Jump to content
Arthur P.

Hacking the I2C interface of Spektrum DX and AR ?

Recommended Posts

I have to dive into this again, and get you the straigh facts in one post. I was wondering how to set up the failsafe on this becaue at the moment I have none.

I will go home and check everything next chance I get.

Assuming you have the V3 board, any random order would definitely not be a good thing. For sure, the Rx's last logical channel (rudder on the AR7000) must be on the PPM board's P1-7 pin (see schematic). Also, if the failsafe PIC is used, the Rx's throttle channel must go to the correct place too, per the PPM board's JP1 jumper configuration that you used. Beyond that, any physical channel order is fine and you can mix as you please without affecting the logical channel order.

Since the same channel is missing, regardless of how you plug in the channel order on the PPM board, then I think it is safe to say that you don't have a PPM board hardware problem. So, there must be something weird on the MK side. BTW, which channel is missing?

Share this post


Link to post
Share on other sites

Mr.RC-Cam,

I received today the card of the PWM2PPM V3 converter. Many thanks! I will search somebody to performe the placement of the missing components. Would you be so kind please to repeate some explanations concerning the inclusion of the converter in the total schem:

a) From the AR7K (the 7 channels) the signal comes in R/C P1 to the numbered 1 to 8 points of the converter? In which point must be the soldered wires?

cool.gif The sum signal (the 7 channels) gets out from the 3 points just under the letters "AR7K-PPM";

c) The 4 points (N,P,+,-) are used for what?

d) The connector JP11 to JP18 is not on the card? Where must I install the jumper across JP1?

e) Where is going the outgoing signal?

Yours sincerely,

Georges

Assuming you have the V3 board, any random order would definitely not be a good thing. For sure, the Rx's last logical channel (rudder on the AR7000) must be on the PPM board's P1-7 pin (see schematic). Also, if the failsafe PIC is used, the Rx's throttle channel must go to the correct place too, per the PPM board's JP1 jumper configuration that you used. Beyond that, any physical channel order is fine and you can mix as you please without affecting the logical channel order.

Since the same channel is missing, regardless of how you plug in the channel order on the PPM board, then I think it is safe to say that you don't have a PPM board hardware problem. So, there must be something weird on the MK side. BTW, which channel is missing?

Share this post


Link to post
Share on other sites
a) From the AR7K (the 7 channels) the signal comes in R/C P1 to the numbered 1 to 8 points of the converter? In which point must be the soldered wires?

The 7 channels (signal line only) go to the 1-7 marked on the converter board. The 'RUDD' channel on the AR7000 must go to the 7 on the converter.

b )The sum signal (the 7 channels) gets out from the 3 points just under the letters "AR7K-PPM";

The three pads under the letters "AR7K-PPM" only 2 of those pads are used. the '+' is for +5V and the '-' is for ground and they get connected to the AR7000 for its power.

c) The 4 points (N,P,+,-) are used for what?

These are what get connected to the MK (this is the output). + and - are for the power comming from the MK (5V, gnd) and the 'N' is for the negative PPM and the 'P' is for the positive PPM out signal. For the MK it does not matter what one you use but you only need to use one of them.

d) The connector JP11 to JP18 is not on the card? Where must I install the jumper across JP1?

Right above the letters "AR7K-PPM" you will two pads for each channel (the other side of the board has it labled 1 through 8) you put a solder jumper on the one channel that lines up with your 'Throttle channel. This is to sellect the failsave input.

e) Where is going the outgoing signal?

See c) above

Richard

Edited by brashley

Share this post


Link to post
Share on other sites
Given that you don't have PC software to configure it, my guess is that you will need to have the Flight Controller loaded with different firmware that supports the channel order of your R/C system.

Hello again,

I have borrowed a Futaba 7C (2.4Ghz) with a R617FS RX. Is it possible for you to confirm my connections to the PPM? I am just not sure about which is the last logical channel on the receiver. I think all other connections are OK.

using the Futaba TX with the Dammar controller should solve all issues with channel order. (I hope)

Thanks in advance.

post-4929-1232510412_thumb.jpg

Share this post


Link to post
Share on other sites

The PPM V3 board is for Spektrum systems (typically using the AR7000). For a Futaba FAAST system, you might need to research Old Man Mike's converter sold by 13brv2 on rcgroups.

Share this post


Link to post
Share on other sites

Thank you Richard! It leaves maybe the will to have a global connexion schem.

Could you please repeat methodicaly the fonctionement and the setup of the fail safe channel in SPEKTRUM.

