Aeroace - Low Speed Throttle Modification

[Martyn McKinney]

My Aeroace even with one wing will climb at the lowest throttle setting.

For this reason I wanted to reduce the throttle at low speed settings even further.

Some have put resistors or diodes in series with the motors, but this is inefficient and also reduces full throttle power.

By using a 555 timer and pulsing the throttle to set it to zero with a 50% duty cycle it is possible to fly even slower.

The circuit below pulses at approximately 2 times per second.

The 2N3904 transistor disables the circuit at the higher throttle settings by means of a signal from P12 on the throttle stick.

It is not necessary to cut any traces. The only modification to the board is to remove the green wire from P10 and the blue wire from P11 and add the two 10K resistors. Wires are then soldered from the circuit to the appropriate places on the board.

I made the circuit "Manhattan style" and just placed it inside the case near the throttle stick.

It took me approximately 2 hours from start to finish.

The 2 diode zero throttle turning mod is no longer connected to P11, but is connected to the blue wire coming from the throttle.

I haven't flown it yet, but I would be grateful for any feedback from those adventuresome souls who might try it.

[iter]

I haven't tried it with the AA yet, but I tried to do low frequency throttle PWM from the transmitter on a WAMF. (had it driven from a parallel port, too). WAMF uses the TX2 chip, and that won't do more than about 3Hz. What I and RCcam (http://members.aol.com/rchelicam/microszr/microszr.htm) found out was that it's too slow for smooth flight. The plane has enough time between pulses to drop the nose, and it flies in series of violent PIO (PWM Induced Oscilations). But the AA may be different. Let us know how it works out.

Ari.

[widgeteer]

Wow, you people rule. I can almost understand what you say. (Stick around this site long enough, you learn lots of stuff). I gotta try that throttle thing, seems that could help to make this bird into a real 'L R F' , just takes time to put the pieces together. Gimme more, please.

[caseyjholmes]

Nice "circuit bend" without harming whats there already. Good job ;)
Casey

[WhiteHot]

I am sure the answer is somewhere in the thousands of posts about the AA, but what is the update rate of radio?

[quax]

Iter is right, I think. We (RolfPW + I) made some tests and first you should synchronize the toggling with the tx-frames and second the standard frames per second are is to slow for smooth control.
With 10fps (= 5 toggles per second) it starts to become a good feeling.

But its worth a try, everything concernig the AA is worth a try :)

Quax

[Martyn McKinney]

Quax:

When you designed your system did you find a microcontroller output on the stock transmitter that could be used for synchronizing on the transmit frames ?

If not, it might be possible to create one by lowpassing the output of the microcontroller driving Q4 through R17.

Whitehot:

Channel A 4 frames per second
Channel B 6 frames per second
Channel C 5 frames per second

http://home.versanet.de/~b-konze/xtwin/info/info_en.htm

[loc nar]

If you want to fly slower, wouldn't it be easier to just turn the props around backwards?...or trim a little off of them....or bend some of the pitch out of them?

[Martyn McKinney]

I tried putting the props on backwards but there wasn't enough thrust.

If they are trimmed you lose full throttle performance.

Others have suggested adding weight, but I want the plane to fly slower, not faster.

[quax]

Martyn McKinney,

someone shortened the props and made them a little bit smaller the same time. Maybe the best solution to lower thrust and being stingy with energy ;) .

If not, it might be possible to create one by lowpassing the output of the microcontroller driving Q4 through R17.
Should work, I would toggle a flip/flop with the falling edge, means the end of each frame. And then using channel B should go quite well.

Quax

[Martyn McKinney]

Quax:

Thanks for your input.

After playing with this for a while and reviewing your comments, I'm going to go back to the drawing board.

I realize now that to ensure reliability a 50% duty cycle will require toggling on alternate frames which is going to require some type of clocking flipflop.

I had qualms about posting this prematurely, but I wanted to get some feedback which both you and Iter provided. Thanks again.

I'll be back !

[Martyn McKinney]

Here is the circuit that I came up with to reduce the throttle by setting it to zero by synchronizing the pulsing with the transmitter frame rate.

Although the original 555 timer circuit worked, it was not consistant.

The circuit below turns the throttle off at the lower throttle settings with a duty cycle of 50%.

The pulses used to modulate the transmitter, trigger a D flip-flop, the output of which turns off the throttle.

If there were an odd number of transitions in each frame, only one flip-flop would be necessary. There are an even number, however.

Adding the second flip-flop gives an output which is high for 50% of the time.

At the higher throttle settings, this circuit is inhibited by a signal from the throttle stick P12 (violet wire) which turns on the transistor, thus setting the flip-flops to a high state.

The 2 diode zero throttle turning mod remains at P11.

Because the frame rate of this transmitter on Channel A is quite low, the motor is turned on only twice per second. I would expect to see an improvement in performance if a Channel B transmitter were used.

The only modification to the board is to remove the yellow COMMON wire from the throttle encoder. This yellow wire is then used to provide power to the circuit.

The COMMON on the throttle encoder (where the yellow wire was removed) is connected to the diode coming from the output of the second flip-flop.

By ORing other diodes from the output of the flip-flops, other duty cycles are possible.

This might require deriving a different clock by using an extra transistor, however.

Purists might note that I have failed to include pull up resistors on some of the inputs. Because this is a TTL device the inputs of which use current sinking and simplicity was a factor, I have chosen to leave them out.

If a CMOS device were used they would be required.

EDIT:

I just flew both my stock Aeroace and my extended wing Aeroace with this mod and it works quite well even with a Channel A system. No sign of porpoising even at the 2Hz pulsing rate.

At low throttle it just barely holds altitude and is over-ridden by the turning mod when turns are made. Previously I had problems flying at any throttle setting because the plane would gain too much altitude.

