pascallanger
DSM2 surface
Hello :-)
Is it possible, to ad DSM2 for surface (16,5 ms)?
Would be great to use my Mini-Z :-)
Greets
I need to have access to a RX first to try things out and if I can't get it to work I would need a TX. If you are ok to send it to France let me know.
Is it ok if I send you the money for such a recipient? They are not expensive. I would send the money to them via PayPal friends and then they can keep the receiver. I think otherwise the shipping is more expensive than the RX ;-)
Here is a link to the Reciever: https://campaign.aliexpress.com/wow/gcp/tesla-pc-new/index?UTABTest=aliabtest344316_486351&_randl_currency=EUR&_randl_shipto=DE&src=google&src=google&albch=shopping&acnt=494-037-6276&slnk=&plac=&mtctp=&albbt=Google_7_shopping&albagn=888888&isSmbAutoCall=false&needSmbHouyi=false&albcp=8632818872&albag=89514576449&trgt=295204014427&crea=de1005003607189968&netw=u&device=c&albpg=295204014427&albpd=de1005003607189968&gclid=Cj0KCQiA1ZGcBhCoARIsAGQ0kkrGO1kQ4pSVjaKK427gTSIPW4ptK67eJb02CrN6lULINnKqPqsmzRwaAmtNEALw_wcB&gclsrc=aw.ds&aff_fcid=3a66352ca1f94f7eb21ee0198936e15e-1669675410477-06003-UneMJZVf&aff_fsk=UneMJZVf&aff_platform=aaf&sk=UneMJZVf&aff_trace_key=3a66352ca1f94f7eb21ee0198936e15e-1669675410477-06003-UneMJZVf&terminal_id=785307d784a94138941d93a47e197581&wh_weex=true&wx_navbar_hidden=true&wx_navbar_transparent=true&ignoreNavigationBar=true&wx_statusbar_hidden=true&bt_src=ppc_direct_lp&scenario=pcBridgePPC&productId=1005003607189968&OLP=1084300508_f_group2&o_s_id=1084300508
Greets Sascha
Hey :-)
So it seems not to be ok for you? Maybe, you could explain how it works to sniff a protocoll? Or is there a thread where i can read about this? Looking forward to hear from you :-)
Greets Sascha
Sorry I forgot to reply. Sure send me a paypal at pascal _ langer @ yahoo . fr and I'll place the order right away. Just looked at your link, it's 16€ delivery included.
Send it :-) Hope, you are succesfull :-)
But, you would not tell me how it works? I know, i need a "sniffer", did read this at RC-Groups, but are not able to find the thread anymore :-( Then, in future, i could make my own protocolls, specialy for Cars and boats :-)
Greets Sascha
Order placed. In that case there is nothing to sniff since I don't have the Spektrum TX. I'm thinking they haven't made huge changes to their DSM2 protocol so I'm willing to try to play around with timing of the frames mostly. That will be a lot of trial and errors which is why it's difficult to do remotely. If you want to participate, purchase a cheap 8 channels logic analyzer and solder it to the RF part of you Spektrum TX. We will see what's being sent and may be the timing. Pascal
The Problem is: I have no more DSM2 TX :-( So i need it for my multimodule. I tried to bind the multi to dsm2-surface, it was binding, but functionally not correct. Maybe, it's because DSM2-surface uses 16,5 ms?
So you see you have nothing to sniff since you don't have a TX... So you can do same as I'm planning to do, open the DSM code and modify the timings to see if it will work. If I remember well I already started the implementation of 16.5ms, it started to give something if I recall but the person asking for it and testing disappeared...
Unfortunately I am not able to edit these codes. I therefore hope that you will make it, but I believe in you ;-)
Hello...
Are there news about sniffing out?
Greets :-)
Gesendet: Samstag, 03. Dezember 2022 um 10:45 Uhr Von: "pascallanger" @.> An: "pascallanger/DIY-Multiprotocol-TX-Module" @.> Cc: "Guden79" @.>, "Author" @.> Betreff: Re: [pascallanger/DIY-Multiprotocol-TX-Module] DSM2 surface (Issue #769)
So you see you have nothing to sniff since you don't have a TX... So you can do same as I'm planning to do, open the DSM code and modify the timings to see if it will work. If I remember well I already started the implementation of 16.5ms, it started to give something if I recall but the person asking for it and testing disappeared...