Yours sincerely,

Georges

The 7 channels (signal line only) go to the 1-7 marked on the converter board. The 'RUDD' channel on the AR7000 must go to the 7 on the converter.

The three pads under the letters "AR7K-PPM" only 2 of those pads are used. the '+' is for +5V and the '-' is for ground and they get connected to the AR7000 for its power.

These are what get connected to the MK (this is the output). + and - are for the power comming from the MK (5V, gnd) and the 'N' is for the negative PPM and the 'P' is for the positive PPM out signal. For the MK it does not matter what one you use but you only need to use one of them.

Right above the letters "AR7K-PPM" you will two pads for each channel (the other side of the board has it labled 1 through 8) you put a solder jumper on the one channel that lines up with your 'Throttle channel. This is to sellect the failsave input.

See c) above

Richard

Share this post


Link to post
Share on other sites

Hi Guys,

I just finished putting together a V3 ppm converter, however I can only see three channels move in the Mikrokopter Tool software. So I decided to check it on an Oscilloscope and here are the pictures I got. It seems to be that I can see only six channels, but actually when I move the inputs of the DX7 controller it seems to be that there are seven, only the voltage for the signal 4 is low. It looks like I am missing a channel in between all the 6 signals. Can someone look at the oscilloscope picture and give me some advice on what is going on? Or how can I trouble shoot the board.

Thanks I really appreciate it

Jesse

PD. I attached another picture showing the signal after the diode array showing the same signal, only together.

post-5270-1233729708_thumb.jpg

post-5270-1233729833_thumb.jpg

Share this post


Link to post
Share on other sites

Channel 4 is completely missing.

Check the Ch-4 wire to the Rx. Plus, look carefully at the diode array (use a 4X magnifier) that Ch-4 is connected to. If necessary, grab the ohmmeter to help diagnose the problem on Ch-4.

Keep in mind that the missing Ch-4 is the logical position, not the physical position. According to chart on post #115, this would be the elevator/nick channel on a AR7000 (which is physical servo position 3). So, this would hint that the diode array input that is connected to the Rx's elevator channel is an open circuit. That leaves the problem with the wire connecting them or the solder connection on the diode.

Share this post


Link to post
Share on other sites

Mr. RC-Cam

Problem solved...!!!! I did what you recommended I added more flux to the diode arrays and reheat them, and this allowed me to see more channels on the Mikrokopter Tools. However I was still not able to see the seventh channel on the oscilloscope. But then, I realized that the DX7 was set to the wrong Heli mode, changing it to the airplane mode allowed me to see the missing channel... :P he he he he

Thanks again

Jesse

Channel 4 is completely missing.

Check the Ch-4 wire to the Rx. Plus, look carefully at the diode array (use a 4X magnifier) that Ch-4 is connected to. If necessary, grab the ohmmeter to help diagnose the problem on Ch-4.

Keep in mind that the missing Ch-4 is the logical position, not the physical position. According to chart on post #115, this would be the elevator/nick channel on a AR7000 (which is physical servo position 3). So, this would hint that the diode array input that is connected to the Rx's elevator channel is an open circuit. That leaves the problem with the wire connecting them or the solder connection on the diode.

Share this post


Link to post
Share on other sites

Hi all, I too am wondering how to set up fail-safe with this... Here is my guess.

When I bind I have the throttle set to a negative value so that when the board sees this negative value it stops sending signals thus putting the MK FC into fail safe mode. Is this how it works?

On another note I plugged my MK in for a flight tonight and left the FC switch on (this I have done many times before without issue) when I do this it causes a spark at the battery connection but this time did some damage.

My ppm converter from then on had the LED stay on continuously. I checked the board for any obvious blown components and found none so I figured the little programmed chip may have gotten flashed... I replaced it (sure glad I orderd two) and the LED goes out and the MK is happy again!!! So am I.

Now I figure I just won't plug it in with the MK board turned on hopefully it does not happen again.

Ps I have a park bec supplying voltage to my lights, the receiver, the ppm converter and it also supplies the MK FC as well because the tilt servo draws enough juice to put the MK into a death spin, thus the bec now feeds the board too.

Anyway I need another chip for back up! I will pm.

Share this post


Link to post
Share on other sites
When I bind I have the throttle set to a negative value so that when the board sees this negative value it stops sending signals thus putting the MK FC into fail safe mode. Is this how it works?

The general idea is to set the Throttle's low stick Travel Adjust to an exaggerated value before you bind. Then after binding you restore the Travel Adjust to 100%. I think Brashley posted the settings he used earlier in this thread.