It might be worthwhile trying 75% duty cycle, which would require an extra flip-flop and diode or a different clock and extra diode.

[sideburn]

couldnt you use a transistor to turn off the zero throttle diode mod when you are NOT at zero throttle? Then you wouldnt effect the turning speeds on all 7 throttle settings.

couldnt you have p12 trigger a transistor to disconnect the diodes?

[Martyn McKinney]

Yes, but the circuits that I came up with were more complicated than the 2 diode mod and the benefit didn't seem worth it.

The motor speeds used for turning are determined by the software in the transmitter encoder and it was difficult to notice any difference between turning with the diode mod and without it.

[sideburn]

Yes, but the circuits that I came up with were more complicated than the 2 diode mod and the benefit didn't seem worth it.

The motor speeds used for turning are determined by the software in the receiver decoder and it was difficult to notice any difference between turning with the diode mod and without it.

yeah, I understand the new low throttle circuit.. I also did the diode mod and like it.. I was just kinda concerned that it was effecting the turning speeds at diffferent throttle settings since that p11 bit is always on when turning now.. but if it does make much difference than no since in making it more complicated..

Did this low throttle mod turn out to be worthwile? I guess you dropped the 555 version.

[Martyn McKinney]

Did this low throttle mod turn out to be worthwile? I guess you dropped the 555 version.
I like the low throttle modification. It makes it easier to keep the plane out of trees, I can grease it in for landings or catch it more easily.

The 555 timer mod worked, but it wasn't synchronous with the transmitter.
This meant that it would pulse at 2 Hz whether on Channel A, B or C.

Because it is synchronous, the flip-flop version will increase the pulse rate to match the transmitter channel with hopefully improved performance.

I was hoping to get some feedback from some indoor flyers who might try it in order to determine the optimum duty cycle or to find out whether other improvements could be made.

The mod is very simple to do, it took me less than 1/2 hour by building it dead bug style and placing it in the left hand side of the case near the throttle encoder.

Don't forget (it's not shown on the schematic), Pin 14 is Vdd (yellow wire), Pin 7 is ground. (for ground I used the black wire at the top end of the transmitter battery compartment)

[Hotflyer]

martyn-

You have NO IDEA how bad I want to do this mod! After doing the diod mod, I began researching other electronic mods for other toys I have. I have opened up my xmod tx to remove r20, a resister that limits the range of the TX, I have upgraded the fets on a bitcar (finally!) and am eager to do more! Just one thing tho..... I do not know EXACTLY what to do and at this point, I love my little plane too much to start doing something to it that may render it grounded! I know you have already made a schematic and explaned what needs to happen in the end BUT: Could you please please please make a step by step and materials needed? The diod mod was self explanitory because of the mere three steps needed but this one seems a bit more in depth :o

I will happily do it, I just need to feel like I am entering the project with a clear vision of the steps needed to reach success. :)
Help a bruthu out!

sam

ps. I have an A channnel so we could see how the flip flop works on it too

[Martyn McKinney]

Glad to see someone enthusiastic about this !

I hope this helps.

Parts needed:

74LS74 Dual D Flip-Flop
2N3904 transistor
100K resistor
1N914 diode

1. Build the circuit dead bug style, joining the pins together as shown on the schematic. With the chip upside down, the pins are numbered clockwise from 1 to 14. The 2N3904 transistor is soldered to the appropriate pins on the chip.
(Collector to pins 4 and 10 of the chip, emitter to ground)

2. Unsolder the yellow wire from the bottom of the throttle encoder.

3. Connect this yellow wire to pin 14 of the chip. (It provides 5V power)

4. Connect Pin 7 (GND) of the chip to the black wire on the battery compartment

5. Connect the cathode of the 1N914 diode from pin 9 of the second flip-flop to the bottom terminal of the throttle encoder (where the yellow wire was removed - COMMON)

6. Connect the clock wire from the first flip-flop (pin 3) to the junction of R16 and R17 on the main board.

7. Connect the 100K resistor from the base of the 2N3904 transistor to P12 (Violet wire at the top of the throttle encoder)

[sideburn]

this is off topic but i figured you would know the answer...

Do NIMH batterys (as in rechargable AA Energizers for the TX) need to be fully discharged before recharging them? I seem to only get about 5-10 lipo charges out of the TX before the green light just stays on and I have to recharge my TX batteries.. but they aren't very low at that point.

[Martyn McKinney]

The recharging problem you are experiencing is likely due to the fact that the voltage of NIMH cells is only 1.25V per cell, not the 1.5V of alkaline cells.

On my transmitter the green light will not turn on if the batteries are too low.

The charger in the transmitter has the provison for lowering the endpoint charge voltage of the receiver battery.

If both jumpers P1 and P2 are shorted, the charger will shut off at a higher voltage than if they are both open.

The voltage with them both open is approximately 8 X Vbe -0.6 or 8 X 0.6-0.6= 4.2V. (there is a diode in series with the battery in the receiver when charging).

Circuit tolerances may alter this, however. Shorting one or both of the jumpers will raise the endpoint voltage if it is too low.

If the green light on your transmitter doesn't turn off, that means that the lipoly cell in the receiver is not reaching the endpoint voltage (nominally 4.2V).

This is most likely due to the fact that you are using NIMH cells.

Opening P1 and/or P2 (if they are already shorted), would lower the voltage at which the green light would turn off.

It would be necessary to measure whether this is the proper 4.2V.

[sideburn]

yeah I wouldnt want to risk it not fully charging the lipo. What about the rechargable AA's? is it bad to not let them completely discharge before recharging them?