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
@Guden79, I have to do the sniffing to work on the Forward programming.. is not trivial. You have to get a Multi-Channel digital analizer (hardware), and then solder wires to an original Spektrum module (in a spektrum radio) to do the capture. After you have the raw frames, you have to filter on a lot of numbers to try to understand what they mean.
Just to give you an example, here is one of my captures: AR636A_FP.csv
Ok, definitely nice to hear that something is happening :-)
LG
Gesendet: Samstag, 18. März 2023 um 16:50 Uhr Von: "Frankie Arzu" @.> An: "pascallanger/DIY-Multiprotocol-TX-Module" @.> Cc: "Guden79" @.>, "Author" @.> Betreff: Re: [pascallanger/DIY-Multiprotocol-TX-Module] DSM2 surface (Issue #769)
@guden, I have to do the sniffing to work on the Forward programming.. is not trivial. You have to get a Multi-Channel digital analizer, and then solder wires to an original Spektrum module to do the capture. After you have the raw frames, you have to filter on a lot of numbers to try to understand what they mean.
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
Hello, is there any news? I saw that you can now bind the TRX-4m... Greetings
Gesendet: Samstag, 03. Dezember 2022 um 10:45 Uhr Von: "pascallanger" @.> An: "pascallanger/DIY-Multiprotocol-TX-Module" @.> Cc: "Guden79" @.>, "Author" @.> Betreff: Re: [pascallanger/DIY-Multiprotocol-TX-Module] DSM2 surface (Issue #769)
So you see you have nothing to sniff since you don't have a TX... So you can do same as I'm planning to do, open the DSM code and modify the timings to see if it will work. If I remember well I already started the implementation of 16.5ms, it started to give something if I recall but the person asking for it and testing disappeared...
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
I'm new to Logic Analyzers but can you tell me if this SPI trace looks correct?
The SR3100 rx works with latest test build but having trouble with the ORX GR300.
In the snapshot its a Spektrum DX3R (DSM2 and DSM, 2009 era) tx connected to a ORX GR300 rx. Frame Rate is 16.5ms
From what I've traced, CS1 is connected to TX module and CS0 is connected to the ADC chip for input.
@pascallanger: is this a bind packet?
SPI result 0.84624725 0.00017375 0x05230FD32FC34FE3
SPI result 0.85174225 0.00017375 0x05230FD32FC34FE5
SPI result 0.8572375 0.00017375 0x05230FD32FC34FE5
SPI result 0.8627325 0.00017375 0x05230FD12FC34FE5
SPI result 0.86822775 0.00017375 0x05230FD12FC34FE5
SPI result 0.873723 0.00017375 0x8523A7A76B6B4FE5 <------
SPI result 0.879218 0.00017375 0x05230FD32FC34FE5
SPI result 0.884713 0.00017375 0x05230FD12FC34FE5
SPI result 0.890208 0.00017375 0x05230FD12FC34FE5
I have no idea of what I'm looking at. If I look at what you have highlighted compared to other lines, it looks like the starting 8523 can be removed at the beginning and 4FE5 at the end which are the same as the other lines, leaving only A7 A7 6B 6B which doesn't look like bind. We would need to compare what the radio is sending compared to what the module is receiving to understand what's going on.
If I look at what you have highlighted compared to other lines, it looks like the starting 8523 can be removed at the beginning and 4FE5 at the end which are the same as the other lines, leaving only A7 A7 6B 6B which doesn't look like bind. We would need to compare what the radio is sending compared to what the module is receiving to understand what's going on.
As I stated I'm a bit of a notice at the hardware side. Jumping in the deep end as they say :) It more like this as I had the CPHA on the down and not the rise of the clock.
0x021107E817E127F2
0x42115353353527F2 <---
0x021107E917E127F2
I presume it is MSB and not LSB?
Where can I source some info to hook up the RX so I can get you the data?
The SPI pin?
I would scrap the paint on the traces as indicated and solder the 4 wires to dump the SPI. Don't forget to connect GND too.
