's way down the outside. And quite probably an explosive fuel/air mixture in the confined space of the rear fuselage.............(
  11. Moneybox, as I recall the fittings in my tank are as you describe: threaded brass 'nuts', moulded in.
  12. As I recall, the two main ports were threaded females that looked to be moulded into the tank, and the fittings that went into them had some sort of 'rubber' ring to make the seal. I wonder how that third port was fitted to make it fuel-tight? My tank developed seepage from from one of those tops ports, which wasn't evident initially. I had to pull the tank and trim some of the plastic round the fitting to get it to seal properly. May be worth keeping an eye on initially, Marty.
  13. I was wondering that too: the standard ICP unit comes with just 2 ports (plus the level switch) in the top, so far as I know. The VG model I have seen was built with each of those ports connected to one of the two wing tanks. To give venting they then brought both/all tanks into one port, and used the other for the vent line.
  14. Marty, at cruise, if I valve off all fuel to the receiver, the low fuel indicator comes on after 2.5minutes. That's about 0.7L. Which is to say it takes a fair sized bubble to give false alarms. But I have seen it in an unvented VG.
  15. Ops....certainly not got him! Got the message....and added the method he outlined to my quite narrow fund of knowledge. Thanks, Blue.
  16. Gotcha............)
  17. Blue, if the blade height is 1100mm, then 1 deg of pitch is about 20mm of laser spot movement. Actually a bit more, but not hugely so: 1100 x 2 x 3.14 = 6908 Divide by 360 = 19.2
  18. After that, I think smooth running comes down to carb synchronisation. That starts with the standard Rotax adjustment of the idle throttle stops and idle jets, which is a quick and simple job. I took particular care installing my throttle linkages and routing the cables, then adjusted the cables so that both carbs come off the full throttle stop at the same time when when throttling back. Since much of flying is done at higher engine speeds, this seemed to me to be the most important end of the range. I do check at the other end of the range...that both carbs come off the idle throttle stops simultaneously, and in my case they do that too. I made minor adjustments to maintain the synchronisation over the first 50hrs. Since then I have checked periodically but no further adjustment has been necessary. I also fitted a simple idle position stop to the throttle bar that runs across the firewall: without this, if the pilot pulls back hard on the throttle, the load goes onto the carb throttle arms, and once they hit the carb idle stops, the arms will flex or distort.
  19. Marty, for the stick with clamp, I just hold it vertical to start with and put a masking tape marker on the floor where it lands. I then position the stick on that marker for each blade. I also use masking tape on the floor where the laser hits, so I can make precise marks with a pen. As for tracking, I can't remember clearly how I did it. I think I had the aircraft facing a wall, brought each blade round onto the stick then measured to the wall. Or something like that. I certainly didn't go to the bother of removing the prop. I'm not sure if you have measured and marked in from the tips for the laser location on the blade. If not, I would definitely recommend that too. I guess it all sounds like a lot of mucking around, but I have found it simple and not a long job once you have the tools and method sorted. An additional check as to how well you've done can be made in flight: at certain sun angles, the sun flashes off the back of the blade and all the flashes will look the same if the pitching is accurate.
  20. Marty......much tidier than the thing I cobbled up! Apart from very accurate measurements from floor to blade (I use a stick with a clamp on it, bring the each blade down onto the clamp) and very accurate measurement from blade tip to laser position (I use a sliding square and mark the location with masking tape).........there is one other thing to watch for using this method, and that is blade tracking: One of my blades tracks about 2mm further forward at the tip than the other two, so I factor that in. If I were using a similar method, but with an inclinometer on the blade, I wouldn't have to be concerned with the tracking. The only proviso then would to be sure I was using a good quality inclinometer, as they are not all born equal........ For myself, I will carry on with what I've got: done with care it does get very accurate results. PS I tighten all bolts fully then check all blades again, as sometimes the tightening does shift the pitch a bit.
  21. Skippy, I have to disagree with 'completely ineffective'. For instance, you don't think landing light/s make an approaching aircraft more obvious when you are waiting to enter a runway??? What I would agree with is that typical/traditional wingtip and beacon lights are very poor in bright sunlight. Having said that, I have seen, a beacon so bright in daylight that I walked the 700M across the airfield to see what it was. That was on the ground. I'm pretty sure it was one of the Aveo products. And you may have noticed that cautionary lights on road maintenance and emergency vehicles are now a whole lot more visible at distance than they used to be? The technology is getting better and brighter, and that will percolate through to aviation kit.
  22. Hi Lyndon, I can't answer as to what is mandatory......but I'm sure others here will know. What I would say is that, now we are better understanding the limitations of the eye in 'see and be seen', they are definitely not something I would skimp on.
  23. IBob

    Total profiles 31122024.jpg

    Red, it would be interesting to know what proportion of those went on to cover the costs of development in production and sales.
  24. I guess they probably do that fine at the factory......where they've had the practise, have worked out their methods, and will have jigs and tools to expedite everything. So they'll be on the tools and building all the time. Whereas, on a first build, there is probably a great deal of time spent figuring things out, getting things right, and generally proceeding cautiously. I have struck the same thing with house renovations/modifications: a builder mate of mine would drop by and do more in 3 hours than I got done in a day. It puzzled me a bit, so I began watching myself......and realised just how much time I was spending trying to think things out etc, while he knew just what he was doing and his 3hrs was on the tools full time.
  25. sfG I can't speak for the Ventura, but for the Savannah I've come across two numbers: 400hrs and 700hrs. It would be really interesting to know where they came from, as an ex-airforce sheetmetal worker here took about 1700hrs on his. And while I did not log the hours, mine would easily have taken that long too.
×
×
  • Create New...