[micro_builder]

hi Martyn,
i had only skimmed through this thread in the past, as the lowest throttle setting always seemed to work ok for me, especially with the diode mod for steering. however, now that i'm working on trying to make the AA into an LRF, i'm finding that even on low throttle, the plane is still faster than it needs to be.

so, i just re-read this thread, and am getting pretty interested, thanks for sharing it with us. it may be one of the last things i need to really get this thing going :) i've got the B channel, and i'll start gathering parts/equipment for this mod on friday probably, and hope to have it done by the end of the weekend. this could prove to be a very handy mod.

nick

[Martyn McKinney]

What about the rechargable AA's? is it bad to not let them completely discharge before recharging them?
Depends on how you charge them.

Safest and best way is to trickle them at 0.1 C. They may be left in this state indefinitely and they will be balanced.

If they are being charged with a delta-peak charger and one or more cells is lower than the others, the pack may not fully charge.

All in all, I don't think that it is something to worry about.

[Martyn McKinney]

micro_builder:

Looking forward to hearing of your progress.

I was interested in receiving some feedback from indoor flyers because there are a number of changes that could be implemented relatively easily to this circuit.

These changes range from altering the duty cycle from 50 to 75 percent or pulsing the high end of the throttle range.

Whether or not these are worthwhile would be dependent on information from those flying with this modification.

[dangitman]

These changes range from altering the duty cycle from 50 to 75 percent or pulsing the high end of the throttle range.

I'm building your low throttle mod, and the two diode mod on my Tx today. Looks pretty straightforward. I think the 75 percent duty cycle would be a nice idea. Would it be possible to make it switchable from 50% to 75%?

Thanks for your work!

[Martyn McKinney]

Yes it could be switchable.

Because in the future I wanted to have the possibility of going to 75%, I made a mod to the mod.

At the present time I have the 2 flip-flop circuit connected to a pushbutton switch on my transmitter. When I want to go into low throttle mode, I push the switch, turning the pulsing circuit on. When I let go, it returns to normal.

I may try a toggle switch and have it set up similar to the dual rate switches on most transmitters.

[dangitman]

Cheers Martyn,

I did both the 2-diode steering mod, and the low-throttle mod. Both were easy, and my Tx works perfectly afterwards - even though I didn't fork out the cash for a small soldering pencil, and just used my chunky old 25 watter.

it's a bit too windy to fly at the moment, a few test flights in my backyard didn't amount to much in the gusty conditions. But the mods work great.

[dangitman]

P.S:

At the present time I have the 2 flip-flop circuit connected to a pushbutton switch on my transmitter. When I want to go into low throttle mode, I push the switch, turning the pulsing circuit on. When I let go, it returns to normal.

What does this button switch? Does it just take the power (yellow wire) away?

[Martyn McKinney]

I removed the high throttle disabling connection from the transistor collector to pins 4 and 10 (flip-flop presets).

I then connected a normally closed pushbutton between the flip-flop presets (pins 4 and 10) to ground.

When not pushed, the flip-flops are preset high and the transmitter behaves normally.

When pushed, the flip-flops are no longer preset and they pulse the throttle, not only through the low range, but the high range as well.

[Bohh33]

Hi Martyn,

I'd really be interested in doing this mod too, but would like to do it where I could turn it on or off as desired like you described.

Do you happen to have updated schematics showing the mod with the switching ability?

Thanks a lot!

Mike

[Martyn McKinney]

Schematic below:

[dangitman]

Wow, this thing is great.

I think the low throttle mod is helping me get longer flights (as is the ability to turn without throttle input)

I wish I had someone to operate a video camera this morning. Clear blue skies, and only a hint of a breeze. The ideal flying conditions. Flew for about 15 minutes before I had to leave for an appointment - the batteries were still going strong! And it's so much smoother to handle with the low-speed mod.

Oh, and Nitro, if you are reading this - you HAVE to try the biplane. I'm very disappointed in my monoplane. The biplane flies brilliantly with either two or one wing. With a single-wing biplane, it just loves wind, can fly in surprisingly windy conditions, with the battery pack moved forwards depending on wind speed. I just cut a slot in the fuselage in front of the wing - I can then either push the battery inside the body to the original position, or further forward.

I also think the bipe looks better - it's funkier. Don't like that huge wing on the monoplane. Most fun I've ever had with something from a toy store.

[Bohh33]

Thanks a ton Martyn! That makes things a lot easier for me.

[amitysanimal]

My AA drops like a rock when I drop the throttle back to 0. Can you guys fly yours with 0 power? I have a thumbtack in the nose to stop the porpoising, is that too heavy? I understand the diode mod is so you can turn while you are at 0 throttle, but mine only stays in the air about 1 second at 0 throttle. What could be my trouble?

[crcasey]

Hi all, I did the zero power turning mod and the switched version of the low power turning mod today, along with adding a couple of DIP switches for A/B/C channel select.

I have the Aeroace running a mono/extended upper wing configuration that just was a pain to handle indoors. You know allways climbing, not wanting to turn... Well this is just a beauty to fly indoors now. It holds a nice level flight on channel C with pulsed power. If I want to start climbing out I just flip the small toggle switch back on and I am back at standard level one power.

With the upper wing extended you realy start to get more of a glide out of this little plane, so the zero throttle diodes are a true NEED if you fly this model as a monoplane, less so if you use an upper gull wing extension with the full size lower wing attached for drag.

You may ask yourself where I got the parts to extend the upper wing surface and still have the lower wing to pop off and back on? Well my first AA is now donating it electronics to a homebuild that I have in progress, so the lower wings from the first AA have found themselves hot glued to the tips of the upper wings of the second AA.

