rrTV-PHOTO   New HD TV
HOME   rrTV-PHOTO   GALLERIES   MY GALLERY   HELP-FAQ
myHOME PM pmRR MEMBERS 722 ONLINE 42 EVENTS SEARCH REGISTER  START HERE
 
6 pages [ <<    <    ( 1 )     2      3     NEXT    >> ]14928 viewsPOST REPLY
Autography FlightPower . Advantage Hobby . Revolution Models

.
.
Radio - Servo - Gyro - Gov - Batt > TX/RX eCCPM Output Test Results
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

Hello,
In this thread: 9303-eCCPM-Channel 6 Behind/Lagging - Confirmed, I mentioned making a movie of how the swash plate behaves on each 13 radio setups (found on page 2 about half way down). I decided I could take collects of the pulse outputs and provide those much faster than filming and posting videos.

I have collected data for five of the setups thus far and will present them here. The data was collected while flipping the flight mode switch between a mode with the pitch curve set at 0 and a mode with the curve set 100 (-100 and 100 on the 14MZ). This presents an instantaneous desired change of pulse widths.

9) Futaba 14MZ with G3 PCM rx with same group channels
This is the best behaved of all the setups. All three channels (4, 5, and 6) have output pulses at the same time and change completely in one frame.


10) Futaba 14MZ with G3 PCM rx with "old" Futaba channels
This test uses the G3 receiver but using the "old" channel assignments of 1, 2, and 6. As can be seen in the graph, this would explain why those who have tried this configuration are disappointed with the eCCPM performance.


11) Futaba 14MZ with R149DP PCM rx with sequential channels
This and the next setup show a rather strange behaviour. It takes five full frames for all three channels to become the new pulse width. It will be very interesting to see if a similar result is obtained from a 9C and 9C Super.


12) Futaba 14MZ with R149DP PCM rx with "old" Futaba channels
Again, a strange result which shows it taking five frames to transition from the old pulse width to the new pulse width. Only the three "middle" frames are shown in this graph.


13) Airtronics Stylus with PCM rx
The Stylus is well behaved visually although these results show that channel 6 changes before channels 2 and 3. Notice also that the Stylus with the PCM receiver shows the fastest servo update rate of these five data collects.


I will add data as I collect it. For each of the above and future collects, data is taken several times to test for consistent behaviour.

- John

Protos -- Logo 14 -- Logo 10
04-10-2005 Over year old.
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

I looked at the data stream from the trainer/DSC port of the Stylus and the 14MZ with the radios in FM modulation mode. The Stylus and 14MZ data is extremely consistent from frame to frame and the same transition as tested above occurs in one frame.

I'll test the FM modulaton output of the other radios (9C, 9CS, and 9303) as well.

- John

Protos -- Logo 14 -- Logo 10
04-10-2005 Over year old.
 
 
I3DM
rrProfessor
Location: Israel

Interesting... the Stylus really excels in that field !

Team pilot for Radix, Helidirect, Logictech, ThunderPower, Fromeco
www.liorzahavi.com
04-10-2005 Over year old.
HOMEPAGE  
 
 
TMoore
rrProfessor
Location: Cookeville, TN

Not bad for a 10 year old radio, eh?

TM
04-10-2005 Over year old.
 
 
I3DM
rrProfessor
Location: Israel

my thoughts exactly..

when they do finally come out with a newer Sanwa radio - its going to be something else !

Team pilot for Radix, Helidirect, Logictech, ThunderPower, Fromeco
www.liorzahavi.com
04-10-2005 Over year old.
HOMEPAGE  
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

More test results. I got my hands on a 9C and an FM receiver.

5) Futaba 9C with PCM rx
Same as with the 14MZ with the PCM 1024 receiver, it takes four or five frames to transition from one pulse width to the other.


Futaba 9C with FM rx
Full transition in one frame.


Futaba 14MZ with FM rx
Full transition in one frame. Functions could also be assigned to put the three swash channels on sequential channels.


From the tests on the 9C and 14MZ it appears better to use FM then PCM 1024 for eCCPM purposes. Does anyone know why it takes four or five frames for the transition to occur with PCM 1024?

I did figure out why I was getting fishy results when looking at the DSC port output of the 14MZ in PPM mode. It was a signal level problem going into the logic analyzer. This became apparent when I could not get any recorded signal with the 9C. Upon looking at it with a scope, I realized I needed to do a bit of level shifting to get good collects.

- John

Protos -- Logo 14 -- Logo 10
04-11-2005 Over year old.
 
 
TMoore
rrProfessor
Location: Cookeville, TN

Winner and still Champion.

ATX Stylus, the best 10 year old radio on the market, bar none.

TM
04-11-2005 Over year old.
 
 
TMoore
rrProfessor
Location: Cookeville, TN

Shameless Bump
04-12-2005 Over year old.
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

> Shameless bump for an MC-24 and that's not even mentioned.

If I had one to test, I would. Gregory, perhaps you could do the test? PM me and I'll tell you how I did it and what software I used. The software is free and all you need is a parallel cable.