I checked the board for any obvious blown components and found none so I figured the little programmed chip may have gotten flashed.

It's hard to say why the PIC was damaged. Component R6 (220 ohms) provides some protection from the outside world. I can't say it would actually help, but you can increase this resistor to 1K ohms. Also, be sure that the voltage to the PIC is 3.3V to 5.2VDC (some BEC's produce higher voltages than that, which would damage it).

Share this post


Link to post
Share on other sites
Now I figure I just won't plug it in with the MK board turned on hopefully it does not happen again.

Personally I think the opposite way, I've never fitted the switch on the MKs I've worked on. And I still wonder how can someone sane deliberately fit a cheap 2A or 6A-rated switch (OK, x2 as there are 2 contacts, but still) in the primary supply of an aircraft when perfectly aware that it will be treated with 40+ amps...

I have seen 2 of them just get permanently welded closed (with unknown contact quality of course) by the spark created inside when switching them on.

Share this post


Link to post
Share on other sites

Just picking a nit here, but, at the voltages we use, the spark is generated when switched off and interrupting the current. The welding of contacts will occur when an excess current flows through the switch contacts heating them up and consequently welding them together due to the small resistance between the contacts.. When opening and interrupting a large current, normally the contacts will vaporize resulting in, everything being equal, an open circuit. Damage to the electronics will usually occur at this point due to spikes being generated and feeding back into the equipment at the moment of sparking.

Edited by W3FJW-Ron

Share this post


Link to post
Share on other sites
And I still wonder how can someone sane deliberately fit a cheap 2A or 6A-rated switch (OK, x2 as there are 2 contacts, but still) in the primary supply of an aircraft when perfectly aware that it will be treated with 40+ amps...

I had casually wondered what that little toggle switch was used for. Indeed, if it is being used to switch the main supply then it would be very wise to replace it with a permanently soldered heavy gauge jumper. Ignoring the obvious issues, do it for the sake of ensuring the lowest possible v-drop under full throttle conditions.

Share this post


Link to post
Share on other sites
Just picking a nit here, but, at the voltages we use, the spark is generated when switched off and interrupting the current.

To nitpick back, not in this case... as switching off is done with no/light load (only the few mA drawn by the electronics as the motors are off). However, when switching on you have the short but very high current spike created by connecting the discharged input capacitors of the 4 ESCs simultaneously to a very low impedance source... In the case of the MK it's not all that bad altogether given the rather small caps and long wires, but still for the poor little switch it's a bit excessive.

Now try connecting a 5s/4000 pack to a good quality 80A ESC (with adequately sized, very low ESR input caps and 4mm2 supply wires) like I did earlier today, and you'll understand what I mean. Sounds pretty much like a small cracker with a bright spark, and will make you do a decent jump if you're not used to it (and sometimes even when you are), leaving very visible marks on the connectors. Usually in that case you'd precharge the caps through a resistor, and only after that connect the low-resistance feed to avoid frying the ESC PCB traces and/or caps, and replacing your connectors every 5 flights due to the several hundred amp spikes...

Edited by Kilrah

Share this post


Link to post
Share on other sites

Ron is correct in saying the spark should occur when switching off but as Kilrah says I get a big spark when connecting my battery. I can only think that I don't get it connected quickly enough, it dose worry me to see it though.

Terry

Share this post


Link to post
Share on other sites

If you ask me, everybody is right on this one. Besides the common arcing that occurs when switches/relays are opened, arcing upon contact closure is also real. Electromechanical switching devices are not perfect -- a tiny bit of contact bounce can cause the arc issue. I will agree that the arc occurs during the brief periods the circuit is open as the contacts settle, but to the observer the arcing happens while they are in the process of closing the switch. Most switches settle down in 20mS or so, but some can take longer.

The problems that occur during relay/switch open and closure have haunted designers since Edison. There are all sorts of design guidelines to be observed to help minimize it. Some more effective than others. Honestly, we could debate the details forever (we wouldn't be the first to do that).

But I think the main point here is that the mini toggle switch on the MikroKopter is something that could easily be replaced with a very reliable piece of heavy gauge wire. I can't imagine any downsides to doing that. The switch is not needed, since the battery pack can be unplugged instead, which is the way electric modelers do it on other e-models.

But I have to ask. Does that little toggle switch really control the power to everything, including the four brushless motors? It seems so out of character for the brilliant fellows that created the MK. But, sometimes these little gotcha's sneak in while more important design details are being hammered out.