As for the 'gull wing' shape, that is just the way it happened to mate up. I have found that you can add or remove a bit of the dihederal but with little effect. I have to turn the motors inward a bit (inward being pointed more to the tail to me) to get more authority on turning. In the mono wing config it gets at least a 6 to 1 glide ratio with a dime in the nose as ballast. Battery and all other electronics are in the stock positions as I have not had this plane open yet. Lastly all tail feathers are at netural, no up or down trim required to remove that nasty stall bob.

Thanks for the great hacks! Happy flying all.

P.S. We are considering doing a bit of indoor pylon racing with these at Mikes over in north Dallas. Come all!

[Martyn McKinney]

Glad to see that some have tried this out.

I presently have mine on a pushbutton switch, but I think it would be better if I used a small toggle switch.

[JRuggiero]

Martyn McKinney,

I'm too lazy to think this through, but would it be possible, instead of adding a toggle or pushbutton switch to actuate the low throttle mod, add a switch to the throttle arm, so that when the throttle arm is pulled down the low throttle mod kicks in?

Or maybe you could do a modification to add just one more "step" of low throttle slower than what's available in stock configuration, and actuated by a position on the throttle arm just before the cutoff position? Having to push a switch with the left thumb after bringing the throttle down (?) to get the new low speed while attempting to steer with my right thumb is more than my middle-aged brain can process in the fraction of a second it takes before crashing my Aero Ace.

Jim R

[JRuggiero]

Martyn,

Another question: Your low throttle mod in Post 12 shows a 2N3904, apparently taking a pulse signal from P12 and using this to clock the 7474 Pins 4 & 10 to ground. But in Post 31 the 2N3904 is replaced by a pushbutton connecting Pins 4 & 10 to ground with no clocking from P12. Is that correct? And does use of the NC switch give, in effect, dual rates to the throttle? Or does the NC switch give only one, low, throttle setting, regardless of throttle arm position?

I'm so confused...

Jim R

[Martyn McKinney]

Post #12 contains a schematic and description of how to do this.

I have tried both and my preference is the latest circuit with a toggle switch.

Perhaps the best combination would be the circuit in Post #12 combined with a toggle switch so that it might be disabled.

[Martyn McKinney]

Oops ! Simultaneous post.

The transistor in Post #12 is not a clock, but it forces the outputs of the flip-flops high (which effectively disables the circuit) when the throttle is in the high positions.

When the throttle is in the low positions, the flip-flops are allowed to toggle.

The actual clock comes from a different portion of the circuit and is applied to Pin 3.

[JRuggiero]

Martyn,

I'll have to try out these mods to really see how they work. Thanks for your help.

Jim R

[rodair]

Martin,
I order the parts here and some more, looks like I found a hobby within the hobby.
you're one of the best on how to explain things!
Thanks.
Roger B.

[crcasey]

You are right Martin, the toggle is the way to go. I have mine drilled into the top of case above the throttle stick, your hand can be there to drop to half power at any time or throttle level. The dip switches for the channel select are located on the base of the case below the throttle.

One note: on the channel select solder pads be carful with the heat, my B pad started lifting off the PCB. I still tagged the wire to it and used hot glue to releve strain and keep all the wires in place. The DIP switch is also held in place with hot glue. Gotta love the stuff, almost as good as duct tape.
-Cecil

[didi]

Martyn,

I just implemented your low throttle mod with two minor modifications.
1. I used an HCT7474 for lower power consumption
2. I used an 5,1 kOhm resistor to pull down inputs 4,10 and a standard microswitch from an old CDROM drive connected with +5 Volt to toggle

Just flew it today with my brand new AA sport (at least that's the name over here in Germany). With the diode mod, it performed admirably. To increase pitch stability, I increased the surface of the horizontal stabilizer by two pieces of 3 mm Depron size 1 x 6 cm.

This is my 3rd X-Twin/ AeroAce and the first time its flying like an airplane, not waggling around like a toy.

Thanks so much for your contributions, which made this possible.

Didi

[Martyn McKinney]

Glad to hear of your success. I chose the 74LS74 because it is TTL logic which works by current sinking and in most cases wouldn't need pullup resistors.

I noticed that you used the word pulldown. In reading your post that means that you have your switch wired to positive as opposed to ground as in the original circuit.

With the HCT7474 as well as the 74LS74, it would also be possible to use a resistor as a pullup and connect the switch to ground.

What did you do with pins 1 and 13 which in the case of the HCT7474 would also need to be pulled up ?

[didi]

@Martyn

You are right with the pull up on pins 4,10. However i don't understand your suggestion to use a pullup rsistor and connect the switch to GND. If you use a closing switch, as I did, this shoud not work!

Surprisingly it works with the HCT 7474 without pullups on pins 1 and 13. To be on the safe side, both pins should be connected directly to Vcc (only possible with HCT74). By the way, also for the LS-version 1,2 kOhm pullups are recommended. But that's theory, in practice it works without pullups, maybe because the circuit is so simple and operating at low frequencies.

Whatever...

Enjoy flying your X-Twins,AeroAces and more

Didi

[Martyn McKinney]

In the schematic in post #31, the switch is connected between pins 4 and 10 and ground.

With TTL logic, when the switch is open it is logic 1 on these pins. Although good design would also dictate the necessity for a pullup resistor, one is not needed with TTL because there is no current sink on these pins when the switch is open.

When the switch is closed, with TTL, the switch provides current sinking to ground which is logic zero, thus setting the flip-flops to the high state.

Your circuit works the opposite way. You are using the resistors for pulldowns (which you can't do with TTL) and using the switch, when closed, to pull these pins high.

Either way is OK with HCT7474, but I was curious why there would be a difference from the schematic.

Did you use a pushbutton or a toggle switch ?

I see that if a normally open pushbutton switch were used, it would be better to use it as a pullup to 5V when closed.

