Davinci Jr Repetier Port (Page 8 of 9)

Hi, pyroball, you really did great job here. But decode the message need lots of experiment, and make as little change as possible so that we can compare the change in code. And it will be helpful if you have more note on what you have done and then what code shows up.

For example, you can do clean nozzle, it should heat up to 200C while move XYZ to right spot. Since step motor don't generate communication here, and clean nozzle don't involve check spool status, we should only see command to raise temperature to 200C, and NTC report temperature change. Once temperature arrived 200C, the NTC report temperature change will become different. Since we have some vocabulary now, it is time to find out where the 200C is in the code and we will have better chance to decode the rest of the code. But it will be very helpful if you can mark in the log when clean nozzle command is send and when temperature riased to 200C and start to maintain 200C

Also, will you be able to fake message to U17 or U2? Once we decode the message, it will be nice if you can fake a command and see if it works.

http://www2.lauterbach.com/pdf/training_jtag.pdf

For debugging any JTAG communication the use of a logic analyzer such as the TRACE32 PowerProbe is
recommended.

By using this tool it is not only possible to record the signals but also decode the JTAG protocol for a better
interpretation. The JTAG decoder is built-in but also available as source-code allowing to extend the analysis
by a higher-level decoder for custom use.

[Never mind, this thing is more expensive than 3d printer]

ppl117, If you like number patterns, I have a challange for you ( after helping out here ).
As you know, the Jr, and mini use a RFID  tag that must be in place for the printer to function.  There are many efforts to work around it, one of which is this thread here.   The tag , among other information, contains a 7 byte UID, a 4 Byte Pasword and a 2 byte PACK ( Password Acknowledge) code. 
Its these 3 pieces of information that keep us from making our own UID and generating the correct Password and PACK.  No one here know thee algorithm use. 
If you want a challenge, I can provide you about 500 valid UID / Password / PACK combinations and perhaps you can see a pattern and help create ( or create ) an algorithm that works.

What do you say?

Let me know, I dont want to hi-jack this thread, but wanted to get this out to you.

JTag communication have certain format, it start with "run test/idle", probably the stuff before 11 EE, Then followed by muti-runs of shift-IR(Instruction Register) and shift-DR(Data Registers). Although I still don't fully understood how JTAG works, my guess is 11EE are shift-IR and 22 DD/0C F3 are shift-DR

"Test Logic Reset sets the Instruction Register to its reset value (IDCODE or BYPASS). This state
can be reached from any other state by shifting five times “1” on TMS.
Some CPUs do also reset the DTAB and/or the on-chip debug logic when this state is entered.
As a result an active debug connection might get lost, if this state is driven by a PRACTICE script
or a custom application.
• Run-Test/ Idle and Select DR-Scan are used by most debuggers as pause parking position.
• In the Shift-IR state the debug tool shifts an instruction into the Instruction Register. The
instruction is activated once the TAP controller reaches the Update-IR state.
• In the Shift-DR state the debug tool shifts data to/from the Data Register selected by the
currently loaded instruction."

Great job! pyr0ball, I will try to break down the communication and see if we can find something. I think it will be very useful if we have a working system to test these commands. If we can record a message, break it down, and fake the same message from code (repeitor-firware or reprap) and see if faked message will be accepted. Then we can manipulate the message say change heat hotend 200C to heat hotend 180C and see if the change is effective. This will greatly accelerate the decoding process.

Only thing I figured out so far is the temperature
11 EE 0C F3 98 01 FA 03 BB 00 00 00 64 64 01 00 03 00
---------------||-----------------------------------------------

98 01 after 11 EE 0C F3 and before 01 FA 03 BB 00 00 00 64 64 01 00 03 00 is the temperature. This number change from F7 03 to 98 01 in the U17 cleaning cycle which is heating up. the first digit change first, so the real number should be 03 F7 to 01 98. It most likely to be resistance of the themistor (1015 to 408). We will need a temperature table for this conversion.

What else information do we need to get out of the communication?

Well not a super necessary thing, but figuring out the trigger to activate the exhaust fan would be good. I think I can probably get captures at different temperatures which would help with the table.

I could also pull direct voltage readings off the divider circuit and map those to the temperature. Based on the way this board is designed, I don't think it can measure thermistor resistance directly. It's taking analog voltage measurements from out-2 on the voltage divider.

Are you certain when you do the bootup log, the exhost fan is not on? If your fan is always on, I can not tell which code is responsible for the fan. The hotend need to cool down below 50C in order for the fan to shut down, Please re-record the U2 bootup log and make sure the fan is not on. Also re-record the U2-bootup clean cycle and make sure temperature start from room temperature so that the fan is off in the beginning and then temperature raise to 50C and 210C.

I suspect 02 FD 20 30 50 AF 40 BF 22 DD is the command for the fan, but it also shows up in bootup log. However, if your hotend temperature is above 50C when you record the bootup log, this will all make sense.

I remember you disconnect NTC, all you get is 91 EE 00 FF and no  02 FD 20 30 50 AF 40 BF 22 DD. I think disconnect NTC you get large resistance which suggest super low temperature, then hotend try to heat up but no need to turn on the fan.

Sorry for some reason I stopped getting notifications on replies for this thread.

Yes 100% no fans are turned on when booting. The fans both come on simultaneously a short time after starting a heating cycle.

What would be the best way to test resistance? Should I solder a test wire onto the thermristor input line directly, or on part of the voltage comparator circuit?

Nah I got a notification this time.

Ok so I'll get a few test wires in place first for various points. Do some heating and measuring, and provide as much data as I can. My main concern is going to be the delay of the printer display to actual temperature

Do you think it will be possible to port it in the nearer future or would it be better to change the boards to Ramps?
Is there any way to support the repetier port without any knwwoledge about how to use an analyzer?

Kind regards and thanks for keeping the project alive!

Have you talked to nilezon about the JTAG hacking, he seems to have lots of experience on JTAG and might provide different angle to it.

http://www.soliforum.com/topic/16879/da … ag-access/