RFID Toys Homepage RFID Toys
A forum for RFID hobbyists and enthusiasts
Forum Home Forum Home > Projects and such > Reader Projects
  New Posts New Posts RSS Feed: Immobilizer bypass project
  FAQ FAQ  Forum Search   Calendar   Register Register  Login Login



Immobilizer bypass project

 Post Reply Post Reply Page  12>
Author
Message / View First Unread Post
Alex View Drop Down
Newbie
Newbie


Joined: 10 January 2010
Location: Italy
Online Status: Offline
Posts: 11
Post Options Post Options   Quote Alex Quote  Post ReplyReply Direct Link To This Post Topic: Immobilizer bypass project
    Posted: 10 January 2010 at 4:17pm
Hi, I've already posted something in the comment section of this blog post http://blog.amal.net/?p=292.
Basically what I'm trying to do is to bypass somehow my bike immobilizer system to be able to start her Big smile with a simple 125khz tag (I'm planning to do an implant on my hand like most of you already did :) )
Amal showed me that immobilizer bypass kits already exist, in particular this type caught my attention:

basically it has a space where you have to put your original key, a coil which reads the tag in the key, and then uses the original coil of the car to read the new tag. if it is a correct one, it uses the original tag and sends it to the original transponder reader.
What I'm trying to realize is something like this, possibly without leaving the original key.
I don't know how exactly the tag reading process works, but I think that it could be complicated to emulate the original key, or nearly impossible if it has some encryption system (variable codes? I don't know anything about this stuff, but I love to learn as much as I can, so I'll try to inform)

I found that immobilizer systems usually use 125khz frequency, so I could use the same coil without adding anything visible on the outside. 

Another solution could be to emulate the signal from the transponder reader to the ECU (like this kit does http://www.bypasskit.com/product.aspx?prodid=CANMAX400&catid=15), maybe this could be simplier with a microprocessor, but I don't know anything about the signal that comes out from my bike's reader, and I don't have enough instruments to study it.

I post the wiring diagram for the immobilizer part, maybe it comes helpful somehow... 
oh, it's in italian, I translate the legend for you:
1. ECU
2. Chassis Ground
3. Dashboard Ground
4. Waterproof junction
5. Fuse Box
6. ECU fuse 10A
7. Start fuse 10A
8. this is the connector that goes to the diagnostic pin
9. Key switch
10. Antenna
11. what they call "immobilizer amplifier", I think it's just a reader who communicates with some communication standard to the ECU
12. Waterproof junction
13. LED on the dashboard
14. dashboard
15. relais box
16. ECU's main relais
17. main fuse 30A
18. 12V 8Ah battery

here I put everything I know so far. If anyone has advices, observations or anything else, just let me know :P


Edited by Alex - 10 January 2010 at 4:17pm
Back to Top
amal View Drop Down
Admin Group
Admin Group
Avatar

Joined: 22 November 2005
Location: United States
Online Status: Offline
Posts: 1530
Post Options Post Options   Quote amal Quote  Post ReplyReply Direct Link To This Post Posted: 10 January 2010 at 4:35pm
Hmm yeah, I'd be very interested to see what happens on the Y, LG, and V lines coming from box 11 do when the key is inserted and rotated to the accessories position. I think you are correct that box 11 is the entire RFID reader and it is only reporting back key data to the ECU, however you might also find that it may be performing the entire authentication process and only reporting basic electrical signals back to the ECU.

If you have an oscilloscope, I would be interested to see what those Y, LG, and V lines do... watch them during each phase of the key rotation from dead to accessories to actually starting the bike.

If you don't have an o-scope, start with a simple volt meter and simply measure voltage on those lines from ground and see if they are simple 0v / 12v signals. If your voltmeter is digital and can handle accurately measuring pulsed dc signal levels, and supports a mix/max function, you may be able to set it to record the max value and catch either a 3v, 3.3v, or 5v signal voltage... which would tend to indicate data of some sort. Only RS232 serial communicates with +12/-12 signals, and that would never be used in equipment like this.
Amal ;)
www.amal.net
Back to Top
Alex View Drop Down
Newbie
Newbie


Joined: 10 January 2010
Location: Italy
Online Status: Offline
Posts: 11
Post Options Post Options   Quote Alex Quote  Post ReplyReply Direct Link To This Post Posted: 10 January 2010 at 11:22pm
LG and V lines go up to 12V when i turn the key, and stay that way (I only used a voltmeter on these two)
on the Y line using a voltmeter i saw that goes on HIGH level for a very short period (it indicates more than 10V, but I don't trust it too much for a signal so short). I only have analog oscilloscopes with no memory, and for what I could see it's definitely a digital signal. (yes, there were no need of an oscilloscope to guess that :P)
as soon as I can put my hands on a digital oscilloscope with memory, I'll dig more.
Back to Top
amal View Drop Down
Admin Group
Admin Group
Avatar

Joined: 22 November 2005
Location: United States
Online Status: Offline
Posts: 1530
Post Options Post Options   Quote amal Quote  Post ReplyReply Direct Link To This Post Posted: 11 January 2010 at 9:06am
Hmm, the Y like spikes over 10v? Is that what the o-scope confirmed? That just seems like an odd voltage for data to me. If you have a decent video camera that you can set to record at a high frame-rate with (many consumer cameras can do this), you could video the screen and replay it (a crude form of "memory") slowly to verify the signal.