I originally used a normally closed pushbutton switch to ground, which would open when pushed and activate the circuit.

After using it this way for a while, I think a toggle switch would be better.

[Speedee]

Hi, has anybody come across a circuit diagram for the receiver in the Aero Ace, I have seen the block diagram and circiut diagram for the transmitter but not the receiver

Thanks, Speedee :)

[didi]

@Martyn

I'm using a pushbutton switch and I'm very happy with it. It is positioned at the front of the Tx (I should make photo's!) where it is easily controlled by my index during flight.

For flying, I use it only for a smooth descent, i.e. for short periods of time. Therefor I think a pushbutton is better than a toggle switch, but this depends of course on personal preferences.

With the circuit I can comment, that for some reason it works also with a 74LS74. I bought one as backup, just in case. Don't ask why, it's the miracle of modern electronics that undefined inputs seem miracously to be stable.

Didi

P.S. A scheme of the Rx would really be interesting. Nobody who has the guts to do some reverse engeneering?

[didi]

I just made two pics to show, how it looks like now.

The red pushbutton is the low throttle switch. The DIP switches are for A, B, C switching

Didi

[CaptEvo]

where did you find small dip switch?

[CaptEvo]

dumb question,
what's the 'flip-flop' are you talking about? switch?
I did the 2 diode mod, but this seems complicated.

[didi]

@captEvo

well, I cut them from a larger one ;-)

works well with a fine saw.

didi

[bronney]

Martyn, I am going to do this tomorrow. But there's just one place I'd need clarification:

In post #31, pin 9 of the LS7474, it connects thru a diode to the throttle PCB right?

1. Do I supply that diode?
2. Does that connect to #10 or #11 on the throttle PCB?

Thanks. I am doing this for flying the eagle wing. The bi-planes are fine as is.

[Martyn McKinney]

Yes, you supply the diode.

It goes to COMMON on the throttle board (where the yellow wire used to go).

The yellow wire is then used to supply Vcc to the chip (pin 14).

You will note on the schematic that I used a NORMALLY CLOSED pushbutton switch.

These are not readily available and after using the circuit for a while I prefer a small toggle switch.

[bronney]

Thanks Martyn. Let me repeat because I am a super noob regarding these :)

1. desolder yellow wire from throttle pcb.
2. solder this into pin 14 of the 555.
3. from pin 9, solder a wire to a diode's anode.
4. cathode goes to where the yellow wire was desoldered from.

Did I get an A?

[Martyn McKinney]

A Plus ! except the circuit I am now using is not a 555.

It is a 74LS74 dual D Flip-Flop.

Don't forget the ground wire to pin 7 of the chip.

[bronney]

Nice!! And thanks for the pin 7, I wouldn't know what happen :)

[Carl T]

Gentlemen, I am brand new to this AA mods. Is there a place where I can go and find all the mods without reading all 70 pages? If not there should be one please.
Where is this two diode mod?
Thanks,
Carl

[Speedee]

Carl T

I think you will find the following page helpful, try the Rx/Tx mod section, I think

http://www.eflightwiki.com/eflightw...Airhogs_AeroAce

[colbyweb]

It looks like the link changed. This one should work:
http://www.eflightwiki.com/eflightwiki/index.php?title=Silverlit_X-Twins_Airhogs_AeroAce
-Rick

[Carl T]

Thank you Rick.

[freddy warbird]

Martyn McKinneyParts needed:
74LS74 Dual D Flip-Flop
2N3904 transistor
100K resistor
1N914 diodeI have a dual JK Flip Flop 16 pin with set and reset. The number on the chip is CD74HCT 109E. Can this chip work.

[freddy warbird]

I just did a search for the HCT109E JK Flip Flop, and it is a CMOS, and not TTL. So, I guess that is not going to work.

74LS74 Dual D Flip-Flop I think I can find this at Fry's.

[freddy warbird]

I guess you omit the 100K ohm resistor if you use the toggle switch.

Ok, I found an NTE74S74 "D" type Flip-Flop. I figure its the same as an LS. Here is the pinout.

[freddy warbird]

Ok. I did the "Low Speed Throttle Mod" with the above chip, and a 1N4150 diode. It works! I cut a tall dip switch in half with a hack saw blade as describe earlier. Hit the switch, and wwwwhhhooo.......wwwwhhhooo......wwwhhhooo......ww whhhooo, at every throttle setting, low, mid, and high. Hit the switch again, and its back to normal operation along with the zero throttle steering. I will test fly it as soon as I can get the switch glued in. Pretty cool Martyn! Thanks for putting in the work, and for starting this thread.

I spoke to a friend of mine today, and he says an ATMEL ( spelling) chip could possibly be used to do mods. But that would require programming, and I know nothing about that stuff. Anyway, I got this mod to work.

[freddy warbird]

I test flew both my Aero Aces with the Low Throttle mod a few minutes ago, and it works pretty good. It give a little more control which is what these little guys need. There is much more control during landings, and altitude can be managed better. Unfortunately, the pulse has a little pause in it. It would be better if the motors would run with out the pulse at a lower rpm. I think a 555 clock curcuit with all the caps, resistors, diodes, and a pot for fine tunning on its own five volt source activated via toggle could also work. this way the trigger can be adjustable.An LED could be installed in the timer circuit so one can see the trigger pulse change with turning the pot, or just listening to the motors. I am no cook book expert, and Know very little about TTL, so maybe not.

[Martyn McKinney]

Ok, I found an NTE74S74 "D" type Flip-Flop.
Glad to hear of your success.

I just wanted to point out that although the pinout diagram in post number 65 is correct, the drawing beside it is not correct.

This is the diagram for a OR gate.

[freddy warbird]

the drawing beside it is not correct.Ok. I edited post 65.

