ARM fastest possible pulse

[eurofox]

I think the original poster has disappear in the hail of controversy.

"fastest pulse" is a different question than "highest frequency"...

On ST microcontroller, don't forget that there are clock prescalers for the relevant bus to set, and a GPIO_Speed setting in the GPIO_InitStructure (or bare registers.)  I don't think I'd trust those to be set for maximum speed by default.

I still watching with really a lot of interest and really appreciate all inputs, I think I will try some of the ideas 

Before I build the hardware I will test it on my STM32FA4 discovery board.

I'm quite experienced with PIC's and other 8 and 16 bits CPU's, the ARM is new for me and I really look forward to start discovering is CPU.

eurofox

[dannyf]

the ARM is new for me

Depending on your ability to read the datasheet, the learning curve can be steep.

[eurofox]

the ARM is new for me

Depending on your ability to read the datasheet, the learning curve can be steep.

 

I have time since I'm retired 

[dannyf]

Time is not as big of an issue as persistence / perseverance - you will eventually get there if you keep trying.

[David Hess]

In the fastest mode, it looks like the output ports of the STM32F344 are good to 50 MHz with 5 nanosecond rise and fall times.  Internally it uses a tapped delay line and 144 MHz clock so there may be other limits to the minimum pulse width with the high resolution timer but I did not find them in the documentation.

[Psi]

I think some of the STM timers can be setup in 'one pulse' mode for generating a narrow pulse.

So you should be able to get a pulse out that's the same as the timer clock rate. Assuming the I/O clock is fast enough.

[westfw]

No one has yet mentioned that if you're going to be generating 50MHz signals, you'd better be paying a lot of attention to the PCB design of that signal and it's connection to whatever it is connected to.  There are a lot of Arduino users (for example) who think that going from Uno (16MHz AVR) to Due (84Mhz ARM) they'll suddenly be able to run their SPI over jumper wires to a breadboard 5x faster, and that doesn't seem very likely to me...

[paulie]

I think the original poster has disappear in the hail of controversy.

I think you are confusing this with the one dollar one minute thread. However that OP will NEVER be scared off.  LOL

"fastest pulse" is a different question than "highest frequency"...

True, I was taking the title literally. For programmable delay line it should have said "highest resolution pulse" or "fastest controllable pulse". XOR would be the path to simply fastest.

[Jeroen3]

No one has yet mentioned that if you're going to be generating 50MHz signals, you'd better be paying a lot of attention to the PCB design

The attenuation of a long (30 cm) two-layer design board with planes is workable up to 60 MHz. Then it becomes a sine and fades out very fast.

[T3sl4co1l]

No one has yet mentioned that if you're going to be generating 50MHz signals, you'd better be paying a lot of attention to the PCB design

The attenuation of a long (30 cm) two-layer design board with planes is workable up to 60 MHz. Then it becomes a sine and fades out very fast.

No, not nearly so bad.  It's about two orders of magnitude better than this.  You do need to be aware of proper signal quality however.  Not because of rounding off, but because of ringing.

Tim

[David Hess]

No one has yet mentioned that if you're going to be generating 50MHz signals, you'd better be paying a lot of attention to the PCB design

The attenuation of a long (30 cm) two-layer design board with planes is workable up to 60 MHz. Then it becomes a sine and fades out very fast.

No, not nearly so bad.  It's about two orders of magnitude better than this.  You do need to be aware of proper signal quality however.  Not because of rounding off, but because of ringing.

It is amazing what can be done with 100+ MHz clocks on FR4 with proper terminations and ground returns.

[dannyf]

-you get 1/2 of the instruction frequency which varies a lot from chip to chip, from tends of Mhz to hundreds of Mhz. -

that assumes a fast bus on gpio. Stm32f4 for example can go to 84mhz - 168mhz / 2 .

on a f0 chip, the fastest is clock / 4 = 16mhz. The same with f1 chips = 18mhz.