Also, I would go to the hardware store and get a physical copy of the bike key made that does not have the RFID transponder and then use that to turn the key switch while monitoring the Y line... or I guess you could try wrapping your factory key head in tinfoil to try and block the signal... just to see what the difference might be in the Y line. If there is a notable difference, then you're probably on the right track to being able to emulate the data signal going to the ECU.
Amal ;)
www.amal.net
Back to Top
Alex View Drop Down
Newbie
Newbie


Joined: 10 January 2010
Location: Italy
Online Status: Offline
Posts: 11
Post Options Post Options   Quote Alex Quote  Post ReplyReply Direct Link To This Post Posted: 11 January 2010 at 9:18am
Originally posted by amal

Hmm, the Y like spikes over 10v? Is that what the o-scope confirmed?
no, that's what the voltmeter indicated, I didn't pay too much attention on the oscilloscope scale, if I remember well it was set on 5V/div and the signal took only one division, but really I'm not sure, I'll try again tonight.

Originally posted by amal

 That just seems like an odd voltage for data to me. If you have a decent video camera that you can set to record at a high frame-rate with (many consumer cameras can do this), you could video the screen and replay it (a crude form of "memory") slowly to verify the signal.
yes, i have a digital camera and it records up to 60fps, not so much but always better than human eye; I'll try that way.

Originally posted by amal


Also, I would go to the hardware store and get a physical copy of the bike key made that does not have the RFID transponder and then use that to turn the key switch while monitoring the Y line... or I guess you could try wrapping your factory key head in tinfoil to try and block the signal... just to see what the difference might be in the Y line. If there is a notable difference, then you're probably on the right track to being able to emulate the data signal going to the ECU.

I'll follow all of this suggestions you gave me and then I'll post the results, thank you Wink
Back to Top
amal View Drop Down
Admin Group
Admin Group
Avatar

Joined: 22 November 2005
Location: United States
Online Status: Offline
Posts: 1530
Post Options Post Options   Quote amal Quote  Post ReplyReply Direct Link To This Post Posted: 11 January 2010 at 9:19am
Cool :) Looking forward to hearing the results!
Amal ;)
www.amal.net
Back to Top
Alex View Drop Down
Newbie
Newbie


Joined: 10 January 2010
Location: Italy
Online Status: Offline
Posts: 11
Post Options Post Options   Quote Alex Quote  Post ReplyReply Direct Link To This Post Posted: 11 January 2010 at 3:35pm

well, I start saying that I absolutely have no idea on how to use an oscilloscope. :)

now that I told all the truth :P , I can post the videos. they're recorded at 60fps, if you want I can post single frame pictures, but they don't seem useful to me.

wire colors are different from the wiring diagram, 2006 was the first year they introduced immobilizer and the last year for that model, so there are no correct diagrams for it. In fact, from 2007 model there is a new ECU and the electrical connections are slightly different, for example the immobilizer amplifier is located under the dashboard.

The yellow wire remains yellow, and so do the brown/white for +12V and the black yellow for gnd.

but the LG (light green) wire on my bike is orange/black, and the Violet wire is Blue.

I made these 3 videos, try to take a look and tell me if you can understand something, i personally do not. :S

From the yellow wire comes a +5V signal, from the other two a 12V signal, I still can't figure out what those 2 wires are there for…

orange/black

blue

yellow

Back to Top
amal View Drop Down
Admin Group
Admin Group
Avatar

Joined: 22 November 2005
Location: United States
Online Status: Offline
Posts: 1530
Post Options Post Options   Quote amal Quote  Post ReplyReply Direct Link To This Post Posted: 12 January 2010 at 2:23am
Nice videos :) It really looks like a nice o-scope too... too bad its got no memory :( Anyway, it looks like a standard 5v data signal is coming out of the reader and heading into the ECU... either that or... hmm, can you by chance disconnect the yellow wire from the ECU and test the data stream from both sides... just to see which direction data is flowing?

You might also be able to just tap yellow wire and dump it into a 5v TTL to RS232 (or TTL to USB) adapter and fiddle with the baud rate to see if you come up with any clear data.

As for the other two wires, I'd play with the timing a bit to see if you can't find out what's going on there. Also, it may be that the ECU is signaling the reader to power up the antenna by pulling one line high. The other line might be an indicator that the ECU is ready to receive data.

Finally, have you been able to try watching the Y line data when a simple key copy (without RFID tag) is used? I'd be interested to see if the Y line stays silent... however it could also send different data in the event a non-authorized key was used to turn the key switch.