[Robster]

Martyn,

I converted your design into a working PIC version. It lent itself well to the conversion and was a fun project. I couldn't have done it without your schematics. Thank you!

Used a MicroChip PIC12F675 8-pin DIP microcontroller to do the business, and MicroChip's PICkit 1 w/MPLAB software to do all of coding, programming, burning etc.. I am still learning this stuff, so I'm sure somebody out there could improve the code and add more bells and whistles to it.

Glued an 8-pin DIP socket on the back of the Tx case, and ran the socket pins, wires and components through 8 little holes in the back of the case. This was to facilitate unplugging & re-programming the chip every time a change was made to the program - there were many, many failed attempts!

The transmitter works normally if the chip is plugged in or not. When I press the N. O. momentary switch, the PIC is enabled and pulses the motors about 2.5 times per second. It sure makes flying my stick and tissue bipe much more controllable.

Rob

[Martyn McKinney]

Smart Guy !

I was raised in the generation when it was necessary to build electronic things using discrete components.

Although I use micros, I am still fascinated by those who are able to quickly create something using them.

I feel like a blacksmith might feel.

They still need blacksmiths don't they? :)

Nice work !

[Robster]

Thanks, Martyn.... Smart guy... me? Naw!... Ah, but the blacksmith's art is one that not many can do!

Martyn, I here ya. I'm an old TTL guy myself. Been into electronics for about 30 years. Got my AAEE in 1986. Those days it was the 8088, TTL logic and lot's of glue chips.

The PIC stuff caught my eye about 2 years ago and has taken me along time to be able to 'do something' with it other than a Pencam shutter control project I did awhile back.

Rob

[Martyn McKinney]

Two suggestions:

If the 2 diode zero throttle mod is used, this will pull up the output of the micro even if it is in the zero state.

http://www.rcgroups.com/forums/showpost.php?p=4826311&postcount=72

An extra diode should be used in series with the output of the microcontroller.

The 1K resistor on the output of the micro could be increased in value to further reduce transmitter current.

I really like this modification because it provides the possibility of a variable duty cycle from 0 to 100% by the addition of some programming changes.

[Robster]

Thanks Martyn, I'll try your suggestions.

How ironic that you mention the zero turning throttle mod, because it WAS installed in this transmitter, but I removed it. The reason was that before I had the low throttle mod in, I was flying my stick and tissue bipe in my living room and when turning with throttle off, it tended to slam it into a wall.

But, when I was testing the throttle mod outdoors (It seemed like there was no wind), my plane seriously almost got away from me, and I thought what an idiot, why didn't I put the turning mod back in!

The zero turning mod is going back in.

I've tried some duty cycle changes with the program, but so far, no luck. Anything other than the 2 out of 4 pulse transmission ended up with the throttle reverting to normal operation. Without an O-scope, I have no explaination as to why.

One thing I would like to try in the future is to tie the P10, 11, & 12 to the three unused inputs of the PIC and program it to sense the 000 position to enable the low throttle and then sense the 111 position to enable the normal throttle operation. In a sense, shifting gears: stick down - low throttle, stick up - normal throttle. But not sure if that is any better than pressing the low throttle enable switch.

Rob

[GregG]

I stopped by Robs house yesterday. He reminds me of a mad scientist on these projects. He gets his teeth set into one and you can't get him away until everything works perfectly! Tried to pry him out of the "PIC Programming Chair" for a few minutes and drag him with me to the new hobby shop in town but the glue had already set. ;) ;) Hey Dude, lets go!!! :p

On the serious side, he demonstrated his modded radio system to me. It really worked great. Can't wait to see it in action with a plane in the air. Hopefully Friday we'll run video on it. Dudester, when do I get mine? ;) :D

[Robster]

Muahaahaaahaha!... Igor, Fetch Me More Chips!

[glencharles]

Has anyone tried pointing the motors down a bit, so that at max power the motors are contributing more lift and less forward speed?

[Robster]

glencharles, welcome to the forum :).

I've never pointed the motors down more than a few degrees, mainly to counteract the tendency to push the nose down on stick and tissue airframes during full throttle. But, it seems possible depending on the airframe that there could be a 'sweet spot' of lift vs forward speed to fly slowly without climbing with a higher downward angle.

Keep in mind that there is a trade off in motor/battery efficiency in using the motors to create more lift. Abrupt throttle changes may cause a little more porpoising.

Might be fun to experiment with this.

Rob

[glencharles]

I am thinking that it might be difficult to do it much without the lower wing interfering with the airflow.

Also, and I have not measured it, I suspect that for optimal effect, the center of the props would have to be at the COG. I dont think they are.

I seriously need to get an extra one to mod. Dont want to sacrifice the one I fly.

[wolfv]

Interesting thread. I really admire the way you all pull together to figure this out. Nice work.

I have had my first AA for 7 days now, and I do believe that this low-power mod is the most significant mod for the AA. Eliminating unwanted wing stalls and porpoising would make the airplane much easier to control. I have an idea to take this mod to the next level, but I am not an electrical engineer. Please tell me if the following idea is viable:

For the children, a single 50%-power step (pulse 100% power half the time) on the left stick would be easier than a toggle. Since the stock transmitter steps through 70% to 100% power, adding a 50%-power step is all that is needed. Is there some way to put a 50%-power step on the stick, between the lowest-stock power and zero power?

Thank you.

[Martyn McKinney]

You can make the lower 3 steps automatically go to lower power without having a switch by having the circuit toggle just 1 bit (P1.2) of the 3 bit throttle encoder instead of all 3 bits.

Another way, without significantly changing the existing circuit is shown below.

The switch in the circuit below allows normal throttle when closed and pulsing when open.

