ADC erratic readings

[Ttelmah]

You'll find that everyone here, actually 'likes' working on threads where the poster is:
1) Reasonably courteous.
2) Involved themselves - reply to questions, do things themselves etc..
3) Something 'interesting' is going on.


I2C, is a little more complex (particularly in the slave), than it at first appears. There are also some chips that have issues (my next move was going to be to check for any errata). I was also going to suggest adding a delay at the start of the master (it could be sending before the supply has really got to a stable level).

The ones we hate, are where people want direct answers, without actually doing any programming themselves, and then (worse) don't actually tell us critical things like 'compiler version', or details of the hardware etc...

Having a break for the problems we are working on can be quite cathartic. Got my USB sorted. Was having problems with a new implementation of a mass storage device, on a different chip to the ones I have used before. Turns out to be a fractional difference in the handling of one status bit. The fun thing was the number of USB connections needed to the test bench:

USB to the unit being developed. This also has to 'pass through' the USB analyser/
USB to ICD.
USB to logic analyser/decoder. Also pass through connection to the unit under development.
USB to programmer (don't ask why but for some reason my ICD on this chip refuses to release program - though it debugs, for programming it complains that it doesn't 'know' the chip).

Spaghetti junction...

[murgui]

I tried my best being courteous. I wonder why people may be rude to somebody who is losing his time to help them with absolutely no reward.

I really found the problem that is 'poorly documented' as you said some posts before.

I will edit the first post and titles with definitive codes to help the rest of people in the future.

Regards,

Murgui.

[temtronic]

re: I2C pullups.

While 4k7 will be fine at 5 volts 'most of the time', you really should use a scope to see the waveforms if possible. Speed ,distance and capacitance will degrade the signal and 'random' or 'odd' results will be seen. Even humidity can be a problem !
I typically use 3k3 on 5v PICs and 2k2 for 3v devices. So far so good....
For any serious PIC projects a scope, even a 20 year old analog one is better than nothing !!

Jay

[PCM programmer]

Murgui,
Ttelmah did not mean his post to be a cut on you. You basically did
the things that he desires to see in a thread.

[murgui]

Oh I know, maybe I explained myself wrong. I said I don't understand why people may be rude to you(Ttelmah, PCM Programmer, Temtronic,...) when you are helping them with absolutely no reward. I'm really glad there's people like all you who can help newbies like me improve the programming skills.

Regards.

[Ttelmah]

Thanks.