New WLToys Protocol V977

More
08 Apr 2014 09:17 #22239 by phantom8
New WLToys Protocol V977 was created by phantom8
WLToys has released a new Power Star series Helicopters, including V977, V966, V930 & V988. Someone posted a few photos for the V977 transmitter & FC in RCGroups . On the FC, it has a BK2423.

Any gurus interested in getting one and add deviation support for it? The V977 costs around $100, while other models are cheaper. I'm planning to buy a V977. I can also buy a Saleae 16 and perform data logging if it helps.



Please Log in or Create an account to join the conversation.

More
08 Apr 2014 09:44 #22241 by SeByDocKy
Replied by SeByDocKy on topic New WLToys Protocol V977

phantom8 wrote: I can also buy a Saleae 16 and perform data logging if it helps.


I think it's the best option. I am feeling that this new protocol is a variation of either the V2x2 or the Hisky. I am sure that victzh can hack it very fast if you can sniff SPI via salae log

Please Log in or Create an account to join the conversation.

More
08 Apr 2014 16:26 #22245 by victzh
Replied by victzh on topic New WLToys Protocol V977
Transmitter looks promising, it most probably has BK2423 daughter board on it. If they're consistent (no reason not to be so - the chip pinout is the same), then the pins are in the same order as in www.inhaos.com/uploadfile/otherpic/DS-RF2400-NANO-V10-EN.pdf

GND, VCC, CE, CSN, SCK, MOSI, MISO, IRQ

We're interested only in GND (obvious), CSN, SCK, MOSI, and MISO.

So if they did not use any nasty tricks, it seems very realistic.

BTW, what is the difference between all these new models? With this numbering scheme it is very hard to tell who's brushless, who's bigger (and what size exactly).

Please Log in or Create an account to join the conversation.

More
08 Apr 2014 16:28 #22246 by victzh
Replied by victzh on topic New WLToys Protocol V977
Sorry for the rant, found a relevant RCGroups post:

www.rcgroups.com/forums/showthread.php?t=2068827

Please Log in or Create an account to join the conversation.

More
09 Apr 2014 17:23 #22281 by phantom8
Replied by phantom8 on topic New WLToys Protocol V977
victzh, thanks for the pin layouts. I think I'll go ahead and order the V977, plus the Saleae 16. Will you be able to help if I can get the logs?

Please Log in or Create an account to join the conversation.

More
09 Apr 2014 17:56 #22283 by PhracturedBlue
Replied by PhracturedBlue on topic New WLToys Protocol V977
Just an FYI:
I've never had a need for more than 8 pins when doing SPI for RC protocols. 8-pins willallow you to snoop both ends of an SPI bus at once if needed (though you'd miss the CE pin, which is generally not an issue).

Nothing wrong with a Salea16, but it is overkill for SPI work. On the other hand, the ability to capture very long samples (minutes worth) is really valuable, and makes the Salea design superior to a trigger/sample based system.

Please Log in or Create an account to join the conversation.

More
10 Apr 2014 05:48 #22312 by victzh
Replied by victzh on topic New WLToys Protocol V977
@phantom8 - sure I will be able to help, but mind you that usually just soldering to and watching the SPI is not enough.

You should be able to buy or borrow a second transmitter (at least, the more the merrier) to watch the differences in packets.

You will need to mock up the TX part of the protocol in controllable environment where you need to modify the implementation fast and possibly add feedback from the RX (I needed it once, but there was almost no way around). PB uses Raspberry PI based setup, I use Arduino - both have faster turnaround than Devo and can be connected easily to other hardware.

Saleae16 is definitely an overkill for SPI, I agree with PB.

Please Log in or Create an account to join the conversation.

More
10 Apr 2014 10:34 #22316 by phantom8
Replied by phantom8 on topic New WLToys Protocol V977
@PhracturedBlue & victzh, thanks for the heads up. I'll get the Saleae 8 channel model then.

@victzh, I do have several Arduino boards. I was using them for home automation and still have a couple spare. I may need your help to set it up tough. Is the 2nd transmitter a must for protocol implementation? I only got one and don't know anyone around has another.

Please Log in or Create an account to join the conversation.

More
10 Apr 2014 14:06 - 10 Apr 2014 14:07 #22320 by PhracturedBlue
Replied by PhracturedBlue on topic New WLToys Protocol V977
It is usually possible to develop the protocol from a single Tx, but it really depends on the protocol. You may end up with a protocol that only binds to exactly one model, or not have robust separation (so that two transmitters cannot be on at once). It can also be difficult to figure out what packets are constsnt, or tx specific. You may need to wire up SPI to the Rx (if possible).

Examples:
Devo: required a few transmitters due to variations between the 6/8/10 (this is the 1st one we did, so we probably could have gotten by without it if we knew more)
DSM: required multiple transmitters and multiple Rx to figure out the challenge/response
J6Pro: only needed one transmitter
Flysky: required multiple tx to determine robust txid and frequency hopping
Hubsan: only needed one tx. A 2nd was used for confirmation though
Frsky1/2: Only needed one tx (though it isn't actually done yet)
Skyartec: Only needed one Tx
Last edit: 10 Apr 2014 14:07 by PhracturedBlue.