Adding the single N-Channel FET in green in the schematic below will short out the open switch (open switch pulsing on) and turn the throttle pulsing off (normal operation) in only the higher 4 throttle positions.

The circuit shows a normally closed pushbutton switch.

I prefer a toggle switch which would be unnecessary if the FET were added.

I haven't tried it yet, but it should work.

[wolfv]

Thanks Martyn. I think I understand what you are saying. Are the power steps in the attached "Toggle P1.2 or single N-Channel FET" correct? The original “Stock steps” are mapped to the “Modified steps”.

“Pulse power” is on half the time.
“power %” is estimated percent of full power.
Power differential for turning should function at all power steps.

Is there a way to use a PIC12F675 or similar chip to map the power steps like in the attached "PIC 12F675". The advantage of this mapping is that the jump from stock to mod is the smallest possible (about 20% power).

[Martyn McKinney]

I think that there might be an error in your first chart.

The FET creates 50% of steps 1 to 3, inclusive, but leaves steps 4 to 7, the same. (Power is proportional to duty cycle)

This implies that the jump with the FET between steps 3 and 4 would be from 40% to 85% or an unacceptable 45%.

I don't quite understand the description "4 pulse power".

--------------------
EDIT: I see! You are jumping power steps. Another way to do this is by varying the duty cycle (or a combination of the two).
--------------------

Because the frame rate is fixed, it is only possible to vary the duty cycle within these limits.

For 25% of whatever throttle setting, for example, this would require one frame on and and 3 frames off for a total of 4 frames. If on channel A (4 frames/sec), this would take 1 sec.

75% duty cycle would be 3 frames on and one frame off.

It should be possible to do this with a modification of the discrete circuit shown by adding another flip-flop and ORing the last 2 flip-flop outputs.

The first flip-flop in the existing circuit is toggling at 1/2 the input clock rate.

Because there are an odd number of clock inputs per frame, the output flip-flop gives a duty cycle of 50%.

Post #70 of this thread demonstrates the use of a PIC.

http://www.rcgroups.com/forums/showpost.php?p=6129511&postcount=70

My experience when flying the biplane with this duty cycle modification is that 50% of maximum throttle is just the right amount for making tight circles around your head.

Anything less and the plane will not climb.

This implies that a duty cycle (or power) greater than 50% would be an improvement.

Your second chart shows a nice way of implementing this, yet it still uses the 50% duty cycle.

A duty cycle greater than 50% would minimize stuttering, however, so a combination of the two might be the optimum solution.

Using a PIC is nice, because it would permit this flexibility.

If you make one up, I would like to be your first customer.

[Robster]

wolfv and Martyn,

I could not get the pic version working at 3 pulses on and 1 pulse off (75%) and I couldn't figure out why - also was unstable. It works great at 50% :D . I gave up after that. Perhaps I need to fire up my pic gear again :) . And, I need a darn oscilliscope! $$$.

Rob

[wolfv]

Martyn,
I corrected the error in the first chart. Thank you for pointing that out. A 45% power step is unacceptable.

Robster,
I am guessing that the receiver needs at least two consecutive pulses of equal value before the receiver decides that the signal is reliable and accepts the new value. So a 2-2 duty cycle would work. And the receiver would filter out the “1” in a 1-3 duty cycle.

I am with Martyn, if you make a PIC for this, I will be your second customer (I send you a check, you mail me the programmed PIC, and I solder the PIC into my transmitter). How many customers do you need to pay for an oscilloscope? :)

[Martyn McKinney]

I am guessing that the receiver needs at least two consecutive pulses of equal value before the receiver decides that the signal is reliable and accepts the new value.
Hadn't thought of that.

Smart guy!

[Robster]

I hadn't thought of that either.

Very smart guy!

[Hans Wurst]

Im to stupid for that programmed pic stuff.
Is there no way to make a "mechanical" solution, maybe some rotating switch with a motor, or a magnetic relais?
Big respect.

[wolfv]

Me smart? If was smart I would understand electronic circuits! How did you guy’s learn circuits? Are you electric engineers?

Robster,
Could the PIC also include a “fast turn mod”, but without the trim limitation as described by Martyn: http://www.rcgroups.com/forums/showthread.php?p=7008800#post7008800

Another idea is to put the PIC on the receiver (not transmitter). Is that plausible?
A PIC on the receiver could have some advantages:
· true zero-power turning (one motor at low power while the other motor is off)
· more evenly spaced power steps (pulse power with 1/2 and 3/4 duty cycles)
· full-power turn for barrel roles? (one motor at full power while the other motor is off )
Disadvantages:
· schematic of the receiver circuit is not available
· adds weight to airplane
· need one mod for each airplane

I don’t know if a full-power turn will result in a barrel role. It could be tested like this:
Disconnect a wire from one motor, climbed onto the roof, and drop the airplane nose down. As it pulls out of the resulting dive, applied full power. If it can’t do a barrel roll, PIC-on-receiver is not worth the trouble.

Hans,
The beauty of PIC is that once the software is written, many people can use it. And it’s easier to solder onto the board because there are fewer components.

[Martyn McKinney]

schematic of the receiver circuit is not available
Here is a simplified schematic of the receiver.

It is accurate enough to implement your own.

The difficulty would be in the mechanics of it all.

[Martyn McKinney]

I do believe that this low-power mod is the most significant mod for the AA.
My Sport AA flies perfectly right out of the box.

Like the Twin, however, it tends to climb at even the lowest power level.

Today, while flying the Sport, I decided to try adding more weight.

Because the Sport flew so well I wanted the extra weight placed at the center of gravity.

I didn't have any tools with me, so I stuck a dime into the foam at the rear of the plastic receiver module.

The dime wouldn't go in very far so I left 90% of it exposed at right angles to the airflow.