Please use logic 1.2.14 for the dumps
- Original TX powered on, start recording and power on RX
- Original TX in bind mode, start recording and power on RX in bind mode Share the 2 capture files
@pascallanger well that was the smallest traces I've ever soldered.
Using logic 1.2.14 as requested. The binding took 3 goes so its at the end of the capture.
16.5ms frame rate.
ORX GR300 rx SPI capture using DX3R tx.zip
Let me know if you want something else.
@idkLogic Could you do a dump but with this time Multi as the TX?
@pascallanger here are more SPIs dumps from the ORX GR300 at 16MS/s with various setups as requested.
16 Mhz SPI ORX GR300 DX3R and Multi 1.3.4.4 - 1.zip 16 Mhz SPI ORX GR300 DX3R and Multi 1.3.4.4 - 2.zip
I wonder if it is only DSM surface and not DSM2, that was my initial hunch. Maybe this could be a new sub protocol to add?
The binding process is to have the RX already in bind mode before activating the TX bind. Doing it the other way like you asked for, TX first, never works on the first or second attempts.
The RX simply crashes after the bind with multi. It doesn't even try to receive anything, just a simple crash... I've updated the code to use the bind code 0x23 of the DX3R which I think will solve the issue.
The RX might also care about the number of channels. You can set it on the GUI. The DX3R advertises 3 channels but I think I've seen RXs which goes up to 5 channels so I'm allowing anything between 3 and 5 to be advertised during bind and sent over the air. Let me know if I'm wrong here.
I still only see one frequency being used by the RX. It never switches to an alternate RF channel making it actually DSM1 and not DSM2. On my side, I'm keeping the DSM2 implementation (with telemetry) since it will work with 16.5ms and 11ms. You can switch on the fly between 16.5ms and 11ms by changing the "Servo refresh rate" between 22ms and 11ms, not sure if it affects the servo refresh rate or just the transmission. Documentation updated here: https://github.com/pascallanger/DIY-Multiprotocol-TX-Module/blob/master/Protocols_Details.md#sub_protocol-dsm2sfc---6
Please test v1.3.4.6 which integrates all these changes. If it still doesn't work do a bind dump so I can see what's going on.
OK, the bind light stays on using v1.3.4.6. The model matching seems to work too. I bound one ORX GR300 on 17 and the test bed on 18. Changing model IDs activates the correct RX led.
Still no servo control. Steering servo seems to be active and trying to stay centred. The THR channel is off and servo can move freely. AUX channel nothing too.
Two SPI dumps in the zip file. One is binding and other is some triggering of the ch0 (steering) servo when you toggle the channel range from 1-4 to 1-5. Swings a Futaba S3003 old analog servo about 45 degrees. I couldn't change the upper range down to 3 either.
16 Mhz Multi 1.3.4.6 - ORX GR300 SPI.zip
I see what's they are doing to the channels. I can do the same but the channel code is currently common with DSMR. If I change the channel code, can you test DMSR too?
I see what's they are doing to the channels. I can do the same but the channel code is currently common with DSMR. If I change the channel code, can you test DMSR too?
No I don't have any DSMR receivers. I was still using my trusty DX3R and a DX2s till I got the MT12 as most of my models are kit builds, not RTRs.
re v1.3.4.7. Seems to be working on both the GR300 and SR3100.
Range on the SR3100 and GR300 using the multi seems normal (for track racing) like before with the DX3R.
Here are 2 dumps of binding and normal operation using the GR300 for reference. 16 Mhz Multi 1.3.4.7 - ORX GR300 SPI now working.zip
Thanks Pascal :)
As an aside, I tried binding the DSM1 SR3300T receiver, some flickering of the LED but then nothing. I'm not particularly worried as I don't use them, more to see if DSM1 worked. I remember it had poor range, 30 to 40 metres if I recall.
Also are you thinking of supporting the 5.5ms frame rate? I presume the DSMR users would like that too.
When I had the DX3R on the logic analyser, I noticed the timing of the frames coming from the Atmel MCU to the Spektrum TX module were clocked at 5.5ms the whole time. Is the TX module doing some averaging or windowing after receiving to get 11ms and 16.5ms before broadcasting? (guessing)