Edited by amal - 12 January 2010 at 2:23am
Amal ;)
www.amal.net
Back to Top
Alex View Drop Down
Newbie
Newbie


Joined: 10 January 2010
Location: Italy
Online Status: Offline
Posts: 11
Post Options Post Options   Quote Alex Quote  Post ReplyReply Direct Link To This Post Posted: 12 January 2010 at 5:31am
Originally posted by amal

Nice videos :) It really looks like a nice o-scope too... too bad its got no memory :( Anyway, it looks like a standard 5v data signal is coming out of the reader and heading into the ECU... either that or... hmm, can you by chance disconnect the yellow wire from the ECU and test the data stream from both sides... just to see which direction data is flowing?


it's impossible to disconnect a single wire unless I cut it, I can disconnect the amplifier connector and see which wires are outputs from the ecu.

Originally posted by amal


You might also be able to just tap yellow wire and dump it into a 5v TTL to RS232 (or TTL to USB) adapter and fiddle with the baud rate to see if you come up with any clear data.
I tried with an arduino and I came up with no result at all, but I was using some components to step down the voltage (I was thinking that it was a 12V signal). I'll try tonight with a direct connection.

Originally posted by amal


As for the other two wires, I'd play with the timing a bit to see if you can't find out what's going on there. Also, it may be that the ECU is signaling the reader to power up the antenna by pulling one line high. The other line might be an indicator that the ECU is ready to receive data.
Finally, have you been able to try watching the Y line data when a simple key copy (without RFID tag) is used? I'd be interested to see if the Y line stays silent... however it could also send different data in the event a non-authorized key was used to turn the key switch.
No, I forgot. I'm going to try this too. :)

Back to Top
Alex View Drop Down
Newbie
Newbie


Joined: 10 January 2010
Location: Italy
Online Status: Offline
Posts: 11
Post Options Post Options   Quote Alex Quote  Post ReplyReply Direct Link To This Post Posted: 16 January 2010 at 9:47am
Got something!
no luck with the serial connection, I tried all rates but nothing came out.
Then I tried to use the analog/digital converter pin, and came out this:
the values are read with 1ms of delay

0
0
0
0
0
0
0
816
834
824
831
825
831
822
824
828
826
829
827
836
829
826
816
832
824
834
825
835
825
835
820
809
73
739
777
779
795
801
27
64
59
43
36
20
805
23
805
798
835
828
831
816
64
0
0
0
0
0
0
0
789
821
818
827
824
832
823
835
824
822
831
823
830
821
832
825
837
829
834
819
826
817
832
827
829
67
781
48
0
0
0
0
0
0
0
0
0
0

now I'm going to make other tries and then I'll go home (more warm than here :D) to see if I can to figure out something!


EDIT:
Now I tried with a port logger on the pc and no delay (about 100us of adc conversion), set the transmission from arduino to pc at 256000 baud and came out this, it seems a bit more complex than before, probably it's the tag code!


0
810
815
821
821
809
819
812
821
823
812
817
822
814
821
814
819
817
826
817
825
809
815
818
814
823
828
817
820
816
825
809
827
819
822
819
820
817
819
820
829
828
813
828
815
817
817
822
815
821
817
822
822
812
813
823
816
825
823
810
821
816
814
812
827
822
815
814
823
811
823
819
820
820
820
824
820
816
820
828
828
816
827
817
817
819
821
819
818
816
822
822
812
812
820
815
818
818
812
817
820
809
825
814
814
819
816
817
810
812
0
807
0
0
821
0
807
821
0
818
0
0
820
0
822
807
819
0
807
0
0
814
0
815
0
0
816
0
0
0
818
0
0
0
812
0
814
812
0
0
805
0
0
0
814
0
816
803
0
0
807
0
0
809
815
0
0
814
0
808
0
0
817
0
0
0
817
0
0
0
819
0
804
820
0
0
802
0
0
817
0
0
815
0
0
0
806
0
0
806
0
0
0
817
0
0
0
809
0
0
0
808
0
818
0
0
813
0
0
0
820
0
805
0
817
0
0
815
815
807
822
822
817
817
812
811
818
815
823
810
822
812
820
815
817
819
816
819
817
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
818
807
805
819
811
814
814
820
815
813
815
821
808
824
815
826
816
817
817
819
816
814
824
811
807
822
811
816
811
818
816
818
811
818
820
806
820
825
812
812
819
816
817
809
817
820
810
823
823
811
811
821
824
813
822
818
818
810
823
813
819
820
816
815
821
818
814
821
820
809
821
810
814
808
811
820
823
812
813
813
814
813
824
811
817
814
823
819
810
811
815
815
819
827
814
815
812
822
808
826
815
821
816
816
818
816
812
821
0
0
815
0
0
0
806
0
813
808
0
808
812
0
0
0


sorry for the lenght :P


Edited by Alex - 16 January 2010 at 11:10am
Back to Top
 Post Reply Post Reply Page  12>



Forum Jump Forum Permissions View Drop Down



This page was generated in 0.047 seconds.