To my surprise, the Sport now flew without climbing at half throttle and climbed very slowly at full throttle.

I believe that it wasn't the extra weight that altered the performance, but the fact that the dime was creating extra drag and not allowing the Sport to increase speed at high throttle.

Drag brakes !

Just something that you may want to try which will obviate the need for any major transmitter modifications.

[wolfv]

The difficulty would be in the mechanics of it all.

Martyn,
Thank you for the receiver schematics. The MICROCONTROLLER is interesting. Are the micro controller inputs/outputs digital or analog?

What does "mechanics" mean? Parts too small to solder? :confused:

A dime sticking out the bottom is well below the thrust line, so it helps keep the nose down while it slows airspeed. What a great idea! I will try the drag brakes tomorrow morning.

[Martyn McKinney]

Are the micro controller inputs/outputs digital or analog?The receiver output is a low amplitude (300mV P-P) signal which likely is amplified in the microcontroller before going into a data slicer.

All the other inputs and outputs appear to be digital. Apparently there are programmable versions of this microcontroller available.

By mechanics I meant that even if the stock microcontroller were removed, replacing it with another would be a difficult proposition.

[Robster]

Robster,
Could the PIC also include a “fast turn mod”, but without the trim limitation as described by Martyn: http://www.rcgroups.com/forums/showthread.php?p=7008800#post7008800


There may be some coding and I/O challenges to get the low throttle AND the fast turn mod on one chip. However, it would be pretty easy to program the fast turn mod on a pic.

Rob

[wolfv]

There may be some coding and I/O challenges to get the low throttle AND the fast turn mod on one chip. However, it would be pretty easy to program the fast turn mod on a pic.


Rob,
Are your comments for the transmitter PIC or the receiver PIC?

I wrote some pseudo code for PIC AA Low Power (see attachment). I have not written assembly in 15 years, but the logic is there. This code was intended for the transmitter PIC because I don not know how or what the receiver’s microcontroller codes. I don’t know how to code the “fast turn mod”. Can you give me a hint?

wolfv

[wolfv]

On the stock AA radio schematic, where are the left & right stick steps converted (mapped) to left & right motor power settings? Could that part be replace with programmable version?

Thank you.

[Martyn McKinney]

The information at this site might be of interest.

http://www.webx.dk/rc/xtwin/xtwin.htm

[Robster]

Rob,
Are your comments for the transmitter PIC or the receiver PIC?

I wrote some pseudo code for PIC AA Low Power (see attachment). I have not written assembly in 15 years, but the logic is there. This code was intended for the transmitter PIC because I don not know how or what the receiver’s microcontroller codes. I don’t know how to code the “fast turn mod”. Can you give me a hint?

wolfv

wolfv,

My comments were referring the transmitter PIC. I think it would be way too difficult to hack a PIC into the receiver.

I'm guessing that your code is similar to what is already programmed into the transmitter's microcontroller. Each throttle position (0-7) sets the motors to run at a specific rpm. The whole idea here is to get less power than the '1' (lowest throttle) position. Unfortunately, we can't modify what is hard coded into the microcontroller, but we can 'fool' some inputs to the microcontroller to try to make it do what we want.

The low throttle mod works by disabling the throttle coder (turn off the motors) every 2 out of 4 pulses coming from the output of the microcontroller. We're not changing any motor rpm's we're just not letting the motors run half of the time quickly. That's why the motors pulse on and off with this mod.

A PIC based fast turn mod would work by having 2 inputs of the PIC reading the turn switches and 2 outputs of the PIC setting the transmitter microcontroller trim inputs to maximum left (00) or maximum right (11) depending on the left or right stick position.

Check out www.digikey.com and look up the PICKit 1 and the data sheets for the pic12f675 chips. This is very cheap and fun stuff! :D, but it could also lead to madness - I've been there.

Edit: It's possible to remap the throttle settings with a PIC between the throttle coder and the microcontroller, but it still would not solve the problem of too much power at the lowest setting.

Another edit: Ah, but maybe pulsing the transmitter's microcontroller throttle inputs. hmmmmm........ let me thimk about this......

Rob

[wolfv]

I think I found an elegant way to implement low power :eek:. Please tell me if this is plausible. But first some background.

The rectangle labeled EM55100 on the transmitter schematic, inputs stick controls and outputs power settings for left and right motors.
inputs:
* 3 bit throttle step
* 2 bit steering
* 3 bit steering balance
outputs:
* 1 bit lead in
* 2 bit zeros
* 4 bit fly ID (A,B,C)
* 4 bit L motor setting
* 4 bit R motor setting
* 2 bit check sum
* 1 bit stop
Input-output mappings are listed on http://www.webx.dk/rc/xtwin/xtwin.htm (thank you for the link Martyn).
Motor settings are expressed in 4 bits.
* The highest motor setting is 1111 (100% of max power).
* The motor off setting is 0000 (0% of max power).
* The lowest motor setting is 0011 (70% of max power).

Here is how to achieve less than 70% of max power.
Find a programmable version of the EM55100; a copy of the code would be helpful.
Assume the airplane’s speed controller is not capable of power lower than 70% of max power.
To achieve 50% of max power, program the chip to alternate between 1111 and 0000 motor settings.

Map other power settings such that power settings are spaced uniformly as possible.
Mappings for fast turns, zero-power turns, and barrel rolls are straight forward.

Is a programmable version of the EM55100 available? What is it called? Do you think it would work?

Thank you for all your support.

wolfv.

[Martyn McKinney]

Is a programmable version of the EM55100 available?
This site might be useful.

http://home.versanet.de/~b-konze/xtwin/xufo_mod/xufo_mod_en.htm

Main Site:

http://home.versanet.de/~b-konze/xtwin/xtwin_en.htm