JB16 ICP Question

[dmi]

Hello, I have a question regarding ICP with JB16. According to AN2399, two 470K resistors are needed to pull down D+ and D-. However, according to the disassembly of the monitor code, only pulldown on D+ is needed, and only when the JB16 is blank (enters monitor mode at 0xFE10). If I provide a method to reenter USB ICP, by jumping from my code to address 0xFA19 (as in AN2399), I think the 470K resistor(s) is(are) not needed. Has anyone tested this?.

Also, can I leave the 470K resistor(s) connected if I want to use the USB module in normal operation (user mode)?

I appreciate your comments. Thanks a lot!

Diego.

[peg]

---

dmi wrote:

Hello, I have a question regarding ICP with JB16. According to AN2399, two 470K resistors are needed to pull down D+ and D-. However, according to the disassembly of the monitor code, only pulldown on D+ is needed, and only when the JB16 is blank (enters monitor mode at 0xFE10). If I provide a method to reenter USB ICP, by jumping from my code to address 0xFA19 (as in AN2399), I think the 470K resistor(s) is(are) not needed. Has anyone tested this?.

Also, can I leave the 470K resistor(s) connected if I want to use the USB module in normal operation (user mode)?

I appreciate your comments. Thanks a lot!

Diego.

---

Hi,

All of what you say is correct, yes I have tested this to death and spent many hours disassembling the monitor. Take a look at this thread:
http://forums.freescale.com/freescale/board/message?board.id=8BITCOMM&message.id=737

You only need the pulldown on D+

The 470k resistor will have very little effect on the bus!

If you jump in there you don't need any resistors. But you may need to get in this way if something goes wrong after you erase it for example.

Have you noticed the unreliable startup into blank part ICP?

Regards David

[dismirli]

David,

Thanks a lot for you reply. It appears to me you are an expert in the subject! I haven't had the chance to try the JB16 exhaustibly yet, because I'm still designing the PCB for my project. Why do you say the startup is unreliable in blank parts?

Diego.

PS: When I finish I will post my results (if I have something new/original to share).

PS2: I will look into the issue you point out, to see if I have the same problem.

[peg]

Hi Diego,

The JB16 often fails to power up into blank part USB ICP mode without some capacitance on the reset pin. This is acknowledged as an issue by Freescale including the team responsible in Hong Kong. But there is no "official" word on it. Or explanation as to why.

Regards David

[dismirli]

Hello, David, at last I've had the chance to work on the JB16. Sorry for the delay. I have come across the issue you mentioned, but in my case it dissappeared only if 1) Both 470K resistors were placed (on D+ and D-) and 2) A MC34064 supervisor with a 100nF capacitor were used. Now I can enter (blank flash) ICP mode, and reenter ICP jumping into 0xFA19 when the Flash is not blank.

Also, I had no problems with USB communication if I left the 470K pull downs in D+ and D-.

[peg]

Hi,

Not sure now if I actually tried it with only the one pulldown. Or it just looked like you only needed one from the code.

Do you really need the 34064 there?

Regards David

[dismirli]

Hi, David, I dont know if I *really* need the 34064. With both pulldowns the JB16 entered ICP correctly, but I feel safer with the 34064. I will try without the 34064, to see what happens.

Diego.