Now try connecting a 5s/4000 pack to a good quality 80A ESC (with adequately sized, very low ESR input caps and 4mm2 supply wires) like I did earlier today, and you'll understand what I mean.

Yup, plugging in such a pack can get a bit scary. A couple hundred amps can flow for a brief period. It's a good way to mess up perfectly good connectors. :)

Share this post


Link to post
Share on other sites

Guys,

My 10uF capacitor broke from its soldering place when I was putting the heat shrink; it was a Mini Electrolytic 16V 10uf Surface Mount 4MM.

I found this other capacitors (Digikey part# 587-1295-1-ND) they are multilayer ceramic capacitors 0805, I don't see in them any polarization mark, can I use them?

Thanks

Jesse

Share this post


Link to post
Share on other sites
But I have to ask. Does that little toggle switch really control the power to everything, including the four brushless motors?

Yes...

Stromleitungen1.jpg

And that small PCB trace on the battery side of the switch (single-sided too...) is bearing all of it as well.

Personally I connect all the ESC wires together and to the main supply off-board, just send 2 small wires to the controller and don't fit the switch.

Share this post


Link to post
Share on other sites

whoa mine is not wired like that at all... my two main power wires go to the board direct (huge solder pads). also the power feeds to the ESC's are off the main borad (huge solder pads). Far as I know or can tell the switch is just used to turn power on to the board itself.

If you solder it up like the picture I could see that being a problem.

If all the power went through that little 6A switch it would have died long ago seeing as there is likely 30-40 amps being drawn.

Yes...

Stromleitungen1.jpg

And that small PCB trace on the battery side of the switch (single-sided too...) is bearing all of it as well.

Personally I connect all the ESC wires together and to the main supply off-board, just send 2 small wires to the controller and don't fit the switch.

Share this post


Link to post
Share on other sites

This is a photo they have made of an old board version (actually the one I have), and that's still referred in the wiring guide. I have however assembled some current ones for other people and seen the newer large pads. But it still seemed to me that the whole supply was supposed to go through (and I've seen someone with the latest version get his switch stuck). Haven't particularly looked at that detail though as I knew I wouldn't be wiring it that way anyway.

Edited by Kilrah

Share this post


Link to post
Share on other sites

ah yes I would not think you would wire as per this picture.... yeah the new board has huge pads that all the wires are soldered but I suppose that does not mean the switch is not the bottle neck like you say....

mmm... I may go home and ditch my switch too... it is just annoying anyway ;-) and I'll save some weight!

I sitll don't know why my PIC died but I don't care... so long as it's working again!

This is a photo they have made of an old board version (actually the one I have), and that's still referred in the wiring guide. I have however assembled some current ones for other people and seen the newer large pads. But it still seemed to me that the whole supply was supposed to go through (and I've seen someone with the latest version get his switch stuck). Haven't particularly looked at that detail though as I knew I wouldn't be wiring it that way anyway.
Edited by saabguyspg

Share this post


Link to post
Share on other sites

I am having some issues...

My 10uF polarized capacitor broke and I replaced it for the ceramic 10uF capacitor and checked the signal on the oscilloscope where I was able to see 7 signals on DX7 Aircraft mode.

Now the last signal on the oscilloscope moves when I move the Rudder in the radio, but it also moves when I move the throttle, so they are somehow together in the last signal.

Furthermore I am not able to see more than 4 signals on the MK tools. I can only see Elev, Ale, Gear and Aux2 but no signal for my Throttle and Rudder?.

When I change to Heli mode on the DX7, channel 4 is missing like in my previous post #307, and in the MK Tools Rudd and Ale, doesn’t work but I can see two channels responding for throttle and one for the Elevator, plus Gear and Aux2.

So what am I missing? Do I have to set up the DX7 radio in a particular way? Or did I damage a component on the PPM V3 converter that is giving me this behavior?

Thanks

Jesse

Share this post


Link to post
Share on other sites
Now the last signal on the oscilloscope moves when I move the Rudder in the radio, but it also moves when I move the throttle, so they are somehow together in the last signal.

With everything connected, does your o-scope show correct waveforms on the Rx inputs into the PPM board?

Are you sure you have the rudder channel going to D3-3? Is R4 installed?

When I change to Heli mode on the DX7, channel 4 is missing

I could be wrong, but I believe the heli mode involves special mixing that disrupts the AR7000 channel sequencing; some channels may be sent at the same time. The PPM board would not like that if such a thing occurred.

So what am I missing? Do I have to set up the DX7 radio in a particular way?

I don't have a DX7 so I can't help. Brashley would have the details you need.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...