Logo
Builders' Plan Gallery  |  Hip Pocket Web Site  |  Contact Forum Admin  |  Contact Global Moderator
October 23, 2017, 11:28:02 AM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with email, password and session length
 
Home Help Search Login Register
Pages: 1 2 3 4 5 [6]   Go Down
Print
Author Topic: E-36 "Puzzle"  (Read 11249 times)
0 Members and 1 Guest are viewing this topic.
che
Bronze Member
***

Kudos: 0
Offline Offline

United Kingdom United Kingdom

Posts: 69



Ignore
« Reply #125 on: March 21, 2017, 09:00:45 AM »

(Apologies in advance to Tapio ...............)

Alan,

Thanks for your time in considering this question. I suggested a multiple pulse avoidance delay (MPAD) a few responses back so we seem to be gelling on what MAY work (legality ignored, after all this is for trimming and can be disabled).

I do see circumstances where those of use without the skill, experience and general craftiness of people such as Prof. Thompson would need to and can successfully stop an engine and then DT (via separate button pushing) without breaking the wings, so it is a realistic scenario. I also think that data on the Bauer system at least might be available. The question is would you want to provide a 2nd system for what might be a small market ?

The issue of multiple pulses being sent and their legality is I suspect a red herring as this was used in the original (?) RCDT system back in the mid 80s, in fact it was a requirement of the receiver to get, if I remember correctly, two coded pules within 0.25s or else the DT wouldn't operate. If not legal then a switch would be required to send just a single pulse when the unit is being inspected I guess................

CHE

Logged
afb
Copper Member
**

Kudos: 0
Offline Offline

United Kingdom United Kingdom

Posts: 4



Ignore
« Reply #126 on: March 21, 2017, 01:29:50 PM »

Che

The Bauer instructions give insufficient detail for *my* purposes - I've mailed Ken for further clarification.

Alan
Logged
rivers
Bronze Member
***

Kudos: 1
Offline Offline

United States United States

Posts: 38



Ignore
« Reply #127 on: March 21, 2017, 04:44:32 PM »

Hi all, For what it's worth I build my own RDT systems which closely follow the Bauer format, except I use spread spectrum (LoRa) transceivers for greater range. When the transmit button is down RDT pulses are generated at an 8 per second rate continuously. The pulses are negative going 50 ms wide.

I think it's a good idea to have continuous pulsing. At long range the target model may be low, with the line of sight not so good, as well as RF propagation difficulties. Multiple shots are a help.

I don't think the Bauer system was originally conceived with motor stop on mind. Nevertheless, it works well with timers configured to stop the motor with RDT. My own timers are designed to stop the motor then automatically DT 1.4 seconds later. I think the FAI rules for contest flying are pretty clear on this, any use of radio must end in DT.

 
Logged
rogermorrell
Silver Member
****

Kudos: 6
Offline Offline

United States United States

Posts: 100



Ignore
« Reply #128 on: March 22, 2017, 01:00:59 PM »

As one of the originators of this RDT stuff along with Ken Bauer I support Dick Ivers comments - I think the RDT Tx should send multiple pulses
"to make sure it gets through". I have tried different other "clever" combos of pulse sequences etc and with the simple radio protocols we use they are just not reliable.   To reliably have some more complex system you need to have a 2 way transmission between the two radios so they know the signal was successfully received.    Ken is a F1A flyer and at the time we did the original work most power flyers would not entertain using any form of electronics so the possible need to stop the motor then activate the after a delay D/T did not arise.

Because in my MAGIC timers I let the sportsman decide what actions to take on the receipt of the RDT signal he can decide what actions and sequences are appropriate for his particular model - shut off motor/engine, pop stabilizer and/or wing, move rudder etc possibly with delays.  I also agree with Dick that rules say the object is to terminate the flight of  the model i.e. D/T so that has to happen. 

In the early days of F1Q in the USA I observed some people in a FF competition using  a regular R/C system to cut the motor and later D/T the model. That is clearly outside of the rules.  You might want to use this during testing.

I have built timers that would let you use a regular R/C Rx as the RDT system - where for example using the throttle stick to do from full throttle to full slow would D/F the model, the timer making the decision based on the length of the pulse.  The ideas was to see RDT costs coul;d be reduced by using an inexpensive regular R/C set instead of the more specialized and expensive RDT. Nice theory but cheap, small R/C sets don't have the range required to RDT  a performance FF model.

