LPC17xx can't enter ISP mode

Did you program any code read protect?

Without code read protect, the ISP interface should always be avaiable and it doesn't care about any incorrect PLL configuration.

If you have too high code read protect, then the boot loader will ignore the 2.10 state.

Next thing - when experimenting with PLL or other things that may lock up the processor, you should try to wait until you reach main() and preferably have a bit of delay before the experiment code. That gives you a time gap where the JTAG interface may manage to catch the processor before it locks up.

You can also try to step up the JTAG clock frequency to try to get the JTAG interface to be quicker to interact with the chip after it is released from reset to see if you have time enough to catch the processor. The PLL activation is two-step. First configure the PLL parameters and wait until the PLL gets a lock. First after the PLL lock will the code switch to actually use the PLL output frequency - at least if the PLL code is correctly written.

Did you program any code read protect?

Without code read protect, the ISP interface should always be avaiable and it doesn't care about any incorrect PLL configuration.

If you have too high code read protect, then the boot loader will ignore the 2.10 state.

Next thing - when experimenting with PLL or other things that may lock up the processor, you should try to wait until you reach main() and preferably have a bit of delay before the experiment code. That gives you a time gap where the JTAG interface may manage to catch the processor before it locks up.

You can also try to step up the JTAG clock frequency to try to get the JTAG interface to be quicker to interact with the chip after it is released from reset to see if you have time enough to catch the processor. The PLL activation is two-step. First configure the PLL parameters and wait until the PLL gets a lock. First after the PLL lock will the code switch to actually use the PLL output frequency - at least if the PLL code is correctly written.

Thanks for your respond,
I didn't program code read protect(even don't know how to do it) or maybe accidently I did it... I don't know. I set up the max JTAG TCK in H-JTAG but it didn't connect. It gives me an error something like "Can't stop in DEBUG mode. If you are using USB try to slow down TCK" but I'm using LPT and of course I tryied to slow it down but it didn't work also. Maybe you got some other hints that may work?

I did a quick google search, just for curiousness.

It seems the H-JTAG Can't debug CORTEX-M3 with KEIL.

www.hjtag.com/.../viewthread.php

IMPORTANT NOTES:
1 - To debug CORTEX-M3, please use IAR 5.11 and above;
2 - Can't debug CORTEX-M3 with KEIL, because the RDI interface of KEIL is not updated to supported CORTEX-M3;
3 - For STM32F, please configue TAP according to the following pic, otherwise, can't debug;
4 - To use auto flash download, please disable Use Flash Loader in IAR (shown below)

It's possible I've done it when everything was working. You just have to check in keil "Make hex file" then you going in to H-JTAG Flasher and you send the code to board. After that you have to click debug and reset debuging and uC. It's not working to great and it often make some small troubles but it's working.

No other sugestions?

Sounds like this problem:

http://www.keil.com/forum/docs/thread16503.asp

Nop it's not working for me. I can't read the device signature via Flash Magic via COM. I can read sygnature via JTag in H-JTag