Help! Having trouble speaking JETI

[cbdane]

I'm usually pretty good at this sort of thing but I'm struggling here. I've had some projects on the backburner that involve interfacing to the JETI telemetry system. I hadn't worried that it would be too hard since the JETI website provides a pretty good interface and communication protocol description. After having little luck, I've finally lowered my sights and I'm just trying to get any response from the communication link.

I'm using an R3 linked to a Profi-box in telemetry mode. I know I have that link set up fine since I can plug in a JETI temperature diagnostic module and it shows up with no problem. After many unsuccessful attempts, I've lowered my expectations and I'm simply transmitting the EX data packet examples in the JETI interfacing documents but just can't get any reaction from the Profi. When I look at the EXT pin on a scope, I see 9 50ms serial packets, spaced by about 25ms and repeating about every 2 seconds. I've tried transmitting my sample packets at all temporal relationships that I can think of relative to these 9 packets with no success.

As specified in the document (http://www.jetimodel.com/en/show-file/26/), I'm using 9600 baud, 1 start bit, 9 data bits, 1 parity bit (odd), and 2 stop bits. I'm confident that I'm doing this correctly (bit banging with a microprocessor) since I can decipher the receivers transmit packets on the scope and can clearly see the expected 0XFE character at the beginning of each packet. The sample packets that I'm using are shown on the last couple of pages of the document.

I'm out of ideas and I know that I'm missing something very simple. I have the half-duplex single-line interface set to listen until the 9 packets complete then I switch to transmit and send out the sample packets. No data or available devices show up in the Profi menu. As I said, I've tried a large range of time intervals between the 9 repeating packets and the one(s) that I'm trying to transmit. I don't have my DS-16 with me since I'm working in a hotel on the road right now.

Once I establish the basic communications link, encoding the data should be very straightforward as detailed in the interfacing guide. Has anyone successfully done this before? I believe that flymustangs here on the forum has been successful but I thought I'd put out a general question to the forum in case I can't reach him.

Thanks in advance!

[s car go]

You say you are transmitting 9 data bits. Yes the spec does say that on page 1 but in all the examples shown they show data it as 8 data bits.

Try 8 and see if you have success.

What kind of scope are you using to decode the serial stream? Does it actually turn the scope traces into a decoded character or are you trying to examine each bit to determine what is being observed?

This shows the scope decoding the packet.

[cbdane]

Thanks! That's very helpful! Yes, the spec says 9-bit and the examples are, as you say, 8-bit but it specifically notes at the top of the example pages that the 9th bit isn't shown. Is the scope trace you show actually from the JETI system? The reason I thought I was doing it right is that, when I look at the first byte from the R3, it is 200us low, 700us high, 200us low, then back to high. This fits well with 1 start bit and parity low which corresponds to 0XFE; the standard text start byte. If it were 8-bit, the stop bit position would be low which doesn't seem to fit but I could obviously be confused.

I'll definitely try 8-bit and see how it works.

[s car go]

The scope example shown is actually DMX protocol. I just grabbed the first example I could find.

However it would not take much to connect the scope to one of my models.

[cbdane]

Thanks to all of the help from everyone, especially the PMs from flymustangs. After fiddling with this system in frustration for over a week, I finally figured it out. It turns out that JETI doesn't use the SAE standard CRC-8 (cyclic redundancy checksum) polynomial that I was familiar with but the biggest issue was properly setting bit 8 (ninth serial bit) in the message structure. It's right there in the JETI document, with a separate data column dedicated to bit 8, but I had just missed it. The confusing thing is that the sample packets on the final pages only show 8-bit values (as described above) and don't indicate the correct state of the ninth bit.

Here's my breakthrough photo from this evening showing that I've created a telemetry device named "cbdane" as can be seen on my Profi. The device that I'm using to generate it is the Robart Pressure Guard which I'm reprogramming to maintain its high speed landing gear failsafe functions as well as provide inflight pressure telemetry for both gear pressure and brake pressure (a separate Pressure Guard for each). Establishing this communications link is the hard part; the rest is easy and will fall right together! I work directly in machine code instead of the more usual approach using compiled C++ so the communication will be very efficient and easily handled by the little microcontroller in the Robart device: http://robart.com/collections/monthl...pressure-guard This also means that the "sky is now the limit" and I'll be able to design all kinds of custom telemetry devices for anything that I'm interested in.