Please Log in or Create an account to join the conversation.

More
10 Apr 2014 15:43 #22322 by victzh
Replied by victzh on topic New WLToys Protocol V977
All protocols I did I used many TXs - it is just much easier. As soon as you have an idea where the essential part of packets go - TX id, controls, checksums, how binding works - you can guess the rest and check your guess by writing TX code, and go by without a second TX if you're lucky.

Initial implementation can be done with one TX definitely - you just replicate TX id/frequency hopping sequence (if any) from the TX you've got. So yes, it worth trying anyway!

Also, people developing these protocols copy each other, so many parts can be resolved by educated guess - comparing with the past.

I think that in currently implemented protocols there are all the tricks they have. You can devise, e.g. encoded protocol, but they try to save on models and model MCU does not have the space or the power to decode cryptographycally strong protocols.

Please Log in or Create an account to join the conversation.

More
11 Apr 2014 15:49 #22343 by phantom8
Replied by phantom8 on topic New WLToys Protocol V977
Hmm, this seems to be much more complicated than I initially anticipated. I can solder the trace wires to the transmitter. If soldering to the FC (receiver) is required, I don't think I can do it. The soldering points on the FC are just too small and are beyond my capabilities. I'm worrying we may get stuck if the SPI trace on the single TX is not sufficient. What do you guys think?

Please Log in or Create an account to join the conversation.

More
11 Apr 2014 16:26 #22346 by victzh
Replied by victzh on topic New WLToys Protocol V977
You'll have a version which works alone, so you will not be able to run 2 or more models from the same code simultaneously. It's not that bad, and these efforts of initial research are needed anyway, so it's not gonna be wasted effort.

I can solder to such chips, I soldered once when there was a 2mm gap between RF chip and MCU. It took me several attempts of 2-4 hours each - I don't have special equipment, just FX-888 soldering iron and binocular magnifying glasses.

It's unpleasant, but is not required for the first step.

If you plan to enjoy V977 anyways it definitely worth a try to reverse it from the transmitter side first.

There is also a chance that the protocol is a variation of already existing one and this will be enough.

Please Log in or Create an account to join the conversation.

More
12 Apr 2014 14:57 #22362 by phantom8
Replied by phantom8 on topic New WLToys Protocol V977
@victzh, I've just ordered a V977 & Saleae 8. I'll let you know when they arrive. Hopefully, we can get it working without the 2nd transmitter or soldering the receiver board. Thanks in advance for your help. :)

Please Log in or Create an account to join the conversation.

More
13 Apr 2014 04:27 #22375 by victzh
Replied by victzh on topic New WLToys Protocol V977
Sure, anyways it's gonna be fun!

Please Log in or Create an account to join the conversation.

More
13 Apr 2014 07:07 #22380 by SeByDocKy
Replied by SeByDocKy on topic New WLToys Protocol V977

Please Log in or Create an account to join the conversation.

More
24 Apr 2014 10:54 #22684 by joeclone
Replied by joeclone on topic New WLToys Protocol V977
Hey Victzh, Banggood has agreed to send you a V977 to do protocol cracking to work with Deviation, please keep us updated on how it goes.

Please Log in or Create an account to join the conversation.

More
24 Apr 2014 12:42 #22688 by SeByDocKy
Replied by SeByDocKy on topic New WLToys Protocol V977

joeclone wrote: Hey Victzh, Banggood has agreed to send you a V977 to do protocol cracking to work with Deviation, please keep us updated on how it goes.


Nice attitude from Jacky ;) Thanks to him

Please Log in or Create an account to join the conversation.

More
24 Apr 2014 14:59 #22692 by phantom8
Replied by phantom8 on topic New WLToys Protocol V977

joeclone wrote: Hey Victzh, Banggood has agreed to send you a V977 to do protocol cracking to work with Deviation, please keep us updated on how it goes.


That's great news. It should make cracking the protocol much easier. I've received the V977, but still waiting for the Saleae to arrive.

Please Log in or Create an account to join the conversation.

More
24 Apr 2014 15:34 #22693 by victzh
Replied by victzh on topic New WLToys Protocol V977
I'll definitely try hard, but it is often a collaborative effort - no one still knows what surprises we will face. For instance, I still can not recognize the signal of RCLogger's last Eye One Xtreme - it looks like SPI, but they use very weird version.

The more reverse engineers, the merrier.

Please Log in or Create an account to join the conversation.

More
24 Apr 2014 22:18 #22701 by SeByDocKy
Replied by SeByDocKy on topic New WLToys Protocol V977

victzh wrote: I'll definitely try hard, but it is often a collaborative effort - no one still knows what surprises we will face. For instance, I still can not recognize the signal of RCLogger's last Eye One Xtreme - it looks like SPI, but they use very weird version.

The more reverse engineers, the merrier.



You should male some YT tutorial :)

Please Log in or Create an account to join the conversation.

Time to create page: 0.074 seconds
Powered by Kunena Forum