Accuracy of 28 X1 without external resonator?

Community General Discussions
, , , , , , ,
1

In my circuit, I have 2 Picaxe 28X1s. The only suitable external ceramic resonator I have is on the 28 pin project board and I don't want to ruin it by accident during desoldering. I am making my own board for each Picaxe, and need to know how accurate the timing of the Picaxes will be only using the internal resonator. Picaxe A is using 2 PWM channels driving motors on a separate power supply handled by an L293D, all other outputs to drive other ICs, will read 1 ADC channel, 2 digital sensors, and 1 digital input. It will be running a continuous program that runs a robot. Picaxe B just uses 2 PWM outputs to drive LEDs and reads 1 ADC channel. 

Since it's controlling a robot, the PWM channels have to be pretty precise since I want the motors to go the correct speeds, and the second Picaxe is modulating a constant 38khz pulse to drive some IR LEDs. PWM is mainly the only thing I'm worried about, and my motors are on a separate power supply.

My questions are:

Most importantly, does only using the internal resonator sound accurate enough for this?

If not, I do have some salvaged 4 MHz 2-pin crystals that look like they're in good shape, and have no damage from soldering heat. If I were to use these instead, what would be the best capacitor values?

Lastly, If I had to resort to removing the resonator from the project board, what would be the best way to remove it without damaging it from heat? I use a medium heat 25 watt, small tipped iron, and only have some desoldering braid to use (and from RadioShack too, so sometimes it doesn't even work).

Thanks.

2

Won’t notice a difference

I would be surprised if you notice any difference at all.

The next question would be why you don’t simply get another resonator? We’re talking about $.59 here.

3

Don’t worry so much

The internal resonator is perfectly suitable for pwm.

You will only run into problems if you use your 28x1 as a clock, because it will drift several seconds each day, possibly minutes.

You really don’t need precise timing for pwm, you just need consistency, so that a 50% duty cycle is always 50% (it really doesn’t matter if the motor is on for 2 milliseconds and off for 2, or on for 3 and of 3, it is still 50 percent.

4

I would but I’m in high

I would but I’m in high school, and my parents aren’t that fond of ordering components for me online, or letting me do it myself.

5

Ok, and even with a lot of

Ok, and even with a lot of things going on in the program, with pauses and loops and such, nothing would really be thrown off? My only worry is that the second Picaxe needs to continuously PWM at 38khz for proper operation.

6

The only concern of internal oscillators

that I have heard about was due to wanting high speed serial communication. 1200 and probably even 2400 baud should be reliable with internal oscillators. I believe, also, that your 38kHz frequency can probably have some drift to it, if you are using it for something like IR communications. Otherwise, maybe you could look at constructing a 555 based timer circuit for your 38kHz frequency.