What??? Doesn't everyone's hotel room look like this while on business travel? Fortunately, this Marriott knows me and the maids didn't call the NSA...

[s car go]

Glad you got it figured out.

[Esprit Model]

I guess you will hate our new product being introduced next week. The pressure sensor is ready to be shipped from Jeti.

Zb/Jeti USA

[cbdane]

Originally Posted by Esprit Model

I guess you will hate our new product being introduced next week. The pressure sensor is ready to be shipped from Jeti.

Zb/Jeti USA

Nope; the joy is in the journey, not the destination! Besides, mine will have high rate (kHz) internal pressure analysis to deploy gear on small pressure changes, remember state of gear, etc. You can cut a hose with scissors and the gear will be down before the tank is empty. JETI communication is just a bonus, not the primary purpose.

[flymustangs]

Glad you got it figured out. Looking forward to what you come up with. That servo tester I got from you at IRCHA that one year is one of my most used tools.

[Puttputt maru]

Originally Posted by flymustangs

That servo tester I got from you at IRCHA that one year is one of my most used tools.

I am always looking for new cool tools, could you enlighten me on it please?

[Mercuriell]

Originally Posted by flymustangs

Glad you got it figured out. Looking forward to what you come up with. That servo tester I got from you at IRCHA that one year is one of my most used tools.

Me too! Love the hotel room!

[cbdane]

Originally Posted by flymustangs

That servo tester I got from you at IRCHA that one year is one of my most used tools.

Originally Posted by Puttputt maru

I am always looking for new cool tools, could you enlighten me on it please?

Haven't built any of these for a long time. Still unrivaled, IMHO. Only device that will measure peak current, minimum transient voltage, deadband, and true servo speed (in ms) on the bench or installed in a model. Plus measure receiver output: frame rate, pulse width, etc. I use it to setup every new model. It can generate 720us narrow pulses for testing rudder servos at frame rates up to 333Hz (3ms) and can do automatic cycling and testing under load. Manual here. https://www.helifreak.com/attachment...9&d=1361839416

[Puttputt maru]

Thanks cbdane, looks like a great tool.
Not sure if this is the place to ask more about i,t with helifreak rules, so you can PM me if it is better.
I am not the greatest with electronics, if it was originally a do it ourself project, would you still build them? And in the affirmative, how much? I would not mind having two (I will talk to a friend who might want one too).

[cbdane]

I love my JETI! Project complete! Full pressure telemetry with all of the Pressure Guard high speed protection fully functional! I only have one test tank so only the Gear channel (not the brake channel) is showing a reading.


Here's a video showing how the Pressure Guard works. The LED display in the clip is now replaced by JETI telemetry!

[s car go]

My hat is off for you. Great work!.

[cbdane]

Originally Posted by s car go

My hat is off for you. Great work!.

Thanks! Spent 7 years in Houston; got a degree from Rice.

[flymustangs]

Brent, looks like fate got you just like it got me with the Xicoy and Jetcat ECU interface.

Look what ZB just announced over on RCG, http://www.rcgroups.com/forums/showp...ostcount=11210

It's still fun building these things, though.

[cbdane]

Originally Posted by flymustangs

Brent, looks like fate got you just like it got me with the Xicoy and Jetcat ECU interface.

Look what ZB just announced over on RCG, http://www.rcgroups.com/forums/showp...ostcount=11210

It's still fun building these things, though.

Heh, heh... No worries Ken. Like I said in an earlier post, the joy is in the journey! I guess they do read their email where I suggested earlier this year. Anyway, it's going to be great fun to begin to adapt all kinds of electronic gizmos to my JETI telemetry!

[Ray66]

Hi,

Searching for the protocol of xicoy ecu's

Cheer Ray

[gunny_jeeves]

That's awesome!

I'd love to be able to get the time to do some coding to mess with some of this stuff...

My bet is that all of the "open source" flight controllers and stabilizers can be modified to do Jeti integration. (Full configuration integration like Spirit, etc... ) couple folks do something like that then commit it to cleanflight, and bang! New world order... Lol.