Actually, the same offer goes for anyone with a radio I haven't listed that would like to test it.

I'll try to get the 9303 and 9C Super testing done within in a week. If I can get my hands on a 10X, I will do that too. A 9ZWCII would be nice too.

- John

Protos -- Logo 14 -- Logo 10
04-12-2005 Over year old.
 
 
Andi G
Veteran
Location: Switzerland

Hi,

the mc24 needs a complete frametime to compute the next frame, so you'll have about 50ms delay (10ms average, 20ms computing, 20ms transmitting).
mc22 is way faster...

Your test method won't show this delay, you should also measure the point where you flicked the switch - but that would require even more soldering work.

I think your test shows the limitations of Futaba PCM, but I doubt this is a huge problem as the servos can't move that fast anyway.

The Stylus isn't available on 35MHz is it?

I'd love to see the 10X heli results as everybody here says how exact the ccpm is (no cyclic->pitch interaction and so on).

Actually there are 3 important points
-delay from reading in the sticks until transmitting the next frame (usually 14-20ms)
-delay until the complete values are transmitted (pcm1024 lacks here, but it should look better when you move the stick slower... I don't have collective on a switch hehe)
-the mixing itself, eg does the swash dance up and down with cyclics only?

Andi
04-12-2005 Over year old.
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

Andi,
I would LOVE to measure the total latency of each setup. I would want to do it such that the input looks like a stick movement. I have a good idea of how to do it and be able to measure both the minimum and maximum latency but it requires opening each radio. I'm not sure I want to open up a 14MZ.

> but I doubt this is a huge problem as the servos can't move that fast
> anyway.

Actually, I would have to disagree a bit. Looking at the faster servos we have available now, the transition time of almost 60 ms is about as long as the servos take to move for a full collective change. It is actually longer than some servos would take (such as the KO PROPO PDS-2123).


- John

Protos -- Logo 14 -- Logo 10
04-13-2005 Over year old.
 
 
bob00
Veteran
Location: Toronto, ON

Excellent info

Very nice work John.

Any chance we can see how the 10x and 8103 weigh in? I'm real curious to see how the 10x stands up.

Thanks for the info,
- Rob
04-13-2005 Over year old.
HOMEPAGE  
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

> Any chance we can see how the 10x and 8103 weigh in?

I know someone who flies a 10X, but I don't see him vary often. I'll try, but I can't promise how soon.

Someone has volunteered to test their 9ZHP WC2 and 9C Super.

- John

Protos -- Logo 14 -- Logo 10
04-13-2005 Over year old.
 
 
Andi G
Veteran
Location: Switzerland

John,

you're right about the servo speeds... my fault. On the other hand digital servos should start to run with their full power when the first different impulse comes in and not when all pulses are right



and looking at this it explains why one channel lags. I just bought my first eccpm heli and I'm tempted to try a fm receiver (9C radio).

Regards,
Andi
04-13-2005 Over year old.
 
 
Papa
Heliman
Location: Hungary

Hi,
JKos,Andi G
Witch kind of software do you use for the data collection?
R
04-13-2005 Over year old.
 
 
I3DM
rrProfessor
Location: Israel

Andi G, i have a 35Mhz model \ Rx for my stylus.

Team pilot for Radix, Helidirect, Logictech, ThunderPower, Fromeco
www.liorzahavi.com
04-13-2005 Over year old.
HOMEPAGE  
 
 
Tim B
Senior Heliman
Location: Pasco, Washington

What, no tests results from the 9303 that you started all this with?
04-14-2005 Over year old.
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

Tim,
I know, I know. It is hard to test a radio I don't have anymore.

Someone else has volunteered to test the whole JR line up.

- John

Protos -- Logo 14 -- Logo 10
04-14-2005 Over year old.
 
 
JKos
Elite Veteran
Location: Town of California in the state of Maryland

Gregory,
Above you made the statement:
> Clearly shows the un-sync (no not the group) of the channels on the
> 14MZ. Very strange ...

Could you please explain what you mean? Which scenario are you referring to? The 14MZ with the G3 receiver and same group channels is dead on synched, so I don't think it is that one you are referring to.

Thanks,
John

Protos -- Logo 14 -- Logo 10
04-20-2005 Over year old.
 
 
I3DM
rrProfessor
Location: Israel

If anyone needs 35Mhz for Stylus:

http://www.runryder.com/helicopter/t173720p1/

Team pilot for Radix, Helidirect, Logictech, ThunderPower, Fromeco
www.liorzahavi.com
04-21-2005 Over year old.
HOMEPAGE  
 
 
6 pages [ <<    <    ( 1 )     2      3     NEXT    >> ]14928 viewsPOST REPLY
CarbonXtreme . Midland Helicopters . HeliProz

.
.
Radio - Servo - Gyro - Gov - Batt > TX/RX eCCPM Output Test Results
  UPDATE SCREEN   PRINT TOPIC Advertisers 

Subscribe to This Topic

Thursday, July 24 - 4:21 am - Copyright © 2000 - 2008 runryder.com | email | link to rr | runryder needs cookie