Roger
Logged
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #129 on: April 17, 2017, 12:38:19 PM »

Let's return, from RDT, back to Puzzle. I have been lately toying the idea of trying a single-bladed prop. eCalc suggests that a 7x6 single-blader should produce slightly more thrust than 6x6. I would assume that the prop is more efficient, with the area being in one wider blade than two, and also the increased prop disp diameter should add to the efficiency. Also single-balder should be straightforward to make a folder.

So I took an APC prop, saw off one blade. Drilled a hole for the pivot pin, filed the root to fit in a folder hub. Made counter-weight holder from aluminum, and added some brass nuts for counter weight. I still would need to find a means to properly balance the setup (and most of all to figure out how to balance. I recall reading a paper in Aeromodeller suggesting that the counter-weight should be a bit lighter than the blade. But cannot find the paper.) I did some rotation tests, and the vibration is not bad, also the blade can hang on at 13000 rpm. But as you can see, my counter-weight holder is too weak, and starts to bend. Gives you some impression of the forces involved.... Also, the one blade of 7x6 weights over 5 grams, as much as the whole 6x6 prop. Too much, as I already have problems with the CG being too far forward. The 7" APC blade is quite wide and thick, so I suppose I could make the blade much lighter out of carbon fiber and foam/balsa core. But maybe I first try to find good balance for the blade and test if I can get more thrust out of it. Also if I make a mold I need to move the pivot point, with this one the blade does not fold past the motor, but remains sticking at 45 degrees...
Attached files Thumbnail(s):
Re: E-36 "Puzzle"
Re: E-36 "Puzzle"
Logged
JohnOSullivan
Silver Member
****

Kudos: 4
Online Online

Canada Canada

Posts: 178


WWW

Ignore
« Reply #130 on: April 17, 2017, 01:58:58 PM »

Might be worth using already usable Graupner CAM prop blades. eg 7.5"x4" at 3.2 gm per blade.  
Adding the counterweight is going to add considerably to the total prop weight.
Also consider moving the counterweight outwards on a wire as per the old Rubber model single blade Props.
Logged

John O'Sullivan
MAAC 5401
MACI 26
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #131 on: April 17, 2017, 11:55:14 PM »


Those CAM blades do not have enough pitch, I need that 5.5 to 6" for a DYS motor turning 13,000 rpm for enough forward speed to climb enough. With 4" pitch the motor should be turning 20,000 rpm, which would need a kV like 4000 I guess. Have not seen such motors.

 
Logged
che
Bronze Member
***

Kudos: 0
Offline Offline

United Kingdom United Kingdom

Posts: 69



Ignore
« Reply #132 on: April 18, 2017, 10:16:09 AM »

Tapio,

I've been using the 7x6 APC modified to a folder and suspect with a single blade you could go to an even larger diameter. One issue is that the balancing would only be correct at one rpm but if you're still intending to use a governor function then maybe that would work. Agree that you need less mass at a longer moment to keep the wight down.

The reason the aluminium bracket bent is quite obvious I'm afraid.

CHE
Logged
RayE
Bronze Member
***

Kudos: 2
Offline Offline

United Kingdom United Kingdom

Posts: 56



Ignore
« Reply #133 on: April 18, 2017, 07:35:00 PM »

Tapio,

I have been playing around, on and off, with the use of large diameter single bladed propellors for F1B and F1G. In doing this I came  across 3 papers which you may find of interest in relation to what you are doing;

- Design considerations for one bladed propellers   by Hewitt Phillips   ( NFFS Symposium 1971)

- Single blade propellers for F1K, why?  by R Hoebinger      ( Free Flight Quarterly  January 2004)

- F1B Enjoyment and the single bladed prop   by George Matherat and Jean Wanzenreiter     ( NFFS Symposium 2005 )

Phillips proposed that, compared to a 2 bladed prop, the diameter and pitch should be increased by 10% and the width by 75%.  I increased the diameter and pitch as suggested but only increased the blade width by a small amount. I have since produced a much wider blade but have not yet tried it.

Ray
 


Logged
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #134 on: April 27, 2017, 12:35:26 AM »

The mailman brought this yesterday.eCalc suggests 10 more Watts than my previous DYS motor. Maybe 180 meters on 10 second motor run with this one?
Attached files Thumbnail(s):
Re: E-36 "Puzzle"
Logged
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #135 on: September 12, 2017, 10:43:09 AM »


I have not attached that motor into a model. Yet. I got some doubts whether a Turnigy would be as efficient as DYS. Ah well.

