I've just bought from Farnell a LPCXpresso board, LPC11C24 (OM13012) board for using as a programmer and playing with 11C24
OM13012 - NXP - Evaluation Board, For Microcontroller, LPCXpresso, CorteX-M0 | Farnell element14
Schematic here
https://www.embeddedartists.com/sites/default/files/docs/schematics/LPCXpressoLPC11C24revB.pdf
Now while playing, we noticed that MCU is directly powered from 5V (USB) through a single diode (D2), so receiving about 4V6 at LPC11C24 Vdd pin. Absolute maxim ratings is 4V6... hmm... do you want us to buy every week a new board ?!?
The problem is that every ext. peripheral will receive 4V6 too instead of 3V3 so I presume NXP wants us to burn all competitor chips like memories, sensors, etc. and buy from them 4V6 "resistant proof" chips. Joking....
The question is why don't you just use the (already) existing 3V3 from USB controller or put a cheap LDO ?!
Another issue that could be solved would be why don't you modify the schematic to be able to program target MCU at any voltage, like low power situations (Vdd = 1V8). Take a look on schematic, U4, U5 and U6 are level translators and in actual schematic, you always power them from internal 3V3.
The smart way will be to power only U6 from internal 3V3 so any incoming signals like SWD (from TARGET to DEBUGGER) will be converted to 3V3. Now outgoing signals, needs to be adapted to TARGET voltage so powering U4 and U5 from TARGET Vdd will just convert those into 1V8 for example.
Just remove D6 diode, de-solder U6 pin 14 and tie him to some internal 3V3, for example pin 6 of U3.
Use VIO_3V3X as V_TARGET.
See attached image that I use for a old LPC1114 target board that I broke and use as a separate programmer.
Regards,