It has been a quiet summer. Did not have a chance to attend Swedish competitions, and besides I heard that they were not too successful. The F1S Nordic championships flown in heavy winds, and Small Classes of Scandinavian FF week cancelled altogether. I have not been flying here in Finland either, as my local fields are so small. Good enough for F1H and P-30, but F1S would land in the woods... but now the harvest season is here, and bigger fields will shortly be available again.

Inspired by the discussion in Facebook, where I learned that Stan B uses surprisingly low power motor on his 200 meters climbing model, I started to think if straight climb would be more efficient than spiral? Maybe not totally, as cambered airfoils have their drag minimum at lift coefficient larger than zero, which means that they have lowest drag when they lift slightly. But naturally too much lift is bad. So maybe wide open spiral would be the way to go? From previous experience with Puzzle, I found that the tall pylon makes the model to roll ( -> spiral ) in climb, while shoulder wing mount results in straight climb (or in my case ot was straight loop, did not sort it all the way out as I wanted to make the model spiral). Thus maybe a not-so-tall pylon would mean less roll and more open spiral? That needs to be tried.

Thus, after a crash with my first model and a need to re-trim anyway, I modified one pylon to half height, and replaced the motor with the same DYS version that I have in the second model (climbing to 170 meters). Now I'm sitting tight waiting for the winds to calm down and the rain to end, so that the farmers can harvest their grain and then I get to test my new setup! :-)

(old pylon on bottom, new on top)
Attached files Thumbnail(s):
Re: E-36 "Puzzle"
Logged
DerekMc
Platinum Member
******

Kudos: 50
Offline Offline

United States United States

Posts: 1,269



Ignore
« Reply #136 on: September 12, 2017, 11:02:27 AM »


Inspired by the discussion in Facebook, where I learned that Stan B uses surprisingly low power motor on his 200 meters climbing model, I started to think if straight climb would be more efficient than spiral?


Last Feb. I watched Stan do a 10 sec. trim flight. The plane did a 1/2 turn during the climb and transitioned perfectly at the top. It was his 'safe' plane for the 10 sec. rounds.  I don't know for sure without altimeter data, but it looked as high as my plane gets on a 15 sec. run. Mine does the safe spiral.

I plan on trying a straighter climb on my next plane now that i have fast RDT :-)
« Last Edit: September 12, 2017, 11:21:19 AM by DerekMc » Logged

They fly better when you smile!
Derek
PeeTee
Titanium Member
*******

Kudos: 48
Offline Offline

United Kingdom United Kingdom

Posts: 2,159



Ignore
« Reply #137 on: September 12, 2017, 11:33:06 AM »

Derek

I thought the rules now had 10 secs for flights ( and 5 secs for flyoffs ) ??

Peter
Logged
DerekMc
Platinum Member
******

Kudos: 50
Offline Offline

United States United States

Posts: 1,269



Ignore
« Reply #138 on: September 12, 2017, 11:40:20 AM »

Derek

I thought the rules now had 10 secs for flights ( and 5 secs for flyoffs ) ??

Peter

Yes, no more 15 secs runs. I mentioned the 15 sec. altitude for comparisons sake.

 
Logged

They fly better when you smile!
Derek
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #139 on: October 01, 2017, 05:45:24 AM »

After some crashes and repairs, I have had two days to trim my Puzzle models. First on Thursday the older, red-yellow one with the lower pylon, and yesterday the red-blue one, which has the same layout as before, but I replaced the 6x5.5 prop with 6x6. And made some increase in altitude, the previous prop gave 160 meters in March, and now the higher pitch add the 10 second climb altitude to 174 meters.

Trimming was also educational. I had forgotten the main thing, you trim the climb pattern with decalage. On Thursday that came to my mind again, and remembering that, trimming was quite straightforward again.

With lower pylon, the yellow model indeed rolls less in the climb, so it seems to confirm my previous conclusion that pylon is needed to make the model roll. The yellow model makes maybe 1 circle per 10 second climb. Yhis makes the pull-put rather critical, and the model quite easily stops nose-high and stalls. With minimal decalage the stall easily loses a lot of altitude. The blue model, on the other hand, turns tight, maybe 3 full turn in the 10 second climb. This seems to help keep the plane banked to the right, so it slips sideways on motor stop and settles to glide easily.

But, when I put the models together to measure decalage and CoG, I made an interesting observation. On the blue model, the pylon is slightly asymmetrical, so that it sets the wing inclined, right wing tip forward of the left. This induces effective extra incidence to the right wing, and less to the left tip. And indeed, the pylon of the yellow model is asymmetrical in the other way. Maybe, indeed, the roll is not caused by pylon height, but wing warps? I think the next thing now is to modify the low pylon for the yellow model for other direction tilt, and test fly the model again. Luckily that is easy to do, as the pylon is attached with two bolt to the main fuselage section.

Another interesting thing is that the DYS motor on the blue model makes a loud rattling noise, and shakes when spooling up. Once model gets to full speed the rattling and shaking disappears. I thought that this might be due to ESC and motor mismatch. On some occasions with improper advance, electric motor may be quite shaky on partial throttle. But now that I have installed similar DYS motor also to the yellow model, it spools up smoothly. So maybe I have a bearing issue on the former motor? It does not seem to affect the power, though.

Measured the models. declage 0.9 degrees between the bottom tangents of the wing and tail. As wing LE radius is larger, the actual decalage maybe is a it larger. CoG 98 to 100 mm from leading ege of the wing, 70 to 71% of the root rib.

The lines show three consecutive test flights I always decreased the decalage a little for the next flight, which increased the climb altitude.
Attached files Thumbnail(s):
Re: E-36 "Puzzle"
Logged
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #140 on: October 12, 2017, 09:18:10 AM »

Yet another calm morning (before the rain came), so a good chance to test some new things.

First I had noticed that the red-yellow model, with half-height pylon, maybe had the wing rotated the right tip aft a bit. Which might explain the lack of roll in the climb. So I adjusted the pylon position a bit, moving right tip forward. Result: no impact on the climb roll, which still is much more open than with the red-blue model. But the glide turn (to left) seemed to tighten a lot. So I still conclude that the higher pylon is needed for the rolling climb, and slight wing rotation is the reason that encourages the model to glide to the left. I'll replace the high pylon again to the yellow model. The blue one still climbs in tightly rolling pattern, but it climbs so well that there is no need to try to remove the roll. Helps a lot on the transition.

The other thing I tested was different batteries. 370mAh which I have used since summer 2016. 450mAh which is a more recent one. And most recently 500mAh High Voltage battery (can be charged to 4.35V per cell, but my Orbit charger only goes to 4.3 volts). All batteries are labeled 65-130C. The 370 gave 170 meters altitude. 450 climbed to 174m as it did two weeks ago. The HV gave 183 meters of altitude, with a nice pattern and perfect transition. Looks like the battery can maintain power better to the end of the climb.
Attached files Thumbnail(s):
Re: E-36 "Puzzle"
Logged
che
Bronze Member
***

Kudos: 0
Offline Offline

United Kingdom United Kingdom

Posts: 69



Ignore
« Reply #141 on: October 13, 2017, 07:13:40 AM »

Very interesting results Tapio which have encouraged me to try some of the Hv batteries as well.

The question I wanted to ask was, if you are using the ESC governor function the same on all the batteries, why do you get higher with one particular battery type ? Is it simply that the RPM setting is too high to allow some/all batteries to deliver the set RPM throughout then run, or something else ?

CHE
Logged
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #142 on: October 13, 2017, 07:25:38 AM »

Just as you say. My target RPM is higher than the motor can turn the 6x6 prop, thus in practice it is running full power all the time.

Looks like with a tight turning trim, it is not necessary to keep constant RPM, but fading power is no problem. On the other hand, the HV battery that has highest capacity and thus probably maintains voltage (-> RPM) best, had the best transition. All the other flights stalled at least some while settling to glide, but the HV showed a glider-like "zoom" transition straight into glide. The model trim is such that if the battery fades, the climb turns less banked and the motor cut results in a stall. At higher power, even when banked, the climb speed is not compromised!
Logged
rivers
Bronze Member
***

Kudos: 1
Offline Offline

United States United States

Posts: 38



Ignore
« Reply #143 on: October 14, 2017, 11:33:59 AM »

Tapio,
I love your altitude charts with the superposed graphs with different colors and line styles.

Are you sure your 370 mah 65C battery is not a 350 mah 65C from Hobby King?

Have you looked at the new Hyperion high voltage batteries? They have a 600 mah HV about the same weight as the HK Bolt 500 HV.
Hyperion G7 2S 600mAh Si-Graphene HvLi 60Cmax (4.35V)

Dick

Logged
Tapio Linkosalo
Platinum Member
******

Kudos: 27
Offline Offline

Finland Finland

Posts: 1,086

Topic starter


Ignore
« Reply #144 on: October 16, 2017, 03:14:16 AM »

Correct, it is 350 not 370. And no, I have not seen those Hyperion batteries. I shall take a look. Thx!

Logged
Pages: 1 2 3 4 5 [6]   Go Up
Print
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!